An improved, regioselective synthesis of the radiolabelling precursor for the translocator protein targeting positron emission tomography imaging radiotracer [18F]GE-180

Article abstract of DOI:10.1016/j.tetlet.2014.07.090

[¹⁸F]GE-180 has been demonstrated to be a promising new positron emission tomography radiotracer for targeting translocator protein. PET imaging of TSPO will enable measurement of neuroinflammation and microglia activity in vivo. The synthetic route used in the initial discovery of GE-180, whilst enabling the rapid evaluation of the structure-activity relationships (SAR) in this molecular class, was not high yielding and not suitable for scale-up. Here we present an optimised route towards GE-180 and the radiolabelling precursor of [¹⁸F]GE-180 with significantly improved yields due to a strategy which improves the regioselectivity of the key indole formation step of the synthesis.

Full text of DOI:10.1016/j.tetlet.2014.07.090

Tetrahedron Letters 55 (2014) 5141–5143
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
An improved, regioselective synthesis of the radiolabelling precursor
for the translocator protein targeting positron emission tomography
imaging radiotracer [18F]GE-180
Véronique Morisson-Iveson a, Harry Wadsworth a, Joanna Passmore a, Amanda Ewan a, Sondre Nilsen b,
Mikkel Thaning b, William Trigg a,
⇑
a GE Healthcare Ltd, The Grove Centre, White Lion Road, Amersham, Bucks. HP7 9LL, UK
b GE Healthcare AS, Nycoveien 1-2, Postboks 4220 Nydalen, Oslo 0401, Norway
a r t i c l e i n f o
a b s t r a c t
18F]GE-180 has been demonstrated to be a promising new positron emission tomography radiotracer for
Article history:
[
Received 15 April 2014
Revised 19 June 2014
Accepted 23 July 2014
Available online 30 July 2014
targeting translocator protein. PET imaging of TSPO will enable measurement of neuroinflammation and
microglia activity in vivo. The synthetic route used in the initial discovery of GE-180, whilst enabling the
rapid evaluation of the structure–activity relationships (SAR) in this molecular class, was not high yield-
ing and not suitable for scale-up. Here we present an optimised route towards GE-180 and the radiola-
belling precursor of [18F]GE-180 with significantly improved yields due to a strategy which improves the
regioselectivity of the key indole formation step of the synthesis.
Keywords:
Positron emission tomography
Target translocator protein
Indole
Ó 2014 Elsevier Ltd. All rights reserved.
Regioselectivity
GE-180
The ability to image inflammation and microglial activation in
the central nervous system (CNS) using positron emission tomog-
raphy has long been exploited using [11C]-(R)-PK111951,2 as the
radiolabelled ligand to target translocator protein (TSPO), which
is upregulated in activated microglia and macrophages.3 Whilst
many productive studies have been completed with [11C]-(R)-
PK11195, there remains a need to develop improved radiotracers
for TSPO as [11C]-(R)-PK11195 can be challenging to synthesise,
has relatively high non-specific binding and is difficult to model,
which makes analysis and interpretation of the data challenging.4
As [11C]-(R)-PK11195 is labelled with carbon-11, there is also a
need for local production of the tracer due to the short half-life
(20 min). The perceived problems with [11C]-(R)-PK11195 has led
to a large number of new radiotracers targeting TSPO,5–7 especially
those labelled with fluorine-18, which has a more suitable half-life
for central manufacturing and distribution.
to high activity and can be carried out to GMP standards using
the GE FASTlab™ radiochemistry platform.10
The chemistry we utilised in the discovery and evaluation of
[
18F]GE-180 was suitable for the rapid synthesis of a series of com-
pounds based on our chosen pharmacophore (Scheme 1) and could
be readily modified to enable the synthesis of the corresponding
mesylated radiolabelling precursors.
However, in the specific case of GE-180, indole formation with
the 3-methoxy-substituted aniline led to a mixture of regioisomers
with the desired product only as a minor component. At the early
stage of the programme, it was feasible to deliver the required
quantities of the unlabelled molecule and the corresponding radio-
labelling precursor, but as GE-180 had been shown to be a promis-
ing lead compound, a more robust and high yielding synthesis was
required.
Scheme 2 highlights the key problematic step within the origi-
nal synthetic route (carried out in this case with the amide in
place). In this step four possible products could be formed and
despite several attempts to modify the indole formation conditions
and/or isolation of the non-cyclised intermediate, the yield of the
desired indole 1 could not be improved beyond 5%.
We recently described a promising novel tracer, [18F]GE-1808
(1) (Fig. 1), which has been shown to have improved biological per-
formance compared to [11C]-(R)-PK11195,9 is fluorine-18 labelled
and has a robust and reliable radiosynthesis which can be scaled
For many of the other anilines used to evaluate the SAR, this
problem did not exist with typical yields of 50–90%, but for our
⇑
Corresponding author.
E-mail address: William.Trigg@ge.com (W. Trigg).
http://dx.doi.org/10.1016/j.tetlet.2014.07.090
0040-4039/Ó 2014 Elsevier Ltd. All rights reserved.

5142
V. Morisson-Iveson et al. / Tetrahedron Letters 55 (2014) 5141–5143
Table 1
Evaluated reaction conditions for the indole formation reactions
O
N
O
Reaction
Lewis acid
Solvent
Results (after reflux for 18 h)
1
2
ZnCl2
BF3ÁEt2O
2-PrOH
THF
Complex reaction mixture
Product could not be isolated
due to THF polymerisation
Complex reaction mixture
Complex reaction mixture
No product formed
Incomplete conversion
Incomplete conversion
No product formed
Complex reaction mixture
24% isolated yield
Complex reaction mixture
Complex reaction mixture
Messy, product not isolated
No product formed
N
3
4
5
6
7
SnCl4
ZnBr2
THF
2-PrOH
2-PrOH
CH2Cl2
CHCl3
MeCN
None
MeCN
2-BuOH
2-Pentanol
Toluene
CH2Cl2
CHCl3
F
Ti(O-iPr)4
BF3ÁEt2O
BF3ÁEt2O
BF3ÁEt2O
ZnCl2
ZnCl2
ZnCl2
ZnCl2
ZnCl2
1
Figure 1. [18F]GE-180.
8
9
10
11
12
13
14
15
16
O
O
O
O
i
Br
O
O
ZnCl2
ZnCl2
ZnCl2
O
O
50% isolated yield
50% isolated yield
3
*
2
Et2O
R1
iii
N
6
F
ii
R1
R1
O
O
O
O
i
F
NH2
Br
N
iv
O
O
H
O
O
5
4
O
2
3
R3
N
iii
O
N
R2
Cl
*
O
15
O
R1
F
ii
N
F
NH2
N
H
Cl
Cl
F
7
13
14
Scheme 1. General synthetic route to tricyclic compounds. Reagents and condi-
tions: (i) Br2, Et2O; (ii) 2-fluoroethyl 4-methylbenzenesulfonate, K2CO3, DMF; (iii)
(a) heat, neat; (b) ZnCl2, IPA, reflux; (iv) (a) NaOH, EtOH; (b) oxalyl chloride, CH2Cl2;
(c) R2R3NH, CH2Cl2.
N
O
O
O
N
O
O
O
O
vi
v
iv
N
N
N
F
F
F
17
16
1
O
N
O
O
Scheme 3. Optimised synthetic route for GE-180. Reagents and conditions: (i) Br2,
Et2O, 85%; (ii) (a) FCH2COCl, CH2Cl2, 78% yield; (b) NaBH4, I2, THF, 90%; (iii) (a)
KHMDS, THF, À40 °C (b) ZnCl2, Et2O, reflux, 80% over two steps; (iv) H2, Pd-C, Et3N,
MeOH, quant; (v) (a) NaOH, EtOH, 90%; (b) oxalyl chloride, CH2Cl2, used crude; (c)
Et2NH2, CH2Cl2, 71% over two steps; (vi) enantiomeric resolution by SFC.
N
N
O
O
N
N
O
O
i, ii
F
F
F
1
Br
10
N
F
N
H
N
O
yield (2%), and therefore the indole formation step required
optimisation.
8
9
N
O
N
O
In contrast to our original synthesis, examples from the litera-
ture11–16 showed that a two-step process led to improved yields.
Deprotonation of the aniline with KHDMS followed by addition
of the bromocyclohexanone ester gave an intermediate which
was then cyclised in the presence of a Lewis acid. We found that
the intermediate could be carried through crude to the cyclisation
step. Different Lewis acids and solvents were evaluated for the
cyclisation and the results are summarised in Table 1. The best
conditions were found to be ZnCl2 in Et2O and this improved pro-
cedure increased the yield from 2% to 80% over the two steps.17
Scheme 3 shows the optimised route towards unlabelled
GE-180, which is required for biological studies, GLP safety studies
and as an analytical standard. Our biological evaluations of GE-180
showed that the S-enantiomer had significantly better perfor-
mance than the racemic mixture, and as described previously,
the final step of the synthesis is a resolution of enantiomers using
supercritical fluid chromatography (SFC).8
O
F
11
12
Scheme 2. Original indole formation reaction. Reagents and conditions: (i) heat,
neat; (ii) ZnCl2, IPA, reflux.
chosen lead candidate, an improved synthetic strategy was
required.
In order to eliminate the possibility of the formation of the
regioisomers, a blocking atom on the aromatic ring of the aniline
was adopted as the strategy of choice. 2-Chloro-5-methoxyaniline
(13) is commercially available and was used as the starting mate-
rial for the preparation of the secondary aniline in the indole-form-
ing reaction. Evaluation of the final molecule showed that there
were no atoms or groups which were susceptible to cleavage by
hydrogenation, so it was felt that the chlorine could be readily
removed later in the synthesis.
This optimised strategy was applied to the synthesis of the
mesylated radiolabelling precursor as shown in Scheme 4. For this
Attempts to perform the cyclisation, starting with 2-chloro-
5-methoxyaniline (13), using the original conditions led to a poor

V. Morisson-Iveson et al. / Tetrahedron Letters 55 (2014) 5141–5143
5143
O
O
O
O
References and notes
i
Br
O
O
O
O
1. Cagnin, A.; Kassiou, M.; Meikle, S.; Banati, R. Neurotherapeutics 2007, 4, 443.
2. Katsifis, A.; Mattner, F.; Zhang, Z.; Dikic, B.; Papazian, V. J. Labelled Compd.
Radiopharm. 2000, 43, 385.
O
2
3
iii
3. Scarf, A.; Kassiou, M. J. Nucl. Med. 2011, 52, 677.
N
Cl
4. Chauveau, F.; Boutin, H.; Van Camp, N.; Dollé, F.; Tavitian, B. Eur. J. Nucl. Med.
Mol. Imaging 2008, 35, 2304.
O
O
OBn
ii
5. Zhang, M.-R.; Maeda, J.; Furutsuka, K.; Yoshida, Y.; Ogawa, M.; Suhara, T.;
Suzuki, K. Bioorg. Med. Chem. Lett. 2003, 13, 201; Takano, A.; Gulyás, B.; Varrone,
A.; Karlsson, P.; Sjoholm, N.; Larsson, S.; Jonsson, C.; Odh, R.; Sparks, R.; Al
Tawil, N.; Hoffmann, A.; Zimmermann, T.; Thiele, A.; Halldin, C. Eur. J. Nucl. Med.
Mol. Imaging 2011, 38, 2058.
19
OBn
NH2
N
H
Cl
13
Cl
18
6. James, M. L.; Fulton, R. R.; Vercoullie, J.; Henderson, D. J.; Garreau, L.; Chalon, S.;
Dollé, F.; Selleri, S.; Guilloteau, D.; Kassiou, M. J. Nucl. Med. 2008, 49, 814;
Arlicot, N.; Vercouillie, J.; Ribeiro, M. J.; Tauber, C.; Venel, Y.; Baulieu, J. L.; Maia,
S.; Corcia, P.; Stabin, M. G.; Reynolds, A.; Kassiou, M.; Guilloteau, D. Nucl. Med.
Biol. 2012, 39, 570.
7. Imaizumi, M.; Briard, E.; Zoghbi, S. S.; Gourley, J. P.; Hong, J.; Musachio, J. L.;
Gladding, R.; Pike, V. W.; Innis, R. B.; Fujita, M. Synapse 2007, 61, 595.
8. Wadsworth, H.; Jones, P. A.; Chau, W.; Durrant, C.; Fouladi, N.; Passmore, J.;
O’Shea, D.; Wynn, D.; Morisson-Iveson, V.; Ewan, A.; Thaning, M.; Mantzilas, D.;
Gausemel, I.; Khan, I.; Black, A.; Avory, M.; Trigg, W. Bioorg. Med. Chem. Lett.
2012, 22, 1308.
N
O
N
O
N
O
O
O
O
vi
v
iv
N
N
N
Cl
22
21
20
OMs
OBn
OH
O
N
O
vii
9. Dickens, A.; Vainio, S.; Marjamäki, P.; Johansson, J.; Lehtiniemi, P.; Rokka, J.;
Rinne, J.; Solin, O.; Haaparanta-Solin, M.; Jones, P.; Trigg, W.; Anthony, D.; Airas,
L. J. Nucl. Med. 2014, 55, 466.
N
10. Wickstrøm, T.; Clarke, A.; Gausemel, I.; Horn, E.; Jørgensen, K.; Khan, I.;
Mantzilas, D.; Rajanayagam, T.; in’t Veld, D.; Trigg, W. J. Labelled Compd.
Radiopharm. 2014, 57, 42.
23
OMs
Scheme 4. Optimised synthetic route for the GE-180 precursor. Reagents and
conditions: (i) Br2, Et2O, 85%; (ii) (a) BnOCH2COCl, CH2Cl2, quant; (b) LiAlH4, I2, THF,
quant; (iii) (a) KHMDS, THF, À40 °C (b) ZnCl2, Et2O, reflux, used crude; (iv) (a)
NaOH, EtOH; (b) oxalyl chloride, CH2Cl2, DMF; (c) Et2NH2, CH2Cl2, 57%; (v) H2, Pd-C,
Et3N, EtOH, 55%; (vi) MsCl, Et3N, CH2Cl2, 33% yield; (vii) enantiomers resolved by
SFC.
11. Kinnick, M. D.; Mihelich, E. D.; Morin, J. J.; Sall, D. J.; Sawyer J. S. WO
2003014082.
12. Ling, R.; Yoshida, M.; Mariano, P. S. J. Org. Chem. 1996, 61, 4439.
13. Goering, B. K. Ph.D. Dissertation, Cornell University, 1995.
14. Haslam, E. Shikimic Acid Metabolism and Metabolites; John Wiley & Sons: New
York, 1993.
15. Buchanan, J. G.; Sable, H. Z. In Selective Organic Transformations; Thyagarajan, B.
S., Ed.; Wiley-Interscience: New York, 1972; Vol. 2, pp 1–95.
16. Lyle, F. R. U.S. Patent 5 973 257, 1985; Chem. Abstr. 1985, 65, 2870.
larger scale synthesis, several of the intermediates were used crude
in the subsequent steps to reduce the number of purifications
required. The step for removal of the directing chlorine atom was
also combined with the benzyl deprotection step prior to mesylate
formation.
In conclusion, robust and reliable synthetic routes have been
developed to a promising new TSPO targeted PET radiotracer and
the corresponding radiolabelling precursor. The synthetic routes
described have been employed to synthesise multi-gram quantities
of both compounds to GMP standards for use in preclinical toxicol-
ogy studies, preclinical efficacy studies and clinical studies.
17.
A
solution of 5-chloro-N-(2-fluoroethyl)-2-methoxyaniline (14) (6.1 g,
30.0 mmol) in THF (170 mL) was cooled to À40 °C. KHDMS (126.0 mL of a
0.5 M solution in toluene, 63.0 mmol) was added dropwise and the reaction
stirred for 30 min at À40 °C.
A
solution of ethyl 3-bromo-2-
oxocyclohexanecarboxylate (3) (7.4 g, 30.0 mmol) in THF (30 mL) was then
added dropwise at À40 °C and the mixture stirred at rt for 4 h. The reaction
was quenched with brine (300 mL) and extracted into EtOAc (2 Â 400 mL),
dried over MgSO4 and concentrated in vacuo to give 12.0 g (quantitative) of a
brown oil which was used crude in the next step. The crude oil (12.2 g,
30.0 mmol) was dissolved in Et2O (250 mL) and ZnCl2 (16.4 g, 120.0 mmol) was
added. The mixture was heated at reflux for 16 h. EtOAc (500 mL) was added.
The solution was washed with 2 M HCl (200 mL), H2O (200 mL) and 10%
aqueous K2CO3 (200 mL), dried over MgSO4 and concentrated in vacuo. The
crude material was purified by silica gel chromatography eluting with petrol
(A): EtOAc (B) (5–20% B, 10 g, 100 mL/min) to afford 8.5 g (80% over 2 steps) of
the desired indole 15 as a white solid. The structure was confirmed by 1H NMR
(300 MHz, CDCl3): dH 1.24 (3H, t, J = 6 Hz, CO2CH2CH3), 1.80–2.10 (4H, m, 2-
and 3-CH2), 2.60–2.80 (2H, m, 1-CH2) 3.78 (3H, s, OCH3), 4.00–4.20 (3H, m, 4-
CH and CO2CH2CH3), 4.40–5.00 (4H, m, NCH2CH2F and NCH2CH2F), 6.35 (1H, d,
J = 4 Hz, 6-CH) and 6.95 (1H, d, J = 4 Hz, 7-CH); 13C NMR (75 MHz, CDCl3): dC
14.4, 20.4, 22.2, 27.4, 40.1, 44.2 (d, JCF = 23 Hz), 55.1, 60.2, 83.9 (d, JCF = 173 Hz),
100.6, 107.9, 108.2, 119.8, 123.1, 131.9, 137.2, 152.7, 175.7; 19F NMR
(283 MHz, CDCl3): dF-222.2.
Acknowledgments
The authors would like to thank Dimitrios Mantzilas, Alan
Clarke, David Grace and Ingvil Gausemel for their chromatography,
purification and analytical support.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2014.07.
090.