Conversion of iodine to fluorine-18 based on iodinated chalcone and evaluation for β-amyloid PET imaging

Article abstract of DOI:10.1016/j.bmc.2018.05.001

In the amyloid cascade hypothesis, β-amyloid (Aβ) plaques is one of the major pathological biomarkers in the Alzheimer's disease (AD) brain. We report the synthesis and evaluation of novel radiofluorinated chalcones, [¹⁸F]4-dimethylamino-4′-flu

Full text of DOI:10.1016/j.bmc.2018.05.001

Accepted Manuscript
Conversion of Iodine to Fluorine-18 Based on Iodinated Chalcone and Evalua-
tion for β-Amyloid PET Imaging
Sho Kaide, Masahiro Ono, Hiroyuki Watanabe, Yoichi Shimizu, Yuji
Nakamoto, Kaori Togashi, Aiko Yamaguchi, Hirofumi Hanaoka, Hideo Saji
PII:
S0968-0896(18)30507-8
https://doi.org/10.1016/j.bmc.2018.05.001
BMC 14343
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
12 March 2018
1 May 2018
2 May 2018
Please cite this article as: Kaide, S., Ono, M., Watanabe, H., Shimizu, Y., Nakamoto, Y., Togashi, K., Yamaguchi,
A., Hanaoka, H., Saji, H., Conversion of Iodine to Fluorine-18 Based on Iodinated Chalcone and Evaluation for β-
Amyloid PET Imaging, Bioorganic & Medicinal Chemistry (2018), doi: https://doi.org/10.1016/j.bmc.2018.05.001
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.

Page 1/34
Conversion of Iodine to Fluorine-18 Based on Iodinated Chalcone
and Evaluation for β-Amyloid PET Imaging
a
a,
a
a,b
b,c
Sho Kaide, Masahiro Ono, * Hiroyuki Watanabe, Yoichi Shimizu, Yuji Nakamoto,
b,c
d
d
a
Kaori Togashi, Aiko Yamaguchi, Hirofumi Hanaoka, & Hideo Saji
a
Department of Patho-Functional Bioanalysis, Graduate School of Pharmaceutical
Sciences, Kyoto University, 46-29 Yoshida Shimoadachi-cho, Sakyo-ku, Kyoto
6
06-8501, Japan
b
Department of Radiology, Kyoto University Hospital, 54 Shogoin Kawahara-cho,
Sakyo-ku, Kyoto 606-8507, Japan
c
Department of Diagnostic Imaging and Nuclear Medicine, Graduate School of
Medicine, Kyoto University, 54 Shogoin Kawahara-cho, Sakyo-ku, Kyoto 606-8507,
Japan
d
Department of Bioimaging Information Analysis, Gunma University Graduate School
of Medicine, 3-39-22 Showa, Maebashi 371-8511, Japan
*
Corresponding author. Tel.: +81-75-753-4608. Fax: +81-75-753-4568. Email:

Page 2/34
ono@pharm.kyoto-u.ac.jp
Abbreviations
Aβ, β-amyloid; AD, Alzheimer’s disease; PET, positron emission tomography;
SPECT, single photon emission computed tomography; DMFC,
-dimethylamino-4’-fluoro-chalcone; FMC, 4’-fluoro-4-methylamino-chalcone; DMIC,
-dimethylamino-4’-iodo-chalcone; IMC, 4’-iodo-4-methylamino-chalcone; FDA,
4
4
Food and Drug Administration; HPLC, high-performance liquid chromatography; %
ID/g, percentage injected dose per gram; BBB, blood-brain barrier; ARG,
autoradiography; Tg, transgenic; KOH, potassium hydroxide; KOAc, potassium
acetate;
K CO ,
potassium
carbonate;
(PdCl (dppf)CH Cl ,
2
3
2
2
2
[
1,1’-bis(diphenylphosphino)ferrocene]dichloropalladium(II),
complex
with
dichloromethane; DMSO, dimethyl sulfoxide; DMF, N,N-dimethylformamide.
Abstract
In the amyloid cascade hypothesis, Aβ plaques are one of the major pathological
biomarkers in the Alzheimer’s disease (AD) brain. We report the synthesis and
evaluation
of
novel
radiofluorinated
chalcones,

Page 3/34
1
8
8
18
[
[
F]4-dimethylamino-4’-fluoro-chalcone
([ F]DMFC)
and
1
18
F]4’-fluoro-4-methylamino-chalcone ([ F]FMC), as β-amyloid (Aβ) imaging
probes. The conversion of iodine directly introduced to the chalcone backbone into
1
8
fluorine was successfully carried out by F-labeling via the corresponding boronate
precursors, achieving the direct introduction of fluorine-18 into the chalcone backbone
1
8
18
to prepare [ F]DMFC and [ F]FMC. In a biodistribution study using normal mice,
18
18
[
F]DMFC and [ F]FMC showed a higher initial uptake (4.43 and 5.47% ID/g at 2
min postinjection, respectively) into and more rapid clearance (0.52 and 0.66% ID/g at
0 min postinjection, respectively) from the brain than a Food and Drug
3
1
8
Administration (FDA)-approved Aβ imaging agent ([ F]Florbetapir), meaning the
improvement of the probability of detecting Aβ plaques and the reduction of
non-specific binding in the brain. In the in vitro binding studies using aggregates of
1
8
18
recombinant Aβ peptides, [ F]DMFC and [ F]FMC showed high binding affinity to
recombinant Aβ aggregates at the K values of 4.47 and 6.50 nM, respectively. In the
d
1
8
in vitro autoradiography (ARG) experiment with AD brain sections, [ F]DMFC and
18
[
F]FMC markedly accumulated only in a region with abundant Aβ plaques,
indicating that they clearly recognized human Aβ plaques in vitro. These encouraging
1
8
18
results suggest that [ F]DMFC and [ F]FMC may be promising PET probes for the

Page 4/34
detection of an amyloid pathology and the early diagnosis of AD with marked
accuracy.
Keywords
Alzheimer’s disease; β-amyloid plaques; positron emission tomography; chalcones;
1
8
F-labeling
1
. Introduction
Dementia affects 47 million people worldwide and this number is expected to
increase to more than 131 million by 2050 as populations age, according to the World
1
Alzheimer Report 2016. Alzheimer’s disease (AD) is the most common cause of
2
dementia worldwide, and the most typical neuropathological hallmarks in the AD
3
brain are senile plaques composed of extracellular β-amyloid peptides (Aβ).
According to the amyloid cascade hypothesis, the accumulation of Aβ is an initiator in
a multi-step pathological process, meaning that Aβ plaques are principal biomarkers
and attractive targets for an early diagnosis of AD, assessment of therapeutic efficacy,
4
and further understanding of the amyloid pathology. In the past, a variety of imaging
probes targeting Aβ plaques were developed for noninvasive imaging, including

Page 5/34
positron emission tomography (PET) and single photon emission computed
5
tomography (SPECT). Among them, only three Aβ imaging agents for PET
18
6
18
7
18
8
(
[ F]Flutemetamol, [ F]Florbetaben, and [ F]Florbetapir ) have been approved by
the FDA, and tested clinically throughout the world. However, some clinical studies
9
, 10
reported the clinical issue that they showed non-specific binding to white matter,
resulting in the overestimation of PET signals and low accuracy of clinical diagnosis.
We previously developed various kinds of Aβ imaging probes based on chalcones
11-17
and studied their structure-activity relationships.
Among them, the radioiodinated
chalcones
with
an
alkylamino
group
at
4-position,
125
125
125
[
I]4-dimethylamino-4’-iodo-chalcone ([ I]DMIC, also abbreviated as [ I]IDP-1
2
6
125
125
in our previous report ) and [ I]4’-iodo-4-methylamino-chalcone ([ I]IMC) (Figure
1
A), displayed favorable characteristics as Aβ imaging probes. They can be easily
formed by a one-pot condensation reaction and exhibit high in vitro affinity for Aβ
plaques and moderate initial in vivo uptake into and rapid washout from the brain of
1
7
normal mice. In addition, iodine-125 has been directly introduced into chalcone
backbones. Chalcones with the iodoallyloxy group at the 4’-position showed modest
1
2
brain uptake into and slow washout from the brain, suggesting that the direct
introduction of iodine allowed favorable in vivo pharmacokinetics in the brain.

Page 6/34
To the present, numerous methodologies to introduce fluorine-18 into organic
1
8
compounds have been devised, aiming at the development of structurally diverse and
clinically relevant radiochemical compounds. In these contexts, some radiochemical
strategies for the direct introduction of fluorine-18, including the preparation of
1
9
radiofluorinated aromatic compounds via aryl iodonium ylides, aryl sulfonium
20
21
salts, and aryl boronic precursors, emerged and attracted much attention in
molecular imaging studies. In addition to these advances of radiofluorination chemistry,
1
8
copper-mediated F-labeling of boronates derived from pinacol has recently been
2
2
reported, which enables the direct incorporation of fluorine-18 into an aromatic ring
and may lead to preferable in vivo pharmacokinetics in the brain, as mentioned above.
1
8
Boronates derived from pinacol are ideal precursors for F-labeling as they are readily
accessible from corresponding iodinated compounds and the commercially available
copper complex [Cu(OTf) (py) ] (OTf = trifluoromethanesulfonate, py = pyridine)
2
4
2
3
were used in this labeling reaction. Therefore, the iodinated compounds can be used
as starting materials in this method for the direct incorporation of fluorine-18 into an
aromatic ring.
In this study, we applied this methodology to our original iodinated chalcones
18
(
DMIC and IMC), and synthesized novel radiofluorinated chalcones, [ F]DMFC and

Page 7/34
18
[
F]FMC (Figure 1B), for the development of more useful Aβ imaging probes for PET
with less non-specific binding in the AD brain. Then, we evaluated their utility as Aβ
imaging probes by focusing on in vitro binding affinity in human brain tissues with AD
pathology and in vivo pharmacokinetics in normal mice.
125
Figure 1. The chemical structural formula of radioiodinated chalcones, [ I]DMIC
125
18
18
and [ I]IMC (A) and radiofluorinated chalcones, [ F]DMFC and [ F]FMC (B).
2
2
. Results and Discussion
.1. Chemistry
The synthesis of nonradioactive chalcones was carried out according to Scheme 1.
4
-Iodoacetophenon
was
reacted
with
4-dimethylaminobenzaldehyde
or

Page 8/34
4
-methylaminobenzaldehyde in the presence of the basic catalyst (10% KOH) in
ethanol at room temperature to form DMIC (1) or IMC (2) at yields of 63.4 and 60.2%,
1
8
respectively. Boronate derivatives 3 and 4 as the precursors of F-labeling were
prepared from 1 and 2 by coupling reactions with bis(pinacolato)diboron in the
presence of the base (KOAc) and palladium catalyst (PdCl (dppf)CH Cl ) at yields of
2
2
2
2
4
3
5.0 and 41.6%, respectively, according to a method reported previously. The
fluorinated derivatives, 4-dimethylamino-4’-fluoro-chalcone (DMFC, 5) and
’-fluoro-4-methylamino-chalcone (FMC, 6), were obtained from 4-fluoroacetophenon
at yields of 65.0 and 76.7%, respectively.
4
Scheme
-methylaminobenzaldehyde, 10% KOH, EtOH; (b) bis(pinacolato)diboron, KOAc,
PdCl (dppf)CH Cl , DMSO.
1.
Reagents:
(a)
4-dimethylaminobenzaldehyde
or
4
2
2
2

Page 9/34
The boronate derivatives (3 and 4) were used as the precursors of the
18
18
copper-mediated nucleophilic F-labeling in the successful preparation of [ F]5
18
18
18
(
[ F]DMFC) and [ F]6 ([ F]FMC) according to Scheme 2, which facilitated the
conversion from iodine directly introduced into the chalcone backbone into fluorine.
1
8 -
Scheme 2. Reagents: (a) F , K CO , Kryptofix 222, Cu(OTf) (py) , DMF.
2
3
2
4
The radiochemical identities of the desired radiofluorinated ligands were verified
by reversed-phase high-performance liquid chromatography (HPLC) analysis. The
1
8
18
radiochemical yields of [ F]DMFC and [ F]FMC were 37.0 and 45.0%, respectively,
with radiochemical purities of >95% after purification by HPLC with specific
activities of 1,256 and 466 GBq/µmol, respectively.
2
.2. In vivo biodistribution study using normal mice
18
18
To evaluate the utility of [ F]DMFC and [ F]FMC as Aβ imaging probes for PET,
we performed a biodistribution study in normal mice (Figure 2 and Table 1). For the

Page 10/34
purpose of the detection of the amyloid pathology in the AD brain, promising Aβ
imaging probes should show high initial uptake into and subsequent rapid washout
2
5
from the brain because normal mice have no Aβ plaques in the brain. The
radioactivity accumulation of compounds is presented as the percentage injected dose
18
18
per gram (% ID/g). [ F]DMFC and [ F]FMC displayed a higher initial brain uptake
(brain2 min = 4.43% ID/g and 5.47% ID/g, respectively) than the corresponding
125
26
125
radioiodinated chalcones, [ I]DMIC (abbreviated as [ I]IDP-1 in our previous
1
25
report) and [ I]IMC (brain
= 2.43 and 3.52% ID/g, respectively) (Figure 2A, 2B,
min
2
and Table 1). In general, some factors including the chemical structure, molecular
charge, lipophilicity, and size have a marked influence on the compound’s
pharmacokinetics in the brain, indicating that the favorable permeability of the
1
8
18
blood-brain barrier (BBB) of [ F]DMFC (log P-value = 2.64 ± 0.01) and [ F]FMC
125
(
log P-value = 2.25 ± 0.01) in comparison with [ I]DMIC (log P-value = 3.16 ± 0.01)
1
25
and [ I]IMC (log P-value = 2.53 ± 0.01) may be due to the suitable lipophilicity and
molecular size by conversion from iodine-125 to fluorine-18 at the 4’-position of the
2
7, 28
chalcone backbone.
pharmacokinetic standards in the brain for clinical evaluations.
Then, clearance from the brain was evaluated by calculating the ratio of the
These novel radiofluorinated chalcones may meet the
2
9, 30

Page 11/34
1
8
18
radioactivity accumulation at 2 and 30 min postinjection. [ F]DMFC and [ F]FMC
exhibited higher brain2 min/brain30
ratios (8.5 and 8.3, respectively) than
min
125
26
125
[
I]DMIC and [ I]IMC (7.4 and 5.6, respectively) (Table 1), indicating that these
radiofluorinated chalcones showed a more rapid clearance from the brain than
radioiodinated forms. This may also be due to the suitable lipophilicity of the
2
7, 28
compound,
resulting in less non-specific binding to white matter.
1
8
18
Then, [ F]DMFC and [ F]FMC were compared with fluoro-pegylated chalcone
with dimethylamino group (FPEG chalcone) previously reported as an Aβ imaging
1
4
18
probe. In the previous study, F-labeled FPEG chalcone displayed high initial brain
14
uptake (brain2 min = 3.48% ID/g) and subsequent clearance (brain30 min = 1.08% ID/g).
However, the ratio of the radioactivity accumulation at 2 and 30 min postinjection was
1
8
18
3
.2, which is much lower than that of [ F]DMFC and [ F]FMC. These results
1
8
18
indicate that [ F]DMFC and [ F]FMC are less likely to show non-specific binding to
1
8
white matter than F-labeled FPEG chalcone. Furthermore, in comparison with a
1
8
FDA-approved Aβ imaging agent, [ F]Florbetapir (4.90% ID/g at 2 min postinjection
3
1
18
18
and 3.0 for brain2 min/brain30 min ratio), [ F]DMFC and [ F]FMC were superior to
18
[
F]Florbetapir from the perspective of in vivo uptake into and washout from the brain
(Figure 2C and Table 1), suggesting that these radiofluorinated chalcones in this study

Page 12/34
may be clinically useful Aβ imaging probes.
1
8
125
Figure 2. Comparison of brain uptake and clearance of [ F]DMFC with [ I]DMIC
1
8
125
18
18
18
(
A), [ F]FMC with [ I]IMC (B), and [ F]DMFC [ F]FMC with [ F]Florbetapir
(C) in normal mice (n = 5).
a
125
18
26, 31
The data of [ I]DMIC and [ F]Florbetapir were previously reported.
1
8
Table 1. Brain uptake of radioactivity after the injection of [ F]chalcones,
125
18
a
[
I]chalcones and [ F]Florbetapir in normal mice and the 2 min/30 min ratio of

Page 13/34
radioactivity accumulation.
Compd
%
ID/g Brain
Ratio
2 min/30 min)
(
2
min
30 min
1
8
[
F]DMFC
4.43 (0.45)
2.43 (0.10)
5.47 (0.52)
3.52 (0.62)
4.90 (0.99)
0.52 (0.07)
0.33 (0.07)
0.66 (0.09)
0.63 (0.19)
1.65 (0.11)
8.5
7.4
8.3
5.6
3.0
1
25
[
I]DMIC
1
8
[
F]FMC
1
25
[
I]IMC
1
8
[
F]Florbetapir
a
Each value represents the mean (SD) of five animals.
1
8
18
The biodistribution of radioactivity in the whole body of [ F]DMFC and [ F]FMC
was evaluated (Table S1). They accumulated initially in the liver (liver2 min = 8.56 and
.96% ID/g, respectively) and kidney (kidney2 min = 8.52 and 9.21% ID/g, respectively)
8
and subsequently in the intestine (intestine60 min = 12.0 and 15.0% ID/g, respectively)
as well as other lipophilic compounds. No marked accumulation in the bone was
observed (bone60 min = 1.65 and 1.86% ID/g, respectively). On the other hand, the
1
8
radioactivity accumulation of F-labeled FPEG chalcone in the bone increased over
1
8
18
time and was higher (bone60 min = 3.58% ID/g) than that of [ F]DMFC and [ F]FMC.

Page 14/34
This difference could be attributable to the different physicochemical characteristics of
their radiometabolites produced through some unclear physiological processes. These
1
8
18
results indicate that [ F]DMFC and [ F]FMC are more stable toward defluorination
18
in vivo than F-labeled FPEG chalcone, and the introduction of fluorine-18 directly
into chalcone backbone without PEG is suitable for the development of chalcone
derivatives as Aβ imaging probes. On the basis of the biodistribution study in normal
18
18
mice, [ F]DMFC and [ F]FMC displayed satisfactory pharmacokinetics in vivo as Aβ
imaging probes for PET.
In order to confirm that these novel radiofluorinated chalcones permeated the
blood-brain barrier (BBB) early followed by fast washout, a small-animal µPET
1
8
imaging study of [ F]DMFC was carried out in a normal mouse (Figure S1 and S2).
1
8
The PET imaging data demonstrated that [ F]DMFC rapidly passed through the BBB,
penetrated at 2 min postinjection, and no radioactivity accumulation in any brain
regions was observed at 30 min postinjection (Figure S1). The time activity curves
1
8
(
TACs) showed that [ F]DMFC subsequently faded away from the brain to the level
of a background by 30 min postinjection (Figure S2A). The PET images and TACs
18
indicate that [ F]DMFC showed little or no non-specific binding in the brain, leading
to complete washout from the brain. TACs of other important tissues are shown in

Page 15/34
Figure S2B. These results of the PET imaging study were consistent with those of the
biodistribution study mentioned above, strongly suggesting that novel radiofluorinated
chalcones may function well as useful Aβ imaging probes for PET in vivo.
2
.3. Binding studies using Aβ aggregates in solution
1
8
18
Binding studies of [ F]DMFC and [ F]FMC to Aβ
aggregates were carried out
-42
1
in order to calculate their Bmax and dissociation constant (K ) values. The B and Kd
d
max
1
8
values of [ F]DMFC were 0.41 pmol/mg protein and 4.47 nM, respectively. The B
max
1
8
and K values of [ F]FMC were 0.42 pmol/mg protein and 6.50 nM, respectively. The
d
3
32
K values of both compounds were as low as that of [ H]PiB (K = 4.7 nM),
d
d
suggesting that their binding affinity to Aβ1-42 aggregates is high enough to function as
Aβ imaging probes. In the previous report, inhibition constant (K ) values of DMFC
i
1
4
and FMC for Aβ1-42 aggregates in solution were 49.8 and 234.2 nM, respectively ,
indicating that they showed moderate binding affinity for recombinant Aβ1-42
aggregates. These discrepancies are probably due to the difference of experimental
conditions such as competitive ligands, buffer materials, and Aβ1-42 aggregates.
18
18
Moreover, the Bmax values of [ F]DMFC and [ F]FMC were almost same, indicating
18
18
that [ F]DMFC and [ F]FMC bound to the same binding sites in recombinant Aβ
1
-42

Page 16/34
aggregates.
2
.4. In vitro autoradiography (ARG) with AD brain sections
1
8
18
For the clinical application of [ F]DMFC and [ F]FMC as Aβ imaging probes to
precisely detect Aβ plaques in the human brain, we investigated their binding affinity
to Aβ plaques by performing an in vitro ARG experiment with AD brain sections
1
8
18
(
Figure 3). In the in vitro autoradiograms, radioactivities of [ F]DMFC and [ F]FMC
accumulated exactly along the gray matter of the frontal lobe (Figure 3A and 3B,
respectively). Conversely, marked radioactivity accumulation was not observed in
1
8
18
other regions such as the white matter, suggesting that [ F]DMFC and [ F]FMC
1
8
selectively accumulated in the gray matter of the frontal lobe. [ F]DMFC and
18
[
F]FMC showed no radioactive accumulation in the normal human brain sections
(Figure S3). To confirm that this selectivity is derived from Aβ plaques,
immunohistochemical staining with anti-Aβ antibody was carried out using adjacent
sections, and the extensive accumulation of Aβ plaques was observed in the gray
matter of the frontal lobe (Figure 3C), in accordance with the accumulation pattern of
1
8
18
18
radioactivities of [ F]DMFC and [ F]FMC. These results indicate that [ F]DMFC
18
and [ F]FMC sensitively recognized Aβ plaques in the AD brain in vitro.

Page 17/34
1
8
Figure 3. In vitro autoradiograms of AD brain sections labeled with [ F]DMFC (A)
18
and [ F]FMC (B). Immunohistochemical staining using an anti-Aβ
antibody (C).
–42
1
We also carried out in vitro ARG with the Tg2576 brain section to investigate their
18
binding ability against Aβ plaques in mice brain. [ F]DMFC with higher binding
1
8
affinity against Aβ aggregates than [ F]FMC was used in this experiment. Compared
1
8
to the marked labelling of Aβ plaques in human brain, [ F]DMFC slightly showed
binding affinity to Aβ plaques in the Tg2576 mouse brain (Figure S4). These
discrepancies may be partially due to the difference in conformation of the aggregates
3
3, 34
formed by recombinant Aβ peptides, Aβ plaques in mice and human
; however,
details remain unclear. Irrespective of this, the encouraging data on AD brain sections
support the potential clinical use in humans of these novel radiofluorinated chalcones.

Page 18/34
3
. Conclusion
In conclusion, novel radiofluorinated chalcones ([ F]DMFC and [ F]FMC) were
18
18
designed as PET probes to detect Aβ plaques with low non-specific binding by the
translation of iodine into fluorine. In order to achieve the objective, our original
iodinated chalcones (DMIC and IMC) were used as starting materials, and the
incorporation of fluorine-18 directly into the 4’-position of the chalcone backbone was
successfully performed via novel copper-mediated radiofluorination chemistry. In the
biodistribution study and PET imaging study with normal mice, they showed desirable
[
in vivo pharmacokinetics in the brain, partially due to the adjustment of the
lipophilicity and molecular size by conversion from iodine into fluorine. In the binding
studies using Aβ aggregates, their K values against Aβ aggregates were very low,
d
indicating their high binding affinity to recombinant Aβ aggregates. In the in vitro
ARG experiment using human brain sections with AD pathology, they intensely
labeled Aβ plaques and did not markedly accumulate in regions without Aβ plaques.
18
18
These encouraging in vitro and in vivo data suggest that [ F]DMFC and [ F]FMC
may be useful for accurate PET imaging of Aβ plaques in the AD brain.

Page 19/34
4
4
. Experimental section
.1. General information
All reagents were commercial products used without further purification. All
compounds were purified by Smart Flash EPCLC W-Prep 2XY (Yamazen Corporation,
1
13
Tokyo, Japan) unless indicated otherwise. H and C NMR spectra were recorded on a
JNM-ECS400 (JEOL, Tokyo, Japan) with tetramethyl silane (TMS) as an internal
standard. Coupling constants are reported in Hertz. Multiplicity was defined as singlet
(s), doublet (d), triplet (t), or multiplet (m). ESI mass spectrometry was conducted with
a SHIMADZU LCMS-2020. High-resolution mass spectrometry (HRMS) was carried
out with a JEOL JMS-700 (JEOL, Tokyo, Japan). HPLC was performed with a
Shimadzu system (an LC-20AD pump with an SPD-20A UV detector, λ = 254 nm)
using a Cosmosil C column (Nacalai Tesque, COSMOSIL 5C -MS-II 4.6 mm I.D. ×
1
8
18
1
50 mm). All key compounds were proven by this method to show > 95% purity. Male
ddY mice were purchased from Japan SLC, Inc. (Shizuoka, Japan), and were fed
standard chow and had free access to water. We made all efforts to minimize suffering.
Animal experiments were conducted in accordance with our institutional guidelines
and were approved by Kyoto University. Experiments involving human subjects were
performed in accordance with relevant guidelines and regulations and were approved

Page 20/34
by the ethics committee of Kyoto University. Postmortem brain tissues from an
autopsy-confirmed case of AD (73-year-old male) and a control subject (82-year-old
female) were purchased from BioChain Institute Inc (California, USA).
4
4
3
.1.1. (E)-3-(4-(Dimethylamino)phenyl)-1-(4-iodophenyl)-2-propen-1-one (1).
-Iodoacetophenon (738 mg, 3.00 mmol) and 4-dimethylaminobenzaldehyde (447 mg,
.00 mmol) were dissolved in 24 mL of ethanol. The mixture was stirred for 10 min in
an ice bath. A 10-mL aliquot of 10% aqueous potassium hydroxide solution was then
slowly added dropwisely to the reaction mixture. The reaction solution was stirred at
room temperature for 9 h. A precipitate was collected and washed with 50% ethanol to
1
give 717 mg of 1 (63.4%). H NMR (400 MHz, CDCl ) δ 7.85-7.77 (m, 3H), 7.72 (d, J
3
=
8.4 Hz, 2H), 7.55 (d, J = 8.8 Hz, 2H), 7.25 (d, J = 5.6 Hz, 1H), 6.69 (d, J = 9.2 Hz,
+
2
H), 3.05 (s, 6H). MS (ESI) m/z 378.3 [MH ].
4
.1.2. (E)-3-(4-(Methylamino)phenyl)-1-(4-iodophenyl)-2-propen-1-one (2). The
same reaction as described above to prepare 1 was used, and 183 mg of 2 (60.2%) was
1
obtained from 4-iodoacetophenon and 4-methylaminobenzaldehyde. H NMR (400
MHz, CDCl ) δ 7.84 (d, J = 8.0 Hz, 2H), 7.78 (d, J = 15.6 Hz, 1H), 7.72 (d, J = 8.4 Hz,
3

Page 21/34
2
H), 7.51 (d, J = 8.8 Hz, 2H), 7,25 (d, J = 8.8 Hz, 1H), 6.60 (d, J = 8.4 Hz, 2H), 4.18
+
(
s, 1H), 2.90 (s, 3H). MS (ESI) m/z 364.3 [MH ].
4
.1.3.
E)-3-(4-(Dimethylamino)phenyl)-1-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-y
l)phenyl)-2-propen-1-one (3). To a solution of 1 (487 mg, 1.29 mmol) in DMSO
anhydrous, 10 mL) were added bis(pinacolato)diboron (361 mg, 1.42 mmol),
(
(
potassium
acetate
(380
mg,
3.87
mmol)
and
[
1,1’-bis(diphenylphosphino)ferrocene]dichloropalladium(II),
complex
with
dichloromethane (PdCl (dppf)CH Cl , 148 mg, 0.18 mmol). The mixture was stirred
2
2
2
under reflux for 4 h. After being cooled to room temperature, the mixture was
extracted with ethyl acetate (2×50 mL). The organic layers were combined and dried
over Na SO . Evaporation of the solvent afforded a residue, which was purified by
2
4
silica gel chlomatography (hexane/ethyl acetate = 4:1) to give 60.0 mg of 3 (35.0%).
1
H NMR (400 MHz, CDCl ) δ 7.98 (d, J = 2.8 Hz, 2H), 7.91 (d, J = 8.4 Hz, 2H), 7.75
3
(d, J = 8.8 Hz, 1H), 7.55 (d, J = 9.2 Hz, 2H), 7.33 (d, J = 15.6 Hz, 1H), 3.05 (s, 6H),
+
1
.37 (s, 12H). MS (ESI) m/z 378.3 [MH ].

Page 22/34
4
.1.4.
(E)-3-(4-(Methylamino)phenyl)-1-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)
phenyl)-2-propen-1-one (4). The same reaction as described above to prepare 3 was
1
used, and 45.3 mg of 4 was obtained (41.6%) from 2. H NMR (400 MHz, CDCl ) δ
3
7
.97 (d, J = 7.2 Hz, 2H), 7.92 (d, J = 7.2 Hz, 2H), 7.76 (d, J = 15.2 Hz, 1H), 7.50 (d, J
=
8.0 Hz, 2H), 7.32 (d, J = 15.6 Hz, 1H), 6.58 (d, J = 7.6 Hz, 2H), 4.22 (s, 1H), 2.88 (s,
+
3
H), 1.36 (s, 12H). MS (ESI) m/z 364.3 [MH ].
4
.1.5. (E)-3-(4-(Dimethylamino)phenyl)-1-(4-fluorophenyl)-2-propen-1-one (5,
DMFC).
4-Fluoroacetophenon
(414
mg,
3.00
mmol)
and
4
-dimethylaminobenzaldehyde (447 mg, 3.00 mmol) were dissolved in 24 mL of
ethanol. The mixture was stirred for 10 min in an ice bath. A 15-mL aliquot of 10%
aqueous potassium hydroxide solution was then slowly added dropwisely to the
reaction mixture. The reaction solution was stirred at room temperature for 9 h. A
precipitate was collected and washed with 50% ethanol to give 525 mg of 5 (65.0%).
1
H NMR (400 MHz, CDCl ) δ 8.06-8.02 (m, 2H), 7.80 (d, J = 15.6 Hz, 1H), 7.55 (d, J
3
=
8.8 Hz, 2H), 7.30 (d, J = 15.2 Hz, 1H), 7.16 (t, J = 8.8 Hz, 2H), 6.70 (d, J = 8.8 Hz,
1
3
2
H), 3.05 (s, 6H). C NMR (100 MHz, CDCl ) δ 188.9, 166,5, 164.0, 152.1, 146.0,
3

Page 23/34
+
1
35.3, 130.5, 122.4, 116.2, 115.4, 111.8, 40.1. MS (ESI) m/z 270.2 [MH ]. HRMS
(
FAB) m/z calcd. for C H FNO 270.1294, found 270.1291.
1
7
17
4
.1.6. (E)-3-(4-(Methylamino)phenyl)-1-(4-fluorophenyl)-2-propen-1-one (6, FMC).
The same reaction to prepare 5 was used, and 39.1 mg of 6 (76.7%) was obtained from
1
4
8
1
-fluoroacetophenon and 4-methylaminobenzaldehyde. H NMR (400 MHz, CDCl ) δ
3
.06-8.02 (m, 2H), 7.78 (d, J = 15.6 Hz, 1H), 7.52 (d, J = 8.8 Hz, 2H), 7.30 (d, J =
5.6 Hz, 1H), 7.18-7.14 (m, 2H), 6.60 (d, J = 8.8 Hz, 2H), 4.17 (s, 1H), 2.90 (s, 3H).
1
3
C NMR (100 MHz, CDCl ) δ 190.3, 154.1, 151.5, 132.3, 130.6, 126.4, 123.6, 116.4,
3
+
1
15.4, 112.1, 111.4, 30.0. MS (ESI) m/z 256.1 [MH ]. HRMS (FAB) m/z calcd. for
C H FNO 256.1138, found 256.1134.
16
15
4
.2. Radiofluorination
18
The [ F]fluoride was produced by a cyclotron (CYPRIS HM-12, Sumitomo Heavy
18
18
Industries, Tokyo) via an O(p,n) F reaction and passed through a Sep-Pak Light
18
QMA cartridge (Waters) as an aqueous solution in O-enriched water. The cartridge
1
8
was dried by nitrogen gas, and the F activity was eluted with 300 µL of 66 mM
18
K CO solution. Kryptofix222 (8.0 mg) was dissolved in the solution of [ F]fluoride
2
3

Page 24/34
in water. The solvent was removed at 120˚C under a stream of nitrogen gas. The
residue was azeotropically dried with 300 µL of anhydrous acetonitrile three times at
1
20˚C under a stream of nitrogen gas. A solution of the boronate precursor 3 or 4 (5.0
mg) and Cu(OTf) (py) (9.0 mg) in N,N-dimethylformamide (DMF) (200 µL) was
2
4
1
8
added to the reaction vessel containing the [ F]fluoride activity. The mixture was
heated at 110°C for 20 min. After passing the filter, the radiofluorinated ligand was
purified by HPLC on a COSMOSIL 5C -MS-II column with an isocratic solvent of
1
8
acetonitrile/H O (60/40 or 45/55, respectively) at a flow rate of 1.0 mL/min.
2
4
.3. In vivo biodistribution study in normal mice
1
8
18
A saline solution (100 μL) of [ F]DMFC or [ F]FMC (111 - 112 kBq) containing
ethanol (10.0 μL) and Tween 80 (0.100 μL) was directly injected intravenously into the
tails of ddY mice (5 weeks old, male). The mice were sacrificed at various time points
postinjection. The organs of interest were removed and weighed, and the radioactivity
was measured with an automatic γ counter (Wallac WIZARD 1470, PerkinElmer).
4
.4. Binding studies using Aβ aggregates in solution

Page 25/34
Aβ1-42 was purchased from the Peptides Institute, Inc. (Osaka, Japan). Aβ1-42
-1
aggregates were prepared by incubating the Aβ1-42 peptide (0.25 mg mL in PBS, pH
7
.4) at 37˚C for 42 h with gentle and constant shaking. The reaction mixture contained
1
8
18
increasing concentrations of [ F]DMFC or [ F]FMC (1.56-50.0 nM, 100 µL, DMSO),
a fixed concentration of Aβ1-42 aggregates (9.2 nM, 50 µL, PBS), and 10% EtOH (850
µL). Nonspecific binding was defined in the presence of 4 µM nonradioactive DMFC
or FMC. After incubating for 3 h at room temperature, the solution was filtered
through GF/B filters using an M-24 cell harvester. The radioactivities of the bound
18
18
[
F]DMFC and [ F]FMC were measured with a γ counter (Wizard 1470,
PerkinElmer). The Bmax and K values were determined by using GraphPad Prism 5.0
d
(GraphPad Software, San Diego, CA, USA).
4
.5. Immunohistochemical staining using AD brain sections
We performed immunohistochemical staining of brain sections according to a
2
7
method reported previously. Post-mortem brain sections of paraffin-embedded blocks
from an autopsy-confirmed case of AD (73-year-old male) were used for staining. The
sections were subjected to two 15-min incubations in xylene, two 1-min incubations in
1
00% EtOH, one 1-min incubation in 90% EtOH, and one 1-min incubation in 70%

Page 26/34
EtOH to completely deparaffinize them, followed by two 2.5-min washes in water.
They were then autoclaved for 15 min in 0.01 M citric acid buffer (pH 6.0) and
incubated with 90% formic acid for 5 min to activate the antigen. After 2-min
incubations in PBS-Tween 20, the sections were incubated at room temperature with
an anti-Aβ1–42 (BC05, Wako) primary antibody for 1 h. After three 5-min incubations
in PBS-Tween 20, they were incubated with goat anti-mouse IgG (Histofine Simple
Stain Mouse MAX-PO (MULTI), Nichirei Biosciences Inc.) at room temperature for
3
0 min. After three 3-min incubations in PBS-Tween 20 and one 5-min incubation in
TBS, the sections were incubated with DAB as a chromogen for 1 min. After being
washed with water, the sections were observed under a microscope (BZ-9000,
KEYENCE).
4
.6. In vitro autoradiography of AD brain sections
This experiment was performed according to the method reported previously. The
2
7
serial sections used in immunohistochemical staining were used for this study.
Deparaffinization was carried out according to the protocol described above. The
sections were incubated with radiofluorinated chalcones (370 kBq/mL in 10% EtOH)
for 1 h at room temperature. They were then washed in 50% EtOH for 3 min twice.

Page 27/34
1
8
After drying, the F-labeled sections were exposed to a BAS imaging plate (Fuji Film)
for 12 h. Autoradiographic images were obtained by using a BAS5000 scanner system
(Fuji Film).
4
.7. Measurement of log P values
The determination of partition coefficient values were performed in 1-octanol and
PBS (−) at a pH of 7.4. 1-Octanol (3.0 mL) and PBS (−) (3.0 mL) were pipetted into a
5-mL-test tube containing 0.40 MBq of test compounds. The test tube was vortexed
1
for 2 min, and centrifuged (10 min, 2000 rpm). Aliquots (500 μL) from the 1-octanol
and PBS (−) phases were transferred into two test tubes for counting. The remaining
1
-octanol phase was transferred (1.0 mL) into a new test tube. 1-Octanol (2.0 mL) and
PBS (−) (3.0 mL) were pipetted into the test tube. The vortexing, centrifuging, and
counting were repeated. The amount of radioactivity in each tube was measured with a
γ counter (Perkin Elmer, Wizard 1470). The partition coefficient was calculated using
Eq. 1:
(counts/μL in 1-octanol)/(counts/μL in PBS(-)) = r
(1)

Page 28/34
Acknowledgments
This research was supported by JSPS KAKENHI Grant Numbers JP15H01555,
JP17H04260, JP17H05092, and JP17H05694.
Conflicts of interest
The authors declare that there are no conflicts of interest.
Supplementary data
Supplementary data (general information for small-animal µPET and procedure for
radiofluorination, in vivo PET imaging of a normal mouse, and in vitro ARG with
control human brain sections) associated with this article can be found online.
References
1
.
Prince M, Comans-Herrera A, Knapp M, Guerchet M, Karagiannidou M. World
Alzheimer Report 2016: Improving Healthcare for People Living with Dementia,
Alzheimer’s Disease International, London, UK. 2016.
2
.
2017 Alzheimer's disease facts and figures. Alzheimer's & Dementia: The
Journal of the Alzheimer's Association. 2017;13:325-373.

Page 29/34
3
2
4
.
Selkoe DJ. Alzheimer's disease: genes, proteins, and therapy. Physiol Rev.
001;81:741-766.
Selkoe DJ. Alzheimer's disease: a central role for amyloid. J Neuropathol Exp
.
Neurol. 1994;53:438-447.
Villemagne VL, Dore V, Burnham SC, Masters CL, Rowe CC. Imaging tau and
amyloid-β proteinopathies in Alzheimer disease and other conditions. Nat Rev Neurol.
5
.
2
6
018.
.
Koole M, Lewis DM, Buckley C, Nelissen N, Vandenbulcke M, Brooks DJ,
Vandenberghe R, Van Laere K. Whole-body biodistribution and radiation dosimetry of
1
8
F-GE067: a radioligand for in vivo brain amyloid imaging. J Nucl Med.
2
7
009;50:818-822.
.
Rowe CC, Ackerman U, Browne W, Mulligan R, Pike KL, O'Keefe G,
Tochon-Danguy H, Chan G, Berlangieri SU, Jones G, Dickinson-Rowe KL, Kung HP,
Zhang W, Kung MP, Skovronsky D, Dyrks T, Holl G, Krause S, Friebe M, Lehman L,
Lindemann S, Dinkelborg LM, Masters CL, Villemagne VL. Imaging of amyloid β in
1
8
Alzheimer's disease with F-BAY94-9172, a novel PET tracer: proof of mechanism.
Lancet Neurol. 2008;7:129-135.
8
.
Choi SR, Golding G, Zhuang Z, Zhang W, Lim N, Hefti F, Benedum TE,