![Fig. 3. Horizontal (top) and sagittal (bottom) PET images showing distribution of radioactivity in the lungs and liver at 20 – 90 min after intravenous administration of (S)-[11C]-1 in normal control rat (right) and rat with LPS-induced lung inflammation (left).](/figures/fig-3-horizontal-top-and-sagittal-bottom-pet-images-showing-e9yi51ct.webp)
![Fig. 2. In vitro autoradiographic labeling of coronal mouse brain slices with (S)-[11C]-1. Brain samples were derived from PS19 and APP23 transgenics and their wild-type littermates. The images demonstrate radioligand binding without blockade (total binding, TB) and with 10 lM fFML (nonspecific binding, NSB).](/figures/fig-2-in-vitro-autoradiographic-labeling-of-coronal-mouse-2seqeaen.webp)
![Fig. 1. (top) Sagittal PET images showing distribution of radioacticity in the rat brain and surrounding tissues after intravenous administration of (S)-[11C]-1. PET data were generated by summation of dynamic data at 0 – 90 min after radioligand injection, and were merged onto the MRI anatomical template. (bottom) Time-radioactivity curve in the whole brain of a rat after intravenous administration of (S)-[11C]-1.](/figures/fig-1-top-sagittal-pet-images-showing-distribution-of-1buea14r.webp)
FULL PAPER
Radiosynthesis and in vivo Evaluation of Carbon-11 (2S)-3-(1H-Indol-3-yl)-2-{[(4-
methoxyphenyl)carbamoyl]amino}-N-{[1-(5-methoxypyridin-2-yl)cyclohexyl]methyl}
propanamide: An Attempt to Visualize Brain Formyl Peptide Receptors in Mouse
Models of Neuroinflammation
by Enza Lacivita*
a
), Madia Letizia Stama
a
), Jun Maeda
b
), Masayuki Fujinaga
b
), Akiko Hatori
b
), Ming-Rong Zhang
b
),
Nicola A. Colabufo
a
)
c
), Roberto Perrone
a
), Makoto Higuchi
b
), Tetsuya Suhara
b
), and Marcello Leopoldo
a
)
c
)
a
) Dipartimento di Farmacia – Scienze del Farmaco, Universit
a degli Studi di Bari ‘Aldo Moro’, via Orabona, 4, IT-70125,
Bari (phone: +39-080-5442750, fax: +39-080-5442231, e-mail: enza.lacivita@uniba.it)
b
) National Institute of Radiological Sciences, National Institutes for Quantum and Radiological Science and Technology, 4-9-
1 Anagawa, Inage-ku, Chiba, Chiba 263-8555, Japan
c
) BIOFORDRUG s.r.l., Spin-off, Universit
a degli Studi di Bari ‘Aldo Moro’, via Orabona, 4, IT-70125, Bari
Here, we describe the very first attempt to visualize in vivo formyl peptide receptors (FPRs) in mouse brain by positron
emission tomography (PET). FPRs are expressed in microglial cells where they mediate chemotactic activity of b-amyloid
peptide in Alzheimer disease and, thus, are involved in neuroinflammatory processes. To this purpose, we have selected (2S)-
3-(1H-Indol-3-yl)-2-{[(4-methoxyphenyl)carbamoyl]amino}-N-{[1-(5-methoxypyridin-2-yl)cyclohexyl]methyl}propanamide ((S)-
1), that we have previously identified as a potent non-peptidic FPR agonist. (S)-[
11
C]-1 has been prepared in high
radiochemical yield. ( S)-[
11
C]-1 showed very low penetration of blood–brain barrier and, thus, was unable to accumulate into
the brain. In addition, (S)-[
11
C]-1 was not able to label FPRs receptors in brain slices of PS19 and APP23 mice, two animal
models of Alzheimer disease. Although (S)-[
11
C]-1 was not suitable to visualize FPRs in the brain, this study provides useful
information for the design and characterization of future potential PET radioligands for visualization of brain FPRs by PET.
Keywords: Formyl peptide receptors, Neuroinflammation, PET, Radiosynthesis, Ureidopropanamide.
Introduction
Neuroinflammation is a complex, dynamic, multicellular
process that plays a central role in a variety of neurological
diseases including neurodegenerative disorders. Activation
of microglia, the innate immune system of the central ner-
vous system (CNS), represents one of the hallmarks of neu-
roinflammation [1]. In normal condition, activation of
microglia exerts a protective function, providing tissue
repair by releasing anti-inflammatory cytokines and neu-
rotrophic factors [2][3]. Upon neuronal injury or infection,
microglia becomes overactivated leading to the release of
neurotoxic and proinflammatory factors [4][5]. Recent liter-
ature evidences also indicate that microglial activation is a
phenotypically and functionally diverse process, which may
change depending on aging, stage of disease, or presence of
other inflammatory events [2][6].
Several studies indicate that neuroinflammation is an
early and continuous feature of Alzheimer disease (AD).
Amyloid-b (Ab), which is considered central to the patho-
genesis of this disease, provokes the recruitment of micro-
glia and astrocytes to the sites in close proximity to Ab
deposits. Ab is able to interact with numerous surface
receptors that are expressed by microglial cells such as
formyl peptide receptors (FPRs) [7], Toll-like receptors
[8], scavenger receptors [9], and receptors for advanced
glycation end products (RAGE) [10]. Binding of Ab to
these receptors induces proinflammatory gene expression
and subsequent production of cytokines and chemokines
[11].
Molecular imaging techniques, such as positron emis-
sion tomography (PET) and single-photon emission com-
puted tomography (SPECT), can noninvasively visualize
specific targets of the inflammation cascade through speci-
fic and sensitive probes and, as such, can be powerful
tools to track the progression of neuroinflammatory pro-
cesses including those in AD. In particular, in vivo imag-
ing of microglia can offer a measure of the inflammatory
process and a means of tracking the progression of those
pathologies that trigger an immune activation in the brain
and the efficacy of therapeutic treatments over time [12].
Over the past decade, several targets for molecular imag-
ing of neuroinflammation have been studied, including
translocator protein 18 kDa (TSPO), that is considered a
hallmark of activated microglia [13], and endothelial
adhesion molecules, that are involved in the recruitment
DOI: 10.1002/cbdv.201500281 © 2016 Wiley-VHCA AG, Z
€
urich
Chem. Biodiversity 2016, 13, 875 – 883 875
of circulating leukocytes [12]. However, none of these
molecular targets has been fully validated as a biomarker
for the visualization of microglial activation.
The human FPR family belongs to the class A of G-pro-
tein coupled receptors and includes three subtypes, termed
FPR, FPR-like type-1 (FPRL-1), and type-2 (FPRL-2).
FPRs are expressed in several immune cells including
leukocytes, monocytes/macrophages, and microglia, and
are considered to play relevant roles in innate immunity
and host defense mechanisms and chemotaxis [14]. Among
FPRs, the FPRL-1, and its murine homologue FPR2,
appear to be relevant to the proinflammatory aspects of
AD, because FPRL-1 is a chemotactic receptor for Ab
42
,
the 42 amino acid form of Ab, which induces monocytes
migration and activation [15]. FPRL-1 also is involved in
the uptake and fibrillary aggregation of Ab
42
in mononu-
clear phagocytes by rapidly internalizing the Ab
42
-FPRL-1
complexes into cytoplasmic region [16]. FPRL-1 mediates
Ab
42
-induced senescence in neural stem/progenitor cells in
the hippocampus of APP/PS1 mice, an animal model of
AD [17]. Recently, it has been demonstrated that the
expression levels of FPRL-1 in primary microglial cells
increase after exposure to Ab and the recognition of Ab by
the FPRL-1 seems to be the starting point of the signaling
cascade inducing the inflammatory state [18]. Slowick et al.
has evidenced that APP/PS1 mice show higher expression
level of FPRs as compared to wild-type littermates. In par-
ticular, a significant increase in expression of FPR and
FPRL-1 in glial cells was observed in the cortex and hip-
pocampus through immunofluorescence and real-time PCR
analyses [19]. Therefore, FPRL1 could represent an inter-
esting target for monitoring in vivo the onset and progres-
sion of neuroinflammation in AD through molecular
imaging techniques.
To the best of our knowledge, only two studies have
been reported to date dealing with in vivo imaging of
FPRs. Locke et al. [20] have reported the radiosynthesis
and in vivo evaluation of cFLFLFK-PEG-
64
Cu, a peptide
analog of formyl peptide. The radiolabeled peptide was
able to bind to neutrophils in vitro and to accumulate
peripherally at sites of inflammation in vivo. Zhang et al.
[21] have evaluated the ability of cFLFLF-
64
Cu in the
detection of macrophages in pancreatic islets and in
the aorta in comparison with [
18
F]-fluorodesoxyglucose.
The ability of these radiotracers to penetrate and dis-
tribute into the brain was not studied. In general, the
in vivo use of peptides is hampered by short half-life due
to rapid proteolysis in plasma. Moreover, peptides cannot
cross blood–brain barrier (BBB) unless they interact with
specific transport systems [22]. Conversely, non-peptidic
small molecules may have some advantages because they
can be suitably designed to modulate properties, such as
potency, selectivity, lipophilicity, and cell permeability,
that are pivotal for a potential radiotracer [23].
Recently, we have reported on the identification of a
series of non-peptidic agonists for FPRs with 3-(1H-indol-
3-yl)-2-[3-(4-nitrophenyl)ureido]propanamide structure
[24]. Among the studied compounds, we have focused our
attention on (S)-1 (Table) as a PET radiotracer candidate,
because (S)-1 demonstrated potent interaction with FPRs
by inducing Ca
2+
mobilization in transfected HL-60 cells
and in human neutrophils. (S)-1 was also able to potently
induce chemotaxis in human neutrophils (Table) [24].
Importantly, (S)-1 showed a MeO group amenable of easy
labeling of the molecule with
11
C. Therefore, on the basis
of such considerations, we have selected (S)-1 as a candi-
date for in vivo PET imaging of FPRs. To the best of our
knowledge, no study aimed at the visualization of FPRs in
activated microglial cells has been reported to date.
Results and Discussion
Lipophilicity
Lipophilicity is a major factor influencing passive brain
entry and can be determined in various theoretical and
Table. Chemical structure and biological properties of (S)-1
Ca
2+
Mobilization
in transfected HL-
60 cells EC
50
[lM]
a
)
Neutrophils EC
50
[lM] Log k
0
Plasma stability t
1/2
[h] BA/AB
FPR FPRL-1 Chemotaxis Ca
2+
mobilization w/o inhibitor With inhibitor
0.26 0.19 0.030 0.086 0.31 > 4 3.5 3.2
a
) Data taken from [24].
876 Chem. Biodiversity 2016, 13, 875 – 883
www.cb.wiley.com © 2016 Wiley-VHCA AG, Z
€
urich
experimental ways [25]. For this purpose, we assessed the
retention factor of (S)-1, log k
0
as lipophilicity index,
using a reversed-phase HPLC method (Table) [26]. We
determined also the retention factor of PD-176252, a
bombesin antagonist that is structurally related to (S)-1
and that is known to be active in vivo in various behav-
ioral tests after i.p. administration in the rat and guinea
pig [27], suggesting that it is able to accumulate into the
brain. The log k
0
for (S)-1 and PD-176252 were 0.31 and
0.43, respectively, indicating that the two compounds have
similar lipophilic properties and, thus, it is likely that (S)-1
is able to enter into the brain.
Serum Stability and Permeability Evaluation
Before embarking in the radiolabeling of (S)-1, we evalu-
ated two additional features that are of importance for
brain PET radiotracers. First, we examined if (S)-1 was a
P-glycoprotein (P-gp) substrate, because PET tracers that
are not substrates of the P-gp efflux pump have higher
chances for crossing the BBB. Therefore, we evaluated
the efflux ratio between basal-to-apical (BA) and apical-
to-basal (AB) fluxes in Caco-2 cells monolayer (BA/AB)
of compound (S)-1. Generally, a cutoff value of 3 is used
to distinguish P-gp substrate from nonsubstrate [28]. The
efflux ratio was evaluated in the presence or in the
absence of a P-gp inhibitor in order to assess the magni-
tude of the interaction with P-gp interaction. (S)-1
showed an efflux ratio of 3.5 and 3.2 in the absence or in
the presence of a P-gp inhibitor, respectively (Table).
These data indicated that (S)-1 can cross biological mem-
branes and has very weak interaction with P-gp. Second,
we evaluated if (S)-1 was stable in serum for, at least, the
timeframe of a PET scan. (S)-1 showed the desired stabil-
ity with degradation half-life (t
1/2
) longer than 4 h
(Table).
Radiochemistry
As stated above, we selected (S)-1 as a candidate for the
preparation of a PET radiotracer, because it presents a
MeO group that can be easily radiolabeled with a posi-
tron emitter
11
C. We chose to prepare the desmethyl
derivative (S)-5, which bears the OH group on the pheny-
lureidic moiety of the molecule, because of its chemical
accessibility.
The desmethyl precursor for
11
C-radiolabeling (S)-5
was easily prepared as shown in Scheme 1.(S)-Boc-tryp-
tophan was activated with N,N
0
-carbonyldiimidazole and
condensed with the amine 2, prepared as previously
reported [29], to give the Boc-protected derivative (S)-3.
Subsequently, the latter compound was deprotected with
trifluoroacetic acid to give the amine (S)-4, which was
condensed with 4-aminophenol in the presence of N,N
0
-
carbonyldiimidazole to give the ureido derivative (S)-5 in
good yield.
The radiosynthesis of (S)-[
11
C]-1 was successfully per-
formed in high yield using a home-made automated syn-
thesis system [30]. (S)-[
11
C]-1 was prepared by reacting
the precursor (S)-5 with [
11
C]-MeI. [
11
C]-MeI was pre-
pared by reducing cyclotron-produced [
11
C]-CO
2
with
LiAlH
4
, followed by iodination with 57% HI. After distil-
lation and drying, [
11
C]-MeI was transferred into a DMF
solution of (S)-5 and NaOH. The [
11
C]-methylation
Scheme 1
(S)-Boc-tryptophan activated with N,N
0
-carbonyldiimidazole. b) Trifluoroacetic acid. c) 4-Aminophenol, N,N
0
-carbonyldiimidazole. d)[
11
C]MeI.
Chem. Biodiversity 2016, 13, 875 – 883 877
© 2016 Wiley-VHCA AG, Z
€
urich www.cb.wiley.com
proceeded efficiently in 5 min at 70°. Semipreparative
RP-HPLC purification of the mixture gave (S)-[
11
C]-1 in
15 3(n = 5) radiochemical yield (decay-corrected)
based on [
11
C]-CO
2
. Starting from 12.2 to 18.5 GBq of
[
11
C]-CO
2
, 0.73 – 1.41 GBq of (S)-[
11
C]-1 were produced
(31 min of synthesis time from the end of bombardment
(EOB)).
The identity of (S)-[
11
C]-1 was confirmed by coinjec-
tion with unlabeled (S)-1 on analytical RP-HPLC. In the
final product solutions, the radiochemical purity was
higher than 99% at EOS. No significant peak correspond-
ing to unreacted (S)-5 was observed in the final product.
Moreover, the radioligand did not show radiolysis at
room temperature after 180 min from formulation,
showing adequate radiochemical stability within the time-
frame of one PET scan. The analytical results were in
compliance with our in-house quality control/assurance
specifications for radiopharmaceuticals.
PET Studies
Fol low ing a heuristic approach, we decided to evaluate
the ability of (S )-[
11
C]-1 to acc umulate in mouse brain
bypassing in vitro autoradiography studies. While these
studie s can offer important in formati on on the interac-
tion bet ween the radiotracer and the target, they cannot
predict radiotracer brain uptake or other biodistribution
issues. In vivo dynamic P ET scans of normal rat brains
were conducted immediately after bolus intravenous
injection of the radioligand. PET images were generated
by averaging dynamic data at 0 – 90 min after injection
of (S)-[
11
C]-1, and indicated minimal radioactivity in the
brain (top panel in Fig. 1). Time–radioactivity curve for
the whole brain also demonstrated an initial spike cor-
responding to radioactivity in cerebral vessels, followed
by very low retention of radioactivity in the b rain (bot-
tom panel in Fig. 1).Thesedataindicatedthat
(S)-[
11
C]-1 was not able to accumulate into the brain
and, therefore, it is not suitable for imaging of FPRL-1
in living brain. The lack of brain uptake was quite
unexpected based on the permeability study in Caco-2
cell monolayer.
Subsequently, we tested the ability of (S)-[
11
C]-1 to
label FPRs in mouse brain in autoradiography studies. To
this end, we selected the APP23 mice model, a transgenic
mouse strain showing a robust formation of amyloid pla-
ques, and the PS19 mouse model that is characterized by
progressive accumulation of fibrillary tau tangles. Both
animal models are characterized by abundant accumula-
tion of activated microglia in tight association with tau
and amyloid deposits that can be visualized through an
in vitro autoradiography or an in vivo PET with suitable
radioligands [31 – 33]. The brain tissue slices from aged
PS19 and APP23 mice and from their wild-type litter-
mates were incubated for 1 h with (S)-[
11
C]-1 solution at
1n
M concentration (2 mCi/l). Total binding (TB) and
nonspecific binding (NSB) of (S)-[
11
C]-1 were assessed by
challenging the radioligand in the absence or in the pres-
ence of 10 l
M of fMLF. No specific radioligand binding
was observed in the sections from any of the mouse
strains, despite massive gliosis in PS19 and APP23 mouse
brains (Fig. 2). These data indicated that the occupancy
of FPRs by (S)-[
11
C]-1 at 1 nM was very low and was not
sufficient to autoradiographically label FPRs receptors in
the brain.
In parallel, we evaluated the ability of (S)-[
11
C]-1 to
label FPRs in peripheral inflammation. To this end, pneu-
monia was induced in rats by administration of
lipopolysaccharide (LPS). LPS is a bacterial endotoxin
that is used to stimulate the immune system through acti-
vation of macrophage-like cells in peripheral tissues.
Moreover, it has been reported that LPS is able to upreg-
ulate expression of FPRs in both peripheral tissues and
microglial cells [34]. According to established experimen-
tal procedures [35], the animals were administered intra-
tracheally with LPS (1.25 mg, 1 d before the scan) to
induce the inflammatory response. Pulmonary radioactiv-
ity retention in the LPS-treated rats was very low and
Fig. 1. (top) Sagittal PET images showing distribution of radioacticity
in the rat brain and surrounding tissues after intravenous administra-
tion of (S)-[
11
C]-1. PET data were generated by summation of dynamic
data at 0 – 90 min after radioligand injection, and were merged onto
the MRI anatomical template. (bottom) Time-radioactivity curve in
the whole brain of a rat after intravenous administration of (S)-[
11
C]-1.
878 Chem. Biodiversity 2016, 13, 875 – 883
www.cb.wiley.com © 2016 Wiley-VHCA AG, Z
€
urich
equivalent to control levels (Fig. 3). These data indicated
that (S)-[
11
C]-1 was not able to label FPRs in this animal
model. post mortem Examinations of LPS-treated animals
proved profound infiltration of neutrophils in the lungs of
LPS-treated rats. Therefore, it is possible that in vivo (S)-
[
11
C]-1 has low affinity for FPRs. However, high
Fig. 2. In vitro autoradiographic labeling of coronal mouse brain slices with (S)-[
11
C]-1. Brain samples were derived from PS19 and APP23 trans-
genics and their wild-type littermates. The images demonstrate radioligand binding without blockade (total binding, TB) and with 10 l
M fFML
(nonspecific binding, NSB).
Fig. 3. Horizontal (top) and sagittal (bottom) PET images showing distribution of radioactivity in the lungs and liver at 20 – 90 min after intra-
venous administration of (S)-[
11
C]-1 in normal control rat (right) and rat with LPS-induced lung inflammation (left).
Chem. Biodiversity 2016, 13, 875 – 883 879
© 2016 Wiley-VHCA AG, Z
€
urich www.cb.wiley.com