Evaluation of carbon-11 labeled 5-(1-methyl-1H-pyrazol-4-yl)-N-(2-methyl-5-(3-(trifluoromethyl)benzamido)phenyl)nicotinamide as PET tracer for imaging of CSF-1R expression in the brain

Article abstract of DOI:10.1016/j.bmc.2021.116245

Pharmacological targeting of tumor associated macrophages and microglia in the tumor microenvironment is a novel therapeutic strategy in the treatment of glioblastoma multiforme. As such, the colony stimulating factor-1 receptor (CSF-1R) has been identified as a druggable target. However, no validated companion diagnostic marker for these therapies exists to date. Towards development of a CSF-1R PET tracer, a set of six compounds based on recently reported CSF-1R inhibitor 5-(1-methyl-1H-pyrazol-4-yl)-N-(2-methyl-5-(3-(trifluoromethyl)benzamido)phenyl)nicotinamide (Compound 5) was designed, synthesized and evaluated in vitro for potency and selectivity. The highest affinity for CSF-1R was found for compound 5 (IC₅₀: 2.7 nM). Subsequent radiosynthesis of [¹¹C]5 was achieved in 2.0 ± 0.2% yield (decay corrected to start of synthesis) by carbon-11 carbon monoxide aminocarbonylation in 40 min after end of bombardment. In vitro autoradiography with [¹¹C]5 on rat brain sections demonstrated high specific binding, but also strong off-target binding. Ex vivo, only intact tracer was observed in blood plasma at 90 min post injection in healthy rats. PET scanning results demonstrated negligible brain uptake under baseline conditions and this brain uptake did not increase by blocking of efflux transporters using Tariquidar. To conclude, [¹¹C]5 was successfully synthesized and evaluated in healthy rats. However, the inability of [¹¹C]5 to cross the blood-brain-barrier excludes its use for imaging of CSF-1R expression in the brain.

Full text of DOI:10.1016/j.bmc.2021.116245

Bioorg. Med. Chem. 42 (2021) 116245
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Evaluation of carbon-11 labeled
5-(1-methyl-1H-pyrazol-4-yl)-N-(2-methyl-5-(3-(trifluoromethyl)
benzamido)phenyl)nicotinamide as PET tracer for imaging of CSF-1R
expression in the brain
,
Berend van der Wildta b, Madina Nezamb, Esther J.M. Kooijmanb, Samantha T. Reyesa,
,
*
Bin Shena, Albert D. Windhorstb, Frederick T. China
a Stanford University, School of Medicine, Department of Radiology, Molecular Imaging Program at Stanford (MIPS), 1201 Welch Road, PS049, Stanford, CA 94305-
5484, USA
b Amsterdam UMC, Vrije Universiteit Amsterdam, Radiology & Nuclear Medicine, de Boelelaan 1117, Amsterdam, Netherlands
A R T I C L E I N F O
A B S T R A C T
Keywords:
Pharmacological targeting of tumor associated macrophages and microglia in the tumor microenvironment is a
novel therapeutic strategy in the treatment of glioblastoma multiforme. As such, the colony stimulating factor-1
receptor (CSF-1R) has been identified as a druggable target. However, no validated companion diagnostic marker
for these therapies exists to date. Towards development of a CSF-1R PET tracer, a set of six compounds based on
recently reported CSF-1R inhibitor 5-(1-methyl-1H-pyrazol-4-yl)-N-(2-methyl-5-(3-(trifluoromethyl)benzamido)
phenyl)nicotinamide (Compound 5) was designed, synthesized and evaluated in vitro for potency and selectivity.
The highest affinity for CSF-1R was found for compound 5 (IC50: 2.7 nM). Subsequent radiosynthesis of [11C]5
was achieved in 2.0 ± 0.2% yield (decay corrected to start of synthesis) by carbon-11 carbon monoxide ami-
nocarbonylation in 40 min after end of bombardment. In vitro autoradiography with [11C]5 on rat brain sections
demonstrated high specific binding, but also strong off-target binding. Ex vivo, only intact tracer was observed in
blood plasma at 90 min post injection in healthy rats. PET scanning results demonstrated negligible brain uptake
under baseline conditions and this brain uptake did not increase by blocking of efflux transporters using Tar-
iquidar. To conclude, [11C]5 was successfully synthesized and evaluated in healthy rats. However, the inability of
Colony stimulating factor 1 receptor
PET imaging
Carbon-11
[
11C]5 to cross the blood-brain-barrier excludes its use for imaging of CSF-1R expression in the brain.
1. Introduction
disease models of neuroinflammation and –degeneration and are
currently under further investigation in clinical studies.7
Colony stimulating factor 1 Receptor (CSF-1R or CD115) is a tyrosine
kinase that is selectively expressed on microglia in the brain.1 CSF-1R,
which is activated by the endogenous agonists colony stimulating fac-
tor and interleukin-34, regulates the survival, proliferation and differ-
entiation of microglia.2 Outside of the central nervous system CSF-1R is
mainly expressed on macrophages. Both microglia and macrophages
play a pivotal role in brain diseases involving immune responses such as
neuroinflammation, neurodegeneration and glioblastoma.3 Towards
potential treatment these cell types have emerged as targets for therapy.
In fact, pharmacological inhibition of CSF-1R, which results depletion of
microglia from the brain,4–6 has shown beneficial effects in various
Positron emission tomography (PET) is an imaging technique that
allows for the quantitative imaging of receptor expression in living
subjects in a non-invasive manner with good spatiotemporal resolution
and high sensitivity. When an appropriate PET tracer is available, PET
can be used for selecting patients that would benefit from treatment as
well as monitoring disease progression during treatment. As the accu-
mulation of microglia and macrophages at the site of disease in for
example glioblastoma will result in a local increase of CSF-1R density, a
validated PET tracer for imaging of CSF-1R expression would be of great
value in understanding and monitoring CSF-1R targeting therapies.
Unfortunately, previously reported CSF-1R PET tracers (Fig. 1) have
* Corresponding author.
E-mail address: chinf@stanford.edu (F.T. Chin).
https://doi.org/10.1016/j.bmc.2021.116245
Received 25 March 2021; Received in revised form 25 May 2021; Accepted 26 May 2021
Available online 30 May 2021
0968-0896/© 2021 Published by Elsevier Ltd.

B. van der Wildt et al.
Bioorganic & Medicinal Chemistry 42 (2021) 116245
serious shortcomings. [18F]10 binds to CSF-1R with a low affinity (IC50
of 170 nM) and is non-selective with similar affinity for the kinases TrkB
and TrkC.8 Commonly, much lower affinities for receptor binding PET
tracers are preferred, in the low to sub-nanomolar range. Another
attempt for developing a CSF-1R PET tracer has led to [11C]AZD6495,
which in healthy rats and monkeys failed to enter the brain.9 [11C]CPPC
in turn showed good brain uptake in various rodent models of brain
disease and non-human primates, however, this tracer suffered from
high non-specific binding.10 Additionally, although no selectivity data
on CPPC has been reported, selectivity over other kinases might be
minimal as close analogues show high affinity (IC50 < 20 nM) for other
possible. Taken together, compound 5 shows promising characteristics
for successful translation into a PET tracer. Therefore, the aim of this
study is to translate compound 5, or a close derivative, into a PET tracer
for in vivo imaging of CSF-1R expression levels in the brain.
2. Results
2.1. Synthesis and in vitro evaluation
Five analogues of 5 were designed with the goal of identifying a
compound that has a high affinity for CSF-1R, a high selectivity towards
other kinases and allows for convenient labeling strategies with either
carbon-11 or the longer lived (t1/2 of 110 min vs 20 min for carbon-11)
and thus clinically more relevant isotope fluorine-18. Initially, the po-
sition of the nitrogen atom in the pyridine ring was altered from meta- to
para- relative to the carbonyl group, resulting in a structure that is often
found as ATP binding pocket motif in kinase inhibitors (e.g. BLZ945
Fig. 1, but also sorafenib, regorafenib). Towards obtaining CSF-1R in-
hibitors that allow for fluorine-18 labelling, the methylpyrazine group
on lead compound 5 was selected for substitution with a fluo-
roethylpyrazine or a fluoroethyltriazole group. The methylpyrazine
group was chosen for modification as this moiety appears to allow for
modification without negatively effecting the affinity.14 Also, a meth-
ylamide group was introduced as a substitute for methylpyrazine. A
methylamide substitute as in BLZ945 was selected as this would lower
molecular weight as well as polar surface area and thus would increase
the blood brain barrier penetration.
kinases such as Flt3, Kit, Axl and TrkA.11–12
[
11C]BLZ945, a highly se-
lective CSF-1R inhibitor, was shown to be a substrate for efflux trans-
porters at the blood brain barrier and also suffered from high non-
specific binding in the brain.13 As a result, there still is a clear need
for a CSF-1R PET tracer that can overcome the aforementioned
limitations.
Recently, a new inhibitor of CSF-1R was reported (5-(1-methyl-1H-
pyrazol-4-yl)-N-(2-methyl-5-(3-(trifluoromethyl)benzamido)phenyl)
nicotinamide, compound 5 in Fig. 1).14 This compound is a potent CSF-
1R inhibitor with an IC50 of 0.5 nM and a 120-fold selectivity over c-KIT,
a closely related kinase family member.14 Compound 5 has been clas-
sified as a type II tyrosine kinase inhibitor, i.e. its binding extends from
the ATP binding pocket into an adjacent allosteric binding site. As a
result, binding was shown to be largely independent on ATP concen-
tration when compared to another CSF-1R inhibitor BLZ945.14 Given
that ATP concentrations in the intracellular environment are high (in the
millimolar range) and PET tracer concentrations are extremely low, this
non-competitive nature of binding might prove advantageous for in vivo
PET imaging. Another positive characteristic of this compound are the
high metabolic stability in liver microsomes. The PET camera is not able
to measure the chemical context of the tracer, but rather detects the
annihilation photons that originate from positron emission of the
carbon-11 atom, metabolic stability of the PET tracer is important to
ensure that the measured signal in the tissue of interest originates from
the injected compound. Finally, good cell membrane permeability and a
low in vitro efflux ratio in a Caco-2 permeability assay (efflux ratio of 2.6
determined as apical to basolateral / basolateral to apical) was consid-
ered an important feature. Since the goal of this study was to image CSF-
1R expression in the brain, active transport of the PET tracer out of the
brain by efflux transporters in the blood brain barrier should be as low as
The chemistry towards the target molecules is depicted in Scheme 1.
First, methyl nitroaniline was reacted with trifluoromethylbenzoyl
´
chloride to obtain nitrophenyl derivative 1. Bechamp reduction resulted
in aniline derivative 2, which was esterified with either bromonicotinoyl
or bromoisonicotinoyl chloride to afford aryl bromides 3 and 4. These
arylbromides were subjected to a Suzuki cross-coupling reaction with
methylpyrazole boronic acid pinacol ester to obtain compounds 5 and
6.15 When evaluating the CSF-1R inhibitory potential of these com-
pounds it became clear that the nitrogen atom is preferably located at
the meta position (Table 1, entry 3, compound 5) as placing the nitrogen
at the para position results in a decrease in affinity by a factor of 10
(Table 1, entry 2, compound 4). Importantly, the IC50 value of 2.7 ± 0.4
nM for compound 5 corresponds fairly well with the previously reported
value of 0.5 nM.14 The small difference appears to be inherent to the
Fig. 1. CSF-1R PET tracers previously developed8–10,13 and [11C]5 (this work). The position of the carbon-11 labels are depicted with *.
2

B. van der Wildt et al.
Bioorganic & Medicinal Chemistry 42 (2021) 116245
Scheme 1. Synthesis of compound 5 and its closely related analogue 6. Reagents and conditions: i) 3-(trifluoromethyl)benzoyl chloride, TEA, DCM, 3 h, rt, 97%; ii)
HCl, Fe, H2O/EtOH, 4 h, reflux, 94%; iii) First, appropriate bromopyridine carboxylic acid, oxalyl chloride, THF/DCM/DMF, 2 h, rt, then compound 2, TEA, DCM, 16
h, rt, 79–88%; iv) 1-methylpyrazole-4-boronic acid pinacol ester, Pd(PPh3)4, CsF, MeOH/DMF, 16 h, 70 ◦C, 68–88%.
analogues that allow for fluorine-18 labeling have been identified,
Table 1
although no compounds with higher potency than compound 5 (IC50
2.7 ± 0.4 nM) were discovered.
:
IC50 values for tested compound for CSF-1R, PDGFR-β and c-KIT.
IC50
(nM)
CSF-
1R
Ratio
Next, selectivity towards c-Kit and PDGFR-β, kinases with strong
structural homology to CSF-1R, was evaluated. To our surprise, 5
showed only a slight selectivity towards CSF-1R by a factor of 2 over
both c-KIT and PDGFR-β, unlike the 120-fold selectivity over c-Kit that
was previously reported.14 Fluoroethyl analogue 8 showed a lower
selectivity, with IC50 values for PDGFR-β and c-KIT of approximately 4
nM, lower than observed for CSF-1R. Similar results were obtained for
the triazole analogue 12. A good selectivity for CSF-1R over PDGFR-β
was found for compound 9 (10-fold), however, not over c-KIT (0.5-fold).
Overall, the more bulky fluoroethyl group at the pyrazole ring resulted
in lower affinities and the optimal position for the nitrogen atom in the
pyridine ring is on the original meta-position. Unfortunately, no selec-
tivity for CSF-1R was observed for any of the compounds.
Entry
Compound
PDGFR-
c-KIT
CSF-1R/
CSF-
1R/c-
KIT
β
PDGFR- β
1
2
3
4
5
6
7
BLZ945
6.9 ±
0.5
>1000
n.d.
>1000
4
30.6 ±
7.2
n.d.
5
2.7 ±
0.4
6.3 ± 1.2
2.5 ± 0.9
4.5 ±
1.0
2.31
0.06
1.55
1.28
1.64
0.09
0.93
0.04
8
40.0 ±
3.1
3.8 ±
2.3
9
29.5 ±
3.6
45.6 ±
2.3
27.4 ±
3.7
12
20
21.7 ±
1.5
27.8 ±
5.0
1.0 ±
3.5
In previous medicinal chemistry efforts towards compound 5, a
methylpiperazine group on the trifluoromethylbenzene ring had been
removed from the structure due to metabolic instability in mouse liver
microsomes, however, at the cost of a 400-fold decrease in affinity for
CSF-1R and a 2000-fold decrease in selectivity over c-KIT.14 Reintro-
duction of this functionality, which is quite a general motif in approved
drugs (like imatinib, masitinib among other kinase inhibitors), could
result in an increase in affinity as well as in selectivity. The convergent
synthesis towards compound 20 is depicted in Scheme 3.
77.2 ±
5.9
n.d.
n.d.
Values are expressed as the average ± standard deviation of at least 3 experi-
ments.
n.d.: not determined.
used assay as BLZ945 gave an IC50 value of 6.9 nM in our assay
compared to a reported value of 1.0 nM.4
Methyltrifluorobenzoic acid was methylated and subsequently
brominated with NBS in a radical reaction to afford bromomethyl ben-
zoate 14. A nucleophilic substitution using methylpiperazine followed
by saponification resulted in carboxylic acid 16. EDC mediated
condensation of nitromethylaniline and 5-bromonicotinic acid afforded
aryl bromide 17, which was subject to a Suzuki reaction with methyl-
Subsequently, modifications were made to the methylpyrazole group
(Scheme 2). A Suzuki reaction15 with between fluoroethylpyrazole
boronic acid pinacol ester 7 and compound 3 gave fluoroethyl analogue
8. Methylamide analogue 9 was obtained by a 2-step palladium cata-
lyzed carbonylation reaction using trichlorophenyl formate.16 A Sono-
gashira reaction17 between arylbromide 3 and trimethylsilylacetylene
resulted in TMS- protected alkyne 10, which after deprotection, was
reacted with fluoroethylazide in a copper catalyzed 1,3-dipolar cyclo-
addition (‘Click’ reaction18) to obtain triazole 12.
´
pyrazoleboronic pinacol ester to afford methylpyrazole 18. Bechamps
reduction and subsequent condensation with 16 resulted in 20 as a
potential CSF-1R inhibitor with high affinity and selectivity. Subsequent
testing for potency was, however, disappointing with an IC50 value for
CSF-1R of only 77 nM. Therefore, no further testing of this compound for
selectivity was performed.
Substitution of methyl for fluoroethyl on the pyrazole group (com-
pound 8) was not well tolerated and resulted in an IC50 of 40 nM.
Methylamide instead of methylpyrazole (compound 9), a similar struc-
tural motif as used in the well-known CSF-1R inhibitor BLZ945, still
retained a fair affinity but did not result in an improvement (IC50: 30
nM). Fluoroethyltriazole group as substitute for methylpyrazole (com-
pound 12) resulted in a modest decrease with an IC50 of 22 nM. Overall,
2.2. Radiochemistry
The most promising compound in this series, compound 5, was
selected for radiolabeling and further evaluation in vivo. A palladium
3

B. van der Wildt et al.
Bioorganic & Medicinal Chemistry 42 (2021) 116245
Scheme 2. Synthesis of analogues of compound 5. Reagents and conditions: i) Compound 3 and 7, Pd(PPh3)4, CsF, MeOH/THF, 16 h, reflux, 73%; ii) First, compound
4, 2,4,6-trichlorophenyl formate, Pd(OAc)2, xantphos, NaOAc, dioxane, 80 ◦C, 16 h. Then, methylamine⋅HCl, TEA, DMF, 3 h, rt, 35%; iii) trimethylsilylacetylene,
PdCl2(PPh3)2, CuI, TEA, DMF, rt, 16 h, 71%; iv) CsF, THF, rt, 16 h, 99%; v) 2-azido-1-fluoroethane, CuSO4⋅5H2O, Na-ascorbate, DMF/H2O, 80 ◦C microwave
irradiation, 10 min, 55%.
Scheme 3. Synthesis of methyl piperazine analogue 20. Reagents and conditions: i) CH3I, K2CO3, DMF, 16 h, rt, 87%; ii) N-bromosuccinimide, benzoyl peroxide,
MeCN, 16 h, reflux, 74%; iii) 1-methylpiperazine, K2CO3, MeCN, 16 h, reflux, 63%; iv) NaOH, H2O/EtOH, 6 h, rt, 46%; v) 5-bromonicotinic acid, EDC, DMAP, DMF,
16 h, rt, 59%; vi) 1-methylpyrazole-4-boronic acid pinacol ester, Pd(PPh3)4, CsF, DMF, 16 h, 100 ◦C, 59%; vii) NH4Cl, Fe, H2O/EtOH, 2 h, 70 ◦C, 90%; viii) EDC,
DMAP, DMF, 16 h, rt, 71%.
mediated carbon-11 carbon monoxide aminocarbonylation reaction was
chosen as depicted in Table 2. This reaction is supposed to result in [11C]
5, an isotopologue of 5 that will behave similarly with the exception of
positron emission of the carbon-11 atom. Conveniently, precursor 19
was already available as it was used as intermediate in the synthesis of
compound 20.
conditions no product was observed, in parallel with previous reports of
aminocarbonylations on aniline derivatives.22,23 Also, using N,N-diiso-
propylethylamine (DIPEA) as a weak base did not result in any [11C]5
(entry 2. Li-HMDS, a strong base, did result in the conversion of [11C]CO
to non-volatile compounds (entry 3, 98% trapping efficiency), however
again no product was obtained. Success was achieved using 1,8-diazabi-
cyclo(5.4.0)undec-7-ene (DBU), which resulted in a good radiochemical
yield of 31 ± 8% (Entry 4). Variation in the palladium source (entry 5, 6)
resulted in lower yields. Finally, different amounts of amine precursor
19 (entry 7–10) did not result in increased radiochemical yields when
compared to the conditions depicted in entry 4.
To efficiently screen various reaction conditions, we used a low
pressure semi-automated [11C]CO dispensing system using xenon as a
carrier gas.19,20 Xantphos as preferred ligand is carbonylation re-
actions21 was used throughout the screening. Initially, the reaction was
executed in the absence of a base (Table 2, entry 1). Using these
4

B. van der Wildt et al.
Bioorganic & Medicinal Chemistry 42 (2021) 116245
Table 2
Optimization of radiolabeling conditions towards [11C]5.
Entry
Amine (µmol)
Base
Catalyst
Trapping efficiency (%)
Radiochemical purity (%)
Radiochemical yield (%)
n
1
10
10
10
10
10
10
0
–
PdCl2(PPh3)2
PdCl2(PPh3)2
PdCl2(PPh3)2
PdCl2(PPh3)2
Pd2dba3
37
0
0
1
1
1
5
3
3
3
3
3
3
2
DiPEA
Li-HMDS
DBU
DBU
DBU
DBU
DBU
DBU
DBU
15
0
0
3
98
0
0
4
85 ± 14
84 ± 0
87 ± 6
82 ± 9
73 ± 2
77 ± 5
78 ± 5
37 ± 8
5 ± 1
21 ± 1
0 ± 0
24 ± 4
29 ± 7
17 ± 0
31 ± 8
4 ± 0
18 ± 2
0 ± 0
17 ± 3
22 ± 4
14 ± 1
5
6
Pd2[π-cinnamyl]Cl2
7
PdCl2(PPh3)2
PdCl2(PPh3)2
PdCl2(PPh3)2
PdCl2(PPh3)2
8
2.5
5
9
10
20
General method: Reactions were performed with trifluoro-iodobenzene (2.7 mg, 10 µmol), base (20 µmol), xantphos (2.9 mg, 5 µmol) and palladium (2 µmol) in THF
(0.7 mL). Reactions mixtures were heated at 100 ◦C for 5 min. The results are expressed as average trapping efficiency (percentage of non-volatile radioactive products
after reaction), radiochemical purity (percentage of radioactive product area under the curve (AUC) relative to the total AUC), and radiochemical yield (the product of
trapping efficiency and radiochemical purity) (number of experiments ‘n’ is depicted in the last column).
The conditions depicted in entry 4 were considered optimal and were
thus used for full scale synthesis of [11C]5. After preparative HPLC and
solid phase extraction reformulation, [11C]5 was obtained in 2.0 ± 0.2%
decay corrected yield based on the starting amount of [11C]CO2 (109 ±
25 MBq in 2 mL of 10% ethanol in saline, n = 3) in 40 min using a semi-
automated synthesis procedure. The product identity was confirmed by
co-injection of an authentic reference standard of compound 5. No
chemical contaminations were observed and radiochemical purity was
> 99% (see supporting information Fig. 4 for chromatograms). Molar
activity was determined as 181 ± 22 GBq⋅µmolꢀ 1. The formulated
product retained a radiochemical purity of > 95% up to 1 h after end of
synthesis.
2.3. In vitro, ex vivo and in vivo evaluation
The partition coefficient logD of the compound was determined by
shake-flask method to be 2.54 ± 0.02, well within the range for brain
penetration by passive diffusion.
In vitro binding experiments on healthy rat brain sections using
autoradiography were performed. Due to the widespread presence of
microglia in the brain, which express CSF-1R, a homogeneous expres-
sion pattern would be expected. Indeed, [11C]5 shows high binding
throughout the rat brain (Fig. 2A). However, binding was not as ho-
mogeneous as expected, with hippocampus and cortex showing strong
binding at baseline conditions. The total binding could be largely
blocked by co-incubation with excess unlabeled 5 (Fig. 2). Despite this
high specific binding, selective binding to CSF-1R seemed marginal, as
co-incubation with excess BLZ945, a highly selective CSF-1R inhibitor,
only decreased the binding of [11C]5 by approximately 20%. This
Fig. 2. In vitro autoradiography on rat brain using [11C]5. Representative im-
incomplete blocking of [11C]5 binding could very well be attributed to
ages of brain sections are shown at baseline, self-blocking conditions, BLZ945
the low selectivity of compound 5 when compared to other kinases such
and CPPC blocking. CSF-1R inhibitors were used at 10 µM concentration. The
as c-KIT and PDGFR-β. Blocking with the broad spectrum kinase inhib-
itor CPPC [11, 12] resulted in a more profound reduction in [11C]5
binding of approximately 50%.
results are expressed as average normalized binding ± standard deviation (n
= 3).
PET imaging in healthy rats showed poor brain uptake of [11C]5
throughout the entire scan (Fig. 3A and B). After an initial peak directly
after tracer injection, which is the result of perfusion, activity concen-
trations remain extremely low (Fig. 3B). Blocking of efflux transporters
did not significantly increase the brain concentrations of [11C]5, which
excludes that strong efflux is causing the low brain uptake levels.
Metabolism of [11C]5 into polar, non-brain permeable radioactive
compounds was not observed either and thus can be excluded as an
explanation for low brain uptake. In fact, only parent radiotracer was
detected in the blood plasma at 90 min post injection (Supporting in-
formation Fig. 5). Corresponding to the lack of brain uptake, no
5

B. van der Wildt et al.
Bioorganic & Medicinal Chemistry 42 (2021) 116245
Fig. 3. In vivo PET imaging in healthy rats. 3A: representative summed PET images of rats (cross-section, axial view) injected with [11C]5 at baseline (Left) or efflux
transporter blocking conditions (Right, tariquidar, 15 mg⋅kgꢀ 1, i.v. administration 30 min prior to tracer injection). The location of the brain is indicated with dotted
oval shapes in gray. Other colors depict the radioactivity concentrations in %ID/g according to the scale bar. 3B: Whole brain time-activity curves following injection
of [11C]5 under baseline of efflux transporter blocking conditions. Results are expressed as average ± standard deviation (n = 2 per group).
radioactive compounds could be extracted from the brain for radio-
metabolite analysis.
which after conversion to the corresponding acyl chloride can react with
the amine precursor 19 to form [11C]5.30 Clearly, this multistep pro-
cedure with several purification steps is not preferred when working
with the short-lived isotope carbon-11. Exploring the proposed [11C]CO
carbonylation chemistry, it was found that in the absence of a base or
using a mild base (DiPEA) no product was formed and mainly unreacted
3. Discussion
Compound 5 was selected for this study due to its very high affinity
for CSF-1R and its reported 120-fold selectivity over c-KIT, a close CSF-
1R homologue.14 In addition, binding of this compound to CSF-1R was,
in contrast to many kinase inhibitors, non-competitive with ATP. This
could theoretically result in improved imaging characteristics as binding
in vivo is not hampered by ATP, which is present at high concentrations
in the intracellular environment. In our hands, however, the selectivity
of compound 5 for CSF-1R was very minimal and in fact the compound
was found to be equipotent for both other kinases evaluated, c-KIT and
PDGFR-β. The discrepancy between the reported and determined values
is unclear. However, this non-selective behavior might not come as a
complete surprise though, as the structure of compound 5 shows
structural overlap with the tyrosine kinase inhibitor imatinib and
masitinib, which both target c-KIT, PDGFR-β and, to a lesser extent, CSF-
1R. It must be noted that both c-KIT and PDGFR-β have been identified
as drug targets in cancer treatment and are pursued for PET and SPECT
tracer development as well.24–27 Therefore, the non-selective PET tracer
[
11C]CO was found, in correspondence with earlier reports on amino-
carbonylations on aniline derivatives.22,23 Using Li-HMDS resulted in
the formation of many side products. The formation of [11C]5 was only
observed when DBU was used as a base. Together with PdCl2(PPh3)2 as a
preferred catalyst, these conditions were reported to provide optimal
results in [11C]CO carbonylation reactions before.31 The overall isolated
yield was fairly low (2.0 ± 0.2% d.c. based on [11C]CO2,) compared to
the results obtained during test screening reactions (31 ± 8% based on
[
11C]CO). This difference can be partially explained by the fact that the
isolated yield is based on [11C]CO2, whereas during the optimization
reactions the yield was calculated based on [11C]CO. Molybdenum,
although reliable and long-lasting as catalyst, does not convert CO2 to
CO as efficiently as zinc does.32 Additionally, a fairly large loss of [11C]
CO was observed at the full scale experiments when retracting the [11C]
CO-inlet needle from the reaction vial after [11C]CO transfer. This effect
was not obviously seen or perhaps overlooked when dispensing [11C]CO
over several vials manually. Finally, at large scale radiosynthesis the
trapping efficiencies dropped significantly. This is potentially due to the
larger amount of xenon carrier gas in a single vial, although no further
experiments were conducted to confirm this hypothesis. Overall, [11C]5
could be obtained in excellent chemical and radiochemical purity in
sufficient amounts and radioactivity concentrations for our studies
(supporting information Fig. 4). For future studies, or studies performed
in PET centers that do not readily have access to [11C]CO, it could be
interesting to explore labeling of compound 5 with the longer lived
isotope fluorine-18 by means of fluorine-18 trifluoromethylation.33
In vitro autoradiography showed very high specific binding on rat
brain sections, as demonstrated by the 95% reduction in [11C]5 binding
when co-incubating with excess unlabeled 5. Selective binding to CSF-
1R, however, appears limited as evidenced by a 20% reduction in
binding when co-incubating with unlabeled BLZ945, a highly selective
CSF-1R inhibitor.4 The residual tracer binding might be attributed to
binding to other targets such as c-KIT and PDGFR-β, both not blocked by
BLZ945. Blocking with CPPC resulted in a stronger reduction of [11C]5
binding of about 50%. CPPC has previously been claimed to be a se-
lective CSF-1R inhibitor.8 Although no selectivity data of CPPC is
available in literature, structurally close analogs are broad spectrum
kinase inhibitors, displaying IC50 values below 0.1 µM for FLT3, c-KIT,
AXL, TRKA, and LCK.11,12 However, even CPPC did not provide the same
[
11C]5 might very well be useful as predictive tool for identification of
cancer patients who respond to broad spectrum kinase inhibitor
therapy.28
In our efforts to finding compounds with increased affinity for CSF-
1R or improved selectivity over other kinases no better compounds
than lead compound 5 were identified. However, a moderately c-KIT
selective inhibitor was identified (compound 12) that allows for
fluorine-18 labeling and could potentially be used for peripheral c-KIT
imaging.
Three obvious positions for carbon-11 labeling can be identified in
the structure of compound 5. Methylation using [11C]methyl iodide or
[
11C]methyl triflate at the pyrazole group could be performed in the
presence of a strong base.29 However, such strong basic conditions could
easily result in alkylation of either of the two amide groups as well. A
more selective reaction would be palladium mediated [11C]CO amino-
carbonylation between an aniline derivative and aryl halide. Both am-
ides are theoretically accessible for labeling, however the current
approach was chosen due to the commercial availability of 1-iodo-3-tri-
fluoromethylbenzene. The carbon-11 label might also be inserted at this
position by employing carbon-11 trifluoromethylbenzoyl chloride as
radioactive precursor. This radiochemistry would proceed in a multistep
fashion where at first cyclotron produced [11C]CO2 is reacted with a
Grignard reagent to obtain carbon-11 trifluoromethylbenzoic acid,
6

B. van der Wildt et al.
Bioorganic & Medicinal Chemistry 42 (2021) 116245
blocking effect as when coincubating with unlabeled compound 5 (self-
blocking). Therefore, it is likely that [11C]5 binds to more kinases or
other biological targets than tested in this work. To elucidate the
selectivity profile of compound 5, a full spectrum kinase screening
should be performed.
Platinum column (5
μ
m, 250 mm × 4.6 mm using H2O/MeCN/TFA
(60:40:0.1; v/v/v) as a mobile phase at a flow rate of 1 mL⋅minꢀ 1, with a
Jasco UV-2075 UV detector (λ = 254 nm) and a NaI radioactivity de-
tector (Raytest, Straubenhardt, Germany). Chromatograms were ac-
quired with Raytest GINA Star software (version 5.01). Preparative
HPLC for isolation of the radiotracer was performed on a Jasco PU-2089
PET imaging results unfortunately showed no brain uptake for [11C]
5 at any given time-point. Blocking of efflux transporters using tar-
iquidar had no effect on brain uptake, despite blocking at a dose of 15
pump station equipped with a Phenomenex Luna C18 column (10 μm,
250 mm × 10 mm) using H2O/MeCN/TFA (45:55:01; v/v) as eluent at a
flow rate of 5 mL⋅minꢀ 1, a Jasco UV-1575 Plus UV detector (λ = 254
nm), a custom made radioactivity detector and Jasco ChromNAV CFR
software (version 1.14.01) for data acquisition. Wistar Rats (RccHan
Wist, male, 8–10 weeks, approximately 250 g) were obtained from
Envigo and were housed under standard conditions (24 ◦C, 60% relative
humidity, 12 h light/dark cycles) and provided with water and food ad
libitum. Tariquidar was freshly formulated as 3.75 mg/mL tariquidar in
2.5% (g/v) aqueous dextrose solution prior to iv administration. Radi-
oTLC plates were exposed to a phosphorimager screen ((GE Healthcare)
for 30 min in a cassette (Hypercasette, Amersham Biosciences, Little
Chalfont, United Kingdom) and subsequently developed using a
Typhoon Trio Imager (GE Healthcare). Images were analyzed using
ImageQuant TL (GE Healthcare). Dynamic PET imaging was performed
using dedicated small animal NanoPET/CT and NanoPET/MR scanners
(Mediso Ltd., Hungary, Budapest) with identical PET components. PET
scans were acquired in list mode and rebinned into the following frame
sequence: 4 × 5, 4 × 10, 2 × 30, 3 × 60, 2 × 300, 1 × 600, 1 × 900 and 1
× 1200 s. Reconstruction was performed with a fully 3-dimensional (3D)
reconstruction algorithm using four iterations and six subsets, resulting
in an isotropic 0.4-mm voxel dimension. Images were analyzed using
VivoQuant™ (Invicro, Boston, MA, USA) by drawing a region of interest
over the full brain.
mg/kg that blocks the main efflux transporters P-GP and BCRP.34,35
A
blood plasma metabolite analysis excluded the possibility that rapid
metabolism of [11C]5 into more polar and thus brain-impermeable
radiometabolites was the cause of this low brain uptake. In fact, only
intact tracer was observed in blood plasma after 90 min post injection
(supporting information Fig. 5). The reason for the low brain uptake
must therefore be attributed to the physiochemical properties of com-
pound 5. A good measure for scoring compounds for successful PET
imaging is PET-CNS-MPO,36 a CNS PET tracer scoring tool derived from
CNS-MPO.37 Although the CNS-MPO score of compound 5 of 3.3 is close
to the desired value of > 4, the score of 1.3 in the CNS-MPO-PET method
is far below the desired score of > 3 (supporting information Table 1).
Overall, the current work has identified compound 5 as a broad
spectrum inhibitor, despite earlier reports on selectivity. A radiolabeling
procedure was developed to obtain [11C]5 in sufficient radiochemical
yields for further studies. In vitro binding of [11C]5 appears highly se-
lective, although at this point the selectivity profile is unknown and this
should likely be elucidated prior to performing additional imaging
studies, e.g. peripheral tumor imaging studies. The clear lack of brain
uptake excludes this radiotracer from imaging of CSF-1R or other ki-
nases in the brain.
4. Conclusion
5.1.1. (N-(4-methyl-3-nitrophenyl)-3-(trifluoromethyl)benzamide)
11C]5 was successfully synthesized and evaluated in healthy rats.
To a solution of 4-methyl-3-nitroaniline (5.04 g, 33.1 mmol) and TEA
(5.00 g, 49.5 mmol) in DCM (30 mL) was added 3-(trifluoromethyl)
benzoyl chloride (6.90 g, 33.1 mmol) dropwise. After stirring for 3 h at
rt, the solution was diluted with 1 M HCl (30 mL). The solids were
filtered, washed with water and DCM and dried to obtain the title
compound as a white solid (10.4 g, 97%). 1H NMR (400 MHz, DMSO‑d6):
δ = 10.77 (s, 1H), 8.52 (d, 1H, J = 2.3 Hz, 8.32 (s, 1H), 8.27 (d, 1H, J =
7.9 Hz), 7.99 (m, 2H), 7.80 (t, 1H, J = 7.8 Hz), 7.50 (d, 1H, J = 8.9 Hz),
2.50 (s, 3H, overlaps with DMSO signal); 13C NMR (101 MHz,
DMSO‑d6): δ = 164.3, 148.5, 137.7, 135.1, 133.1, 131.9, 129.8, 129.2
(d, J = 32.1 Hz), 128.5 (d, J = 3.5 Hz), 128.0, 125.0, 124.3 (d, J = 3.5
Hz), 123.4 (d, J = 272.6 Hz), 115.7, 19.3; ESI-HRMS: m/z calculated for
[
The inability of [11C]5 to cross the blood–brain-barrier in rats, however,
excludes its use for imaging of CSF-1R expression in the brain. Com-
pound [11C]5 could be useful for in vivo imaging of kinase expression in
tumors in the periphery, although at first a thorough kinase selectivity
study seems warranted. Future work towards identifying a PET tracer for
imaging of CSF-1R expression in the brain will be focusing on struc-
turally distinct CSF-1R inhibitors.
5. Materials and methods
5.1. General
C
15H11F3N2NaO3: 347.0619; found: 347.0615 [M + Na]+.
All chemical reagents and solvents were obtained from commercial
suppliers and used as received. Microwave reactions were performed on
a CEM Discover Legacy (Matthews, NC, USA). Reaction monitoring was
performed by thin layer chromatography on pre-coated silica 60 F254
aluminium plates (Merck, Darmstadt, Germany). Spots were visualized
with UV light (254 nm), KMnO4 or ninhydrin staining. NMR spectros-
copy was performed using an Agilent 400 MR (Agilent Technologies,
Santa Clara, CA, USA) with chemical shifts (δ) reported in parts per
million (ppm) relative to the solvent (CDCl3 1H: 7.26 ppm, 13C: 77.16
ppm; DMSO‑d6 1H: 2.50 ppm, 13C 39.52 ppm; CD3OD 1H: 3.31 ppm, 13C:
49.00 ppm). High resolution mass spectrometry was performed on a
Bruker Daltonics - apex-Qe (Bruker, Billerica, MA, USA) in either posi-
tive or negative ion mode. Enzyme inhibition assay kits (Z-LYTETM
assay) and human recombinant enzymes were from ThermoFisher Sci-
entific (Waltham, MA, USA). Fluorescent readout of 384-well plates was
performed on a fluorimeter (Safire, Tecan, XFluo). Carbon-11 was pre-
5.1.2. (N-(3-amino-4-methylphenyl)-3-(trifluoromethyl)benzamide)
To a solution of compound 1 (6.00 g, 18.5 mmol) in ethanol/H2O
(4:1, 30 mL) was added 1 mL conc. HCl and iron powder (5.16 g, 92.5
mmol). The mixture was refluxed for 4 h followed by filtration over
Celite. After concentration, the mixture was partitioned between DCM
(50 mL) and sat. NaHCO3 (50 mL). The organic layer was washed with
NaHCO3 (2 × 50 mL) and brine (50 mL). After drying on Na2SO4,
filtration and concentration, the product was obtained as a white solid
(5.30 g, 97%).1H NMR (400 MHz, DMSO‑d6): δ = 10.14 (s, 1H), 8.24 (m,
2H), 7.93 (d, 1H, J = 7.8 Hz), 7.77 (t, 1H, J = 7.8 Hz), 7.11 (s, 1H), 6.84
(m, 2H), 4.88 (s, 2H), 2.03 (s, 3H); 13C NMR (101 MHz, DMSO‑d6): δ =
163.5, 146.6, 137.2, 136.1, 131.7, 129.7, 129.6, 129.1, 127.8 (d, J = 3.6
Hz), 124.1 (d, J = 3.9 Hz), 124.0 (d, J = 272.6 Hz), 117.2, 108.8, 106.5,
17.0; ESI-HRMS: m/z calculated for C15H14F3N2O: 295.1053; found
295.1055 [M + H]+.
pared by the 14N(p,
α)
11C nuclear reaction using a mixture of 0.5% O2/
5.1.3. (5-bromo-N-(2-methyl-5-(3-(trifluoromethyl)benzamido)phenyl)
nicotinamide)
N2 as target gas on an IBA Cyclone 18/9 cyclotron (IBA, Louvain-la-
Neuve, Belgium) and was delivered as [11C]CO2 to the experimental
set up using helium as a carrier gas. Analytical HPLC was performed on a
Jasco (Easton, MD, USA) PU-2089 pump station equipped with a Grace
To a solution of 5-bromonicotinic acid (1.1 g, 5.5 mmol) in THF (5
mL) and DCM (5 mL) and DMF (0.2 mL) was added oxalic chloride (0.47
7

B. van der Wildt et al.
Bioorganic & Medicinal Chemistry 42 (2021) 116245
mL, 5.5 mmol). The resulting solution was stirred for 2 h at rt. After
concentration in vacuo, the residue was taken up in DCM (10 mL) and a
solution of compound 2 (1.5 g, 5.0 mmol) and TEA (1.5 mL, 11 mmol) in
DCM (10 mL) was added. The resulting mixture was stirred overnight
and diluted with DCM (50 mL) and 1 M HCl (50 mL). The solids were
collected by filtration, rinsed with water and DCM to obtain the product
as a white solid (2.1 g, 88%). 1H NMR (400 MHz, DMSO‑d6): δ = 10.51
(s, 1H), 10.24 (s, 1H), 9.13 (s, 1H), 8.92 (s, 1H), 8.57 (s, 1H), 8.32 (s,
1H), 8.28 (d, 1H, J = 7.9 Hz), 7.95 (d, 1H, J = 7.8 Hz), 7.90 (s, 1H), 7.78
(t, 1H, J = 7.8 Hz), 7.63 (d, 1H, J = 8.3 Hz), 7.28 (d, 1H, J = 8.4 Hz),
2.25 (s, 3H). 13C NMR (101 MHz, DMSO‑d6): δ = 163.9, 162.5, 152.8,
147.3, 137.8, 136.9, 135.7, 135.7, 131.8, 131.7, 130.4, 129.7, 129.3,
129.2 (q, J = 32.1 Hz), 128.1 (q, J = 3.6 Hz), 124.2 (q, J = 3.9 Hz), 124.0
(q, J = 272.4 Hz), 120.1, 118.7, 118.6, 17.4; ESI-HRMS: m/z calculated
for C21H15BrF3N3NaO2: 500.0193; found 500.0193 [M + Na]+.
137.2, 136.9, 135.8, 135.7, 131.9, 130.4, 129.9, 129.7, 129.4, 129.2 (q,
J = 32.1 Hz), 128.2, 128.1, 124.2 (q, J = 3.8 Hz), 124.0 (q, J = 272.4
Hz), 122.4, 118.7, 118.6, 117.0, 25.1, 17.4; ESI-HRMS: m/z calculated
for C25H20F3N5NaO2: 502.1467; found 502.1461 [M + Na]+.
5.1.7. (1-(2-fluoroethyl)-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-
1H-pyrazole)
A suspension of 4-pyrazoleboronic acid pinacol ester (0.97 g, 5.0
mmol) and NaH (0.18 g, 7.5 mmol) in DMF (20 mL) was stirred for 30
min prior to the addition of 1-fluoro-2-iodoethane (0.45 mL, 5.5 mmol)
and the resulting mixture was stirred overnight at rt. The mixture was
diluted with Et2O (100 mL) and washed with sat. NaHCO3 (3 × 50 mL),
1 M HCl (3 × 50 mL) and brine (1 × 50 mL). The organic fraction was
dried on Na2SO4, filtrated and concentrated in vacuo. Flash column
chromatography (hexanes/EA 1:1) afforded the product as a colorless oil
(0.35 g, 14%). 1H NMR (400 MHz, CDCl3): δ = 7.89 (s, 1H), 7.78 (m,
1H), 4.30 (t, 1H, J = 5.3 Hz), 3.85 (t, 1H, J = 5.4 Hz), 3.63 (t, 1H, J = 6.7
Hz), 3.16 (t, 1H, J = 6.7 Hz), 1.32 (d, 12H, J = 5.3 Hz); 13C NMR (101
MHz, CDCl3): δ = 145.8, 140.7, 137.5, 83.5 (d, J = 10.3 Hz), 70.7 (d, J =
260.1 Hz), 50.3, 24.9; ESI-HRMS: m/z calculated for C11H18BFN2NaO2:
263.1343; found: 263.1341 [M + Na]+.
5.1.4. (4-bromo-N-(2-methyl-5-(3-(trifluoromethyl)benzamido)phenyl)
picolinamide)
To a solution of 4-bromopicolinic acid (1.1 g, 5.5 mmol) in THF (5
mL) and DCM (5 mL) and DMF (1.0 mL) was added oxalic chloride (0.47
mL, 5.5 mmol). The resulting solution was stirred for 2 h at rt. After
concentration in vacuo, the residue was taken up in DCM (10 mL) and a
solution of compound 2 (1.5 g, 5.0 mmol) and TEA (1.5 mL, 11 mmol) in
DCM (10 mL) was added. The resulting mixture was stirred overnight
and diluted with DCM (50 mL) and 1 M HCl (50 mL). The solids were
collected by filtration, rinsed with water and DCM to obtain the product
as a white solid (1.9 g, 79%). 1H NMR (400 MHz, DMSO‑d6): δ = 10.52
(s, 1H), 10.29 (s, 1H), 8.74 (d, 1H, J = 5.3 Hz), 8.27 (m, 3H), 8.17 (d, 1H,
J = 2.1 Hz), 7.96 (d, 1H, J = 7.8 Hz), 7.86 (dd, 1H, J = 2.1, 5.3 Hz), 7.80
(t, 1H, J = 7.8 Hz), 7.62 (dd, 1H, J = 2.2, 8.2 Hz), 7.27 (d, 1H, J = 8.4
Hz), 2.29 (s, 3H) 13C NMR (101 MHz, DMSO‑d6): δ = 163.8, 160.9,
151.4, 150.2, 145.0, 137.0, 135.7, 135.6, 131.9, 130.2, 129.7, 129.1 (q,
J = 32.0 Hz), 128.1 (q, J = 3.7 Hz), 126.9, 126.3, 124.0 (q, J = 272.6
Hz), 124.2 (q, J = 3.8 Hz), 122.3, 117.7, 115.8, 17.0; ESI-HRMS: m/z
calculated for C21H15BrF3N3NaO2: 500.0193; found 500.0201 [M +
Na]+.
5.1.8. (5-(1-(2-fluoroethyl)-1H-pyrazol-4-yl)-N-(2-methyl-5-(3-
(trifluoromethyl)benzamido)phenyl)nicotinamide)
A mixture of compound 3 (95 mg, 0.20 mmol), compound 7 (42 mg,
0.20 mmol), Pd(PPh3)4 (23 mg, 20 µmol) and CsF (50 mg, 0.33 mmol) in
MeOH/THF (1:1, 3 mL) was refluxed for 16 h. After concentration in
vacuo and flash column chromatography (EA/hexanes 3:1), the product
was obtained as a yellow solid (75 mg, 73%). 1H NMR (400 MHz,
DMSO‑d6): δ = 10.51 (s, 1H), 10.17 (s, 1H), 9.06 (s, 1H), 8.96 (s, 1H),
8.50 (m, 2H), 8.32 (m, 2H), 8.16 (s, 1H), 7.95 (m, 2H), 7.79 (t, 1H, J =
7.8 Hz), 7.64 (d, 1H, J = 8.2 Hz), 7.31 (d, 1H, J = 8.2 Hz), 4.82 (dt, 2H, J
= 47.3, 4.7 Hz), 4.50 (dt, 2H, J = 27.8, 4.7 Hz), 2.26 (s, 3H); 13C NMR
(101 MHz, DMSO‑d6): δ = 164.4, 164.3, 149.0, 146.6, 137.5, 137.3,
136.4, 136.2, 132.3, 131.9, 131.5, 130.8, 130.7, 130.2, 129.8, 129.6 (q,
J = 32.8 Hz), 129.1, 128.7, 128.6 (q, J = 3.8 Hz), 124.7 (q, J = 3.8 Hz),
124.4 (q, J = 272.4 Hz), 119.2, 119.0, 118.5, 81.9 (d, J = 168.8 Hz),
52.6 (d, J = 19.9 Hz); ESI-HRMS: m/z calculated for C26H21F4N5NaO2:
534.1529; found 534.1524 [M + Na]+.
5.1.5. (5-(1-methyl-1H-pyrazol-4-yl)-N-(2-methyl-5-(3-(trifluoromethyl)
benzamido)phenyl)nicotinamide)
A solution of compound 3 (96 mg, 0.20 mmol), 1-methylpyrazole-4-
boronic acid pinacol ester (42 mg, 0.20 mmol), Pd(PPh3)4 (23 mg, 20
µmol) and CsF (50 mg, 0.33 mmol) in MeOH/DMF (1:1, 2 mL) was
reacted for 16 h at 70 ◦C. After concentration in vacuo and flash column
chromatography (hexanes/EA 1:3 -> 0:1) the product was obtained as a
white solid (65 mg, 68%). 1H NMR (500 MHz, CDCl3): δ = 9.29 (s, 1H),
8.90 (m, 2H), 8.67 (s, 1H), 8.17 (s, 1H), 8.05 (s, 1H), 7.94 (d, 1H, J = 7.8
Hz), 7.81 (s, 1H), 7.72 (s, 1H), 7.64 (m, 2H), 7.42 (t, 1H, J = 7.8 Hz),
7.24 (d, 1H, J = 8.3 Hz), 6.90 (d, 1H, J = 8.4 Hz), 3.85 (s, 3H), 1.93 (s,
3H); 13C NMR (101 MHz, DMSO‑d6): δ = 164.4, 164.3, 149.0, 146.4,
137.3, 137.0, 136.4, 136.2, 132.3, 131.4, 130.8, 130.7, 130.2, 129.8,
129.5, 129.1, 128.8, 128.5 (q, J = 3.8 Hz), 124.7 (q, J = 3.8 Hz), 124.4
(q, J = 272.4 Hz), 119.2, 119.0, 118.4, 39.2, 17.9; ESI-HRMS: m/z
calculated for C25H20F3N5NaO2: 502.1467; found 502.1461 [M + Na]+.
5.1.9. (N3-methyl-N5-(2-methyl-5-(3-(trifluoromethyl)benzamido)
phenyl)pyridine-3,5-dicarboxamide)
To a solution of compound 3 (95 mg, 0.20 mmol), Pd(OAc)2 (2.2 mg,
25 µmol), xantphos (12 mg, 50 µmol), NaOAc (41 mg, 0.50 mmol) in
dioxane (4 mL) was added a solution of 2,4,6-trichlorophenyl formate
(77 mg, 0.30 mmol) in toluene (1 mL). The resulting solution was heated
◦
at 80 C in a closed vial for 16 h. After dilution with EA (10 mL) the
mixture was filtered. The filtrate was concentrated in vacuo and resus-
pended in DMF (5 mL). Subsequently, TEA (0.28 mL, 2.0 mmol) and
methylamine⋅HCl (0.13 g, 2.0 mmol) were added and the mixture was
reacted for 3 h at rt. After dilution with EA (50 mL), the organic fraction
was washed with 1 M HCl (50 mL) and brine (50 mL). After drying on
Na2SO4, filtration and concentration, the product was purified using
flash column chromatography (2% MeOH in EtOAc) to yield the product
as a white solid (32 mg, 35%). 1H NMR (400 MHz, DMSO‑d6): δ = 10.50
(s, 1H), 10.29 (s, 1H), 9.24 (s, 1H), 9.16 (s, 1H), 8.84 (d, 1H, J = 4.7 Hz),
8.71 (s, 1H), 8.30 (m, 2H), 7.98 (d, 1H, J = 7.8 Hz), 7.88 (s, 1H), 7.79 (t,
1H, J = 7.8 Hz), 7.61 (d, 1H, J = 8.4 Hz), 7.30 (d, 1H, J = 8.2 Hz), 2.85
(d, 3H, J = 4.3 Hz), 2.24 (s, 3H); 13C NMR (101 MHz, DMSO‑d6): δ =
164.6, 163.9, 163.5, 150.6, 136.9, 135.9, 135.7, 134.3, 131.9, 130.4,
129.9, 129.8, 129.7, 129.3, 129.0, 128.2, 128.1, 124.2 (q, J = 3.8 Hz),
124.0 (q, J = 272.4 Hz), 118.7, 118.6, 26.3, 17.5; ESI-HRMS: m/z
calculated for C23H19F3N4NaO3: 479.1307; found: 479.1303 [M + Na]+.
5.1.6. (2-(1-methyl-1H-pyrazol-4-yl)-N-(2-methyl-5-(3-(trifluoromethyl)
benzamido)phenyl)isonicotinamide)
A solution of compound 4 (96 mg, 0.20 mmol), 1-methylpyrazole-4-
boronic acid pinacol ester (42 mg, 0.20 mmol), Pd(PPh3)4 (23 mg, 20
µmol) and CsF (50 mg, 0.33 mmol) in THF/DMF (1:1, 3 mL) was reacted
for 16 h at 70 ◦C. After concentration in vacuo and flash column chro-
matography (2% MeOH in EA) the product was obtained as a white solid
(85 mg, 88%). 1H NMR (400 MHz, DMSO‑d6): δ = 10.49 (s, 1H), 10.19
(s, 1H), 8.68 (d, 1H, J = 5.1 Hz), 8.35 (s, 1H), 8.29 (m, 2H), 8.10 (s, 1H),
8.05 (s, 1H), 7.96 (d, 1H, J = 7.8 Hz), 7.87 (s, 1H), 7.77 (t, 1H, J = 7.8
Hz), 7.62 (m, 2H), 7.29 (d, 1H, J = 8.2 Hz), 3.89 (s, 3H), 2.23 (s, 3H); 13
C
NMR (101 MHz, DMSO‑d6): δ = 164.0, 163.9, 152.5, 150.1, 142.4,
8

B. van der Wildt et al.
Bioorganic & Medicinal Chemistry 42 (2021) 116245
5.1.10. (N-(2-methyl-5-(3-(trifluoromethyl)benzamido)phenyl)-5-
((trimethylsilyl)ethynyl)nicotinamide)
(q, J = 30.6 Hz), 128.3, 127.3 (q, J = 5.7 Hz), 124.2 (q, J = 273.8 Hz),
52.5, 19.7.
A mixture of compound 3 (0.48 g, 1.0 mmol), PdCl2(PPh3)2 (70 mg,
0.10 mmol), CuI (19 mg, 0.10 mmol), TEA (0.6 mL, 4.0 mmol), and
trimethylsilylacetylene (0.6 mL, 4.0 mmol) in DMF (5 mL) were stirred
at rt for 16 h. After concentration in vacuo and purification by flash
column chromatography (EA/hexanes 1:2) the product was obtained as
a yellow solid (0.35 g, 71%). 1H NMR (400 MHz, CDCl3): δ = 8.95 (m,
2H), 8.75 (s, 1H), 8.44 (s, 1H), 8.18 (s, 1H), 8.05 (s, 1H), 7.95, (d, 1H, J
= 7.8 Hz), 7.82 (s, 1H), 7.70 (d, 1H, J = 7.6 Hz), 7.49 (t, 1H, J = 7.7 Hz),
5.1.14. (methyl 4-(bromomethyl)-3-(trifluoromethyl)benzoate)
To a solution of compound 13 (1.09 g, 5.00 mmol) and N-bromo-
succinimide (1.78 g, 10.0 mmol) in acetonitrile (20 mL) was added
benzoyl peroxide (0.12 g, 0.50 mmol) and the resulting solution was
heated at reflux temperature for 16 h. After concentration in vacuo, the
residue was taken up in ethyl acetate (50 mL) and washed with saturated
NaHCO3 (3 × 50 mL) and brine (50 mL). The organic fraction was dried
on Na2SO4, filtered and concentrated in vacuo. After flash column
chromatography (5% EA in hexanes) the product was obtained as a
white solid (1.10 g, 74%). 1H NMR (400 MHz, CDCl3): δ = 8.32 (d, 1H, J
= 1.5 Hz), 8.20 (dd, 1H, J = 1.7, 8.1 Hz), 7.70 (d, 1H, J = 8.1 Hz), 4.64
(s, 2H), 3.96 (s, 3H); 13C NMR (101 MHz, CDCl3): δ = 165.4, 140.9,
133.3, 133.1, 130.5, 128.7 (q, J = 31.5), 127.7 (q, J = 5.6 Hz), 123.7 (q,
J = 274.2 Hz), 52.7, 27.5.
7.24 (d, J = 8.0 Hz), 6.97 (d, J = 8.1 Hz), 2.00 (s, 3H), 0.27 (s, 9H); 13
C
NMR (101 MHz, CDCl3): δ = 164.9, 164.2, 155.0, 147.1, 138.3, 136.3,
135.5, 134.9, 131.1 (q, J = 32.7), 130.9, 130.9, 129.7, 129.2, 128.5,
128.3 (q, J = 3.6 Hz), 124.5 (q, J = 3.8 Hz), 123.3 (q, J = 272.5 Hz),
120.8, 119.7, 118.2, 100.3, 100.3, 17.3, ꢀ 0.2; ESI-HRMS: m/z calcu-
lated for C26H24F3N3NaO2Si: 518.1488; found 518.1481 [M + Na]+.
5.1.11. (5-ethynyl-N-(2-methyl-5-(3-(trifluoromethyl)benzamido)phenyl)
nicotinamide)
5.1.15. (methyl 4-((4-methylpiperazin-1-yl)methyl)-3-(trifluoromethyl)
benzoate)
To a solution of compound 10 (0.10 g, 0.20 mmol) in THF (2 mL) was
added CsF (46 mg, 0.30 mmol) and the resulting solution was stirred
overnight at room temperature. After filtration and concentration in
vacuo, the product was purified by flash column chromatography
(hexanes/EA 1:1 -> 1:2) yielding the product as a white solid (85 mg,
quant.). 1H NMR (400 MHz, CD3OD): δ = 9.09 (s, 1H), 8.82 (s, 1H), 8.43
(s, 1H), 8.25 (s, 1H), 8.20 (d, 1H, J = 8.2 Hz), 7.89 (d, 1H, J = 7.8 Hz),
7.83 (s, 1H), 7.74 (t, 1H, J = 7.8 Hz), 7.58 (dd, 1H, J = 2.0, 8.2 Hz), 7.31
(d, 1H, J = 8.3 Hz), 3.93 (s, 1H), 2.29 (s, 3H); 13C NMR (101 MHz,
CD3OD): δ 167.0, 165.9, 155.5, 148.9, 139.8, 138.1, 137.3, 136.7,
132.3, 132.3, 132.0 (q, J = 32.5 Hz), 131.9, 131.6, 130.6, 129.3 (q, J =
3.5 Hz), 126.0 (q, J = 271.7 Hz), 125.6 (q, J = 3.9 Jz), 121.3, 121.0,
120.5, 83.9, 80.1, 17.7; ESI-HRMS: m/z calculated for C23H16F3N3NaO2:
446.1092; found 446.1087 [M + Na]+.
A mixture of compound 14 (0.89 g, 3.0 mmol), 1-methylpiperazine
(0.90 g, 9.0 mmol) and K2CO3 (0.83 g, 6.0 mmol) in acetonitrile (20
mL) was heated at reflux temperature for 16 h, prior to concentration in
vacuo. The residue was diluted with ethyl acetate (50 mL) and washed
with saturated NaHCO3 (3 × 50 mL). The organic fraction was
concentrated in vacuo and the residue purified by flash column chro-
matography (5% MeOH in DCM) to obtain the product as a yellow solid
(0.60 g, 63%). 1H NMR (400 MHz, CDCl3): δ = 8.28 (d, 1H, J = 1.4 Hz),
8.16 (dd, 1H, J = 1.5, 8.1 Hz), 7.92 (d, 1H, J = 8.2 Hz), 3.93 (s, 3H), 3.69
(s, 2H), 2.51 (m, 8H), 2.29 (s, 3H); 13C NMR (101 MHz, CDCl3): δ =
166.0, 143.4, 132.8, 130.6, 129.0, 129.0 (q, J = 31.0 Hz), 127.3 (q, J =
6.2 Hz), 124.1 (q, J = 274.0 Hz), 58.1, 55.3, 53.3, 52.5, 46.1; ESI-HRMS:
m/z calculated for C15H20F3N2O2: 317.1471; found: 317.1473 [M +
H]+.
5.1.12. (5-(1-(2-fluoroethyl)-1H-1,2,3-triazol-4-yl)-N-(2-methyl-5-(3-
(trifluoromethyl)benzamido)phenyl)nicotinamide)
5.1.16. (4-((4-methylpiperazin-1-yl)methyl)-3-(trifluoromethyl)benzoic
acid
To a 0.5 M solution of 1-azido-2-fluoroethane in DMF (0.30 mL, 0.15
mmol) was added 13 (30 mg, 71 µmol). Both a solution of sodium
ascorbate (0.10 g, 0.51 mmol) in water (0.1 mL) and CuSO4⋅5H2O (12
mg, 50 µmol) in water (0.1 mL) were added and the mixture was reacted
for 10 min at 80 ◦C using microwave irradiation. The resulting mixture
was diluted with ethanol, filtered and purified by flash column chro-
matography (EA) to obtain the product as a white solid (20 mg, 55%). 1H
NMR (400 MHz, DMSO‑d6): δ = 10.49 (s, 1H), 10.26 (s, 1H), 9.25 (d, 1H,
J = 2.0 Hz), 9.09 (d, 1H, J = 1.9 Hz), 8.86 (s, 1H), 8.76 (t, 1H, J = 2.1
Hz), 8.29 (m, 2H), 7.96 (d, 1H, J = 7.8), 7.88 (d, 1H, J = 2.1 Hz), 7.79 (t,
1H, J = 7.8 Hz), 7.62 (dd, 1H, J = 2.2, 8.2 Hz), 7.29 (d, 1H, J = 8.5 Hz),
4.94 (m, 1H), 4.82 (m, 3H), 2.24 (s, 3H); 13C NMR (101 MHz, DMSO‑d6):
δ = 163.9, 163.7, 148.7, 147.9, 143.1, 136.9, 135.9, 135.7, 131.9,
131.6, 130.4, 130.4, 129.7, 129.3, 129.2 (q, J = 32.0 Hz), 128.1 (q, J =
3.8 Hz), 126.4, 124.2 (q, J = 3.9 Hz), 124.0 (q, J = 272.5 Hz), 123.1,
118.7, 118.6, 81.9 (d, J = 81.86 Hz), 50.4 (d, J = 19.5 Hz), 17.5; ESI-
To a solution of compound 15 (0.47 g, 1.5 mmol) in ethanol (2 mL)
was added 2 M NaOH (2 mL). The solution was stirred at rt for 6 h, prior
to concentration in vacuo to approximately 2 mL. The solution was
diluted to 10 mL with H2O and the pH was adjusted to pH 5 using 1 M
HCl. The aqueous phase was extracted with tBuOH (5 × 10 mL) and the
combined organic fractions were dried on Na2SO4, filtered and
concentrated to yield a white solid (0.21 g, 46%) that was used in the
subsequent reaction without any further purification. ESI-HRMS: m/z
calculated for C14H18F3N2O2: 303.1315; found: 303.1315 [M + H]+.
5.1.17. (5-bromo-N-(2-methyl-5-nitrophenyl)nicotinamide)
A solution of 2-methyl-5-nitroaniline (1.52 g, 10.0 mmol), 5-bromo-
nicotinic acid (2.02 g, 10.0 mmol), EDC (2.11 g, 11.0 mmol) and DMAP
(1.83 g, 15.0 mmol) in DMF (20 mL) was stirred at rt for 16 h. After
concentrating in vacuo the residue was purified by flash column chro-
matography (DCM – 10% MeOH in DCM) to obtain the product as a
white solid (2.00 g, 59%). 1H NMR (400 MHz, DMSO‑d6): δ = 10.40 (s,
1H), 9.10 (d, 1H, J = 1.7 Hz), 8.94 (d, 1H, J = 2.0 Hz), 8.57 (s, 1H), 8.38
(d, 1H, J = 2.1 Hz), 8.06 (dd, 1H, J = 2.2, 8.4 Hz), 7.59 (d, 1H, J = 8.4
Hz), 2.41 (s, 3H); 13C NMR (101 MHz, DMSO‑d6): δ = 163.1, 153.0,
147.5, 145.7, 141.6, 138.0, 136.7, 131.6, 131.4, 120.7, 120.6, 120.0,
18.2; ESI-HRMS: m/z calculated for C13H10BrN3NaO3: 357.9803; found:
357.9799 [M + Na]+.
HRMS: m/z calculated for
535.1476 [M + Na]+.
C25H20F4N6NaO2: 535.1482; found:
5.1.13. (13) (methyl 4-methyl-3-(trifluoromethyl)benzoate)
To a mixture of 4-methyl-3-(trifluoromethyl)benzoic acid (2.04 g,
10.0 mmol) and K2CO3 (2.13 g, 15.0 mmol) in DMF (10 mL) was added
methyl iodide (0.93 mL, 15 mmol) and the resulting mixture was stirred
for 16 h at rt. After concentrating in vacuo, the residue was diluted with
ethyl acetate (50 mL) and washed with subsequently H2O (3 × 50 mL)
and brine (50 mL). The organic fraction was dried on Na2SO4, filtered
and concentrated in vacuo to obtain the title compound as a yellow oil
(1.90 g, 87%). 1H NMR (400 MHz, CDCl3): δ = 8.27 (s, 1H), 8.07 (d, 1H,
J = 7.9 Hz), 7.36 (d, 1H, J = 7.9 Hz), 3.93 (s, 3H), 2.53 (s, 3H); 13C NMR
(101 MHz, CDCl3): δ = 166.1, 142.2 (q, J = 1.6 Hz), 132.7, 132.3, 129.4
5.1.18. (5-(1-methyl-1H-pyrazol-4-yl)-N-(2-methyl-5-nitrophenyl)
nicotinamide
A mixture of compound 17 (0.67 g, 2.0 mmol), 1-methylpyrazole-4-
boronic acid pinacol ester (0.42 g, 2.0 mmol), Pd(PPh3)4 (58 mg, 50
µmol) and CsF (0.60 g, 4.0 mmol) in DMF (10 mL) was heated at 100 ◦C
9

B. van der Wildt et al.
Bioorganic & Medicinal Chemistry 42 (2021) 116245
for 16 h. The mixture was concentrated in vacuo, diluted with DCM (50
mL) and washed with saturated NaHCO3 (3 × 50 mL) and brine (50 mL).
The organic fraction was concentrated in vacuo and purified using flash
column chromatography (gradient: DCM – 3% MeOH in DCM) to obtain
5.3. Radiosynthesis
[
11C]CO2 was produced by the 14N(p,
α)
11C nuclear reaction per-
formed in a 0.5% O2/N2 gas mixture using an IBA Cyclone 18/9 cyclo-
tron (IBA, Louvain-la-Neuve, Belgium). Radioactivity levels were
measured using a Veenstra (Joure, The Netherlands) VDC-405 dose
calibrator. After irradiation, [11C]CO2 was concentrated on a silica trap
(ꢀ 196 ◦C, 50 mg silica gel, 100/80 mesh). When the activity reached a
maximum, the trap was heated and [11C]CO2 was passed over a gas
purifier column (400 × 4 mm, silica gel, 100/80 mesh) using helium (30
mL⋅minꢀ 1) as carrier gas. The purified [11C]CO2 was passed over a
molybdenum reductor column heated at 850 ◦C (Sigma Aldrich, < 150
1
the product as a yellow solid (0.20 g, 59%). %). H NMR (400 MHz,
DMSO‑d6): δ = 10.34 (s, 1H), 9.04 (d, 1H, J = 2.1 Hz), 8.94 (d, 1H, J =
2.0 Hz), 8.46 (m, 1H), 8.40 (d, 1H, J = 2.4 Hz), 8.37 (s, 1H), 8.06 (m,
2H), 7.61 (d, 1H, J = 8.5 Hz), 3.91 (s, 3H), 2.43 (s, 3H); 13C NMR (101
MHz, DMSO‑d6): δ = 164.5, 148.7, 146.0, 145.7, 141.6, 137.0, 136.5,
131.6, 131.2, 130.0, 129.7, 128.4, 120.6, 120.5, 117.8, 18.3; ESI-HRMS:
m/z calculated for C17H16N5O3: 338.1248; found: 338.1251 [M + H]+.
5.1.19. (N-(5-amino-2-methylphenyl)-5-(1-methyl-1H-pyrazol-4-yl)
nicotinamide
μ
m, 99.99%) after which unreacted [11C]CO2 was trapped on an ascarite
column and [11C]CO was collected on a cooled silica trap (ꢀ 150 ◦C, 1
mg silica gel, 100/80 mesh). The transfer gas was switched from helium
to xenon (Fluka, ≥ 99.995). [11C]CO was released by heating of the trap
and transferred by a gentle xenon flow (2.0 mL ⋅ minꢀ 1) [20] to the
previously charged and sealed reagent vial. For reaction optimization,
A mixture of compound 18 (0.10 g, 0.30 mmol), NH4Cl (80 mg, 1.5
mmol), iron powder (84 mg, 1.5 mmol) in EtOH/H2O (7:3, v/v, 5 mL)
◦
was heated at 70 C for 2 h. The mixture was filtered over celite and
dried in vacuo. The residue was diluted with DCM (25 mL) and washed
with brine (25 mL). The organic fraction was dried on Na2SO4, filtered
and concentrated in vacuo to obtain the product as a yellow solid (83
mg, 90%). 1H NMR (400 MHz, DMSO‑d6): δ = 9.84 (s, 1H), 8.99 (d, 1H,
J = 2.1 Hz), 8.89 (s, 1H), 8.41 (s, 1H), 8.35 (s, 1H), 8.05 (s, 1H), 6.92 (d,
1H, J = 8.3 Hz), 6.61 (d, 1H, J = 2.1 Hz), 6.43 (dd, 1H, J = 2.5, 8.2 Hz),
4.94 (s, 2H), 3.90 (s, 3H), 2.07 (s, 3H); 13C NMR (101 MHz, DMSO‑d6): δ
= 163.9, 148.5, 146.1, 146.4, 136.7, 136.2, 131.1, 130.7, 130.6, 128.9,
128.5, 120.7, 118.1, 112.7, 112.5, 37.3, 17.1; ESI-HRMS: m/z calculated
for C17H18N5O 308.1506; found: 308.1508 [M + H]+.
[
11C]CO was transferred to a collection syringe mounted on a automated
syringe pump and subsequently dispensed over up to ten reaction vials
[19], whereas for production runs the full batch of [11C]CO was trans-
ferred to a single reaction vial. For production runs, the vial was charged
with a mixture of precursor 19 (3.0 mg, 10 µmol), trifluoro-iodobenzene
(2.7 mg, 10 µmol), PdCl2(PPh3)2 (1.4 mg, 2.0 µmol), xantphos (2.9 mg,
5.0 µmol), DMAP (3.0 mg, 20 µmol) in THF (0.7 mL). After transfer of
[
11C]CO to this reaction vial the reaction solution was heated at 100 ◦C
for 5 min. Radioactivity levels in the reaction vial were measured after
heating and after degassing of the reaction solution to determine trap-
ping efficiency of [11C]CO. A sample was taken for HPLC analysis to
determine radiochemical purity. The reaction mixture was diluted with
1 mL of mobile phase and [11C]5 was isolated by preparative HPLC.
Formulation for injection was performed by diluting the HPLC product
fraction with H2O (40 mL) and passing over a Seppak C18 light solid
phase extraction cartridge (Waters, Milford, MA, USA). The cartridge
was then washed with 10 mL of water followed by elution of the product
from the cartridge with 1 mL of ethanol. The ethanol fraction was
concentrated to 0.2 mL in vacuo at 80 ◦C with a helium flow of 50
mL⋅minꢀ 1 and diluted with 1.8 mL of saline to obtain a final concen-
tration of 10% ethanol in saline. The identity of [11C]5 was confirmed by
analytical HPLC by co-injection of both labeled and unlabeled
compounds.
5.1.20. 5-(1-methyl-1H-pyrazol-4-yl)-N-(2-methyl-5-(4-((4-
methylpiperazin-1-yl)methyl)-3-(trifluoromethyl)benzamido)phenyl)
nicotinamide
To a solution of compound 16 (30 mg, 0.10 mmol), compound 19
(31 mg, 0.10 mmol), DMAP (37 mg, 0.30 mmol) in DMF (1 mL) was
added EDC (29 mg, 0.15 mmol). The resulting solution was stirred for
16 h at rt. After concentration in vacuo the mixture was purified using
flash column chromatography (5% MeOH in DCM) to obtain the product
as a white solid (42 mg, 71%). 1H NMR (400 MHz, DMSO‑d6): δ = 8.96
(m, 2H), 8.54 (s, 1H), 8.25 (s, 1H), 8.20 (s, 1H), 8.16 (d, 1H, J = 8.1 Hz),
8.01 (s, 1H), 7.98 (d, 1H, J = 8.2 Hz), 7.85 (d, 1H, J = 1.9 Hz), 7.57 (dd,
J = 2.1, 8.3 Hz), 7.32 (d, 1H, J = 8.5 Hz), 3.97 (s, 3H), 3.76 (s, 2H), 2.58
(bs, 8H), 2.35 (s, 3H), 2.32 (s, 3H); ESI-HRMS: m/z calculated for
C
31H33F3N7O2: 592.2642; found: 592.2648 [M + H]+.
5.4. Partition coefficient LogD
5.2. In vitro IC50 determination
The partitioning of [11C]5 between 1-octanol and 0.2 M phosphate
buffer (pH = 7.4) was determined by vigorously mixing [11C]5 (100 µL,
10 MBq) with a solution of 0.2 M phosphate buffer (2 mL, pH 7.4) and 1-
octanol (2 mL) for 1 min using a vortex apparatus. After a settling period
Affinities for designed compounds were determined using a
commercially available Z’-LYTE™ assay according to the manufac-
turer’s instructions. Briefly, in a 384 well plate were added the respec-
tive kinase (either CSF-1R, PDGFR-β or c-KIT in a final concentration of
0.5, 1.0 and 1.0 ng/µL, respectively) in supplied kinase buffer, ATP (in a
final concentration of 50 µM for CSF-1R and PDGFR-β, 100 µM for c-
KIT), the respective inhibitor (2-fold dilution series in a concentration
range from 1 µM to 0.5 nM), and the assay FRET-peptide. The resulting
solutions (total volume of 10 µL per well) were incubated at rt for 1 h,
followed by the addition of protease solution (5 µL per well). After
careful mixing, the solutions were incubated for 1 h followed by the
addition of stop buffer (5 µL per well). Readout of the plate was per-
formed using excitation wavelength of 400 nm and an excitation
wavelength of 445 (coumarin) and 520 nm (fluorescein) with a 12 nm
bandwidth. Negative control and positive control reactions were per-
formed by omission of ATP and inhibitor, respectively. The obtained
values are the result of three independent experiments and are expressed
as average IC50 value ± standard deviation.
of 1 h, three samples of 100 μL were taken from both layers. Samples
were counted for radioactivity and the Log D values were calculated
according to the following formula: Log Doct,7.4 = Log (Aoct/Aphosphate
buffer), where Aoct and Aphosphate buffer represent average radioactivities
of three 1-octanol and three phosphate buffer samples, respectively. The
result is expressed as mean ± standard deviation (n = 3).
5.5. Radiometabolite analysis
Rats (14–16 weeks old, 250–300 g, n = 2) were injected with [11C]5
(250 µL, approximately 25 MBq) under isoflurane anesthesia (2% in O2
at 1 L/min). After 1.5 h, animals were sacrificed and blood was collected
by arterial puncture (approximately 0.5 mL) and transferred to a Hep-
arin coated Eppendorf tube. In addition, the left hemisphere was
collected. Blood samples were centrifuged for 5 min at 4600 RPM to
separate blood plasma from cells. The plasma (100 µL) was diluted in
acetonitrile (200 µL at 0 ◦C) and centrifuged for 5 min at 15,000 RPM for
removal of proteins. Brains were homogenized in acetonitrile (200 µL at
10

B. van der Wildt et al.
Bioorganic & Medicinal Chemistry 42 (2021) 116245
0 ◦C) and centrifuged for 5 min at 15,000 RPM. An aliquot of each su-
pernatant (10 µL) was transferred to a TLC plate, which was subse-
quently dried at room temperature for 5 min and ran in a solution of
DCM/MeOH (90:10, v/v). The radioTLC plate was transferred to a
phosphorimager storage screen and left for 1 h. Readout was performed
on a Typhoon phosphorimager and subsequent analysis was performed
using ImageQuant.
4 Pyonteck SM, Akkari L, Schuhmacher AJ, et al. CSF-1R inhibition alters macrophage
polarization and blocks glioma progression. Nat Med. 2013;19:1264–1272.
5 Elmore MR, Najafi AR, Koike MA, et al. Colony-stimulating factor 1 receptor
signaling is necessary for microglia viability, unmasking a microglia progenitor cell
in the adult brain. Neuron. 2014;82:380–397.
6 Spangenberg E, Severson PL, Hohsfield LA, et al. Sustained microglial depletion with
CSF1R inhibitor impairs parenchymal plaque development in an Alzheimer’s disease
model. Nat Commun. 2019;10:3758.
7 Peyraud F, Cousin S, Italiano A. CSF-1R inhibitor development: current clinical
status. Curr Oncol Rep. 2017;19:70.
8 Bernard-Gauthier V, Schirrmacher R. 5-(4-((4-[(18)F]Fluorobenzyl)oxy)-3-
methoxybenzyl)pyrimidine-2,4-diamine: A selective dual inhibitor for potential PET
imaging of Trk/CSF-1R. Bioorg Med Chem Lett. 2014;24:4784–4790.
9 Tanzey SS, Shao X, Stauff J, et al. Synthesis and initial in vivo evaluation of [11C]
AZ683 - A novel PET radiotracer for colony stimulating factor 1 receptor (CSF1R).
Pharmaceuticals. 2018;11:136.
5.6. PET scanning
Dynamic PET imaging was performed using dedicated small animal
NanoPET/CT and NanoPET/MR scanners (Mediso Ltd., Hungary,
Budapest) with identical PET components. Rats (n = 2 per group) were
anaesthetized with 4 and 2% isoflurane in 1 L⋅minꢀ 1 O2 for induction
and maintenance, respectively. Rats were positioned on the scanner bed,
and the respiratory rate was monitored for the duration of the scan,
adjusting anaesthesia when required. A dynamic PET scan was acquired
immediately after intravenous (i.v.) administration (tail vein) of 25 MBq
10 Horti AG, Naik R, Foss CA, et al. PET imaging of microglia by targeting macrophage
colony-stimulating factor 1 receptor (CSF1R). PNAS. 2019;116:1686–1691.
11 Illig CR, Chen J, Wall MJ, et al. Discovery of novel FMS kinase inhibitors as anti-
inflammatory agents. Bioorganic Med Chem Lett. 2008;18:1642–1648.
12 Illig CR, Manthey CL, Wall MJ, et al. Optimization of a potent class of arylamide
colony-stimulating factor-1 receptor inhibitors leading to anti-inflammatory clinical
candidate 4-cyano-N-[2-(1-cyclohexen-1-yl)-4-[1-(dimethylamino)- acetyl]-4-
piperidinyl]phenyl]-1H-imidazole-2-carboxamide (JNJ-28312141). J Med Chem.
2011;54:7860–7883.
[
11C]5. For efflux blocking experiments, rats received tariquidar 30 min
prior to tracer injection (15 mg⋅kgꢀ 1). Data was analyzed using Viv-
oQuant, by drawing a region of interest around the full brain. Results are
expressed as percentage injected dose per gram (%ID/g). Error bars
indicate standard deviation.
13 Van der Wildt B, Miao Z, Park JH, et al. Carbon-11 labeled BLZ945 as PET tracer for
Colony Stimulating Factor 1 Receptor imaging in the brain. J Labelled Compd
Radiopharm. 2019;62:S487–S488.
14 Ramachandrana SA, Jadhavara PS, Miglania SK, et al. Design, synthesis and
optimization of bis-amide derivatives as CSF1Rinhibitors. Bioorg Med Chem Lett.
2017;27:2153–2160.
15 Miyaura N, Yamada K, Suzuki A. A new stereospecific cross-coupling by the
palladium-catalyzed reaction of 1-alkenylboranes with 1-alkenyl or 1-alkynyl
halides. Tet Lett. 1979;20:3437–3440.
5.7. In vitro autoradiography
16 Ueda T, Konishi H, Manabe K. Trichlorophenyl formate: highly reactive and easily
accessible crystalline CO surrogate for palladium-catalyzed carbonylation of aryl/
alkenyl halides and triflates. Org Lett. 2012;14:5370–5373.
Autoradiography was performed in flash frozen rat brain sections
(10 µm thickness). Sections were washed with 50 mM Tris-HCl buffer
(pH 7.4) for 15 min. After drying under a gentle air flow the section were
incubated with [11C]5 (0.5 MBq⋅mLꢀ 1) in 50 mM Tris-HCl, pH 7.4 in the
absence or presence of a CSF-1R inhibitor at 1 µM concentration for 30
min. Washing was performed with cold Tris-HCl (5 mM, 4 ◦C, two times)
followed by dipping in ice cold water. After drying in an air stream, rat
brain sections were exposed to a phosphorimaging screen (GE Health-
care, Buckinghamshire, UK) for 10 min and developed on a Typhoon
FLA 7000 phosphor imager (GE Healthcare, Buckinghamshire, UK).
Visualisation of binding was performed using ImageQuantTL v8.1.0.0
(GE Healthcare, Buckinghamshire, UK).
17 Sonogashira K. Development of Pd-Cu catalyzed cross-coupling of terminal
acetylenes with sp2-carbon halides. J Organomet Chem. 2002;653:46–49.
18 Rostovtsev VV, Green LG, Fokin VV, et al. A stepwise Huisgen cycloaddition process:
copper(I)-catalyzed regioselective “ligation” of azides and terminal alkynes. Angew
Chem Int Ed. 2002;41:2596–2599.
19 Van der Wildt B, Shen B, Chin FT. A [11C] CO dispensing system for rapid screening
of carbonylation reactions. J Label Compd Radiopharm. 2018;61(14):1110–1114.
20 Eriksson J, Van den Hoek J, Windhorst AD. Transition metal mediated synthesis
using [11C]CO at low pressure—a simplified method for 11C-carbonylation. J Label
Compd Radiopharm. 2012;55:223–228.
21 Dahl K, Schou M, Amini N, et al. Palladium-mediated [11C]Carbonylation at
atmospheric pressure: a general method using xantphos as supporting ligand. Eur J
Org Chem. 2013:1228–1231.
¨
22 Karimi F, Langstrom B. Synthesis of 11C-labelled amides by palladium-mediated
carboxamination using [11C]carbon monoxide, in situ activated amines and
1,2,2,6,6-pentamethylpiperidine. Eur J Org Chem. 2003;11:2132–2137.
23 Van der Wildt B, Wilhelmus MMM, Bijkerk J, et al. Development of carbon-11
labeled acryl amides for selective PET imaging of active tissue transglutaminase. Nuc
Med Biol. 2016;43(4):232–242.
Funding
The project was supported, in part, by The Ben and Catherine Ivy
Foundation (FTC), the National Cancer Institute: R21 CA205564 (FTC),
the National Institutes of Health: S10 OD018130 (FTC) and the Dutch
Research Council (BvdW, NWO-VENI grant (VI.Veni.192.229).
24 Peng Z, Maxwell DS, Sun D, et al. Imatinib analogs as potential agents for PET
imaging of Bcr-Abl and c-KIT expression at a kinase level. Bioorg Med Chem. 2014;22:
623–632.
25 Kil KE, Ding YS, Lin KS, et al. Synthesis and positron emission tomography studies of
carbon-11-labeled imatinib (Gleevec). Nucl Med Biol. 2007;34:153–163.
26 Cai H, Shi Q, Tang Y, et al. Positron emission tomography imaging of platelet-derived
growth factor receptor β in colorectal tumor xenograft using zirconium-89 labeled
dimeric affibody molecule. Mol Pharmaceut. 2019;16:1950–1957.
27 Tolmachev V, Varasteh Z, Honarvar H, et al. Imaging of platelet-derived growth
factor receptor beta expression in glioblastoma xenografts using affibody molecule
111In-DOTA-Z09591. J Nucl Med. 2014;55:294–300.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
28 Slobbe P, Poot A, Windhorst AD, et al. PET imaging with small-molecule tyrosine
kinase inhibitors: TKI-PET. Drug Discovery Today. 2012;17:1175–1187.
29 Wakabayashi Y, Telu S, Dick RM, et al. Discovery, radiolabeling, and evaluation of
subtype-selective inhibitors for positron emission tomography imaging of brain
phosphodiesterase-4D. ACS Chem Neurosci. 2020;11:1311–1323.
Appendix A. Supplementary material
30 Mathews WB, Burns HD, Dannals RF, et al. Carbon-11 labeling of a potent,
nonpeptide, at1-selective angiotensin-II receptor antagonist: MK-996. J Labelled
Compd Radiopharm. 1995;36:729–737.
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bmc.2021.116245.
31 Van der Wildt B, Shen B, Chin FT. Efficient synthesis of carbon-11 labelled
acylsulfonamides using [11C]CO carbonylation chemistry. Chem Commun. 2019;55:
3124–3127.
References
32 Dahl K, Itsenko O, Rahman O, et al. An evaluation of a high-pressure 11CO
carbonylation apparatus. J Label Compd Radiopharm. 2020;58:220–225.
33 Van der Born D, Sewing C, Herscheid JDM, et al. A universal procedure for the [F-18]
trifluoromethylation of aryl iodides and aryl boronic acids with highly improved
specific activity. Angewandte Chem-Int Ed. 2014;53:11046–11050.
34 Bankstahl JP, Kuntner C, Abrahim A, et al. Tariquidar-induced P-glycoprotein
inhibition at the rat bloodbrain barrier studied with (R)-11C-verapamil and PET.
J Nucl Med. 2008;49:1328–1335.
1 Akiyama H, Nishimura T, Kondo H, et al. Expression of the receptor for macrophage
colony stimulating factor by brain microglia and its upregulation in brains of patients
with Alzheimer’s disease and amyotrophic lateral sclerosis. Brain Res. 1994;639:
171–174.
2 Chitu V, Gokhan S¸, Nandi S, Mehler MF, Stanley ER, et al. Emerging roles for CSF-1
receptor and its ligands in the nervous system. Trends Neurosci. 2016;39:378–393.
3 Li Q, Barres BA. Microglia and macrophages in brain homeostasis and disease. Nat
Rev Immunol. 2018;18:225–242.
11

B. van der Wildt et al.
Bioorganic & Medicinal Chemistry 42 (2021) 116245
¨
35 Bankstahl JP, Bankstahl M, Romermann K, et al. Tariquidar and elacridar are dose-
dependently transported by P-glycoprotein and Bcrp at the blood-brain barrier: a
small-animal positron emission tomography and in vitro study. Drug Metab Dispos.
2013;41:754–762.
36 Rankovic Z. CNS drug design: balancing physicochemical properties for optimal
brain exposure. J Med Chem. 2015;58:2584–2608.
37 Wager TT, Hou X, Verhoest PR, et al. Moving beyond rules: the development of a
central nervous system multiparameter optimization (CNS MPO) approach to enable
alignment of druglike properties. ACS Chem Neurosci. 2010;1:435–439.
12