Synthesis and evaluation of novel radioiodinated anthranilate derivatives for in vivo imaging of vascular endothelial growth factor receptor with single-photon emission computed tomography

Article abstract of DOI:10.1007/s12149-020-01475-6

Objective: Angiogenesis facilitates tumor survival and promotes malignancy. The vascular endothelial growth factor (VEGF)/VEGF receptor (VEGFR) tyrosine kinase (TK) signaling pathway is a key factor mediating angiogenesis, suggesting that this pathway may

Full text of DOI:10.1007/s12149-020-01475-6

Annals of Nuclear Medicine
https://doi.org/10.1007/s12149-020-01475-6
ORIGINAL ARTICLE
Synthesis and evaluation of novel radioiodinated anthranilate
derivatives for in vivo imaging of vascular endothelial growth factor
receptor with single‑photon emission computed tomography
Masahiko Hirata · Akihiko Asano · Yasuhiro Magata · Yoshiro Ohmomo · Takashi Temma1,3
1
1
2
1,4
Received: 7 February 2020 / Accepted: 28 April 2020
©
The Japanese Society of Nuclear Medicine 2020
Abstract
Objective Angiogenesis facilitates tumor survival and promotes malignancy. The vascular endothelial growth factor (VEGF)/
VEGF receptor (VEGFR) tyrosine kinase (TK) signaling pathway is a key factor mediating angiogenesis, suggesting that
this pathway may be a target for diagnosis and therapy. In this study, we aimed to develop small molecule radioiodinated
probes applicable for in vivo VEGFR imaging considering the versatility and usefulness of single-photon emission computed
tomography (SPECT).
Methods We designed and synthesized four radioiodinated anthranilate compounds (6a–d) based on the structure of an
anticancer drug targeting VEGFR-TK. The inhibitory potencies of corresponding cold compounds 4a–d and in vitro stabil-
ity of compounds 6a–d were assessed by cellular proliferation inhibition assays and radio thin-layer chromatography after
incubation in neutral solution. In vivo biodistributions were evaluated by determining radioactivity in tissues of interest after
intravenous injection of test compounds in tumor-bearing mice. In vitro and in vivo blocking experiments using a selective
VEGFR-TK inhibitor and SPECT/computed tomography (CT) imaging were performed in tumor-bearing mice.
Results The radioiodinated compounds 6a–d were obtained with more than 68.0% radiochemical yield and more than 95%
radiochemical purity. Because compounds 4a–d showed high inhibitory potencies and compounds 6c and 6d showed high
1
25
125
in vitro stability, 6c ([ I]m-NPAM) and 6d ([ I]p-NPAM) were further evaluated. Analysis of the in vivo biodistribution
1
25
revealed a tumor to blood radioactivity ratio of greater than 4 at 24 h after [ I]p-NPAM administration. Accumulation of
1
25
radioactivity in cultured tumor cells and tumor xenografts after [ I]p-NPAM administration was signiꢀcantly blocked by
1
25
inhibitor pretreatment. Tumors were clearly imaged at 24 h after [ I]p-NPAM injection with SPECT/CT in comparison to
that in inhibitor-pretreated tumor-bearing mice.
1
25
Conclusion [ I]p-NPAM may have potential applications as a lead compound for future development of a clinically usable
VEGFR imaging probe for SPECT.
Keywords Vascular endothelial growth factor · Single-photon emission computed tomography · Radioiodine ·
Angiogenesis · Anthranilate
*
*
Yoshiro Ohmomo
Education and Research Center, Hamamatsu University
School of Medicine, 1-20-1 Handayama, Higashi-ku,
Hamamatsu 431-3192, Japan
ohmomo@gly.oups.ac.jp
Takashi Temma
3
4
ttemma@gly.oups.ac.jp
Department of Biofunctional Analysis, Osaka University
of Pharmaceutical Sciences, 4-20-1 Nasahara, Takatsuki,
Osaka 569-1094, Japan
1
2
Osaka University of Pharmaceutical Sciences, 4-20-1
Nasahara, Takatsuki, Osaka 569-1094, Japan
Education and Research Center for Pharmaceutical Sciences,
Osaka University of Pharmaceutical Sciences, 4-20-1
Nasahara, Takatsuki, Osaka 569-1094, Japan
Department of Molecular Imaging, Institute for Medical
Photonics Research, Preeminent Medial Photonics
Vol.:(0
1
123456789)
3

Annals of Nuclear Medicine
Introduction
have been developed [14], some of which have been evalu-
ated in clinical trials. However, few reports have described
the successful development of VEGF imaging probes for
single-photon emission computed tomography (SPECT)
[15, 16]. Furthermore, to the best of our knowledge, no
small molecule imaging probes have been used for in vivo
SPECT imaging of VEGF/VEGFR expression to date.
Thus, the development of small molecule imaging probes
that enable in vivo SPECT imaging of VEGF/VEGFR
expression is urgently needed owing to the versatility and
usefulness of SPECT.
Various receptors are expressed on the surfaces of tumor
cells and are related to signaling cascades that control the
expression of genes and proteins to regulate tumor charac-
teristics, such as growth, invasion, metastasis, and therapy
resistance. Therefore, nuclear medical imaging probes tar-
geting such receptors expressed in tumor cells are essential
for elucidation of therapy-related tumor properties.
Angiogenesis plays an important role in the eꢁcient
supply of nutrients and oxygen to tumors, promoting
tumor survival and malignancy. During angiogenesis, the
number of abnormal capillary vessels increases; this is a
common event that occurs in many malignant tumors and
is reportedly related to metastasis and prognosis, suggest-
ing its potential applications as a target for diagnosis and
therapy [1–3]. Vascular endothelial growth factor (VEGF)
proteins are a family of mitogenic glycoproteins that acti-
vate VEGF receptors (VEGFRs) via a tyrosine kinase (TK)
signaling pathway. VEGF produced by tumor cells and
macrophages is a key player in the angiogenic process in
malignant tumors [4–6]. Indeed, therapeutic drugs, such
as a humanized anti-VEGF monoclonal antibody (bevaci-
zumab), a Fab fragment of bevacizumab (ranibizumab),
and small molecule VEGFR TK inhibitors (sunitinib and
sorafenib), have been developed as examples of molecular
targeted drugs and are now widely utilized in the clinical
setting. Moreover, a variety of imaging probes for posi-
Anthranilate derivatives have attracted attention owing to
their major anticancer eꢂects [17]. Among these compounds,
2-([pyridin-4-ylmethyl]amino)-N-(3-[triꢃuoromethyl]phe-
nyl)benzamide (AAL-993) is an anticancer drug targeting
VEGFR-TK [18, 19]. Based on the structure of AAL-993,
in this study, we introduced an iodine atom into the benzene
ring connected to the anthranilic acid moiety via an ester or
amide linker to produce the compounds 4a–d, as shown in
Scheme 1. Since several derivatives having halogen atom
(ꢃuorine or chlorine), alkoxy group, or methylthio group
on the benzene showed high potencies as such drug candi-
dates in addition to AAL-993 having triꢃuoromethyl group
at m-position [17–19], we decided to introduce an iodine
atom at m- and p-positions. Besides, considering the control
of stability and blood clearance in vivo recognized as one
of important factors of molecular imaging probe, ester and
amide linkers were tested. In this study, we ꢀrst synthesized
compounds 4a–d and radiosynthesized compounds 6a–d,
as shown in Scheme 2, and then performed cellular prolif-
eration inhibition assays and stability estimation in neutral
solutions. Subsequently, we further evaluated the potential
of selected compounds as VEGFR imaging probes by in vivo
8
9
tron emission tomography (PET), such as Zr-labeled
8
9
89
antibody derivatives ( Zr-bevacizumab [7, 8] and Zr-
1
8
ranibizumab [9]), an F-labeled sunitinib derivative [10],
1
8
11
and F- and C-labeled sorafenib derivatives [11–13],
Scheme 1 Synthesis of com-
pounds 4a–d
1
3

Annals of Nuclear Medicine
Scheme 2 Radiosynthesis of
compounds 6a–d
biodistribution experiments, in vitro and in vivo blocking
experiments, and SPECT/CT imaging in tumor-bearing
mice.
Male ddY mice and male BALB/c-nu mice were obtained
from Japan SLC (Shizuoka, Japan) and were housed for
at least 1 week with free access to food and water before
the experiments. The animal experiments in this study
were conducted in accordance with the animal experi-
ment guidelines of Osaka University of Pharmaceutical
Sciences (approval number, 06025).
Materials and methods
Reagents and instruments
The melting points (mp) were determined on a Yanagimoto
micro-melting point apparatus MP-500D (Yanaco, Kyoto,
Synthesis and radiosynthesis (Schemes 1 and 2)
1
Japan) and were uncorrected. The H-nuclear magnetic
We synthesized compounds 4a–d following a previous
resonance (NMR) spectra were recorded on a Mercury-300
report [19] with slight modiꢀcations.
(
Varian, Tokyo, Japan) spectrometer, and the chemical
shifts were recorded in δ (ppm) downꢀeld from an internal
tetramethylsilane standard. High-resolution mass spectra
Methyl 2‑([pyridin‑4‑ylmethyl]amino)benzoate (2)
(
HRMS) were measured with JMS-700 [2] (JEOL, Tokyo,
Japan) or DIP-181 (JASCO, Tokyo, Japan). The high-per-
formance liquid chromatography (HPLC) system included
a 600C pump (Nihon Waters, Tokyo, Japan), a 2487 dual
wavelength absorbance detector (Nihon Waters), a 170 NaI
radioactivity detector (Beckman Coulter, Tokyo, Japan),
and a Cosmosil 5C18-AR column (10 × 250 mm; Nacalai
Tesque, Kyoto, Japan). Radioactivity was measured using
an ARC-300 or ARC-380 NaI (Tl) gamma counter (Aloka,
Tokyo, Japan). All chemicals used were of reagent grade and
were used without puriꢀcation.
Methyl anthranilate (1) (9.06 g, 60 mmol) and isonico-
tin aldehyde (10.26 g, 60 mmol) were stirred in distilled
methanol (100 ml), followed by addition of acetic acid
(3.6 ml) and further stirring for 18 h at room temperature.
Sodium cyanotrihydroborate (7.2 g) dissolved in a small
amount of methanol was added dropwise to the solution
over 2 h and stirred for 2 h. After evaporating the solvent
under reduced pressure, ethyl acetate was added to the
residue. The solution was washed with saturated sodium
bicarbonate and brine and dried over sodium sulfate. After
evaporating the solvent, the residue was separated and
puriꢀed by silica gel column chromatography using ethyl
acetate/n-hexane (1/1, v/v) as an eluate to obtain 2: yield,
53.3%; mp, 86.1–86.3 ℃.
Human umbilical vein endothelial cells (HUVECs) and
a medium set for cultivation were obtained from Kurabo
Industries (Osaka, Japan). PC-3 cells were provided by the
Cell Resource Center for Biomedical Research, Institute
of Development, Aging and Cancer, Tohoku University.
1
3

Annals of Nuclear Medicine
1
2
‑([Pyridin‑4‑ylmethyl]amino)benzoic acid (3)
429.0337; H-NMR (CDCl ) δ: 4.45 (d,2H), 6.55 (d,1H),
3
6
.70 (t,1H), 7.10 (t,1H), 7.25–7.55 (dt,1H + 3H), 7.7
Addition of 10% sodium hydroxide to 2 (1.0 g, 4.12 mmol)
was followed by reꢃux for 3 h. After cooling, 20% hydro-
chloric acid was added to the solution until the pH reached
(s,1H), 8.19 (d,2H), 8.65 (d,3H).
3
–4, with constant stirring on ice. The precipitate was ꢀl-
N‑(4‑Iodophenyl)‑2‑([pyridin‑4‑ylmethyl]amino)benzamide
tered and recrystallized with methanol to obtain 3: yield,
(4d)
1
9
8.1%; mp, 215.5–215.6 ℃; H-NMR (CDCl ) δ: 4.51
3
(
(
d,2H), 6.50 (dd,1H), 6.60 (dd,1H), 7.18 (dd,1H), 7.32
Using a method similar to that described above for the syn-
thesis of compound 4a, p-iodoaniline (1.05 g, 4.79 mmol)
was used instead of m-iodophenol to obtain 4d: yield,
dm,2H), 7.82 (dd,1H), 8.20 (qd,1H), 8.60 (de,2H).
3
1.0%; mp, 70.6–70.8 ℃; HRMS m/z 429.0338, Found
1
3‑Iodophenyl 2‑([pyridin‑4‑ylmethyl]amino)benzoate (4a)
429.0337; H-NMR (CDCl ) δ: 4.42 (d,2H), 6.55 (d,1H),
3
6
.73 (t,1H), 7.10 (t,1H), 7.25–7.50 (st,1H + 3H), 7.7
m-Iodophenol (1.21 g, 5.5 mmol) and 4-dimethylami-
(s,1H), 8.19 (d,2H), 8.65 (d,3H).
nopyridine (0.672 g, 5 mmol) were mixed in chloroform
(
50 ml) and stirred for 30 min at room temperature. Com-
pound 3 (1.135 g, 5 mmol) and N,N’-dicyclohexylcarbo-
diimide (1.135 g, 5.5 mmol) were added to the solution
and stirred for 24 h at room temperature. After ꢀltering
the precipitated impurities, the solvent was evaporated
under reduced pressure. The residue was resolved in 1 M
hydrochloric acid and extracted with chloroform three
times. The organic layer was washed with brine and dried
over sodium sulfate. After evaporation and recrystalli-
zation with ethanol, silica gel column chromatography
using chloroform/methanol (20/1, v/v) as an eluate was
Tributylstannyl precursors (5a–d)
After dissolving 4a–d (0.5 g, 1.16 mmol) in tolu-
ene (25 ml), bis(tributyltin) (0.43 ml, 0.85 mmol)
and tetrakis(triphenylphosphine) palladium (0.002 g,
0.01 mmol) were added and reꢃuxed for 24 h under argon
atmosphere. After cooling to room temperature, the impu-
rities were removed using celite ꢀltration, and the ꢀltrate
was evaporated under reduced pressure. The residue was
separated and puriꢀed by silica gel column chromatogra-
phy using chloroform/methanol (50/1, v/v) as an eluate to
obtain 5a–d as an oily matter.
performed to obtain 4a: yield, 35.8%; mp, 135.5–135.6℃;
1
HRMS m/z 430.0178, Found 430.0183; H-NMR (CDCl )
3
δ: 4.49 (d,2H), 6.55 (d,1H), 6.71 (t,1H), 7.20 (t,1H),
3-(Tributylstannyl)phenyl 2-([pyridin-4-ylmethyl]
+
7
.25–7.40 (dt,3H + 1H), 7.60 (t,2H), 8.19 (d,2H), 8.65
amino)benzoate (5a): yield, 49.4%; MS-EI m/z: 595[M] ;
1
(
d,2H).
H-NMR (CDCl ) δ: 0.91 (t,9H), 1.18 (t,4H), 1.33 (tt,6H),
3
1
.57 (dt,7H), 4.49 (d,2H), 6.55 (d,1H), 6.73 (t,1H), 7.10
(
d,1H), 7.20–7.40 (st,4H + 2H), 8.21 (t,2H), 8.65 (d,2H).
4‑Iodophenyl 2‑([pyridin‑4‑ylmethyl]amino)benzoate (4b)
4-(Tributylstannyl)phenyl 2-([pyridin-4-ylmethyl]
+
amino)benzoate (5b): yield, 39.7%; MS-EI m/z: 595[M] ;
1
Using a method similar to that described above for the syn-
H-NMR (CDCl ) δ: 0.91 (t,9H), 1.18 (t,4H), 1.33 (tt,6H),
3
thesis of compound 4a, p-iodophenol (1.21 g, 5.5 mmol) was
1.57 (dt,7H), 4.49 (d,2H), 6.55 (d,1H), 6.73 (t,1H), 7.10
(d,1H), 7.20–7.40 (st,4H + 2H), 8.21 (t,2H), 8.65 (d,2H).
2-([Pyridin-4-ylmethyl]amino)-N-(3-[tributylstannyl]
used instead of m-iodophenol to obtain 4b: yield, 19.5%; mp,
1
1
97.1–197.2 ℃; HRMS m/z 430.0178, Found 430.0180; H-
NMR (CDCl ) δ: 4.52 (d,2H), 6.55 (d,1H), 6.75 (t,1H), 6.98
phenyl)benzamide (5c): yield, 29.0%; MS-EI m/z:
3
1
(
d,1H), 7.25–7.40 (st,1H+3H), 7.76 (d,2H), 8.19 (dd,2H),
535[mono desbuthyl fragment]; H-NMR (CDCl ) δ:
3
8
.65 (d,2H).
0.91 (t,9H), 1.18 (t,4H), 1.33 (tt,6H), 1.57 (dt,7H), 4.46
(
(
d,2H), 6.51 (d,1H), 6.70 (t,1H), 7.10 (d,1H), 7.20–7.40
tt,2H + 2H), 7.45 (s,1H), 7.58 (d,1H), 7.68 (d,1H), 7.79
N‑(3‑Iodophenyl)‑2‑([pyridin‑4‑ylmethyl]amino)benzamide
4c)
(s,1H), 8.12 (t,1H), 8.54 (d,2H).
(
2-([Pyridin-4-ylmethyl]amino)-N-(4-[tributylstannyl]
phenyl)benzamide (5d): yield, 24.0%; MS-EI m/z:
1
Using a method similar to that described above for the syn-
thesis of compound 4a, m-iodoaniline (1.05 g, 4.79 mmol)
was used instead of m-iodophenol to obtain 4c: yield,
535[mono desbuthyl fragment]; H-NMR (CDCl ) δ:
3
0.91 (t,9H), 1.18 (t,4H), 1.33 (tt,6H), 1.57 (dt,7H), 4.49
(d,2H), 6.55 (d,1H), 6.73 (t,1H), 7.10 (d,1H), 7.20–7.40
(st,4H + 2H), 8.21 (t,2H), 8.65 (d,2H).
6
3.8%; mp, 178.6–178.8 ℃; HRMS m/z 429.0338, Found
1
3

Annals of Nuclear Medicine
Radioiodinated compounds (6a–d)
v/v) as an eluent was performed to measure Rf values of the
radioactivity to calculate percent intact rates of tested com-
pounds 6a–d in comparison to Rf values of UV absorbance
for control nonradioactive compounds 4a–d.
In a sealed vial, 0.1 M hydrochloric acid solution (25 μl),
1
25
[
I]NaI (3.7–37 MBq/1–10 μl), and 30 w/v% hydrogen
peroxide solution (10 μl) were sequentially added to pre-
cursor tributylstannyl compounds 5a–d (25 μl, 1.0 mg/ml
methanol solution). The mixture was then allowed to react at
room temperature for 15 min. The target radioiodinated com-
pounds 6a–d were puriꢀed by HPLC using methanol/0.01 M
phosphate buꢂer (90/10 or 85/15, v/v) as the mobile phase
at a ꢃow rate of 3 ml/min, followed by radiochemical purity
analysis. After evaporating the collected HPLC fractions,
saline was added to the vial to obtain ꢀnal formulation for
in vitro and in vivo experiments.
Cellular uptake study
PC-3 cells were cultured in RPMI 1640 medium contain-
ing 10% FCS supplemented with penicillin–streptomycin
solution in a CO incubator (5% CO at 37 ℃, humidiꢀed)
2
2
in accordance with the instruction provided by the supplier
(IDAC, Tohoku University).
5
PC-3 cells (1.0 × 10 ) were suspended in RPMI 1640
medium (serum free, 800 μl), and 100 μl of aqueous solution
containing 0.05% TritonX and 1% DMSO with or without
selective VEGFR-TK inhibitors (VEGFR-TK inhibitor II and
SU5614 at a ꢀnal concentration of 10 μM) were added to
the suspension. Samples were then incubated for 30 min at
37 ℃. To the mixed solution, 100 μl of 10% DMSO contain-
In vitro estimation of inhibitory potencies
HUVECs were cultured in HuMedia-EB2 medium sup-
plemented with 2% fetal calf serum (FCS), human epider-
mal growth factor, human ꢀbroblast growth factor (basic),
hydrocortisone, gentamicin, amphotericin B, and heparin in
a CO incubator (5% CO at 37 ℃, humidiꢀed) in accord-
1
25
ing [ I]p-NPAM (1.85 kBq) was added to a total volume
of 1 ml and incubated for 1 h at 37 ℃. After the reaction, the
solution was suction ꢀltered using a glass ꢀlter and washed
with 3 ml of 0.2 mM phosphate buꢂer (pH 7.2). The radioac-
tivity was measured by a gamma counter to estimate dimin-
2
2
ance with the manufacturer’s instruction. The medium was
exchanged every other day and cultured to reach subconꢃu-
ent conditions for the following experiments. The purities of
the compounds 4a–d used for biological experiments were
estimated by HPLC in advance and conꢀrmed to show only
single peak detected.
1
25
ished rates of cellular uptake of [ I]p-NPAM following
inhibitor treatment.
First, HUVECs collected from dishes by trypsin treat-
4
ment, inoculated at 1.0 × 10 cells/well in 24-well plates,
In vivo biodistribution study
and incubated for 24 h at 37 ℃. The medium was exchanged
with MCDB131 medium supplemented with 0.5% FCS and
further incubated for 24 h. Ten μl of VEGF (1 μg/ml) and
7
A suspension of PC-3 cells (1.0×10 cells in 200 μl RPMI
medium containing 10% FCS) was subcutaneously injected
into the thighs or the ꢃank of 4-week-old male BALB/c-nu
mice (20–25 g). Mice were then raised for 2 weeks. Subse-
1
00 μl of 1% dimethyl sulfoxide (DMSO) containing 4a–d
or SU5614 (ꢀnal concentrations: 1 nM–10 μM) were added,
and samples were incubated for 72 h at 37 ℃. The cells were
then collected by trypsin treatment, and cell counting was
performed to estimate the eꢂects of each tested compound
on HUVEC growth. The inhibitory potencies of the com-
pounds were evaluated by determining half-maximal inhibi-
tory concentration (IC ) values taken from inhibition curves
1
25
125
quently, [ I]m-NPAM or [ I]p-NPAM (100 μl, 18.5 kBq
in saline) was injected into tumor-bearing mice via the
tail vein. The mice were sacriꢀced at various time points.
Samples of blood and the tissues of interest were excised
and weighed. Radioactivity was measured using a gamma
counter. Accumulation of radioactivity in each tissue was
expressed as a percentage of the injected dose per gram tis-
sue (%ID/g tissue).
50
plotted as the relationship between the ꢀnal cell number and
the ꢀnal concentration of each compound added [20].
1
25
To evaluate the speciꢀcity of [ I]p-NPAM accumu-
lation in tumors, an in vivo pretreatment study was per-
formed. Initially, VEGFR-TK inhibitor II (1.0 mg/kg) or
SU5614 (1.0 mg/kg) was intravenously injected into the
In vitro stability assay
Compounds 6a–d (9.25 kBq, 100 μl), 0.01 M phosphate
1
25
buffer (pH 7.4, 890 μl), and 1% bovine serum albumin
mice, followed by intravenous injection of [ I]p-NPAM
(
10 μl) were mixed and incubated for 1, 4, and 24 h at 37 ℃.
(100 μl, 18.5 kBq in saline) 5 min later. At 30 min after
1
25
At each time point, radioactivity was extracted from the
solution with ethyl acetate three times, followed by evapo-
ration under reduced pressure. Normal-phase thin-layer
chromatography (TLC) using chloroform/methanol (10/1,
[
I]p-NPAM administration, the tumors were excised
and weighed, and the radioactivity was measured to obtain
%ID/g tissue. The data were presented in comparison to the
results in the non-pretreated group.
1
3

Annals of Nuclear Medicine
Table 1 Summary of the
No
Name
HPLC eluent
Retention time
(min)
Radiochemical Radio-
radiosynthesis of compounds
a
A/B
yield
%)
chemical
purity
6
a–d
(
(
%)
1
1
1
1
25
25
25
25
6a
6b
6c
6d
[
[
[
[
I]m-NPAE
I]p-NPAE
I]m-NPAM
I]p-NPAM
10/90
10/90
15/85
15/85
12.8
13.0
9.6
90.3
97.9
95.2
68.0
>95
>95
>95
>95
9.3
aA, 0.01 M phosphate buꢂer; B, methanol
SPECT/CT
Table 2 Inhibitory potencies
IC ) of compounds 4a–d in
No
4a
Name
IC50
(
5
0
cell proliferation assays
1
25
m-NPAE
p-NPAE
m-NPAM
p-NPAM
SU5614
61.4 nM
13.2 nM
2.53 μM
12.5 nM
0.44 μM
[
I]p-NPAM (100 μl, 11.1 MBq in saline) was injected
4
4
4
b
c
d
into a tumor-bearing mouse via the tail vein. A static SPECT
scan acquired over 60 min (60 s×64 projections, 360°) using
a low-energy multi-pinhole collimator (1 mm, energy win-
1
25
dow; 20–50 keV) was started at 24 h after [ I]p-NPAM
injection using a PET/SPECT/CT FX system (Gamma Med-
ica Ideas, Northridge, CA) [21]. Subsequently, CT images
were acquired at 256 projections for 1 min. The mouse was
anesthetized under isoꢃurane during the scans. Following
a reconstruction process by OSEM (iteration, 5; subset,
Each value represents the mean
4
), the SPECT image was fused with the CT image using
Vivid software (Gamma Medica Ideas). A pretreatment
study was also performed in a similar way using a tumor-
bearing mouse subjected to intravenous injection of SU5614
(
1.0 mg/kg) 5 min prior to [125I]p-NPAM administration.
Statistics
Data are shown as means± standard deviations. Statistical
analyses were performed using Dunnett’s multiple compari-
sons tests as post hoc tests, following one-way analysis of
variance. Results with two-tailed p values of less than 0.05
were considered statistically signiꢀcant.
Results and discussion
Synthesis (Scheme 1) and radiosynthesis (Scheme2)
Nonradioactive targeted compounds 4a–d, which were
designed based on the structure of the selective VEGFR-TK
inhibitor AAL-993 as a candidate for in vivo SPECT imag-
ing of VEGFR-TK in tumors, were successfully obtained
at overall yields of 10.2–33.4% through a three step syn-
thesis, as shown in Scheme 1. The compounds 4a–d were
designated m-NPAE, p-NPAE, m-NPAM, and p-NPAM,
respectively.
Fig. 1 In vitro stability assay. Each value represents the mean± stand-
ard deviation for three samples at each time point
reaction from corresponding tributylstannyl precursors
5a–d, as shown in Scheme 2. Successful syntheses were
conꢀrmed by the coincidence of retention times between
UV absorbance peaks of nonradioactive 4a–d and radioac-
tivity peaks of 6a–d in chromatograms from reverse-phase
HPLC analysis (Table 1). The radiochemical yield and the
radiochemical purity of 6a–d were more than 68.0% and
1
25
Radioiodinated compounds 6a–d ([ I]m-NPAE,
1
25
125
125
[
I]p-NPAE, [ I]m-NPAM, and [ I]p-NPAM) were
obtained by an organotin–radioactive iodine exchange
1
3

Annals of Nuclear Medicine
more than 95%, respectively, as summarized in Table 1. The
Table 4 Radioactivity biodistribution after intravenous administration
125
of [ I]m-NPAM in tumor-bearing mice
molar activity of 6a–d was approximately 81.4 TBq/mmol.
Time after administration (h)
In vitro estimation of inhibitory potency (Table 2)
and stability (Fig. 1)
0
.5
1
2
24
0.67± 0.06
0.36± 0.06
0.38± 0.05
2.89± 0.90
Blood
Pancreas
Spleen
1.21± 0.11
1.63± 0.19
0.97± 0.13
0.81± 0.09
0.77± 0.04
0.62± 0.04
0.83± 0.04
0.64± 0.03
0.50± 0.04
4.42± 0.91
Inhibitory potencies of 4a–d toward VEGFR-TK activity
in HUVECs were measured by cell proliferation assays
and compared with the results for SU5614, a representa-
tive VEGFR-TK inhibitor (Table 2). As a result, m-NPAE,
p-NPAE, m-NPAM, and p-NPAM showed high inhibitory
potencies, with IC s of 12.5 nM–2.53 μM, which were com-
Stomach 11.47± 1.44 11.91± 1.49
Intestine 20.00± 1.19 26.12± 1.55 30.78± 3.12 14.11± 1.36
Liver
11.36± 1.50
4.13± 0.38
1.14± 0.14
2.97± 0.40
0.75± 0.05
1.32± 0.34
1.19± 0.53
3.50± 0.37
1.67± 0.18
0.62± 0.05
1.60± 0.27
0.21± 0.02
0.90± 0.14
0.44± 0.05
2.44± 0.99
1.21± 0.09
0.44± 0.01
1.40± 0.12
0.11± 0.01
0.77± 0.20
0.57± 0.24
0.51± 0.16
0.47± 0.07
0.28± 0.02
0.86± 0.22
0.04± 0.01
0.75± 0.29
0.17± 0.03
Kidney
Heart
Lung
Brain
Tumor
Muscle
50
parable to that of SU5614.
Next, to estimate the in vivo availability of the radi-
oiodinated compounds 6a–d, stability was assessed by
normal-phase TLC analysis following incubation for sev-
eral hours in neutral phosphate buꢂer. As shown in Fig. 1,
1
25
125
[
I]m-NPAM and [ I]p-NPAM stably existed in an intact
Data are presented as %ID/g tissue. Each value represents the
mean± standard deviation for four animals at each interval
1
25
form in the solution for over 24 h, whereas [ I]m-NPAE
1
25
and [ I]p-NPAE were rapidly degraded. Therefore, on the
1
25
basis of the above two in vitro experiments, [ I]m-NPAM
1
25
and [ I]p-NPAM were selected for further evaluation in
in vivo experiments owing to their high inhibitory potencies
and stabilities.
intravenous administration. The tumor radioactivity was then
slowly cleared but retained to some extent to give 0.85%ID/g
at 24 h (Fig. 2a). The blood radioactivity was the highest
at 30 min (1.98%ID/g) and then rapidly cleared to achieve
0.20%ID/g at 24 h. Thus, the radioactivity ratios of tumor/
muscle and tumor/blood, important indexes for determin-
ing the feasibility of the approach for in vivo imaging, were
2.32 and 1.41, respectively, at 30 min and rose to 5.55 and
4.16, respectively, at 24 h (Fig. 2b). It is previously reported
that radiolabeled humanized monoclonal antibody bevaci-
zumab showed high blood radioactivity retention (9.54 and
In vivo biodistribution study (Tables 3 and 4, Fig. 2)
The pharmacokinetics of [125I]p-NPAM and [ I]m-NPAM
were evaluated in biodistribution experiments in PC-3
tumor-bearing mice, as shown in Tables 3 and 4, respec-
125
1
25
tively. Notably, [ I]p-NPAM showed high and fast radio-
activity accumulation in tumors (2.77%ID/g) at 30 min after
8
9
111
1
1.87%ID/g for Zr and In labeled bevacizumab, respec-
89
tively) and low tumor/blood ratio (0.51 and 0.52 for Zr and
111
Table 3 Radioactivity biodistribution after intravenous administration
In labeled bevacizumab, respectively) at 24 h post-admin-
istration in tumor bearing mice [22]. Thus, it is indicated that
125
of [ I]p-NPAM in tumor-bearing mice
1
25
Time after administration (h)
[
I]p-NPAM succeeded in faster targeting and clearance
1
25
compared with antibody probes.[ I]m-NPAM also showed
the highest tumor radioactivity (1.32%ID/g) at the earliest
time point over the experimental period; however, this value
0
.5
1
2
24
Blood
Pancreas
Spleen
1.98± 0.20
1.54± 0.27
1.39± 0.08
1.62± 0.18
1.14± 0.08
1.03± 0.06
1.27± 0.24 0.20± 0.01
0.64± 0.12 0.17± 0.01
0.66± 0.16 0.21± 0.04
8.58± 0.73 1.17± 0.34
125
was only approximately half that of [ I]p-NPAM (Fig. 2a).
The blood radioactivity was not high at 30 min (1.21%ID/g),
but slow clearance thereafter retained blood radioactivity
as 0.67%ID/g at 24 h. Thus, the tumor/blood radioactiv-
ity ratios were 1.10 and 1.13 at 30 min and 24 h (Fig. 2b),
respectively, whereas the tumor/muscle radioactivity ratio
Stomach 20.48± 0.80 12.69± 0.71
Intestine 13.79± 1.18 19.61± 1.12 23.37± 0.95 9.03± 0.51
Liver
12.24± 1.19
3.68± 0.29
1.24± 0.18
3.55± 0.50
0.75± 0.13
2.77± 0.45
1.14± 0.20
3.93± 0.57
1.84± 0.10
0.89± 0.07
2.29± 0.18
0.23± 0.01
1.73± 0.14
0.71± 0.23
2.24± 0.34 0.24± 0.03
1.07± 0.11 0.18± 0.01
0.45± 0.08 0.11± 0.02
1.20± 0.21 0.32± 0.09
0.09± 0.01 0.03± 0.01
1.21± 0.16 0.85± 0.03
0.40± 0.05 0.15± 0.08
Kidney
Heart
Lung
Brain
Tumor
Muscle
1
25
reached the same range as for [ I]p-NPAM.
125
Analysis of the excretory routes of [ I]p-NPAM and
125
[
I]m-NPAM from the body indicated that the liver and
intestine played important roles owing to high radioactivity
accumulation in the liver and intestine in comparison with
that in the kidneys (Tables 3 and 4). This was expected
based on the high lipophilicity of the compounds (4.2
Data are presented as %ID/g tissue. Each value represents the
mean± standard deviation for four animals at each interval
1
3

Annals of Nuclear Medicine
Fig. 2 Comparison of biodis-
A
B 5.0
125
3
3
.5
.0
tributions of [ I]p-NPAM and
125
[
I]m-NPAM in tumor-bearing
4
4
3
.5
.0
.5
mice. a Accumulation of radio-
activity in the tumor. b Tumor/
blood (T/B) ratio. Each value
represents the mean± standard
deviation for four animals at
each interval
1
1
25
25
125
125
[
[
I]m-NPAM
I]p-NPAM
[
[
I]m-NPAM
I]p-NPAM
2.5
3.0
2
1
1
0
0
.0
.5
.0
.5
.0
2
2
1
1
0
.5
.0
.5
.0
.5
0.0
0
6
12
18
24
0
6
12
18
24
Time aꢀer injecꢁon (h)
Time aꢀer injecꢁon (h)
clogP value estimated by ChemBioDraw software; Perki-
nElmer Japan (Tokyo, Japan)). Notably, there was some
radioactivity detected in the stomach (Tables 3 and 4),
indicating that the deiodination reaction occurred to some
extent in vivo; this result was inconsistent with the in vitro
stability assay (Fig. 1). Although this was a potential draw-
back of the study and should be addressed in future stud-
In vitro and in vivo blocking study (Fig. 3)
1
25
To estimate the specificity of [ I]p-NPAM targeting
VEGFR-TK expressed in cultured cells and inoculated
tumor tissues, blocking experiments were performed using
selective VEGFR-TK inhibitors in vitro and in vivo (Fig. 3).
The results showed that pretreatment with selective VEGFR-
TK inhibitors, such as SU5614 and VEGFR-TK inhibitor
1
25
ies, we further evaluated the potential of [ I]p-NPAM
in blocking assays and in vivo imaging assays because
II, signiꢀcantly decreased radioactivity accumulated after
1
25
125
[
I]p-NPAM showed superior biodistribution, including
[
I]p-NPAM administration in comparison with those of
high tumor targeting and tumor/blood ratios, in tumor-
corresponding control groups both in cultured PC-3 cells
(Fig. 3a) and inoculated tumor tissues (Fig. 3b). Therefore,
these ꢀndings clearly suggested that speciꢀc interactions
bearing mice in comparison to [125I]m-NPAM.
A
B
3
.5
3
100
80
60
40
20
0
*
*
2.5
2
*
*
1
.5
1
0
.5
0
VEGFR-TK
inhibitor II
VEGFR-TK
inhibitor II
Control
SU5614
Control SU5614
Fig. 3 In vitro and in vivo blocking study. a Radioactivity accumu-
Radioactivity accumulation (%ID/g tissue) in the tumor 30 min after
1
25
125
lation (%) in PC-3 cells 1 h after [ I]p-NPAM addition with or
without pretreatment using selective VEGFR-TK inhibitors. Each
value represents the mean± standard deviation for four samples. b
[
I]p-NPAM injection with or without pretreatment using selective
VEGFR-TK inhibitors. Each value represents the mean± standard
deviation for 4 animals. *p<0.05 versus the control
1
3

Annals of Nuclear Medicine
Fig. 4 Fused SPECT/CT images in a transaxial view obtained 24 h
were performed over approximately 60 and 1 min, respectively, under
isoꢃurane anesthesia. The white arrows indicate tumor site
1
25
after injection of 11.1 MBq [ I]p-NPAM with or without pretreat-
ment using SU5614 in tumor-bearing mice. SPECT and CT scans
1
25
between [ I]p-NPAM and VEGFR-TK would signiꢀcantly
lipophilicity of the probe. Accordingly, such reꢀnement of
1
25
contribute to the accumulation of tumor radioactivity after
[
I]p-NPAM may lead to the development of promising
1
25
[
I]p-NPAM administration both in vitro and in vivo.
VEGFR imaging probes for use with SPECT in the future.
Among the four anthranilate derivatives synthesized in this
1
25
SPECT/CT (Fig. 4)
study, [ I]p-NPAM may have potential applications as a
lead compound for future reꢀnement of in vivo VEGFR
imaging with SPECT.
Lastly, SPECT/CT images were obtained in tumor-bearing
mice with a PET/SPECT/CT FX system 24 h after intrave-
1
25
Acknowledgements The authors have no competing interests to
declare. This work was partly supported by JSPS KAKENHI (Grant
nos. 19591437 and 19H03606). The funding bodies had no role in
study design, data collection and analysis, decision to publish, or manu-
script preparation.
nous injection of 11.1 MBq [ I]p-NPAM with or without
pretreatment using SU5614 (Fig. 4). The results showed that
radioactivity accumulation was observed in the tumor of a
control mouse, whereas radioactivity was negligible in the
tumor of a pretreated mouse, supporting the ꢀndings of the
blocking study shown in Fig. 3.
125
In summary, [ I]p-NPAM, which was designed based on
the selective VEGFR-TK inhibitor AAL-993, showed high
aꢁnity and stability in vitro and a favorable biodistribution
with high radioactivity accumulation in tumors. This com-
pound also exhibited rapid blood clearance, with a tumor to
blood radioactivity ratio of 4 at 24 h, enabling in vivo imag-
ing of the tumor tissue in living mice with SPECT. To the
best of our knowledge, this is the ꢀrst report describing the
development of a small molecule imaging probe for VEGFR
expression in rodent tumors with SPECT. Further reꢀnement
References
1
.
Folkman J. Tumor angiogenesis: therapeutic implications. N Engl
J Med. 1971;285(21):1182–6.
2. Leenders WP, Kusters B, de Waal RM. Vessel co-option: how
tumors obtain blood supply in the absence of sprouting angiogen-
esis. Endothelium. 2002;9(2):83–7.
3
.
Terry SY, Rijpkema M, Abiraj K, van der Graaf WT, Oyen
WJ, Boerman OC. Radiolabeled imaging probes targeting
angiogenesis for personalized medicine. Curr Pharm Des.
2
014;20(14):2293–307.
125
4. Ferrara N. Vascular endothelial growth factor: basic science and
clinical progress. Endocr Rev. 2004;25(4):581–611.
of [ I]p-NPAM will encourage the development of superior
imaging probes for VEGFR SPECT in a variety of clinical
situations. The in vivo instability noted in our biodistribu-
tion study could be addressed by further modiꢀcations of the
chemical structure, and improvement in the pharmacokinet-
5
.
Goldman CK, Kendall RL, Cabrera G, Soroceanu L, Heike
Y, Gillespie GY, et al. Paracrine expression of a native solu-
ble vascular endothelial growth factor receptor inhibits tumor
growth, metastasis, and mortality rate. Proc Natl Acad Sci USA.
1
998;95(15):8795–800.
1
25
ics of [ I]p-NPAM may be necessary to achieve SPECT
6
.
Prewett M, Huber J, Li Y, Santiago A, O’Connor W, King K,
et al. Antivascular endothelial growth factor receptor (fetal liver
imaging within several hours after injection by reducing the
1
3

Annals of Nuclear Medicine
kinase 1) monoclonal antibody inhibits tumor angiogenesis
and growth of several mouse and human tumors. Cancer Res.
evaluation of 5-[125I]Iodo-sunitinib. Bioorg Med Chem Lett.
2012;22(8):2850–5.
1
999;59(20):5209–18.
16. Rainer E, Wang H, Traub-Weidinger T, Widhalm G, Fueger B,
1
23
7
8
.
.
Gaykema SB, Brouwers AH, Lub-de Hooge MN, Pleijhuis RG,
Chang J, et al. The prognostic value of [ I]-vascular endothelial
8
9
123
Timmer-Bosscha H, Pot L, et al. Zr-bevacizumab PET imaging
in primary breast cancer. J Nucl Med. 2013;54(7):1014–8.
van Es SC, Brouwers AH, Mahesh SVK, Leliveld-Kors AM, de
growth factor ([ I]-VEGF) in glioma. Eur J Nucl Med Mol Imag-
ing. 2018;45(13):2396–403.
17. Cocco MT, Congiu C, Lilliu V, Onnis V. Synthesis of new N-(2-
(triꢃuoromethyl)pyridin-4-yl)anthranilic acid derivatives and
their evaluation as anticancer agents. Bioorg Med Chem Lett.
2004;14(23):5787–91.
8
9
Jong IJ, Lub-de Hooge MN, et al. Zr-Bevacizumab PET: poten-
tial early indicator of everolimus eꢁcacy in patients with meta-
static renal cell carcinoma. J Nucl Med. 2017;58(6):905–10.
Chopra A. 89Zr-Labeled N-succinyldesferrioxamine-ranibizumab.
Molecular Imaging and Contrast Agent Database (MICAD).
Bethesda: National Center for Biotechnology Information (US);
9
.
18. Furet P, Bold G, Hofmann F, Manley P, Meyer T, Altmann K-H.
Identiꢀcation of a new chemical class of potent angiogenesis
inhibitors based on conformational considerations and database
searching. Bioorg Med Chem Lett. 2003;13(18):2967–71.
19. Manley PW, Furet P, Bold G, Brüggen J, Mestan J, Meyer T, et al.
Anthranilic acid amides: a novel class of antiangiogenic VEGF
receptor kinase inhibitors. J Med Chem. 2002;45(26):5687–93.
20. Xin X, Yang S, Kowalski J, Gerritsen ME. Peroxisome
proliferator-activated receptor gamma ligands are potent
inhibitors of angiogenesis in vitro and in vivo. J Biol Chem.
1999;274(13):9116–211.
2
004.
1
8
1
1
0. Wang JQ, Miller KD, Sledge GW, Zheng QH. Synthesis of [ F]
SU11248, a new potential PET tracer for imaging cancer tyrosine
kinase. Bioorg Med Chem Lett. 2005;15(19):4380–4.
1. Asakawa C, Ogawa M, Kumata K, Fujinaga M, Kato K, Yamasaki
11
T, et al. [ C]sorafenib: radiosynthesis and preliminary PET study
of brain uptake in P-gp/Bcrp knockout mice. Bioorg Med Chem
Lett. 2011;21(8):2220–3.
1
1
2. Ilovich O, Åberg O, Långström B, Mishani E. Rhodium-mediated
21. Sakai T, Saito Y, Takashima M, Ogawa M, Magata Y. Develop-
ment of radioiodinated lipophilic cationic compounds for myocar-
dial imaging. Nucl Med Biol. 2015;42(5):482–7.
1
1
[
C]carbonylation: a library of N-phenyl-N′-(4-(4-quinolyloxy)
11
phenyl)-[ C]urea derivatives as potential PET angiogenic probes.
J Label Compd Radiopharm. 2009;52(5):151–7.
22. Nagengast WB, de Vries EG, Hospers GA, Mulder NH, de Jong
JR, Hollema H, et al. In vivo VEGF imaging with radiolabeled
bevacizumab in a human ovarian tumor xenograft. J Nucl Med.
2007;48(8):1313–9.
3. Ilovich O, Jacobson O, Aviv Y, Litchi A, Chisin R, Mishani E.
Formation of ꢃuorine-18 labeled diaryl ureas–labeled VEGFR-2/
PDGFR dual inhibitors as molecular imaging agents for angiogen-
esis. Bioorg Med Chem. 2008;16(8):4242–51.
1
1
4. Kniess T. Radiolabeled small molecule inhibitors of VEGFR -
recent advances. Curr Pharm Des. 2012;18(20):2867–74.
5. Kuchar M, Oliveira MC, Gano L, Santos I, Kniess T. Radi-
oiodinated sunitinib as a potential radiotracer for imaging
angiogenesis-radiosynthesis and first radiopharmacological
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional aꢁliations.
1
3