Development of a High-Affinity PET Radioligand for Imaging Cannabinoid Subtype 2 Receptor

Article abstract of DOI:10.1021/acs.jmedchem.6b00554

Cannabinoid receptors type 2 (CB₂) represent a target with increasing importance for neuroimaging due to its upregulation under various pathological conditions. Encouraged by preliminary results obtained with [¹¹C](Z)-N-(3-(2-methoxyethyl)-4,5-dimethylthiazol-2(3H)-ylidene)-2,2,3,3-tetramethyl-cyclopropanecarboxamide ([¹¹C]A-836339, [¹¹C]1) in a mouse model of acute neuroinflammation (induced by lipopolysaccharide, LPS), we designed a library of fluorinated analogues aiming for an [¹⁸F]-labeled radiotracer with improved CB₂ binding affinity and selectivity. Compound (Z)-N-(3-(4-fluorobutyl)-4,5-dimethylthiazol-2(3H)-ylidene)-2,2,3,3-tetramethyl-cyclopropanecarboxamide (29) was selected as the ligand with the highest CB₂ affinity (K_i = 0.39 nM) and selectivity over those of CB₁ (factor of 1000). [¹⁸F]29 was prepared starting from the bromo precursor (53). Specific binding was shown in vitro, whereas fast metabolism was observed in vivo in CD-1 mice. Animal PET revealed a brain uptake comparable to that of [¹¹C]1. In the LPS-treated mice, a 20-30% higher uptake in brain was found in comparison to that in nontreated mice (n = 3, P < 0.05).

Full text of DOI:10.1021/acs.jmedchem.6b00554

Subscriber access provided by Northern Illinois University
Article
Development of a high affinity PET radioligand
2
for imaging cannabinoid subtype 2 receptor (CB)
Rares-Petru Moldovan, Rodrigo Teodoro, Yongjun Gao, Winnie Deuther-Conrad, Mathias
Kranz, Yuchuan Wang, Hiroto Kuwabara, Masayoshi Nakano, Heather Valentine, Steffen
Fischer, Martin G Pomper, Dean F Wong, Robert F. Dannals, Peter Brust, and Andrew G. Horti
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00554 • Publication Date (Web): 08 Aug 2016
Downloaded from http://pubs.acs.org on August 15, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Development of a high affinity PET radioligand for
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
imaging cannabinoid subtype 2 receptor (CB2)
1
,
1
2
1
RareꢀꢁPetru Moldovan *, Rodrigo Teodoro , Yongjun Gao , Winnie DeutherꢁConrad , Mathias
1
2
2
2
2
Kranz , Yuchuan Wang , Hiroto Kuwabara , Masayoshi Nakano , Heather Valentine , Steffen
1
2
2
2
1
Fischer , Martin G. Pomper , Dean F. Wong , Robert F. Dannals , Peter Brust , Andrew G.
2
,
Horti *
1
HelmholtzꢀZentrum DresdenꢀRossendorf e. V., Institute of Radiopharmaceutical Cancer
Research, Leipzig, Germany
2
Johns Hopkins School of Medicine, Department of Radiology, Baltimore, USA
ABSTRACT: Cannabinoid receptors type 2 (CB ) represent a target with increasing importance
2
for neuroimaging due to its upregulation under various pathological conditions. Encouraged by
1
1
preliminary results obtained with [ C](Z)ꢀNꢀ(3ꢀ(2ꢀmethoxyethyl)ꢀ4,5ꢀdimethylthiazolꢀ2(3H)ꢀ
1
1
11
ylidene)ꢀ2,2,3,3ꢀtetramethylꢀcyclopropanecarboxamide ([ C]Aꢀ836339, [ C]1) in a mouse
model of acute neuroinflammation (induced by lipopolysaccharide, LPS), we designed a library
1
8
of fluorinated analogs aiming for an [ F]ꢀlabeled radiotracer with improved CB binding affinity
2
and selectivity. Compound (Z)ꢀNꢀ(3ꢀ(4ꢀfluorobutyl)ꢀ4,5ꢀdimethylthiazolꢀ2(3H)ꢀylidene)ꢀ2,2,3,3ꢀ
tetramethylꢀcyclopropanecarboxamide (29) was selected as a ligand with the highest CB affinity
2
18
(
K = 0.39 nM) and selectivity over CB (factor 1000). [ F]29 was prepared starting from the
i
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
bromo precursor (53). Specific binding was shown in vitro whereas fast metabolism was
1
1
observed in vivo in CDꢀ1 mice. Animal PET revealed a brain uptake comparable to [ C]1. In the
LPS treated mice, a 20ꢀ30% higher uptake in brain was found in comparison to nonꢀtreated mice
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(n = 3, P<0.05).
INTRODUCTION
Since its isolation and elucidation of the structure in 1964 by Mechoulam and coꢀ
1, 2
9
workers,
(–)ꢀtransꢀꢁ ꢀtetrahydrocannabinol (THC) has been the subject of many scientific
3
investigations due to its intriguing biological properties. This led to the discovery of the
endogenous cannabinoid system, which comprises a class of transmembrane proteins that
4, 5
belongs to the superfamily of Gꢀproteinꢀcoupled receptors and their modulatory lipids.
Etymologically, cannabinoid receptors are proteins that are activated by THC, the principal
6
psychoactive constituent of the cannabis plant.
Two types of cannabinoid receptors have been cloned so far, cannabinoid receptor type 1
7
8
(CB ) and cannabinoid receptor type 2 (CB ). There is evidence for further cannabinoid
1
2
9
receptors with unclear molecular identity, collectively known as nonꢀCB /CB receptors. The
1
2
1
0
11
most prominent candidates are GPR55 and GPR119. CB receptors are mainly located at the
1
terminals of central and peripheral neurons, and are the most abundantly expressed Gꢀproteinꢀ
5
coupled receptors in the brain. Their activation is responsible for the psychotropic effect of
5
, 12
THC.
On the other hand, the CB receptor has been categorized as the peripheral cannabinoid
2
receptor due to its presence on the cells and tissues of the immune, reproductive, cardiovascular,
12
gastrointestinal and respiratory systems.
ACS Paragon Plus Environment

Page 3 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Although numerous reports were unable to detect CB receptor in normal healthy brain,
2
recent evidence suggests that CB receptors are present in the brain under normal and in
2
particular under pathological conditions, although to a much lesser extent than the ubiquitously
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
expressed CB receptors. The role and distribution of CB receptors in the living human brain
1
1
1
4, 15
have already been investigated intensively.
Much less is known about the role and
1
3
distribution of the CB receptors in the human brain. Low levels of CB receptors are found in
2
2
1
6ꢀ18
19
microglial cells,
human fetal astrocytes, and human cerebral microvascular endothelial
2
0
cells. Overexpression of the CB receptor has been found on microglia associated with neuritic
2
21
plaques in Alzheimer’s disease. Enhanced CB receptor expression, primarily on activated
2
microglia, has been demonstrated in several other neurodegenerative disorders such as multiple
2
2, 23
24
25
sclerosis,
Down's syndrome, and Huntington’s disease. CB expression pathologies also
2
26
27
28
include traumatic brain injury, neuropathic pain, and HIVꢀinduced encephalitis. Significant
alterations of a balance in the cannabinoid system between the levels of endogenous ligands and
their receptors occur during malignant transformation in various types of cancer including brain
29, 30
tumors, as previously reviewed.
CB ꢀselective ligands have gained increasing importance as therapeutic drugs for the diseases
2
3
0ꢀ38
mentioned above, especially for the prevention of psychoactive side effects.
Molecular
imaging with positron emission tomography (PET) of diseaseꢀspecific expression of CB2
receptors in various disorders of the central nervous system like neuroinflammation,
neurodegeneration and glioma offers the potential of individualized diagnosis and therapeutic
3
9
monitoring.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Recently a substantial effort has been put into the development of radioligands for PET
3
1, 40, 41
imaging of CB receptors (see for reviews:
). Several radiolabeled CB molecules have
2
2
3
1, 32
been developed with most representative ones depicted in Figure 1.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The principle of CB imaging under neuroinflammatory circumstances was proven by our
2
1
1
42
previous work using [ C]1. Recent PET studies revealed a significant increase in the binding
1
1
of [ C]1 in a mouse model of Alzheimer’s disease that was consistent with increased CB ꢀ
2
4
3
11
11
immunoreactivity in astrocytes and microglia. Compound [ C]2 ([ C]NE40) was developed
4
4
45
by Evens and coꢀworkers and a firstꢀinꢀman study in healthy subjects was reported. However,
11
PET studies with [ C]2 did not show the expected increase of binding in subjects with
Alzheimer’s disease as compared to healthy controls, perhaps due to the insufficient binding
4
6
affinity of the radiotracer. The usefulness of CB2 PET radiotracers in models of
neuroinflammation was also demonstrated in dynamic PET studies in primates using
1
8
34, 47, 48
49
radiolabeled triazines scaffolds ([ F]3, Figure 1).
The most recently developed PET
1
1
11
11
11
50
radioligands for CB receptors are [ C]4 ([ C]RSꢀ016) and [ C]5 ([ C]RSRꢀ056) , which
2
demonstrated moderate specific binding in the mouse LPS mouse model of neuroinflammation.
Altogether, the importance of targeting CB receptors has been shown by several groups
2
worldwide and has been characterized in several pathological conditions in animals. However,
the development of a radioligand suitable for the nonꢀinvasive investigation of the CB receptor
2
in healthy and diseased brain remains challenging.
ACS Paragon Plus Environment

Page 5 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
2, 44, 47, 49,50
Figure 1. Representative radiolabeled CB ꢀligands.
2
1
8
51
Our aim was to develop an Fꢀlabeled PET radiotracer based on the scaffold of compound 1
Figure 2). The newly synthesized derivatives contain the basic structural features of the lead
compound 1 with modifications to increase CB binding affinity and selectivity versus CB1
(
2
1
8
combined with a facile incorporation of the fluorineꢀ18 ([ F]) via nucleophilic substitution at
aliphatic position. As part of our efforts, modifications at the thiazole Nꢀalkyl chain, thiazole
position 5 and at the cyclopropyl subunit were performed (Figure 2 and Table 1) and their impact
on CB binding affinity and selectivity towards CB were investigated. Based on the structureꢀ
2
1
activity relationship findings, radiofluorination and biological investigation in a murine model of
neuroinflammation were performed for the most promising derivative.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
N
O
S
N
S
N
N
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Alk(ox)yl
O
18F
1, K (CB ) = 0.7 nM
i
2
K (CB ) = 270 nM
i
1
5
2
Figure 2. Lead compound 1 and proposed fluoroꢀderivatization. For details on the Alk(ox)yl
see Scheme 1.
RESULTS
Synthesis
42
The synthesis of the new derivatives has been performed as earlier described for 1 and shown
in Scheme 1 with minor modifications. Generally, the synthesis is starting with the commercially
available thiazole 6 as free base which is Nꢀalkylated in solvent free, thermal reactions
conditions followed by the BOP reagent mediated coupling with 2,2,3,3ꢀtetramethylꢀ
cyclopropylcarboxylic acid to give compounds 17ꢀ29 (Scheme 1).
For the synthesis of the fluoroꢀtriethylene and fluoroꢀtetraethylene glycols 20 and 21 (see
Scheme 1) tosylates 34 and 35 were synthesized as shown in Scheme 2. Triethylene and
tetraethylene glycol (30 and 31 respectively) were bitosylated then monofluorinated using TBAF
5
3, 54
resulting in a 42% yield over two steps (Scheme 2A).
,5ꢀdimethylthiazol (6) was reacted with the two corresponding tosylates 34 and 35 (Scheme 2)
in a minimum amount of DMF (to solubilize) to provide compounds 7 and 8 respectively as
The commercially available 2ꢀamineꢀ
4
ꢀ
OTs salts as observed by 1H NMR spectroscopy (Scheme 1).
ACS Paragon Plus Environment

Page 7 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
For the synthesis of the compounds 25 and 26 (Scheme 1), mesylates 37 and 40 respectively,
5
5
were synthesized as racemates from the corresponding alcohols (Scheme 2B and C). All the
other Nꢀalkylating agents used for the synthesis of compounds 7ꢀ16 (Scheme 1) were
commercially available. The coupling of derivatives 7ꢀ16 with 2,2,3,3ꢀtetramethylꢀ
cyclopropylcarboxylic acid was performed according to the general procedure previously
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
2
described, by employing BOP as coupling reagent and triethylamine (Et N) in DCM. Yields
3
between 20 and 45% were obtained. Compounds 18 and 19 were synthesized from the alcohol 17
by Williamson ether synthesis. Analogously, compounds 23 and 24 were synthesized from
alcohol 22 (Scheme 1).
O
S
S
N
a)
b)
S
N
NH
NH2
N
N
R
*HX
R
6
7ꢀ16
1, 17ꢀ29
O
O
7
: R =
: R =
: R =
; X = OTs
1: R =
F
2
3
O
OH
O
8
9
F ; X = OTs
; X = Br
17: R =
18: R =
19: R =
c)
c)
O
F
OH
O
O
F
10: R =
; X = Br
OH
; X = Br
2
2
2
2
0: R =
1: R =
2: R =
3: R =
F
11: R =
12: R =
13: R =
14: R =
15: R =
2
3
O
O
F
; X = OMs
; X = OMs
; X = Br
OH
O
F
c)
c)
O
O
F
24: R =
25: R =
; X = Br
F
O
F
16: R =
; X = Br
O
F
26: R =
27: R =
28: R =
F
F
29: R =
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 1. Synthesis of the lead compound 1 and Nꢀalkyl derivatives 17ꢀ29. a) RX, neat, 90 °C,
1
6 h; b) 2,2,3,3ꢀtetramethylcyclopropaneꢀ1ꢀcarboxylic acid, BOP, Et N, DCM, rt, 24 h, 30ꢀ50%
3
42
over two steps; c) NaH, alkyl bromide, DMF, 0 °C to rt, 80ꢀ85%; BOP = (Benzotriazolꢀ1ꢀ
yloxy)tris(dimethylamino)phosphonium hexafluorophosphate.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 2. Synthesis of reagents 34, 35, 37 and 40. a) TsCl, KOH, DCM, 0 °C, 3 h,
5
3
56
quantitative; b) TBAF, THF, rt, 16 h, 42%; c) MsCl, Et N, DCM, 0 °C to rt, 1 h, quantitative;
3
5
5
d) BBr , LiAlH , DCM, 0 °C to rt, 1 h, 90%.
3
4
In a further effort to investigate the pharmacology of the lead compound 1, we introduced a
hydroxyl group and its corresponding methyl ether by substituting the 2,2,3,3ꢀ
tetramethylcyclopropyl partial structure with a 3ꢀ(hydroxymethyl)ꢀ2,2ꢀdimethylcyclopropyl
subunit (compounds 44 and 45 respectively, Scheme 3).
The synthesis of 45 started from the commercially available racꢀethylꢀchrysanthemate (41)
which was first oxidized with KMnO and the resulting carboxylic acid (42, mixture of cisꢀtrans
4
5
7
isomers) was coupled with the 2ꢀaminothiazole derivative 9 in presence of BOP and Et N to
3
ACS Paragon Plus Environment

Page 9 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
give ester 43. At this step, epimerization took place presumably via Et Nꢀmediated
3
deprotonationꢀprotonation at the cyclopropyl subunit. Compound 43 was obtained as a racemic
mixture of transꢀisomers, as determined by 2DꢀNOESY NMR (see Supporting Information).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reduction of the ester 43 to the alcohol 44 was performed with LiAlH in quantitative yield,
4
which was further etherified with MeI under deprotonation via Williamson ether synthesis to
give 45 in 95% yield over two steps.
Scheme 3. Synthesis of the hydrophilic derivatives 44 and 45. a) KMnO , Na SO , H SO , 4 h,
4
2
3
2
4
5
7
3
0%; b) 2, BOP, Et N, DCM, rt, 24 h, 42%; c) LiAlH , THF, rt, 4 h, quantitative; d) NaH, MeI,
3
4
DMF, 0 °C to rt, 4 h, 95%.
The structure of 1 was also modified at the thiazole 5ꢀposition starting from ethyl 3ꢀ
5
8
oxobutanoate which was brominated in the first step and then cyclized with thiourea (Scheme
4). 2ꢀAminothiazole 48 was Nꢀalkylated with 2ꢀbromoethyl methyl ether and then coupled with
2,2,3,3ꢀtetramethylcyclopropaneꢀ1ꢀcarboxylic to give compound 49. Double methylation of 49
with the Grignard reagent MeLi gave the 2ꢀisopropanole derivative 50. Methoxylation of 50 was
performed via Williamson ether synthesis to form 51 in 79% yield.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
8
Scheme 4. Synthesis of compounds 50 and 51. a) KBr, H O , 1 h, quantitative; b) thiourea,
2
2
EtOH, reflux, 1 h, quantitative; c) i. 2ꢀBromoethyl methyl ether, 95 °C, 16 h, ii. 2,2,3,3ꢀ
tetramethylcyclopropaneꢀ1ꢀcarboxylic acid, BOP, Et N, DCM, rt, 24 h, 36% over two steps; d)
3
MeLi, –78 °C, 30 min, 72%; e) NaH, MeI, overnight, rt, 79%.
Structure-activity relationship
18
The main purpose of the study was to develop an [ F]fluoro analog of 1 with greater CB
2
binding affinity. The new fluorinated derivative had to be highly selective for CB versus CB1
2
(
CB /CB >100) because of the high CB receptor density in the healthy human brain. We mainly
2 1 1
focused on compounds 1 Nꢀalkyl chain modifications. The binding affinities of the new
derivatives were determined by CEREP (France) and HZDRꢀ(Leipzig).
The first Nꢀalkyl chain analog of 1 is the fluoroꢀethoxyethane derivative 18 that exhibits
greater CB binding affinity than 1, but its CB /CB selectivity was reduced (<100 fold) (Table
2
2
1
ACS Paragon Plus Environment

Page 11 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
). The increase in the number of ethoxyethane groups (20 and 21) led to a dramatic reduction in
the CB affinity (19 and 1100 nM respectively) compared to 18 and 1 (Table 1).
2
The propoxyethane analogs of 1 and 18 (compounds 22, 23 and 24) exhibited comparable CB2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
and CB binding affinities, but they were slightly more lipophilic. Two derivatives were
1
synthesized by branching a methyl group at the Nꢀalkyl chain (compounds 25 and 26), which led
to a significant decrease in the CB binding affinity.
2
Three further derivatives have been synthesized, 27, 28 and 29, all of them lacking the oxygen
atom at the thiazoleꢀNꢀalkyl chain subunit and clearly showing the correlation between the
thiazoleꢀNꢀalkyl chain length and the binding affinity towards CB . Furthermore, derivatization
2
at both the cyclopropyl and thiazoleꢀ5ꢀposition led to decreased CB binding affinity
2
(compounds 44, 45 and 50, Table 1).
Of this new series of compounds (Table 1), 29 manifests the best combination of high CB2
binding affinity (K = 0.39 nM) and excellent CB /CB selectivity. It was selected for
i
2
1
radiolabeling and animal experiments.
Table 1. Binding affinities of 1 and the new derivatives determined by (A) CEREP (France); (B)
HZDR (Leipzig).
R1
K CB (nM)
K CB (nM)
i 1
N
i
2
1
2
3
N
O
R
R
R
R3
R2
S
A
B
A
B
1
Me
Me
Me
1.2
1.8
N/A
1.2
270
25
N/A
N/A
N/A
18
0.31
0.70
19
33
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
0
Me
Me
Me
Me
Me
Me
Me
Me
Me
19
1100
0.81
0.94
7.0
12.3
1395
1.4
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1600
380
N/A
N/A
N/A
146
21
23
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
24
25
26
N/A
6.9
N/A
N/A
N/A
N/A
3031
95
212
N/A
1.1
27
28
7.2
1.2
2
9
4
0.39
0.90
4
Me
Me
200
N/A
N/A
N/A
4
5
0
22
18
53
N/A
N/A
N/A
N/A
5
N/A
N/A = not available.
Radiochemistry
1
8
The synthesis of the precursor for the Fꢀradiosynthesis (53) was performed starting from 4,5ꢀ
dimethylthiazolꢀ2ꢀamine (6) by coupling with 2,2,3,3ꢀtetramethylcyclopropaneꢀ1ꢀcarboxylic acid
in presence of BOP reagent with a 18% yield. Alkylation of the thiazole 52 was performed by
using a large excess of 1,4ꢀdibromobutane (10 equiv) and deprotonation with NaH (2 equiv).
ACS Paragon Plus Environment

Page 13 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
8
Scheme 5. Synthesis of the precursor for radiolabeling 53 and radiosynthesis of [ F]29. a)
2
,2,3,3ꢀtetramethylcyclopropaneꢀ1ꢀcarboxylic acid, BOP, Et N, DCM, rt, 24 h, 18%; b) NaH,
3
1
8
Br(CH ) Br, DMF, rt, overnight, 56%; c) [ F]FꢀK2.2.2., CH CN, 80 °C.
2
4
3
18
Radiosynthesis of [ F]29 was achieved by aliphatic nucleophilic substitution of the
corresponding bromo precursor (53). Manual optimization of the radiosynthesis was performed
in order to select the best radiolabeling conditions prior to the transfer to an automated module
(e.g. solvent, temperature, heating method and amount of precursor for radiolabeling).
Low labeling yields (≤10%) were obtained when using DMF and DMSO (120ꢀ150 °C) as
solvents under conventional and microwaveꢀassisted radiofluorination (power cycling mode, 100
W, 120ꢀ150 °C) heating methods in presence of the reactive anhydrous K CO /K2.2.2. complex,
2
3
which is probably due to the observed decomposition of the bromo precursor 53 at short reaction
times (≤ 5 min). No significant increases in labeling efficiencies were noticed when a twoꢀfold
increase of precursor amount was used with DMF and DMSO as solvents. The best bromoꢀto
fluoro displacement was obtained when acetonitrile (CH CN) was used as the solvent at 80 °C
3
for 15 minutes (labeling yields ~30%) in the presence of the K CO /K2.2.2. complex (Scheme
2
3
18
ꢀ
5
). It is worth mentioning that under these conditions the maximum [ F]F incorporation yield
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
was reached with 3.5 mg of 53. The radioꢀthinꢀlayer chromatography (radioꢀTLC) and the radioꢀ
HPLC findings pointed out that besides the desired radiotracer only unreacted radiofluoride was
present in the reaction mixture.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
18
For animal experiments, the radiosynthesis of [ F]29 was translated to the automated module
Tracerlab FXFN and the total synthesis time was about 70 min (Scheme 5). For isolation of the
product, the reaction mixture was transferred onto a semiꢀpreparative HPLC column (entry 7,

Figure 3) followed by purification on a preꢀconditioned SepꢀPack C18 light cartridge (entry 11)
using EtOH as eluent. An injectable solution was obtained by evaporation of the solvent under a
gentle stream of nitrogen at 70 °C with a subsequent addition of a salineꢀalcohol mixture
1
8
(
0.9%:10% NaCl:EtOH). [ F]29 was obtained with a 16 ± 8.7% radiochemical yield, high
radiochemical purity (≥98%) and a specific activity of 169.7 ± 11.7 GBq/µmol at the end of the
18
synthesis. The identity of [ F]29 was confirmed by coꢀinjection of the corresponding reference
18
compound as depicted in Figure 4B. In vitro evaluation of stability proved that [ F]29 was stable
in saline and EtOH with more than 98% intact radiotracer remaining up to 90 min. An
18
experimental logD7,4 of 3.22 ± 0.03 was found for [ F]29, which is within the desired range for
compounds that will penetrate the bloodꢀbrain barrier.
ACS Paragon Plus Environment

Page 15 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
3
4
5
9
10
7
1
1
1
12
6
8
Figure 3. Scheme of the synthesis module Tracer Lab FXꢀFN for the automated radiosynthesis
1
8
®
ꢀ
of [ F]29. (1) Chromafix 30 PSꢀHCO ; (2) K CO (1.78 mg in 0.4 mL H O); (3) K2.2.2. (11.2
3
2
3
2
mg in 1 mL CH CN); (4) 3.5 mg 53 in 1 mL CH CN; (5) 3 mL H O/CH CN (1:1); (6) Reactor;
3
3
2
3
(7) ReprosilꢀPur C18ꢀAQ (65% CH CN/20 mM NH OAc; flow rate 4.2 mL/min); (8) 40 mL
3
4

H O; (9) 2 mL H O; (10) 1.25 mL EtOH; (11) Sep Pak C18 light cartridge; (12) Product in
2
2
NaCl 0.9% containing 10% EtOH.
A
B
1
8
[ F]29
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
8
Figure 4. (A) Representative semiꢀpreparative radioꢀHPLC profile of [ F]29 (Conditions:
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ReprosilꢀPur 120 C18ꢀAQ (5 µm, 250x10 mm), 65% CH CN/20 mM NH OAc aq., 4.2 mL/min);
3
4
1
8
(B) Analytical UVꢀ and radioꢀHPLC chromatogram of [ F]29 spiked with the corresponding
reference compound 29 (conditions: ReprosilꢀPur C18ꢀAQ (5 µm, 250x4.6 mm), 70%
CH CN/20 mM NH OAc aq., 1 mL/min). [(solid line, ɣꢀtrace);(grey line, UVꢀtrace)].
3
4
In vitro autoradiography studies: radiotracer characterization
18
[
F]29 was further evaluated in vitro by receptor autoradiography using rat spleen, an organ
59
which expresses high levels of CB under physiological conditions. For a displacement study,
2
18
tissue slices were incubated with [ F]29 without or with the CB ꢀselective inverse agonist Nꢀ
2
[
(1S)ꢀendoꢀ1,3,3ꢀtrimethylbicyclo
[2.2.1]ꢀheptaneꢀ2ꢀyl]ꢀ5ꢀ(4ꢀchloroꢀ3ꢀmethylphenyl)ꢀ1ꢀ[(4ꢀ
methylphenyl)methyl]ꢀ1Hꢀpyrazoleꢀ3ꢀcarboxamide (SR144528, 54), the CB ꢀselective
1
antagonist Nꢀ(piperidinꢀ1ꢀyl)ꢀ5ꢀ(4ꢀchlorophenyl)ꢀ1ꢀ(2,4ꢀdichlorophenyl)ꢀ4ꢀmethylꢀ1Hꢀpyrazoleꢀ
3
ꢀcarboxamide (SR141716A, 55) the CB /CB ꢀselective agonist (–)ꢀcisꢀ3ꢀ[2ꢀhydroxyꢀ4ꢀ(1,1ꢀ
1 2
dimethylheptyl)phenyl]ꢀtransꢀ4ꢀ(3ꢀhydroxypropyl)cyclohexanol (CP55,940, 56) or the nonꢀ
1
8
labelled reference compound 29. The autoradiograms of the binding of [ F]29 without and with
1
µM of the indicated compounds are presented in Figure 5A.
ACS Paragon Plus Environment

Page 17 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
8
Figure 5. [ F]29 selectively binds to CB receptors in vitro. (A) Rat spleen autoradiograms of
2
18
[
F]29 without and with coꢀincubation of the radioligand with 1 µM of CB ꢀspecific 54, CB ꢀ
2 1
specific 55, CB /CB ꢀspecific 56, and 29; (B) Evaluation of the specific binding and
1
2
18
displacement of [ F]29 on sections of rat spleen determined without and with coꢀincubation of
the radioligand with CB ꢀspecific 54 at 10 nM and 1 µM, CB ꢀspecific 55 at 10 nM and 1 µM,
2
1
CB /CB ꢀspecific 56 at 1 µM, and 29 at 1 µM.
1
2
18
The specific binding of [ F]29 in wild type spleen (100%) was displaced by the CB ꢀselective
2
compounds 54 at 10 nM (ꢀ8%) and 1 µM (ꢀ39%) as well as 1 µM of 56 (ꢀ56%), and the nonꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
labelled reference compound 29 (ꢀ70%), but remained unaffected by coꢀincubation with CB1ꢀ
targeting 55 at 10 nM and 1 µM (114% and 107%, respectively) (Figure 5B).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
18
In vivo metabolism of [ F]29
18
In vivo metabolism of [ F]29 in blood plasma and brain tissue was assessed in female CDꢀ1
mice. Blood samples and brain homogenates were obtained at 30 min post radiotracer injection
and proteins removed by organic solvent deproteinization. In general, extraction efficiencies of
radioactivity higher than 90% were achieved using MeOH/H O (9:1, v/v) as the extraction
2
solvent for both tissues.
1
8
The radiometabolite analysis of plasma reflects a rapid metabolism of [ F]29 with only 7% of
intact radiotracer accounting for the total plasma activity at 30 min post injection (Figure 6A).
Brain homogenates (Figure 6B) revealed that 36% of the extracted radioactivity was attributable
to the intact radiotracer.
A
Plasma, 30 min p.i.
B Brain, 30 min p.i.
18
[
F]29
6%
3
18
[
F]29
7%
Retention time [min]
Retention time [min]
18
Figure 6. In vivo metabolism of [ F]29 at 30 min p.i.. (A) Plasma samples [Extraction with
MeOH/H O (9:1); Extraction yield: 94%] and (B) Brain homogenates [Extraction with
2
ACS Paragon Plus Environment

Page 19 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
MeOH/H O (9:1); Extraction yield: 93%]. HPLC conditions: ReprosilꢀPur C18ꢀAQ (250x4.6
2
mm; 5 µm); Gradient mode (10ꢀ90% CH CN/20 mM NH OAc aq.).
3
4
PET imaging in healthy CD-1 mice
Dynamic small animal PET/MR scans under baseline conditions were performed in three
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
8
healthy female CDꢀ1 mice after a bolus i.v. injection of 12.0 ± 4.1 MBq of [ F]29. The in vivo
timeꢀactivity curves (TACs) (Figure 7A) reflect high uptake in liver and kidneys followed by a
fast washout while the activity concentration is constantly increasing in the intestines, urinary
bladder, and gallbladder. The summed whole body PET/MR image (Figure 7B) illustrates these
high uptakes for liver and urinary bladder. As region with high CB expression the spleen was
2
investigated separately. The respective TAC (Figure 8A) shows a clear uptake of activity in the
spleen (mean SUV, n=3 animals) followed by a washout. In addition, the spleen is visible in the
3
0 min PET and T1 weighted MR image (Figure 8B). Finally, blocking studies using the CB2ꢀ
1
8
specific compound 1 were performed to prove the specificity of [ F]29 towards CB . Preꢀ
2
blocking (3 mg 1/kg i.p. at 15 min before tracer administration and PET start) significantly
reduced the uptake of activity in the spleen (p<0.05, students t test) compared to the baseline
conditions (Figure 8C).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
8
Figure 7. Wholeꢀbody pharmacokinetics of [ F]29. (A) Small animal PET/MR derived TACs in
healthy female CDꢀ1 mice (n=3); (B) Left: Summed whole body PET image (0ꢀ60 min p.i.),
Middle: T1 weighted whole body MR image, Right: Fused PET/MR images of one
representative animal.
ACS Paragon Plus Environment

Page 21 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
8
Figure 8. Spleen uptake of [ F]29 (A) TAC of the spleen showing mean SUV (n=3) and
standard deviation; (B) Detailed view of the abdominal region (Left: PET; Middle: T1 weighted
MR; Right: PET/MR fusion) with main organs highlighted by arrows; (C) Spleen uptake under
baseline and preꢀblocking conditions (3 mg/kg 1, i.p. at 15 min before tracer) . Activity uptake at
18
1
0 min p.i. of [ F]29 is signifcantly reduced by preꢀblocking (p<0.05, students t test) .
1
8
PET [ F]29 imaging in a mouse model of neuroinflammation
1
8
PET scans with [ F]29 were performed in a lipopolysaccharide (LPS) mouse model of
neuroinflammation. In this model, mice are injected i.p. with 5 mg/kg LPS that result in a rapid
60
increase of brain TNFα that remains activated for a prolonged period of time.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In our study three pairs of mice (LPSꢀtreated and control) were scanned sideꢀbyꢀside after
1
8
the bolus iv administration of [ F]29 (7.4 – 11.1 MBq, specific activity >370 GBq/µmol) and
timeꢀactivity curves (whole brain) were generated (Figure 9A). When comparing the areaꢀunderꢀ
theꢀcurve (AUC) using the nonꢀparametric MannꢀWhitney test (one tailed), a statistically
significant difference (p=0.05) was observed between the control and LPS groups. The study
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
8
demonstrated significantly increased brain uptake of [ F]29 in the LPSꢀtreated animals versus
control animals. In a separate experiment using the same arrangement, the LPSꢀtreated mice
were preꢀinjected with the highly selective CB binding affinity agonist 1 (1 mg/kg, i.p.) and
2
compared to control animals injected with vehicle (Figure 9B). Almost identical average timeꢀ
activity curves were observed demonstrating no difference between the LPSꢀtreated animals preꢀ
injected with the CB agonist and the control animals. This result further demonstrates that the
2
18
elevated uptake of [ F]29 in the LPSꢀtreated animals in the baseline study is specific and
mediated by the CB receptor.
2
Controls baseline
LPS baseline
A
1.6
1.4
B 0.5
LPS
Control
0
.4
1
.2
1
0.3
0
.8
0.2
0.6
SUV (0 – 0.7 g/ml)
0
.4
.2
0
0.1
0
0
control-vehicle LPS-Block
0
10
20
30
40
Time, min
18
Figure 9. PET whole brain uptake of [ F]29 in LPSꢀtreated (red) and control (green) mice. (A)
1
8
Comparison of [ F]29 baseline uptake in LPSꢀtreated and control mice. There was a statistically
ACS Paragon Plus Environment

Page 23 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
significant difference between the LPSꢀtreated and control animals (p < 0.05, nonꢀparametric
MannꢀWhitney test (one tailed) of the areaꢀunderꢀcurve), Data: mean SUV ± SD (n = 3). Insert:
1
8
representative images of LPS and control mice (0 ꢀ 35 min); (B) Whole brain uptake of [ F]29
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
in the LPSꢀtreated mice is reduced with injection of CB agonist 1 (solution in 50% aq DMSO,
2
i.p.) to the nearly same level as that in control mice injected with vehicle. Data: averaged (10ꢀ30
min) SUV ± SD (n = 3).
DISCUSSION
1
1
Our previous studies with [ C]1 showed the potential of this radioligand to image
42
neuroinflammationꢀinduced increases of CB receptor densities in LPSꢀtreated mice and in a
2
43
mouse model of Aβ amyloidosis.
To achieve a PET radioligand with a longer halfꢀlife for CB receptor imaging in the brain, a
2
new series of 1 derivatives has been synthesized (Scheme 1). First, the influence of the thiazole
Nꢀalkyl chain length and atomic composition on the CB receptors binding affinity and
2
selectivity against CB receptors was investigated for the possibility to introduce the fluorine
1
atom. Large substituents like the hydrophobic triꢀ and tetraethylene glycols have been
implemented to increase hydrophilicity and also Nꢀalkyl ethers which are comparable in size to
1
. The role of the oxygen atom at the Nꢀalkyl chain was investigated by shifting its position and
also by excluding it in the case of the more lipophilic fluoroꢀethyl, ꢀpropyl and ꢀbutyl derivatives
27, 28 and 29 respectively.
The Nꢀalkylation of the 2ꢀaminoꢀ4,5ꢀdimethyl thiazole (Scheme 2) was generally performed at
elevated temperatures in sealed tube, neat, except for compounds 7 and 8, where DMF was used
to solubilize the reaction mixture since tosylates 34 and 35 are solid compounds. The resulting
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Nꢀalkylated compounds (7ꢀ16, Scheme 1) were obtained as salts. The amide formation step was
4
2
mediated by the BOP coupling reagent as earlier described (Scheme 1). In further efforts to
functionalize the structure of 1, modifications at the cyclopropyl and at the thiazole subunit were
performed (Scheme 3 and Scheme 4). As a result, alcohols 44 and 50 have been obtained
(Scheme 3 and Scheme 4 respectively) which are important key building blocks for further
derivatizations at these subunits.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Our structureꢀactivity relationship study shows that implementation of large substituents at the
thiazole nitrogen (20 and 21) cause a sharp decrease in CB binding affinity. By taking
2
advantage of the Nꢀalkyl ether chain of the lead 1, elongation to ethyl and propyl derivatives (15
and 19, respectively) kept the binding affinity while an average 9ꢀfold decrease in selectivity
towards CB was observed. Aside from the branched analogs 25 and 26, a CB binding affinity
1
2
in the same range as the lead compound (along with a decreased selectivity) was found when the
oxygen atom position was changed in the Nꢀalkyl chain (23 and 24).
Derivatization at both the cyclopropyl and thiazole 5ꢀpositions were not well tolerated as
reflected by the average 16ꢀfold decrease in binding affinity for the derivatives 45 and 50. The
replacement of the thiazole Nꢀmethoxyethyl with fluoroalkyl (e.g. fluoro Nꢀethyl (27), ꢀpropyl
(28) and ꢀbutyl (29) derivatives) resulted in a progressive increase in binding affinity and
selectivity with the chain length. As a result, the robust subnanomolar affinity of the thiazole Nꢀ
butylꢀ4ꢀfluoro derivative 29 (CB = 0.39 nM) in combination with the ≥900ꢀfold selectivity
2
towards CB and suitable lipophilicity (LogD = 3.22 ± 0.03) lead us to perform the
1
7.4
1
8
radiosynthesis and biological investigation of [ F]29.
The precursor for the radiosynthesis, (compound 53, Scheme 5) was obtained by first coupling
4,5ꢀdimethylthiazolꢀ2ꢀamine (6) with 2,2,3,3ꢀtetramethylcyclopropaneꢀ1ꢀcarboxylic acid in
ACS Paragon Plus Environment

Page 25 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
presence of BOP reagent followed by deprotonative Nꢀalkylation of the thiazole with 1,2ꢀ
dibromobutane in 10% yield over two steps.
1
8
A fully automated radiosynthesis of [ F]29 was successfully accomplished via aliphatic
nucleophilic substitution of the corresponding halo precursor 53 in high radiochemical purity
(≥98%) and specific activity of up to 1073 GBq/µmol at the end of the synthesis (Scheme 5).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
18
The target specificity of [ F]29 was further evaluated in autoradiographic studies in vitro
8
18
using CB expressing rat spleen. The distribution pattern of [ F]29 corresponds with the
2
3
binding of [ H]56 in the spleen, which is assumed to detect CB ꢀexpressing B lymphocytes in the
2
59
immune system of rats. Blocking experiments with the reference compound 29 and the CB ꢀ
2
18
selective inverse agonist 54 confirmed saturable and specific binding of [ F]29 to CB . Plasma
2
and brain samples were investigated at 30 minutes post injection via radioꢀHPLC and a fast
18
18
metabolism of [ F]29 in vivo was observed. Thus, at this time point ~7% intact [ F]29 was
found in plasma and ~36% in the brain samples.
Smallꢀanimal PET/MR experiments were performed in healthy female CDꢀ1 mice to
18
investigate the peripheral pharmacokinetics of [ F]29. The studies revealed high uptake in liver,
kidneys, intestines, urinary bladder, and gallbladder confirming a hepatic tracer excretion
pathway. Furthermore, a clear and marked uptake was found in the spleen as target specific
tissue. It could be blocked by the CB selective agonist 1.
2
18
Compound [ F]29 was studied with microꢀPET in both healthy CDꢀ1 mice and a murine
neuroinflammation mouse model induced by LPS. An elevated radiotracer uptake in the brain of
the LPSꢀtreated mice compared to control was observed. Moreover, blocking experiments
performed by preꢀinjecting the LPSꢀtreated mice with a CB selective ligand proved the CB2
2
specificity of the herein described radiotracer.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CONCLUSIONS
A novel series of highly CB affine derivatives of 1 has been developed for PET imaging.
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compound 29 demonstrated high CB binding affinity and selectivity, and therefore it was
2
1
8
18
selected for F radiolabeling. [ F]29 was successfully synthesized with high radiochemical
purity and specific activities. In vitro autoradiography findings using rat spleen as well as in vivo
smallꢀanimal PET/MR imaging confirmed CB specific binding. In addition, smallꢀanimal PET
2
1
8
studies showed that [ F]29 readily entered the mouse brain and specifically labeled brain CB
2
receptors in a mouse model of neuroinflammation. Rapid metabolism and the presence of brainꢀ
18
penetrating radiometabolites diminishes the potential of [ F]29 to image CB receptors with
2
PET.
EXPERIMENTAL
All reagents were used directly as obtained commercially unless otherwise noted. Reaction
progress was monitored by thinꢀlayer chromatography (TLC) using silica gel 60 F254 (0.040 −
0.063 mm) with detection by UV. All moistureꢀsensitive reactions were performed under an
argon atmosphere using ovenꢀdried glassware and anhydrous solvents. Column flash
chromatography was carried out using E. Merck silica gel 60F (230–400 mesh). Analytical TLC
was performed on aluminum sheets coated with silica gel 60 F254 (0.25 mm thickness, E.
Merck, Darmstadt, Germany). Melting points were determined with a FisherꢀJohns apparatus and
1
were not corrected. H NMR spectra were recorded with a Brukerꢀ400 NMR spectrometer at
nominal resonance frequencies of 400 MHz, in CDCl or DMSOꢀd (referenced to internal Me Si
3
6
4
at δH 0 ppm). The chemical shifts (δ) were expressed in parts per million (ppm). High resolution
ACS Paragon Plus Environment

Page 27 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mass spectra were recorded utilizing electrospray ionization (ESI) at the University of Notre
Dame Mass Spectrometry facility. All compounds that were tested in the biological assays were
analyzed by combustion analysis (CHN) to confirm a purity of >95%. Elemental analyses were
determined by Galbraith Laboratories (Knoxville, TN). A dose calibrator (Capintec 15R) was
used for all radioactivity measurements. Radiofluorination was performed with a modified GE
MicroLab radiochemistry box. The experimental animal protocols were approved by the Animal
Care and Use Committee of the Johns Hopkins Medical Institutions and by the respective State
of Saxony Animal Care and Use Committee in accordance with the German Law for the
Protection of Animals and the EU directive 2010/63/EU.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chemistry
General procedure 1. 2ꢀaminoꢀ4,5ꢀdimethylthiazole (2, 1 equiv) and appropriate alkylꢀ
bromide, ꢀmesylate or ꢀtosylate (see Scheme 2, 1.5 equiv) were reacted in a sealed vessel at 90
°
C for 16 hours. The resulting residue was triturated with minimum amount of iꢀPrOH, the solid
was filtered, dried under high vacuum, and used in the next step without further purification.
2,2,3,3ꢀtetramethylcyclopropane acid (1 mmol, 1 equiv), Et N (3 mmol, 3 equiv) and BOP (1.3
3
mmol, 1.3 equiv) were added to a suspension of the alkylated 2ꢀaminoimidazole (1 mmol, 1
equiv) in 5 mL DCM at 0 °C, and the mixture was stirred at room temperature for 20 hours. The
reaction was quenched by addition of 2 mL water followed by a 10 mL aqueous saturated
solution of NaHCO and 15 mL EtOAc. The phases were separated and the aqueous phase was
3
washed with 2x10 mL EtOAc. The combined organic fractions were washed with 20 mL brine,
dried over MgSO and concentrated by rotary evaporation. The obtained residue was purified by
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
column chromatography (silica, EtOAc:Petrolether, 1/9 to 1/4 for fluorinated and methoxylated
derivatives and 2/3 to 1/1 for alcohols).
General procedure 2. NaH was added (60% in mineral oil, 2 mmol, 2 equiv) to a solution of
alcohol 17 (1 mmol, 1 equiv), in 0.5 mL DMF, and the mixture was stirred for 5 minutes at room
temperature. Thereafter the alkylating agent RX (5 mmol, 5 equiv) was added, and the reaction
mixture was stirred overnight at room temperature. The reaction was quenched by addition of
H O (2 mL), followed by 15 mL aqueous saturated solution NaHCO and 20 mL EtOAc while
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
stirring. The phases were separated, the organic phase was washed with 20 mL EtOAc and the
combined organic fractions were dried over MgSO and concentrated under reduced pressure.
4
The resulting residue was subjected to column chromatography purification (silica,
EtOAc:Petrolether, 1/9 to 1/4).
(Z)-N-(3-(2-(2-Fluoroethoxy)ethyl)-4,5-dimethylthiazol-2(3H)-ylidene)-2,2,3,3-
1
tetramethyl-cyclopropanecarboxamide (18), General procedure 1, white solid, 48% yield; H
NMR (400 MHz, CDCl ): δ (ppm) = 1.22 (s, 6 H), 1.35 (s, 6 H), 1.53 (s, 1 H), 2.16 (s, 3 H), 2.22
3
(
s, 3 H), 3.65 (t, J=6.06 Hz, 2 H), 3.72 (t, J=5.31 Hz, 2 H), 4.28 (t, J=5.31 Hz, 2 H), 4.44 (t,
J=5.94 Hz, 1 H), 4.56 (t, J=5.94 Hz, 1 H); Elemental analysis: C 59.42, H 7.85, N 8.09, calcd. C
9.62, H 7.95, N 8.18
Z)-N-(3-(2-(3-Fluoropropoxy)ethyl)-4,5-dimethylthiazol-2(3H)-ylidene)-2,2,3,3-
5
(
1
tetramethyl-cyclopropanecarboxamide (19), General procedure 1, white solid, 43% yield; H
NMR (400 MHz, CDCl ): δ (ppm) = 1.22 (s, 6 H), 1.35 (s, 6 H), 1.53 (s, 1 H), 1.83 ꢀ 2.02 (m, 2
3
H), 2.17 (d, J=0.76 Hz, 3 H), 2.21 (d, J=0.76 Hz, 3 H), 3.54 (t, J=6.06 Hz, 2 H), 3.77 (t, J=5.31
13
Hz, 2 H), 4.26 (t, J=5.31 Hz, 2 H), 4.43 (t, J=5.94 Hz, 1 H), 4.55 (t, J=5.94 Hz, 1 H); C NMR
(
100 MHz, CDCl ): δ (ppm) = 11.41, 11.71, 17.00, 24.10, 30.20, 30.63, 30.83, 42.33, 42.35,
3
ACS Paragon Plus Environment

Page 29 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
6.30, 66.88, 68.49, 80.27, 81.90, 112.61, 128.32, 135.04, 181.49; HRMS (ESI+): m/z ( %)
+
+
=
357.2004, calcd. 357.2012 for C H FN O S [M+H] ; Elemental analysis: C 61.09, H 8.14, N
18 30 2 2
7.31, calcd. C 60.64, H 8.20, N 7.86
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
Z)-N-(3-(2-(2-(2-Fluoroethoxy)ethoxy)ethyl)-4,5-dimethylthiazol-2(3H)-ylidene)-2,2,3,3-
1
tetramethyl-cyclopropanecarboxamide (20), General procedure 1, white solid, 30% yield; H
NMR (400 MHz, CDCl ): δ (ppm) = 1.22 (s, 6 H), 1.35 (s, 6 H), 1.53 (s, 1 H), 2.16 (d, J=0.76
3
Hz, 3 H), 2.22 (d, J=1.01 Hz, 3 H), 3.58 ꢀ 3.64 (m, 4 H), 3.64 ꢀ 3.67 (m, 1 H), 3.70 ꢀ 3.78 (m, 1
H), 3.82 (t, J=5.31 Hz, 2 H), 4.27 (t, J=5.31 Hz, 2 H), 4.46 ꢀ 4.51 (m, 1 H), 4.56 ꢀ 4.66 (m, 1 H);
13
C NMR (100 MHz, CDCl ): δ (ppm) = 11.43, 11.70, 17.00 (2 C), 24.11, 30.19, 42.32, 46.27,
3
6
8.72, 70.27, 70.46, 70.57, 70.75, 82.28, 83.96, 112.50, 128.51, 164.37, 181.45; HRMS (ESI+):
+
+
m/z ( %) = 387.2127, calcd. 387.2118 for C H FN O S [M+H] ; Elemental analysis: C 59.53,
1
9
32
2
3
H 8.14, N 6.93, calcd. C 59.04, H 8.08, N 7.25
Z)-N-(3-(2-(2-(2-(2-Fluoroethoxy)ethoxy)ethoxy)ethyl)-4,5-dimethylthiazol-2(3H)-
(
ylidene)-2,2,3,3-tetramethyl-cyclopropanecarboxamide (21), General procedure 1, white
1
solid, 30% yield; H NMR (400 MHz, CDCl ): δ (ppm) = 1.22 (s, 6 H), 1.34 (s, 6 H), 1.53 (s, 1
3
H), 2.16 (d, J=1.01 Hz, 3 H), 2.22 (d, J=1.01 Hz, 3 H), 3.54 ꢀ 3.76 (m, 9 H), 3.76 ꢀ 3.86 (m, 3 H),
1
3
4
.26 (t, J=5.43 Hz, 2 H), 4.47 ꢀ 4.55 (m, 1 H), 4.58 ꢀ 4.68 (m, 1 H); C NMR (100 MHz,
CDCl ): δ (ppm) =11.46, 11.72, 17.01, 24.12 (4 C), 30.20, 42.32, 46.30, 68.71, 70.53, 70.61,
3
70.65, 70.84, 82.33, 84.01, 114.53, 127.26, 165.13, 181.10; HRMS (ESI+): m/z ( %) = 431.2392,
+
+
calcd. 431.2380 for C H FN O S [M+H] ; Elemental analysis: C 58.44, H 8.19, N, 6.71,
2
1
36
2
4
calcd. C 58.58, H 8.19, N 6.51
(Z)-N-(3-(3-Hydroxypropyl)-4,5-dimethylthiazol-2(3H)-ylidene)-2,2,3,3-tetramethyl-
1
cyclopropanecarboxamide (22), General procedure 1, white solid, 39% yield; H NMR (400
ACS Paragon Plus Environment