Development of 11C-Labeled ω-sulfhydryl fatty acid tracer for myocardial imaging with PET

Article abstract of DOI:10.1016/j.ejmech.2017.10.062

[¹¹C]-S-methyl-16-thiopalmitic acid (a) was developed with excellent heart-to-background uptake ratios and higher retention in heart. Myocardial uptake and metabolism of the tracer is markedly higher CPT I dependent. When compared to [¹¹C]-S-methyl-14-thiomyristic acid (b), [¹¹C]-S-methyl-12-thiododecanoic acid (c) and [¹¹C]-palmitate, a showed an early high uptake and a significantly slower late clearance in heart and a prolonged myocardial elimination half-life (30 min). Analysis of heart tissue and urine samples showed that a was metabolized via beta-oxidation in myocardium. Small animal PET images of the accumulation of a in the rat myocardium were clearly superior to [¹¹C]-palmitate. These initial studies suggest that a could be a potentially useful clinical PET tracer to assess myocardial fatty acid metabolism.

Full text of DOI:10.1016/j.ejmech.2017.10.062

Accepted Manuscript
11
Development of C-Labeled ω-sulfhydryl fatty acid tracer for myocardial imaging
with PET
Xiangxiang Wu, Peizhi Wang, Ruixin Liu, Huahui Zeng, Fangfang Chao, Hao Liu,
Caiyun Xu, Haifeng Hou, Qiong Yao
PII:
S0223-5234(17)30864-4
10.1016/j.ejmech.2017.10.062
EJMECH 9854
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 18 August 2017
Revised Date: 21 October 2017
Accepted Date: 21 October 2017
Please cite this article as: X. Wu, P. Wang, R. Liu, H. Zeng, F. Chao, H. Liu, C. Xu, H. Hou, Q. Yao,
11
Development of C-Labeled ω-sulfhydryl fatty acid tracer for myocardial imaging with PET, European
Journal of Medicinal Chemistry (2017), doi: 10.1016/j.ejmech.2017.10.062.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
11
Development of C-Labeled ω-Sulfhydryl Fatty Acid Tracer for
Myocardial Imaging with PET
a,
a,
a,b,
Xiangxiang Wu , Peizhi Wang , Ruixin Liu ∥, Huahui Zeng a,c,*, Fangfang
∥
∥
Chao c, Hao Liu c, Caiyun Xu c, Haifeng Hou c, Qiong Yao c
a Chinese Medicine Immunology Laboratory & Science and Technology Department,
Henan University of Chinese Medicine, Zhengzhou 450046, China;
b Department of Pharmacy, First Affiliated Hospital of Henan University of Chinese
Medicine, Zhengzhou 450000, China;
c Department of Nuclear Medicine, Second Affiliated Hospital of Zhejiang University
School of Medicine, Hangzhou 310009, China.

ACCEPTED MANUSCRIPT
ABSTRACT
[11C]-S-methyl-16-thiopalmitic acid (a) was developed with excellent
heart-to-background uptake ratios and higher retention in heart. Myocardial
uptake and metabolism of the tracer is markedly higher CPT I dependent.
When
compared
to
[11C]-S-methyl-14-thiomyristic
acid
(b),
[11C]-S-methyl-12-thiododecanoic acid (c) and [11C]-palmitate, a showed an
early high uptake and a significantly slower late clearance in heart and a
prolonged myocardial elimination half-life (30 min). Analysis of heart tissue and
urine samples showed that a was metabolized via beta-oxidation in
myocardium. Small animal PET images of the accumulation of a in the rat
myocardium were clearly superior to [11C]-palmitate. These initial studies
suggest that a could be a potentially useful clinical PET tracer to assess
myocardial fatty acid metabolism.
Keywords: Fatty acid; [11C]-palmitate; PET; Beta-oxidation; Myocardial
imaging.

ACCEPTED MANUSCRIPT
1. Introduction
Fatty acid (FA) is the primary energy substance in normal heart, which can
satiate 60-80% of aerobic metabolism via the β-oxidation pathway [ 1].
Disturbances in FA metabolism usually reflect cardiac dysfunction, such as
myocardial ischemia and heart failure [2]. Therefore, the radiolabeled FA
analogs may offer an easy visualization of myocardial energy metabolism
using positron emission tomography (PET) or single-photon emission
computed tomography (SPECT).
Currently, a number of fatty acid-based radiopharmaceuticals have been
developed to estimate myocardial metabolism by PET/SPECT and to delineate
specific metabolic pathways [2]. [11C]-palmitate, a classical 16-carbon FA
labeled with C-11 at the C-1 position, is most commonly used for measuring
myocardial FA metabolism in PET studies [3]. However, fatty acids (e.g.,
palmitic acid) in cardiac tissue are thoroughly degraded in a two-minute
process [4]. Many groups have tried to block the oxidative degradation by
insertion of an S-atom or substitution with a methyl group for a trapping effect
of the tracer. 1-[11C]-beta-R,S-methylheptadecanoic acid ([11C]-BMHA) is the
first β-methyl modified fatty acid that showed the β-oxidation process in
cardiac tissue and its inhibition by the methyl group [5]. [11C]-BMHA has high
retention of activity in the myocardium and excellent imaging properties,
because it is trapped in the cardiomyocyte owing to incomplete metabolism [4].
[123I]-BMIPP, another branched-chain fatty acid, has been investigated

ACCEPTED MANUSCRIPT
extensively and is a commercially successful cardiac SPECT agent for clinical
use [6,7,8]. The thia fatty acid labeled with F-18, such as [18F]-FTHA, [18F]-FTO,
[18F]-FTP, is commonly used to estimate certain details of myocardial FA
metabolism [9,10,11], which demonstrated excellent properties as myocardial
PET tracers. This analog has a sulfur heteroatom substitution near the
carboxyl group to block beta-oxidation of the molecule and an 18F label to allow
quantitative imaging with PET [12]. The technetium-99m-labeled fatty acids
have been developed for many years, but none of them has a potential use in
clinical practice due to its poor heart-to-background uptake ratios.
In general the modified tracers, which are labeled with various
radioisotopes (e.g., F-18 or I-123), or contain other steric or functional
perturbations of the FA structure (e.g., introduction of substituent group), may
not be metabolized in the same way as their natural analogs do and thus may
give one-sided or even misleading results. In addition, an application of FA
labeled with long-lived isotope (18F, 123I) is not really necessary for the potential
of higher image quality because of rapid catabolism of FA. Thus 11C (t1/2 = 20.4
min) labeled FA analogs can be considered as a feasible alternative to
11
radioiodinated or radio-fluorinated FA analogs. Currently, C-labeled palmitic
analog, [11C]-palmitate, has been used extensively in clinical research with
PET [13,14]. Since the C-11 carbon is cleaved in the initial step of β-oxidation,
[11C]-palmitate showed biexponential clearance from cardiac tissue,
significantly complicating data analysis [4,15]. [ω-11C]-palmitate with terminal

ACCEPTED MANUSCRIPT
labeling with C-11 has a higher heart uptake ratio than [11C]-palmitate, due to
avoidance of the C-11 β-oxidation of [ω-11C]-palmitate in the initial step [16]. In
spite of the progress made, the tedious synthesis process and the unstable
Grignard precursor of [ω-11C]-palmitate limits its clinical application. As a result,
discovering new 11C-labeled fatty acid analogs that are easily synthesized and
function similarly as their natural analogs in myocardium is essential for the
development of the next-generation FA tracers, which still presents a major
challenge in this field.
Recently, DeGrado and colleagues demonstrated that uptake of
radiolabeled FA and metabolic mechanism in various tissues can be
modulated by varying the structures of fatty acid. Furthermore, many research
findings revealed that there was a direct relationship between the tissue
distribution of the FA tracer and the calculated solvation energy level of the
radiotracer and its corresponding octanol-water partition coefficients (log P)
[17,18]. This discovery has been shown to serve as a practical guide in design
of radio-labeled fatty acid tracer [18]. We hypothesized that the introduction of
the carbon-11 at the terminal of FA through sulfur atom would make it easy to
produce the 11C-labeled FA analog, which reflects the metabolic mechanism of
its natural counterpart just as [11C]-palmitate or [ω-11C]-palmitate does. Hence,
ω-sulfydryl fatty acids as the precursors were coupled with [11C]-methyl iodide
for the efficient preparation of [11C]-S-methyl-16-thiopalmitic acid ([11C]-MTPA,
a),
[11C]-S-methyl-14-thiomyristic
acid
([11C]-MTMA,
b)
and

ACCEPTED MANUSCRIPT
[11C]-S-methyl-12-thiododecanoic acid ([11C]-MTDA, c). Our present study
provides a simple radiosynthesis route for 11C-labeled FA, and the initial results
measuring myocardial uptake and metabolic mechanism of those tracers in
mice. Small-animal PET imaging studies of a were also performed in rats and
its pharmacokinetics were compared with that of [11C]-palmitate.
Scheme 1. Synthesis of compounds [11C]-MTPA (a), d, e.
Scheme 2. Synthesis of [11C]-MTMA (b) and [11C]-MTDA (c).
2. Results
2.1.Chemistry
The synthesis of reference compounds and radioactive tracers was
1
performed according to Scheme 1 and 2 and characterized by H-NMR,
13C-NMR and ESI-MS. The preparation of reference compounds (a’-e’) was
based on an one-step synthetic route, i.e. methylation of ω-sulfhydryl FA with
methyl iodide or substitution reaction of ω-bromo-fatty acids with NaSMe

ACCEPTED MANUSCRIPT
afforded S-methyl-16-thiopalmitic acid (a’), S-methyl-14-thiomyristic acid (b’),
S-methyl-12-thiododecanoic acid (c’), S-methyl-4-thiobutyric acid (d’) and
S-methylthioacetic acid (e’) in high yield (> 63% yield). The formations of five
reference compounds except for compound e’ were characterized by the
appearance of one single peak and two triplet peaks in the 1H-NMR spectrum.
The single peak at δ 2.10-2.13 ppm denoted three protons (CH3S) and two
triplet peaks respectively denoted two protons (CH2CO) at δ 2.23-2.31 ppm
and two protons (CH2S) at δ 2.45-2.49 ppm. The carbon NMR spectrum
verified the new methyl carbon (CH3S) at δ 17.51-18.10 ppm and the carboxyl
carbon at δ 177.08-179.90 ppm. The product formations were further
confirmed by mass spectrometry. Based on these results mentioned above,
compounds a’-e’ were considered as the acceptable reference standards for
evaluating the corresponding radioactive tracers.
2.2.Radiochemistry
The radiosynthesis for all radiolabeling FAs was performed using a
three-step reaction sequence. First, in-target produced [11C]-CO2 was reduced
using LiAlH4 to [11C]-CH3OH complexes; second, an iodination reaction with
[11C]-CH3OH complexes in the presence of hydrogen iodide aqueous afforded
n.c.a [11C]-CH3I; finally, a methylation of the precursors (1-5) using [11C]-methyl
iodide produced target compounds (a-e), shown in Scheme 1 & 2. After the
methylation of ω-sulfhydryl fatty acids, the tracers were isolated by removal of
solvent using a rotary evaporator, this also ensured the removal of [11C]-methyl

ACCEPTED MANUSCRIPT
iodide and other low molecular weight volatile [11C] compounds. The
evaporated residues were reconstituted with 5 mL 0.1M phosphate buffer
(pH=7) and analyzed by TLC. To further identify the tracers (a-c), the products
and their reference compounds were coinjected into analytical HPLC system
using the mobile phase of acetonitrile and water containing acetic acid (0.5% v)
(90/10 v/v) at a flow rate of 1 mL/min. The radio peaks of tracers are highly
consistent with the UV peaks of reference compounds at the retention time of
14.1 min (a), 10.2 min (b) and 6.5 min (c) (Supporting Information Figure S1),
indicating high radiochemical purity of tracers. HPLC analysis had produced
results essentially in agreement with those using TLC analysis. The overall
radiochemical yields ranged 41.4-53.1% (336-430 mCi, 12.432-15.910 GBq;
decay-corrected), the radiochemical purity was almost 100%, the specific
activity was estimated with about 231-301 GBq/µmol by TLC, and the total
radiosynthesis time was about 35 min, which indicated that tracers had the
necessary characteristics for clinical use.
2.3.Physicochemical properties
The stability of the tracer (a, b or c) was determined in ex vivo and in vivo
serum by using TLC. The percentage of the remaining tracer was over 97.8%
in ex vivo serum and over 94.6% in in vivo serum. This stability studies
demonstrated their high in vitro/in vivo stability to justify further investigation as
myocardial imaging agents.

ACCEPTED MANUSCRIPT
The solubility and biodistribution mechanism were studied by measuring
octanol-water partition coefficients (log P) of the tracers. The log P values of a,
b and c were 5.26 ± 0.03, 4.31± 0.10 and 3.12 ± 0.07, respectively. a showed
higher log P than b and c, which indicated a was more hydrophobic.
Table 1. Preliminary Calculation of log P, Molecular Volume and Solvation Free Energies
of the Fatty Acid Analogs. a
radiotracer
Molecular volume b Solvation energy c log P
[11C]-MTPA (a)
[11C]-MTMA (b)
[11C]-MTDA (c)
1079.35
884.54
797.63
-828.76
-750.15
-632.23
-857.97
-779.36
-700.75
-810.38
-981.84
-1021.15
-888.71
-926.78
-826.07
5.62(5.26)d
4.73(4.31)d
3.52(3.12)d
6.59
Octadecanoic acid 1082.22
Palmitate acid
Myristic acid
[11C]-BMHA
967.33
5.72
855.45
4.90
1045.43
1207.62
1249.55
1022.53
1077.05
978.55
6.41
[
123I]-IPPA
8.27
[
123I]-BMIPP
8.57
[18F]-FTHA
[18F]-FTO
[18F]-FTP
4.68
4.91
4.50
a Calculations for molecular volume, free energy of solvation and log P using Gaussian 03
implemented with b3lyp/lanl2dz.
b The molar volume of the fatty acid analogs (cm3 mol-1).
c The free energy of the fatty acid analogs in water (a.u.).
d experimental value.
Figure 1. Radio-labeled fatty acids analogs.

ACCEPTED MANUSCRIPT
The log P values and volumes of the FAs were calculated using the
Gaussian 03 implemented with b3lyp/lanl2dz (Table 1 and Figure 1). The
calculated values (a: 5.66, b: 4.78, c: 3.57) are approximate to the
experimental data of the (n)octanol-water partition coefficients, while only log P
of a is closely approximate to that of palmitate acid (5.72) but significantly less
than that of octadecanoic acid (6.59). In comparison with the natural FAs
(Octadecanoic acid, Palmitate acid, Myristic acid), the FA analogs (a, b, c,
[18F]-FTHA, [18F]-FTO and [18F]-FTP) had been inserted by sulphur atom which
was only 1.255 times volume of carbon group. The molecular volume of the
radiotracers, such as a, [18F]-FTHA, [18F]-FTO and [18F]-FTP, is approximate to
those of their natural FAs, while the rest is outside the range. The solution free
energies of the FAs in water were calculated using Gaussian 03 implemented
with b3lyp/lanl2dz PCM. The solvation energies of the natural FAs are
between -700.75 to -857.97 a.u., only a (-828.76 a.u.) and b (-750.15 a.u.) had
fallen into this range.
2.4.Biodistribution in mice
Myocardial uptake of a purified by HPLC or not in fasted mice was
statistically unchanged respectively at 0.5, 5 and 30 min p.i., indicating higher
radiochemical purity of a without further HPLC purification (Figure 2,
Supporting Information Table S2). It was also observed that HPLC-treated a
had slightly high uptake in heart at 0.5, 5 and 30 min postinjection (p.i.), while

ACCEPTED MANUSCRIPT
there was no significant difference in the two activity uptakes in liver and kidney,
demonstrating no need for further HPLC purification of a in the animal studies.
Figure 2. Comparison of heart uptake of [11C]-MTPA with/without purification using HPLC
at 0.5, 5 and 30 min p.i. in fasted mice.
Table 2. Biodistribution of 11C-labelled Fatty Acids in Kunming Mice (%ID/g, x±SD, n = 3)
Tissue/organ
0.5 min
5 min
[11C]- MTPA (a)
0.58 ± 0.05
10 min
30 min
60min
Blood
1.92 ± 0.39
0.12 ± 0.01
3.28 ± 0.15
3.42 ± 0.21
0.84 ± 0.10
0.84 ± 0.07
2.87 ± 0.27
0.58 ± 0.08
0.37 ± 0.06
0.51± 0.09
1.71
0.39 ± 0.04
0.31 ± 0.05
2.55 ± 0.16
2.78 ± 0.14
0.69 ± 0.08
0.78 ± 0.14
3.32 ± 0.21
0.53 ± 0.09
0.43 ± 0.06
0.49 ± 0.01
6.54
0.18 ± 0.01 0.08 ± 0.02
0.28 ± 0.06 0.11 ± 0.01
Brain
0.20 ± 0.02
2.89 ± 0.18
3.82 ± 0.19
0.75 ± 0.04
1.09 ± 0.14
3.33 ± 0.45
0.83 ± 0.04
0.49 ± 0.02
0.42 ± 0.02
4.98
Heart
1.82 ± 0.23 0.92 ± 0.17
1.17 ± 0.12 1.00 ± 0.14
0.81 ± 0.20 0.75 ± 0.03
0.35 ± 0.17 0.14 ± 0.04
1.42 ± 0.31 1.20 ± 0.27
0.39 ± 0.02 0.69 ± 0.06
0.39 ± 0.04 0.13 ± 0.02
0.44 ± 0.00 0.14 ± 0.00
Liver
Spleen
Lung
Kidney
small intestine
Muscle
Bone
Heart/Blood
Heart/Liver
Heart/Lung
10.11
1.56
5.20
11.50
0.92
6.57
0.96
0.76
0.92
3.90
2.65
3.27
[11C]-MTMA (b)
Blood
2.11 ± 0.42
0.19 ± 0.01
2.93 ± 0.25
3.58 ± 0.62
0.93 ± 0.08
1.45 ± 0.65
3.87 ± 0.67
0.61 ± 0.02
0.56 ± 0.03
0.17 ± 0.01
2.03 ± 0.21
3.93 ± 0.23
0.80 ± 0.07
1.33 ± 0.08
3.62 ± 0.21
0.48 ± 0.05
0.41 ± 0.07
0.14 ± 0.00
1.82 ± 0.09
3.15 ± 0.21
0.73 ± 0.08
0.74 ± 0.11
3.76 ± 0.26
0.45 ± 0.04
0.20 ± 0.03 0.09 ± 0.00
0.09 ± 0.01 0.02 ± 0.00
Brain
Heart
1.12 ± 0.11
0.51 ± 0.03
Liver
2.10 ± 0.15 1.12 ± 0.09
0.66 ± 0.07 0.54 ± 0.07
0.40 ± 0.05 0.15 ± 0.01
2.12 ± 0.12 1.06 ± 0.09
0.38 ± 0.07 0.24 ± 0.01
Spleen
Lung
Kidney
small intestine

ACCEPTED MANUSCRIPT
Muscle
0.63 ± 0.02
0.41± 0.01
1.39
0.58 ± 0.11
0.44 ± 0.06
3.63
0.51 ± 0.06
0.45 ± 0.05
4.44
0.42 ± 0.06 0.16 ± 0.01
0.32 ± 0.04 0.17 ± 0.02
Bone
Heart/Blood
Heart/Liver
Heart/Lung
5.60
0.53
2.80
5.67
0.46
3.40
0.82
0.52
0.58
2.02
1.53
[11C]-MTDA (c)
2.46
Blood
1.87 ± 0.43
0.22 ± 0.05
2.37 ± 0.75
3.55 ± 0.99
0.97 ± 0.07
1.79 ± 0.11
3.52 ± 0.79
0.88 ± 0.03
0.55 ± 0.05
0.41± 0.04
1.27
0.52 ± 0.02
0.21 ± 0.04
1.84 ± 0.13
3.21 ± 0.12
0.78 ± 0.02
1.29 ± 0.34
3.26 ± 0.32
0.87 ± 0.04
0.50 ± 0.01
0.50 ± 0.02
3.54
0.44 ± 0.00
0.18 ± 0.02
1.73 ± 0.18
3.02 ± 0.13
0.70 ± 0.03
0.87 ± 0.28
3.39 ± 0.21
0.78 ± 0.05
0.46 ± 0.01
0.55 ± 0.01
3.94
0.24 ± 0.01
0.11 ± 0.01
Brain
0.05 ± 0.01 0.05 ± 0.00
0.86 ± 0.22 0.25 ± 0.08
Heart
Liver
1.38 ± 0.11
0.34 ± 0.01
0.36 ± 0.07
0.11 ± 0.00
Spleen
Lung
0.31 ± 0.15 0.17 ± 0.05
1.58 ± 0.28 0.44 ± 0.09
0.36 ± 0.03 0.14 ± 0.01
0.12 ± 0.01 0.12 ± 0.02
0.22 ± 0.02 0.19 ± 0.03
Kidney
small intestine
Muscle
Bone
Heart/Blood
Heart/Liver
Heart/Lung
3.58
0.62
2.77
2.27
0.69
1.47
0.67
0.57
0.57
1.32
1.43
1.99
To evaluate the pharmacokinetics of the tracers in the heart, the
biodistribution experiments were performed in fasted mice. After injecting a
into mice, high levels of radioactivity accumulated in blood, heart, liver, and
kidneys at 0.5 min p.i., and the myocardial uptake washed out slowly
(3.28 %ID/g at 0.5 min, 2.89 %ID/g at 5 min, 2.55 %ID/g at 10 min, 1.82 %ID/g
at 30 min, and 0.92 %ID/g at 60 min p.i.) with an elimination half-life of 30 min.
Furthermore, the myocardial uptakes of a were higher than those of b and c at
all time points, indicating longer myocardial retention due to the longer FA
chain of a (Table 2). In contrast, low uptakes of those tracers were observed in
the other tissues, except for liver and kidney between 10 and 60 min. The
heart-to-blood uptake ratios of a increased from 1.71 at 0.5 min p.i. to 11.50 at
60 min p.i., which was higher than those of b (from 1.39 at 0.5 min p.i. to 5.67

ACCEPTED MANUSCRIPT
at 60 min p.i.) and c (from 1.27 at 0.5 min p.i. to 2.27 at 60 min p.i.). The
heart-to-lung uptake ratios for a were 1.9-, and 4.5-fold higher than b and c
between 30 and 60 min, respectively. Heart-to-liver uptake ratios for a were
2.9- and 2.5-fold higher than b and c at 30 min p.i., respectively. The results
demonstrated that a was superior to b and c in the uptake ratios of
heart-to-tissues at all time points.
Figure 3. Biodistribution of 11C-labelled fatty acids with/without etomoxir-treated at 5 and
30 min p.i. in fasted mice.
It was also observed that mice treated with etomoxir (a CPT I inhibitor) had
relatively higher uptake in blood. Etomoxir-treated mice with a, b, c and
[11C]-Palmitate showed 2.1-, 1.7-, 1.5- and 1.5-fold higher blood uptake at 30
min than control mice, respectively. In contrast, such enhancement with
etomoxir treatment was not observed in heart, liver, lung, kidney, and muscle.
For example, pretreatment of mice with etomoxir decreased the myocardial
uptake of a by 63.7% and 76.9% at 5 and 30 min, respectively (Figure 3 and
Supporting Information Table S2).

ACCEPTED MANUSCRIPT
Figure 4. TLC analyses of aqueous fraction (A), organic fraction (B) and urine samples (C)
obtained at 5 min and 30 min after injecting mice with [11C]-MTPA (a), and TLC profile of d
(D), e (E), a (F).
2.5.Metabolite analysis in mice
Radioactive metabolites of a were analyzed by TLC in heart tissue samples
and in urine at 5 and 30 min postinjection. Analysis of aqueous fractions
showed two radioactive peaks with TLC Rf values of 0.76 and 0.65,
respectively, which were identified as d and e (Figure 4 A, D, E), suggesting
that a was metabolized to d and e after six or seven cycles of β-oxidation.
Analysis of the urine samples also showed the two radioactive peaks with the
same Rf values except for an unidentified tiny peak (Rf = 0.32) (Figure 4C).

ACCEPTED MANUSCRIPT
Analyses of the organic fractions showed the unmetabolized a (Figure 4 B, F)
and an unidentified nonpolar compound (Rf = 0.95). Most of the radioactivity in
control heart tissues was found in tissue pellet (57.10%, 60.51% of total at 5
and 30 min postinjection with a, respectively), which was significantly lower
relative to b (68.88%, 66.61%) and c (65.09%, 62.24%) (Table 3). In addition,
the radioactivity in tissue pellet was reduced to about half when mice were
pretreated with etomoxir at 2 h before administration of a. The radioactivity in
tissue pellet fractions of b and c were similar to a in the control and
etomoxir-pretreated mice (Table 3).
Table 3. Distribution of Radioactivity in Heart Homogenates. (% of Total, n=3)
Heart
5min
5min
etomoxir a
9.85 ± 0.67
30min
control
30min
etomoxir a
6.42 ± 0.74
fraction
control
aqueous 4.30 ± 0.19
4.79 ± 0.86
[11C]-MTPA (a)
[11C]-MTMA (b)
[11C]-MTDA (c)
[11C]-Palmitate
organic
pellet
35.19 ± 1.23 63.21 ± 4.36
60.51 ± 3.21 26.94 ± 2.42
38.11 ± 1.65 70.79 ± 4.86
57.10 ± 3.11 22.79 ± 1.54
aqueous 2.78 ± 0.11
11.54 ± 1.29
3.35 ± 0.33
4.55 ± 0.53
organic
pellet
28.34 ± 0.98 59.31 ± 4.72
68.88 ± 6.21 29.15 ± 1.65
30.04 ± 2.12 65.16 ± 5.21
66.61 ± 2.33 30.29 ± 1.65
aqueous 2.81 ± 0.14
12.01 ± 1.75
32.10 ± 2.21 57.16 ± 3.27
65.09 ± 5.34 30. 83 ± 2.89 62.24 ± 7.32 26.38 ± 1.42
10.23 ± 0.97 4.05 ± 0.51 5.01 ± 0.39
3.72 ± 0.65
5.12 ± 1.11
organic
pellet
34.04 ± 2.75 68.50 ± 2.44
aqueous 3.56 ± 0.29
organic
pellet
26.45 ± 1.67 58.65 ± 3.31
69.99 ± 5.63 31.12 ± 2.54
34.15 ± 3.41 68.43 ± 4.75
61.80 ± 6.65 26.56 ± 3.11
a Pretreated with etomoxir (40 mg/kg ip) 2 h prior.
2.6.Small-animal PET imaging in rats
A preliminary small-animal PET imaging study was performed with a and
[11C]-palmitate in fasted rats (Figure 5). The images clearly demonstrated the
high level of both tracer accumulations in the hearts, with excellent definition of

ACCEPTED MANUSCRIPT
the left ventricle from the lungs, kidneys and blood pool at 30 min p.i. (Figure
5A, B). However, [11C]-MTPA images displayed more excellent quality and
higher activity relative to [11C]-palmitate. Furthermore, there was no significant
accumulation of a in muscle and bone. Initial dynamic images showed rapid
accumulation of both tracers in the heart followed by a rapid washout within 3
min and immediate increase of radioactivity in cardiac blood pool followed by a
rapid decline in approximately 5 min (Figure 5C, D). However, some
differences were noted when the data were compared with the time-activity
curves of [11C]-palmitate. a was retained at a relatively high and stable level in
the myocardium from 10 to 30 min after injection. In addition, the accuracy of
the ROI quantitative analysis was confirmed by biodistribution experiments
after the last PETscan at 30 min p.i., which showed that the PET data agreed
well with the biodistribution data (Supporting Information Table S3). The
results demonstrated good visualization of the heart, with excellent
heart-to-background contrast at all time points, which were roughly the same
as the biodistribution studies in mice, despite the different animal models used.

ACCEPTED MANUSCRIPT
Figure 5. Representative small-animal PET images of [11C]-MTPA (A) and [11C]-palmitate
(B) in fasted SD rats. Images acquired at 30-35 min p.i. are displayed on the coronal,
sagittal, and transverse axis. [11C]-MTPA images show excellent quality and lower
background than the [11C]palmitate images. Time–activity curves for ROIs drawn over left
ventricular wall and cardiac blood pool of rat are shown in panel C for [11C]-MTPA and
panel D for [11C]-palmitate. [11C]-MTPA was retained at constant level in myocardium and
rapidly washed out from blood.
3. Discussion
Long-chain FAs (chain length of 14-18 carbon atoms) are the major
sources of energy in heart, metabolized via β-oxidation under aerobic

ACCEPTED MANUSCRIPT
conditions [19]. PET imaging technique with FA-oxidation specific can improve
greatly the diagnostic level of heart disease by determining myocardial viability.
A variety of palmitic analogs labeled with 11C, 18F, or 123I demonstrate excellent
properties for the high uptake or retention in the myocardium. However, those
tracers involve complex catabolic and anabolic processes and kinetic models
to estimate the FA metabolism [4,20]. Currently, the FA radiotracers have been
mostly labeled with F-18 at the terminal position, where metabolic
defluorination happens in liver [ 21 ]. Conversely, the particular area of
11C-labeled FA has practically been neglected since the clinical application of
[11C]-palmitate, although C-11 has a relatively short half life (t1/2 = 20.4 min)
meaning it naturally becomes safer somewhat quickly.
We report herein the rational design and experimental validation of a mimic
FA tracer (a) for myocardial imaging based on a FA metabolic mechanism by
β-oxidation. The fatty acid chain length of 16 or 17 carbon atoms has been
known to be optimal for myocardial imaging [22,23]. The sulfur atom,
increasing the local polarity of a, substitutes for the methylene group at
17-position of octadecanoic acid, which makes the radiolabeling easier
through reaction with [11C]-methyl iodide. Hence, we first calculated the
solvation energy, molecular volume and log P value of FA and its
corresponding natural FA with density functional theory (DFT) at the
b3lyp/lanl2dz. Results from our calculation revealed that the solvation energy
level of a was between those of palmitate acid and octadecanoic acid, and the

ACCEPTED MANUSCRIPT
log P value of a was closely approximate to that of palmitate acid but much
lower than its counterpart’s, while the molecular volume was approximate to its
counterpart’s. [123I]-IPPA and [123I]-BMIPP with the lowest solvation energy and
the largest molecular volume and log P are different from the other tracers
showed in Table 1. Thus, if the16-thiopalmitic acid as a precursor is covalently
attached to [11C]-methyl group, the tracer should have the similar
physical-chemical properties including log P, and the desirable biocompatibility
and facile synthesis comparing to [ω-11C]-palmitate and [11C]-palmitate.
As a proof of concept, we developed a according to above calculation,
which mimicked octadecanoic acid as a template molecule due to their similar
structures including molecular volume. To understand the structure-bioactivity
relationship of a, three additional radiotracers (b, c, and [11C]-palmitate) were
synthesized and allowed for the study of the effects of chain length (a vs b, c)
and the ω-thia substituent (a vs [11C]-palmitate). It was observed that a was
superior to b and c in myocardial uptake, myocardial retention, and ratios of
heart-to-tissues (blood, liver and lung) at all time points (Table 2). Here, the
retention kinetics may be dependent on the length and myocardial uptake of
fatty acids. In addition, b showed better myocardial uptake and myocardial
retention than [11C]-palmitate at 30 min (Figure 3 and Supporting Information
Table S2), because the C-11 carbon of [11C]-palmitate is cleaved in the initial
step of β-oxidation, and compound b is just the opposite. Similarly, a also
showed the higher myocardial uptake and better myocardial retention than

ACCEPTED MANUSCRIPT
[11C]-palmitate. Those results suggested that extending the chain length of the
analogs (b or c) by two or four carbons to form a or introducing an 11C-labeled
methyl group at the terminal of FA through sulfur atom resulted in similar
uptake and biological properties but lengthened myocardial retention. In
mitochondria the transport of long-chain fatty acids are performed by the outer
membrane proteins, CPT I. Pharmacologic inhibition of CPT I with etomoxir in
mice was found to reduce myocardial uptake of a by >63.7% while reduce that
of [11C]-palmitate by >55.2%, demonstrating excellent sensitivity to change in
myocardial FA oxidation. To further clarify the nature of the myocardial FA
oxidation, a metabolite analysis experiment of a was performed with heart
homogenates and urine samples. Radioactivity was predominantly retained in
tissue pellet, and the remainder extracted was retained in the organic and the
aqueous fractions. The β-oxidation metabolites, short thiafatty acid analogs
such as d and e, were detected in either the aqueous fractions obtained from
heart tissues or urine samples. Furthermore, the metabolite d as a substrate
for the subsequent β-oxidation step was gradually degraded to the metabolite
e (Figure 4A, C). In the organic fractions, a part of a was converted into the
more nonpolar radiolabeled metabolite with time (Figure 4B). Unfortunately,
this nonpolar radiolabeled metabolite in the organic fraction remains to be
identified. The radioactivity in the tissue pellet, markedly higher CPT I
dependent which shows high sensitivity of a to reflect beta oxidation rates

ACCEPTED MANUSCRIPT
(Table 3), may be derived from protein-bound metabolites or other complexes
similar to palmitic acid, though further research is needed to confirm that.
The alkylation reaction of sulfhydryl compound is useful for synthesizing
various S-[11C]-methyl-ω-sulfhydryl fatty acids [9,11]. Our synthesis procedure
has greatly simplified the radiosynthesis and purification for FA tracer. The high
radiochemical yield (41.4-53.1%) and radiochemical purity (approximate 100%)
indicate that the synthesis procedure is an approach of choice to rapidly obtain
a for the animal studies. Small-animal PET image of a in rats showed clear
outline of heart during the time of study, while a defects’ borders of the upper
liver and the inferior heart overlapped in [11C]palmitate PET image (Figure 5A,
B). Due to the short biological half-life of FA in heart, the tissue time-activity
curves of a and [11C]-palmitate were consistently displayed for 0-30 min
(Figure 5C, D), which provided sufficient information for measuring the kinetics
of FA metabolism. a exhibited significant myocardial uptake and enhanced
myocardial retention. However, some slight but significant differences in the
time-activity curves were noted when compared with those of [11C]-palmitate.
For example, a exhibited an early activity plateau followed by a rapid early
clearance in heart (at the first 3 min) and a significantly slower late clearance
(at 10 min later). The time-activity curves reveal a difference in FA metabolism
and washout kinetics of a versus [11C]-palmitate due to the position of the
radiolabel (S-17 position in a vs C-1 position in [11C]-palmitate). The results

ACCEPTED MANUSCRIPT
demonstrated superior myocardial imaging abilities of a, which could be useful
for clinical PET imaging of myocardial diseases.
4. Conclusions
Three 11C-labeled, thia-substituted FA analogs were rationally designed
using computational chemistry and successfully synthesized with high
radiochemical yield (41.4-53.1%) and radiochemical purity (approximately
100%). The tracers had suitable lipophilicity, high stability and specific activity
(about 231-301 GBq/µmol). Biodistribution and PET imaging studies of a were
performed to validate the structure-activity relationship. The enhanced
myocardial uptake and retention of a are primarily attributed to the longer chain
length and the radiolabeling at ω-position of FA. Myocardial uptake and
metabolism of the tracers appear to be markedly higher CPT I dependent.
Radioactive metabolite analysis suggested that a was metabolized via
β-oxidation in myocardium. Preliminary images of the accumulation of a in the
rat myocardium were clearly superior to [11C]-palmitate. In summary, a is a
promising FA-oxidation tracer with the ω-thia-substituted FA analog and
deserves for further preclinical evaluation.
5. Experimental section
5.1.Material and methods

ACCEPTED MANUSCRIPT
Chemicals and solvents were purchased from commercial suppliers and
used without additional purification. Nuclear magnetic resonance (NMR)
spectra were recorded on a Bruker ARX-400 (500 MHz) spectrometer. Mass
spectra (MS) were obtained with a Brucker Apex IV FTM instrument.
Chromatograms of Radio-Thin Layer Chromatography (Radio-TLC) were
recorded with a Bioscan System 200 (Washington, DC, USA) on 250-µm,
silica gel AL SILG/UV plates (Whatman Limited, Kent, UK). High-pressure
liquid chromatography (HPLC) was performed on a Gold HPLC System
equipped with a UV detector (210 nm) and a radioisotope detector.
Radioactivity was measured in a dose calibrator (CRC-15R, Capintec) and
tissue radioactivity in a Bioscan-γ-detector 2D-6000.
Sprague-Dawley rats (250±10 g) and Kunming mice (25±2 g) were
provided by the Slac laboratory animal Co. Ltd., Shanghai. All the animal
protocols were approved by the Institute’s Animal Care and Use Committee
and conformed to the Guide for the Care and Use of Laboratory Animals.
5.1.1. S-methyl-16-thiopalmitic acid (a’). The synthesis of the reference
compound (a’) was conducted according to the reaction procedure illustrated
in Scheme 1. The reaction flask was charged with 16-mercaptohexadecanoic
acid (1, 1.44 g, 5 mmol), sodium hydroxide (0.40 g, 10.20 mmol) and methanol
(30mL). Methyl iodide (0.32 mL, 5.1 mmol) was added to this solution and
stirred at room temperature for 48 h. After its acidity and alkalinity were
regulated by ethylic acid, the solvent was removed under reduced pressure.

ACCEPTED MANUSCRIPT
Flash column chromatography (ethyl acetate-hexane, 10:90) gave a’ as a pale
1
yellow oil (0.95 g, 63.1%). H NMR (500 MHz, CDCl3) δ 1.22-1.31 (m, 22H),
1.54-1.63 (m, 4H), 2.13 (s, 3H), 2.23-2.26 (t, 2H), 2.47-2.49 (t, 2H); 13C-NMR
(CDCl3, 125 MHz) δ 18.10, 24.76, 28.55, 29.01, 29.06, 29.09, 29.66, 35.01,
36.04, 38.35, 177.08; MS (ESI): m/z 302.50 (M+).
5.1.2. S-methyl-4-thiobutyric acid (d’). Compound d’ was synthesized in
87.9% yield using the procedure described for a’. 1H NMR (500 MHz, CDCl3) δ
1.90-1.96 (m, 2H), 2.12 (s, 3H), 2.28-2.31 (t, 2H), 2.47-2.49 (t, 2H); 13C-NMR
(CDCl3, 125 MHz) δ 17.51, 24.36, 34.90, 35.14, 177.58; MS (ESI): m/z 134.04
(M+).
5.1.3. S-methylthioacetic acid (e’). Compound e’ was synthesized in 92.2%
yield using the same procedure described for a’. 1H NMR (500 MHz, CDCl3) δ
2.11 (s, 3H), 3.44 (s, 2H); 13C-NMR (CDCl3, 100 MHz) δ 17.10, 47.35, 173.12;
MS (ESI): m/z 106. 02 (M+).
5.1.4. S-methyl-14-thiomyristic acid (b’). Synthesis of reference compound b’
was performed as described in Scheme 2 [ 24 , 25 , 26 ]. A mixture of
14-bromomyristic acid (4, 1.53 g, 5 mmol) and NaSMe (0.7 g, 10 mmol) in
MeOH (15 ml) was refluxed under nitrogen atmosphere for 12 h. After cooling
to room temperature, the reaction mixture was poured onto ice water (80 ml),
acidified with HCl (1 mol/L) to pH=1, and extracted with CHCl3 (50 ml, 3 times).
The CHCl3 extract was dried with MgSO4 and the solvent was removed under
reduced pressure. Flash column chromatography (ethyl acetate-hexane, 10:90)

ACCEPTED MANUSCRIPT
1
gave b’ as a pale yellow oil (0.98 g, 71.8%). H NMR (500 MHz, CDCl3) δ
1.23-1.30 (m, 18H), 1.54-1.61 (m, 4H), 2.10 (s, 3H), 2.23-2.26 (t, 2H),
13
2.47-2.49 (t, 2H); C-NMR (CDCl3, 125 MHz) δ 17.54, 24.81, 28.32, 28.88,
29.00, 29.35, 29.66, 30.62, 35.44, 36.29, 179.04; MS (ESI): m/z 274.6 (M+).
5.1.5. S-methyl-12-thiododecanoic acid (c’). Compound c’ was prepared as
the synthesis of b’ described above by coupling 12-bromododecanoic acid (5,
1.40 g, 5 mmol) with NaSMe (0.7 g, 10 mmol) in MeOH (15 ml), which afforded
c’ as an oil (0.96 g, 78%).1H NMR (500 MHz, CDCl3) δ 1.22-1.30 (m, 14H),
13
1.54-1.58 (m, 4H), 2.11 (s, 3H), 2.23-2.26 (t, 2H), 2.45-2.47 (t, 2H); C-NMR
(CDCl3, 125 MHz) δ 17.96, 24.66, 28.45, 28.95, 29.03, 29.33, 29.92, 34.01,
34.04, 39.29, 179.90; MS (ESI): m/z 246.2 (M+).
5.1.6. 14-mercaptomyristic acid (2). Synthesis of 14-mercaptomyristic acid
was performed as described in the literature [27,28] (Scheme 2). A mixture of
14-bromomyristic acid (5.00 g, 16.3 mmol) and thiourea (2.10 g, 27.60 mmol)
in ethanol (50 mL) was refluxed under nitrogen atmosphere for 1 day. After
cooling off to room temperature, the solvent was evaporated under reduced
pressure and the residue heated with NaOH (50 mL, 7.5 mol/L) for another 6 h.
The mixture was carefully acidified with HCl (1 mol/L), the organic layer
separated and the aqueous phase extracted as usual. Purification of the
remaining raw material was accomplished by distillation under vacuum. Yield
1
of 14-mercaptomyristic acid: 3.22 g, 76%. H NMR (500 MHz, CDCl3) δ

ACCEPTED MANUSCRIPT
1.28-1.39 (m, 18H), 1.58-1.66 (m, 4H), 2.35 (t, 2H), 2.53 (m, 2H); MS (ESI):
m/z 260.42 (M+).
5.1.7. 12-mercaptododecanoic acid (3). 12-mercaptododecanoic acid was
prepared (3.13 g, 83% yield) using a procedure similar to that used for the
1
synthesis of compound 2 described above. H NMR (500 MHz, CDCl3) δ
1.25-134 (m, 14H), 1.55-1.63 (m, 4H), 2.31-2.34 (t, 2H), 2.47-2.52 (m, 2H); MS
(ESI): m/z 232.40 (M+).
5.2.Radiochemistry
Positron-emitting [11C]-CO2 were produced by 14N(p,a)11C nuclear
reactions, via proton irradiation of N2 containing 5% O2 using the cyclotron
(HM-12; Sumitomo Heavy Industry) with target beam current set to 30 µA and
irradiation time of 20 min. [11C]-CO2 (26-30GBq, 702-810mCi) was catalytically
reduced to [11C]-CH3OH complexes and iodinated in the MeI Microlab
synthesis module (Sumitomo Heavy Industry) to give n.c.a [11C]-CH3I which
was wafted by a helium flow to a glass reaction vial containing corresponding
desmethylated precursor (0.5-1 mg), methanol (0.5mL) and sodium hydroxide
(0.2 mol/L, 0.05 mL) (Scheme 1, 2) [29]. [11C]-Methyl iodide was reacted with
the precursor at 85°C for 3 min. After the reaction mixture was cooled to
〜
20°C and quantitatively transferred into a round flask of rotary evaporator, the
solvent was removed under reduced pressure. The products were
reconstituted with 5 mL 0.1M phosphate buffer (pH=7) and passed through a
0.22-µm membrane filter into a sterile multidose vial for in vitro and in vivo

ACCEPTED MANUSCRIPT
experiments. The total synthesis time was about 35 min starting from
[11C]-carbon dioxide. The radiochemical purity analysis and identification was
performed on instant thin-layer chromatography silica gel strips developed with
a 30:1 ratio of methylene chloride: methanol. Subsequently the
chromatographic strips were scanned on an automatic thin-layer
chromatography device. The pH was determined using pH strips and sterility
and pyrogen tests were analyzed according to the monographs prescribed in
the Chinese Pharmacopoeia [30].
5.3.Physicochemical Properties.
The radiotracer (0.1 mCi, 100 µL) was added to 900 µL of freshly collected
Kunming mice serum in a culture tube and incubated at 37 °C for 60 min. The
radiotracer (0.5 mCi, 100 µL) was injected via the tail vein of Kunming mice.
The blood was collected at 60 min p.i. from eyes and centrifugalized to obtain
radioactivity serum. The mixture was filtered with a 0.22 µm Millipore filter and
the radiochemical purity was analyzed by thin-layer chromatography. All
experiments were done in triplicate.
The partition coefficients (log P) of the radiotracers were measured by
mixing 0.1 mL of radiotracer with n-octanol (5 mL) and phosphate buffer (PBS,
pH 7.4, 4.9 mL) in a test tube. The test tube was vortexed for 3 min at room
temperature, followed by centrifugation for 5 min at 3500 rpm. Weighted
sample from the n-octanol and buffer layers were counted on a γ-counter. The
partition coefficient was expressed as the logarithm of the ratio of the counts

ACCEPTED MANUSCRIPT
per gram from n-octanol versus that of PBS. Sample from n-octanol layer was
repartitioned until consistent partition of coefficient value was obtained. The
measurement was done in triplicate and repeated three times.
5.4.Biodistribution
Female Kunming mice were housed for a week under a 12 h light/12 h dark
cycle with free access to food and water for subsequent tissue distribution
studies. Animals were fasted before experiments for 12 h. The radiotracer (0.1
ml, 100 µCi) was injected via the tail vein of mice (n = 3). The mice were killed
by decapitation at 0.5, 5, 10, 30 and 60 min after injection, and the tissues and
organs of interest were collected, weighed wet, and counted in a γ-counter.
The radioactivity counts were corrected for the background and radioactive
decay. Data are expressed as the percent injection dose per gram of tissue
(%ID/g).
In another set of experiments, female Kunming mice (n=3) were injected
intraperitoneally with a CPT-I inhibitor etomoxir (40 mg/kg) at 2 h prior to a
tail-vein injection of radiotracer (a, b, c, [11C]-palmitate) (0.1 ml, 100 µCi). The
mice were killed by decapitation at 5 and 30 min p.i. The remainder of the
procedure was the same as that described above
5.5.Metabolite Analysis in Mouse Hearts
After injection intraperitoneal with or without etomoxir (40 mg/kg) at 2 hours
ago, mice were injected with tracers (37 MBq, 1 mCi). Samples of the heart