Synthesis and radiofluorination of novel fluoren-9-one based derivatives for the imaging of α7 nicotinic acetylcholine receptor with PET

Article abstract of DOI:10.1016/j.bmcl.2018.03.081

By structure–activity relationship studies on the tilorone scaffold, the ‘one armed’ substituted dibenzothiophenes and the fluoren-9-ones were identified as the most potential α₇ nAChR ligands. While the suitability of dibenzothiophene derivatives as PET tracers is recognized, the potential of fluoren-9-ones is insufficiently investigated. We herein report on a series of fluoren-9-one based derivatives targeting α₇ nAChR with compounds 8a and 8c possessing the highest affinity and selectivity. Accordingly, with [¹⁸F]8a and [¹⁸F]8c we designed and initially evaluated the first fluoren-9-one derived α₇ nAChR selective PET ligands. A future application of these radioligands is facilitated by the herein presented successful implementation of fully automated radiosynthesis.

Full text of DOI:10.1016/j.bmcl.2018.03.081

Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and radiofluorination of novel fluoren-9-one based derivatives
for the imaging of a7 nicotinic acetylcholine receptor with PET
Rodrigo Teodoro a, ,1, Matthias Scheunemann a,1, Barbara Wenzel a, Dan Peters b, Winnie Deuther-Conrad a,
⇑
Peter Brust a
a Helmholtz-Zentrum Dresden-Rossendorf, Institute of Radiopharmaceutical Cancer Research, Permoserstraße 15, Leipzig 04318, Germany
b DanPET AB, Rosenstigen 7, Malmö SE-21619, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
By structure–activity relationship studies on the tilorone scaffold, the ‘one armed’ substituted dibenzoth-
iophenes and the fluoren-9-ones were identified as the most potential a7 nAChR ligands. While the suit-
Received 13 March 2018
Accepted 29 March 2018
Available online xxxx
ability of dibenzothiophene derivatives as PET tracers is recognized, the potential of fluoren-9-ones is
insufficiently investigated. We herein report on a series of fluoren-9-one based derivatives targeting a7
nAChR with compounds 8a and 8c possessing the highest affinity and selectivity. Accordingly, with
[
18F]8a and [18F]8c we designed and initially evaluated the first fluoren-9-one derived
a7 nAChR selective
Keywords:
a7 nAChR
PET ligands. A future application of these radioligands is facilitated by the herein presented successful
implementation of fully automated radiosynthesis.
PET
Radiofluorination
Fluoren-9-one
Dibenzothiophenes
Ó 2018 Elsevier Ltd. All rights reserved.
Nicotinic acetylcholine receptors (nAChR) belong to the super-
family of ligand-gated ion channels and are known to modulate
diverse physiological functions as well as to play a critical role in
the pathophysiology of various diseases including Alzheimer’s dis-
ease, schizophrenia, inflammatory conditions and cancer.1,2 The
broad spectrum of effects is a result of multiple pentameric
nanomolar range to a7 nAChR.7 However, with the exception of
11C]CHIBA-1001 (Ki 7 = 35 nM), none of them fulfilled the require-
ments for advancement to clinical trials (e.g., reasonable metabolic
stability and suitable pharmacokinetics). The discovery of the
[
a
interferon inducer tilorone as potent a7 nAChR antagonists opened
up new avenues for the development of PET ligands.9,10 Structure-
activity relationship studies (SAR) on this tricyclic pharmacophore
a
and b subunit arrangements as conferred by the 17 different
combinations cloned so far.3 Amongst them, the
a7 nAChR subtype
revealed a subnanomolar a7 nAChR binding affinity of the disubsti-
has drawn considerable attention due its involvement in key intra-
cellular signaling cascades eventually affecting cognition, learning,
memory and attention-related processes in the central nervous
system.4 Despite of the poorly understanding of the underlying
mechanisms of a7 nAChR expression during disease progression,
several a7 nAChR-based therapies have reached clinical trials.5,6
tuted fluoren-9-one derivative and the ‘one armed’ substituted
dibenzothiophene (A and B, Fig. 1).10 From the hit A, Horti et al.11
developed the 11C-labelled radiotracer [11C]A-752274 (Fig. 1).
However, pre-clinical evaluation in mice and baboon revealed a
low brain penetration of [11C]A-752274 due to an assumed low
lipophilicity as a result of its dibasic structure. On the other hand,
further insertion of fluorine on the hit B enabled the development
of the corresponding 18F-labelled derivatives (e.g., the ortho- and
para-substituted, [18F]ASEM and [18F]DBT10, Fig. 1).12,13 First-in-
human studies with [18F]ASEM14 have been carried out and a com-
prehensive study comparing both radiotracers in non-human pri-
mates revealed nearly equivalent pharmacokinetic properties.15
Further evaluation of [18F]ASEM in healthy aging volunteers high-
lighted the potential of this class of compounds to monitor changes
in a7 nAChR distribution in the brain.8
However, the inefficacy of the proposed
a7 nAChR-targeted thera-
pies in clinical use underline the need for a validation of
a7 nAChR
as biomarker for particular pathologies in preclinical research.
With this regard, the selection of highly affine and selective new
drug candidates has been greatly facilitated by advancements in
molecular imaging tools such as positron emission tomography
(PET).7,8
Various carbon-11 (11C) and fluorine-18 (18F) labelled quinu-
clidine PET ligands have been developed exhibiting affinity in a
The remarkable potential of tilorone derivatives prompted us to
further investigate this compound class aiming at the development
of a new 18F-labelled a7 nAChR PET tracer. Thus, confirming the
data presented by Schrimpf et al.10, the ‘one armed’ substituted
⇑
Corresponding author.
E-mail address: r.teodoro@hzdr.de (R. Teodoro).
These authors contributed equally to this work.
1
https://doi.org/10.1016/j.bmcl.2018.03.081
0960-894X/Ó 2018 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Teodoro R., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.081

2
R. Teodoro et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
Fig. 1. Structure-activity relationship on the tilorone scaffold and the corresponding tilorone-based a7 nAChR PET tracers. From the lead C, a schematic representation of the
rationale for the development of the small series of fluoren-9-one bioisosteres.
fluoren-9-one
2-((1s,5s)-1,4-diazabicyclo[3.2.2]nonan-4-yl)-9H-
the regioisomeric nitro fluoren-9-ones 4a and 4c. The 2-nitrofluo-
ren-9-one 4b was obtained in moderate yields starting from the
9H-fluoren-9-one 4.19
fluoren-9-one (C, Fig. 1) binds with low-nanomolar affinity to a7
nAChR also in our in house assays.16,17 By inserting fluorine in
ortho- (o-), meta- (m-) and para- (p-) position to the carbonyl func-
tionality we synthesized the fluoren-9-one derivatives 8a-d
(Table 1) and directly compared these novel sulfur-free bioisos-
teres with the corresponding dibenzothiophenes with regard to
Regioselective bromination of 4a-c was accomplished with N-
bromosuccinimide (NBS) in a mixture of H2SO4 and trifluoroacetic
acid to afford compounds 5a-c in high yields. By exploiting the
same fluorodenitration protocol we used for the corresponding
dibenzothiophenes,12 the fluorinated 7-bromo-fluoren-9-ones 6a,
6c and monofluoro derivative 6d were accessed in reasonable
yields in one step. Starting with 5b, a three-step approach [(i) Zinin
reaction, (ii) diazotization, (iii) Balz-Schiemann reaction] was car-
ried out to obtain the corresponding 2-fluoro7-bromofluoren-9-
one 6b in 64% overall yield.
(i) the
a7 nAChR binding affinity and off-target selectivity towards
the 4b2 and
a
a
3b4 nAChR subtypes as well as the serotonin receptor
5-HT3, and (ii) the radiofluorination efficiency. To assess the impact
of the o-substitution on binding affinities, we also synthesized the
so far not reported o-regioisomeric derivatives 12 and 12a
(Table 1). Additionally, since symmetrical substitutions on the flu-
oren-9-one core are known to afford highly affine
a
7 nAChR deriva-
As shown in Scheme 2, the 7-bromo-fluoren-9-ones 6, 6a-c
were then reacted with the bicyclic diamines 7a or 7b under Pd-
catalyzed Buchwald-Hartwig conditions to provide the reported
non-fluorinated lead compound 810 and the novel fluorinated
derivatives 8a-d. Unexpectedly, compound 8a was obtained in a
considerable lower yield (ꢀ12%) and remained slightly impure
according to 1H NMR data. To circumvent this drawback, the car-
bonyl protected 10a was synthesized from o-fluorofluoren-9-one
6a via ketal formation (Scheme 3). After Buchwald-Hartwig cou-
pling of 10a with the 1,4-diazabicyclo[3.2.2]nonane 7a to give
the intermediate 11a, deprotection under acidic conditions led to
the hydrochloride of 8a in 38% overall yield. The good leaving
ability of fluorine substituted in the o-position to the electron
withdrawing carbonyl group of 6a in aromatic nucleophilic substi-
tution (SNAr) appeared to be responsible for the poor chemoselec-
tivity. By contrast, neither impurities nor side products were
observed for the m- and p-substituted derivatives 8b-d according
to the high yields and chemical purities (ꢁ95%). Likewise, the nitro
tives (B, Fig. 1),10 we herein report on a new unsymmetrical dibasic
fluoren-9-one 13, which is o- and m-substituted with the 1,4-diaz-
abicyclo[3.2.2]nonane pharmacophore. Eventually, we developed
and translated to a fully automated procedure the 18F-labelling of
the most potent derivatives.
The four monobasic fluorofluoren-9-one derivatives 8a-d were
synthesized according to published data with minor modifications
(see supporting information for full characterization of the final
compounds).12 As shown in scheme 1 the synthesis was carried
out starting with the bromo-nitro substituted toluenes 1a and 1c
to obtain the regioisomeric 1-nitro-fluoren-9-one 4a and 3-nitro–
fluoren-9-one 4c in three steps.18 The biphenyls 2a and 2c were
obtained via Suzuki coupling of 1a and 1c with phenyl boronic acid
in 93% and 84% yields, respectively. Subsequently, 2a and 2c were
directly converted into their corresponding acids 3a and 3c via oxi-
dation in presence of KMnO4/pyridine. An acid promoted
intramolecular Friedel-Crafts acylation led to the formation of
Please cite this article in press as: Teodoro R., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.081

R. Teodoro et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
3
Table 1
In vitro binding affinities of compounds 8, 8a - 8d, 9a, 9c, 12, 12a, 13 towards human homomeric
comparison to DBT10 and ASEM.
a7, heteromeric
a
4b2 and
a
3b4 nAChR subtypes, and human 5-HT3 receptor, in
Binding affinity ( Ki, nM)
Selectivity (Ki ratio)
Comp.
R1
R2
h
a7a
h
a
4bb2
ha
3bb4
5-HTc3
a7/
a4b2
a7/a3b4
[d]
8
H
2-A1
2-A1
2-A1
2-A1
2-A2
2-A1
2-A1
1-A1
1-A1
1-A1
1.91 0.01
1.12 0.14
2.09 0.83
1.18 0.36
15.4 0.35
1.21 0.30
84 10.7
1660 57.3
1796 231
1315 254
1063 371
>10000
1431 52.3
568 33.2
>10000
3093 155
722 65.8
517 375
211 108
45.6 3.69
33.2 6.93
35.5 1.20
65.4 0.12
307 11.3
63.3 17.7
44.3 7.92
454
251 10.6
486 107
119 29.0
42.3 4.73
49%
49%
23%
49%
n/a
n/a
n/a
n/a
n/a
n/a
2%
104
24
29
17
39
16
53
0.5
1
0.4
28
198
50
8a
8b
8c
8d
9a
9c
12
8-F
7-F
6-F
6-F
8-NO2
6-NO2
H
7-Br
7-A1
–
>1000
626
244
>10000
>1000
7
21
5
42
862
251
466
12a
13
573 254
16.6 8.41
0.60 0.44
0.84 0.16
DBT10e
ASEMe
–
42%
a
Human a7 nAChR in stably transfected SH-SY5Y cells, with radiotracer [3H]methyllycaconitine (0.5–1 nM), KD = 2.0 nM.
b
Human
a
4b2 and
a
3b4 nAChR in stably transfected HEK-293 cells, with radiotracer [3H]epibatidine (0.5–1 nM), KD = 0.025 nM for h
a
4b2 nAChR, KD = 0.117 nM for h
a3b4
nAChR.
c
Percentage of inhibition at 0.1
tration n = 0.69 nM; KD = 0.2 nM).
l
M concentration of test compound; Human 5-HT3 receptor recombinant-HEK293 cells, with radiotracer [3H]GR65630 (working concen-
d
Compound 810
.
e
Binding affinity data.12
Scheme 1. Reagents and conditions: (a) Phenylboronic acid, K2CO3 (2.5 eq), DME/H2O, Pd(PPh3)4, 100–102 °C, 6 h (2a, 93%, 2c, 84%); (b) KMnO4 (5–9 eq), pyridine/H2O, 105–
115 °C, 9–72 h (3a, 58%, 3c, 86%); (c) 160–165 °C, 6 h, PPA (polyphosphoric acid) (4c, 91%), or H2SO4 (4a, 87%); (d) H2O, H2SO4, HNO3, 90 °C, 2.5 h (4b, 46%); (e) TMAF
(azeotropically dried), DMSO/c-hexane, 95 °C, 6 h (6a, 75%, 6c, 54%, 6d, 58%); (f) NBS, TFA/H2SO4, 22 °C, 24 h (5a, 83%, 5b, 94%, 5c, 72%); or (g) for conversion 5b ? 6b (i)
Na2Sꢂ9H2O, NaOH, EtOH/H2O 110 °C, 4 h (2-amino derivative [not shown], 94%), (ii) HBF4 (50% in H2O), NaNO2, H2O, DMSO, 0–20 °C, 1 h; (iii) xylenes, 140 °C, 2 h (6b, 64%
starting from 5b).
Scheme 2. Reagents and conditions: (a) Pd2(dba)3, r-BINAP, Cs2CO3, 7a (or 7b), toluene, 90 °C, 24 h (8, 53%, 8a [containing impurities as detected by 1H NMR], ꢀ12%, 8b, 70%,
8c, 80%, 8d, 81%, 9a, 58%, 9c, 56%).
Please cite this article in press as: Teodoro R., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.081

4
R. Teodoro et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
Scheme 3. Reagents and conditions: (a) Trimethylorthoformate, H2SO4 (cat.), MeOH, 14 d, 22 °C (10a, 57% [from 6a]); (b) Pd2(dba)3, r-BINAP, Cs2CO3, 7a, toluene, 90 °C, 24 h
(11a, 76%); (c) MeOH/H2O, HCl, 22 °C, 10 min (8a, 99%); (d) Cs2CO3, 7a, toluene, 90–100 °C, 30 h (12a, 78% [from 6a], 12, 38% [from 6d]); (e), Cs2CO3, 7a (2.6 eq), Pd2(dba)3, r-
BINAP, (2h after start of reaction), toluene, 90–100 °C, 48 h (13, 40% [from 6a]).
precursors 9a and 9c for radiofluorination were synthesized using
Buchwald-Hartwig coupling of 5a and 5c with the 1,4-diazabicyclo
[3.2.2]nonane 7a in 58% and 56% yields, respectively.
The incorporation of fluorine in o-, m- and p-positions (8a-c)
resulted in a 4b2 nAChR selectivity ratio over 1000. Whilst the
proposed changes in the cationic center were reported to positively
contribute to the selectivity towards the
4b2 subtype12 as herein
a
In order to assess the effect of the o-substitution on binding
affinities, compounds 12, 12a and 13 were synthesized (Scheme 3).
Although 12a and 13 could be already detected by TLC during the
synthesis of the o-fluorofluoren-9-one 8a, their syntheses were
scaled up to enable binding affinity measurements. Compound
12, which is regioisomeric to 8, was prepared from 6d in 38% yield
and compound 12a from 6a in 38% yield by exploiting the above-
mentioned reactivity of the fluorine at the ortho-position to the
carbonyl group. The 1,7-dihalofluorenone 6d was reacted with
amine 7a under Pd-catalyzed Buchwald-Hartwig conditions to give
the disubstituted derivative 13 in 40% yield.
a
exhibited by 8d (Ki 4b2 ꢁ 10000 nM), the diminished binding affin-
a
ity towards the a7 subtype led to the exclusion of this derivative
for further 18F-labelling.
An overlap in the receptor expression of a7 and a3b4 nAChRs, in
particular within autonomic neurons21,22 makes the investigation
of the selectivity of potential ligands towards this receptor subtype
necessary. As summarized in Table 1, compound 8, 8a-c showed
only a modest Ki 7/Ki selectivity ratio of about 24. The
a3b4
a
a3b4
nAChR binding affinities remained in the same order of magnitude
for the fluoren-9-one 8 (Ki 3b4 = 45.6 3.69 nM) compared to the
a
The a7 nAChR binding affinity and selectivity towards the off-
corresponding non-fluorinated dibenzothiophene (Ki 3b4 = 49.6
a
target receptors
a4b2 and
a
3b4 as well as 5-HT3 is depicted in Table 1
14.7 nM12, Fig. 1B). However, insertion of fluorine (8a-c) resulted
(for methodology see supporting information). The
a
7 nAChR bind-
in an increased a3b4 nAChR affinity in comparison, for example,
ing affinity of the lead 8 (Ki 7 = 1.91 0.01 nM) was sevenfold lower
with their bioisosteres DBT10 and ASEM. Interestingly, for com-
a
than the previously reported by Schrimpf et al. (Ki 7 = 0.28 nM).10
pound 8d (Ki 3b4 = 307 11.30 nM) the deshielded N-methyl group
a
a
Although these values cannot be directly compared due to the pre-
viously discussed impact of assay-related differences on binding
affinities (e.g., radioligand, tissue),13 our findings confirmed the
potential of this tricyclic pharmacophore as a7 nAChR agent. The
of the cationic center greatly contributed to the selectivity towards
the a3b4 subtype in comparison to 8a-c.
The in vitro binding affinity to 5-HT3 was investigated by mea-
suring the percentage of inhibition of binding of [3H]GR65630 at
100 nM test compound. This analysis was performed only for com-
sulfonyl (SO2, Ki 7 = 0.51 0.32 nM)12 to carbonyl (CO) exchange
a
resulted in an approximately fourfold diminished affinity of the
pounds possessing the highest a7 nAChR binding affinity and good
lead 8 (Ki 7 = 1.91 0.01 nM), which is less pronounced than the
nAChR off-target selectivity (8, 8a-c). With the exception of the m-
fluoro substituted derivative 8b (23%), the lead 8 and the fluoro
isomers 8a and 8c exhibited percentages of inhibition in the range
of 50%. These values are in the same order of magnitude as for
ASEM. It is worth noticing that the neither non-specific binding
nor undesirable pharmacological effects due to 5-HT3 binding were
reported in pre-clinical15 and in clinical trials with [18F]ASEM.8,14So
we assume that the likelihood of potential 5-HT3 cross-target
issues for the corresponding radioligands of the fluoren-9-one
derivatives 8a and 8c are rather small.
a
tenfold difference reported for the same matched pair under the
same assay conditions (B vs C, Fig. 1).10 Equivalent binding affinities
(Ki in the range of 1.18–2.09 nM) were found for the o-, m- and
a
7
p-fluoro substituted derivatives 8a-c in comparison to the lead 8.
So we can conclude that the introduction of a fluorine atom at this
part of the molecule does not affect the a7 nAChR binding affinity as
it is for the corresponding dibenzothiophene analogs ASEM
and DBT10. The decrease in binding affinity obtained for 8d
(Ki 7 = 15.4 0.35 nM), where we introduced an azatropane ring
a
instead of the 1,4-diazabicyclo[3.2.2]nonane, corresponds also to
In summary, based on the pronounced a7 nAChR binding affin-
the dibenzothiophene bioisosteres.12
ity and sufficient off-target selectivity the fluoren-9-one deriva-
For compound 12, the isomer of 8, in which the 1,4-diazabicyclo
[3.2.2]nonane is substituted at the o-position, we observed a
tives 8a and 8c were selected for further 18F-labelling.
To directly compare the radiofluorination efficiency of the fluo-
ren-9-ones 8a and 8c with the corresponding bioisosteric diben-
zothiophenes ASEM and DBT10, equivalent reaction conditions to
those applied for the radiosynthesis of [18F]DBT10 were investi-
gated (Table S.1, supporting information).12 The initial screening
was performed with regard to the solvent and the heating mode
(conventional vs microwave) at 120 °C using minimal amount of
the nitro precursors for radiolabeling (9a or 9c ffi 0.8 mg).12,23 The
radiochemical yields were determined via radio-TLC and radio-
HPLC analysis of aliquots taken from the crude reaction mixture
at different time points (up to 15 min) unless stated otherwise.
remarkable loss in binding affinity (Ki 7 = 466 nM). We hypothe-
a
size that this decline could be addressed to a steric hindrance
caused by the proximity of the cationic center to the carbonyl
group which act as hydrogen bond acceptor on the tricyclic phar-
macophore.20 The presence of an electron withdrawing group at
the C-7 position of the tricyclic pharmacophore as in 12a did not
improve the affinity. However, as expected, restoring the 1,4-diaz-
abicyclo[3.2.2]nonane as cationic center at the C-7 (meta-) position
(compound 13) of the tricyclic fluoren-9-one yielded a pronounced
increase in binding affinity (Ki 7 = 16.6 8.41 nM).
a
Please cite this article in press as: Teodoro R., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.081

R. Teodoro et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
5
Table 2
Outcomes of the automated radiosynthesis of [18F]8a and [18F]8c and the corresponding experimental logD7.4 values. Data represent mean SD (n = 3).
RCY(%)
RCP(%)
Am(GBq.
l
molꢃ1
)
logD(7.4)
[
[
18F]8a
18F]8c
37
32
5
3
ꢁ98
ꢁ98
44
31
3
6
1.2 0.1
2.2 0.1
Abbreviations. RCY: radiochemical yield; RCP: radiochemical purity; Am: molar activity.
[
18F]8a and [18F]8c were obtained by nucleophilic aromatic sub-
affinity and sufficient selectivity towards off-target receptors, and
stitution of the NO2 leaving group by [18F]fluoride in the presence
of the kryptofixÒ(K2.2.2)/K2CO3 system. In dimethylsulfoxide
(DMSO), SNAr proceeded smoothly resulting in moderate radio-
chemical yields of 39% and 65%, respectively, for [18F]8a and [18F]
8c under conventional heating at 10 min reaction time. While the
use of microwave assisted radiofluorination in DMSO afforded a
roughly twofold increase of radiochemical yield for [18F]8a (73%),
comparable yields were found for [18F]8c (62%). When N,N-
dimethylformamide (DMF) was used as solvent, radiochemical
yields in the range of 70–90% under conventional and microwave
heating were obtained for [18F]8a and [18F]8c. For all reactions,
besides of unreacted [18F]F-, radioactive by products accounted
for less than 3%.
therefore were selected for further 18F-labelling. With [18F]8a and
[
18F]8c we presented the first fluoren-9-one derived
a7 nAChR-tar-
geting PET tracers, obtained in good radiochemical yields and with
high radiochemical purities. The radiosyntheses were successfully
translated to a rapid, versatile and reproducible automated process
which will allow the accessibility of [18F]8a and [18F]8c for wide-
spread production for further pre-clinical investigation.
Acknowledgments
The Deutsche Forschungsgemeinschaft is acknowledged for
financial support (Project DE 1165/2-3). We thank the staff of the
Institute of Analytical Chemistry, Department of Chemistry and
Mineralogy of the University of Leipzig, for the NMR spectra.
Dr. Karsten Franke, Dr. Alexander Mansel and Dr. Steffen Fischer
from Helmholtz-Zentrum Dresden-Rossendorf (HZDR) for provid-
ing [18F]fluoride as well as Tina Spalholz, HZDR, for technical
assistance.
It is worth noticing that the SNAr for the fluoren-9-ones in DMF
reached a plateau in very short reaction times (5 min), remaining
constant up to 15 min independent of the heating method applied.
This rapid nitro-to-[18F]fluoro conversion was also observed for the
corresponding dibenzothiophene matched pairs.12 Interestingly,
the superior electron withdrawing effect of the sulfonyl group of
[
18F]DBT10 and [18F]ASEM did not influence the radiofluorination
A. Supplementary data
efficiency in comparison to the corresponding radiolabeled car-
bonyl bioisosteres [18F]8a and [18F]8c.24
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.bmcl.2018.03.081.
The manual radiosynthesis conditions were successfully trans-
lated to the TracerLab Fx-FN automated module device (Table 2).
For the labelling reaction of [18F]8a and [18F]8c radiochemical
yields of about 67 3% and 82 4%, respectively were obtained.
After isolation via semi-preparative HPLC (Figure S.1 A-B, sup-
porting information), [18F]8a or [18F]8c were trapped on a pre-con-
ditioned Sep-PakÒ C18 light cartridge, and formulated in isotonic
saline containing 10% of EtOH (v/v). Starting with approximately
3–5 GBq, the average decay-corrected radiochemical yields for
References
1. Fujii T, Mashimo M, Moriwaki Y, et al. J Pharmacol Sci. 2017;134:1.
2. Zhao Y. J Oncol. 2016;2016:9.
3. Wu J, Lukas RJ. PETCholinpdf. 2011;82:800.
4. Corradi J, Bouzat C. Mol Pharmacol. 2016;90:288.
5. Walling D, Marder SR, Kane J, et al. Schiz Bull. 2015;42:335.
6. Gault LM, Lenz RA, Ritchie CW, et al. Alz Res Ther. 2016;8:44.
7. Brust P, Peters D, Deuther-Conrad W. Curr Drug Targets. 2012;13:594.
8. Coughlin JM, Du Y, Rosenthal HB, et al. NeuroImage. 2018;165:118.
9. Briggs CA, Schrimpf MR, Anderson DJ, et al. Br J Pharmacol. 2008;153:1054.
10. Schrimpf MR, Sippy KB, Briggs CA, et al. Bioorg Med Chem Lett. 2012;22:1633.
11. Horti AG, Ravert HT, Gao Y, et al. Nucl Med Biol. 2013;40:395.
12. Teodoro R, Scheunemann M, Deuther-Conrad W, et al. Molecules.
2015;20:18387.
13. Gao Y, Kellar KJ, Yasuda RP, et al. J Med Chem. 2013;56:7574.
14. Wong DF, Kuwabara H, Pomper M, et al. Mol Imaging Biol. 2014;16:730.
15. Hillmer A, Li T, Zheng S, et al. Eur J Nucl Med Mol Imag. 2017;1.
16. Scheunemann M, Teodoro R, Wenzel B, Deuther-Conrad W, Steinbach J, Brust P.
J Label Compd Radiopharm. 2017;60:S585.
[
18F]8a and [18F]8c were 37 5% and 32 3%, respectively, calcu-
lated at the end of the synthesis (EOS). Both radiotracers were
obtained with a radiochemical purity of > 98% and molar activities
of 44 3 GBqꢂ
l l
molꢃ1 and 31 6 GBq molꢃ1 for [18F]8a and [18F]
8c, respectively. Analytical radio-HPLC analysis of the final product
co-eluted with the corresponding reference compound (8a or 8c)
confirmed the identity of the radiotracers (Fig. S.1C-D, supporting
information). The distribution coefficient (logD7.4, Table 2) in the
n-octanol-PBS system was experimentally determined by the
shake flask method (for methodology, see supporting information).
With logD7.4 values of 1.2 0.1 and 2.2 0.1, respectively for [18F]
8a and [18F]8c, we assume that a passive diffusion through the
blood–brain barrier is most likely.
17. Scheunemann M, Teodoro R, Wenzel B, Deuther-Conrad W, Brust P.
Nuklearmedizin. 2016;55:A23–A24.
18. Iihama T, Fu Jm, Bourguignon M, Snieckus V. Synthesis. 1989;184.
19. Zhang X, Han J-B, Li P-F, Ji X, Zhang Z. Syn Commun. 2009;39:3804.
20. Horenstein NA, Leonik FM, Papke RL. Mol Pharmacol. 2008;74:1496.
21. Girard BM, Merriam LA, Tompkins JD, Vizzard MA, Parsons RL. Am J Physiol –
Ren Physiol. 2013;305:F1504.
22. Glushakov AV, Voytenko LP, Skok MV, Skok V. Auton Neurosci Basic Clin.
2004;110:19.
23. Teodoro R, Wenzel B, Oh-Nishi A, et al. Appl Radiat Isot. 2015;95:76.
24. Chataigner I, Panel C, Gerard H, Piettre SR. Chem Commun. 2007;3288.
We herein reported on the synthesis of a small series of fluoren-
9-one derivatives exhibiting binding affinities to a7 nAChR and
off-target selectivities equivalent to the corresponding dibenztoth-
iophenes bioisosteres.12 From this series, the ‘one armed’ fluorinated
fluoren-9-ones 8a and 8c exhibited the highest a7 nAChR binding
Please cite this article in press as: Teodoro R., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.081