Identification and Preclinical Development of a 2,5,6-Trisubstituted Fluorinated Pyridine Derivative as a Radioligand for the Positron Emission Tomography Imaging of Cannabinoid Type 2 Receptors

Article abstract of DOI:10.1021/acs.jmedchem.0c00778

Despite the broad implications of the cannabinoid type 2 receptor (CB2) in neuroinflammatory processes, a suitable CB2-targeted probe is currently lacking in clinical routine. In this work, we synthesized 15 fluorinated pyridine derivatives and tested their binding affinities toward CB2 and CB1. With a sub-nanomolar affinity (Ki for CB2) of 0.8 nM and a remarkable selectivity factor of >12,000 over CB1, RoSMA-18-d6 exhibited outstanding in vitro performance characteristics and was radiofluorinated with an average radiochemical yield of 10.6 ± 3.8% (n = 16) and molar activities ranging from 52 to 65 GBq/μmol (radiochemical purity > 99%). [18F]RoSMA-18-d6 showed exceptional CB2 attributes as demonstrated by in vitro autoradiography, ex vivo biodistribution, and positron emission tomography (PET). Further, [18F]RoSMA-18-d6 was used to detect CB2 upregulation on postmortem human ALS spinal cord tissues. Overall, these results suggest that [18F]RoSMA-18-d6 is a promising CB2 PET radioligand for clinical translation.

Full text of DOI:10.1021/acs.jmedchem.0c00778

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Identification and Preclinical Development of a 2,5,6-Trisubstituted
Fluorinated Pyridine Derivative as a Radioligand for the Positron
Emission Tomography Imaging of Cannabinoid Type 2 Receptors
Ahmed Haider, Luca C. Gobbi, Julian Kretz, Christoph Ullmer, Andreas Brink, Michael Honer, Thomas
J. Woltering, Dieter Muri, Hans Iding, Markus Bürkler, Martin Binder, Christian Bartelmus, Irene
Knuesel, Pal Pacher, Adrienne Müller Herde, Francesco Spinelli, Hazem Ahmed, Kenneth Atz,
Claudia Keller, Markus Weber, Roger Schibli, Linjing Mu, Uwe Grether, and Simon Mensah Ametamey
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00778 • Publication Date (Web): 11 Aug 2020
Downloaded from pubs.acs.org on August 17, 2020
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 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 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.
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 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Identification and Preclinical Development of a 2,5,6-Trisubstituted
Fluorinated Pyridine Derivative as a Radioligand for the Positron Emission
Tomography Imaging of Cannabinoid Type 2 Receptors
1
2
2
2
2
2
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
Ahmed Haider , Luca Gobbi , Julian Kretz , Christoph Ullmer , Andreas Brink , Michael Honer , Thomas J.
2
2
3
2
2
2
2
Woltering , Dieter Muri , Hans Iding , Markus Bürkler , Martin Binder , Christian Bartelmus , Irene Knuesel ,
4
1
1
1
2
1
Pal Pacher , Adrienne Müller Herde , Francesco Spinelli , Hazem Ahmed , Kenneth Atz , Claudia Keller , Markus
5
1,6
1,6
2
1
Weber , Roger Schibli , Linjing Mu , Uwe Grether , Simon M. Ametamey *
1Institute of Pharmaceutical Sciences, ETH Zurich, Vladimir-Prelog-Weg 4, 8093 Zurich, Switzerland
2Pharma Research and Early Development, Roche Innovation Center Basel, F. Hoffmann-La Roche, 4070 Basel, Switzerland
3Pharma Technical Development, F. Hoffmann-La Roche Ltd., 4070 Basel, Switzerland
4Laboratory of Cardiovascular Physiology and Tissue Injury, National Institute of Health/NIAAA, 5625 Fishers Lane,
Rockville, Maryland 20852, USA
5Neuromuscular Diseases Unit/ALS Clinic, Kantonsspital St. Gallen, 9007 St. Gallen, Switzerland
6Department of Nuclear Medicine, University Hospital Zurich, 8091 Zurich, Switzerland
Abstract
Despite the broad implications of the cannabinoid type 2 receptor (CB2) in neuroinflammatory
processes, a suitable CB2-targeted probe is currently lacking in clinical routine. In this work, we
synthesized 15 fluorinated pyridine derivatives and tested their binding affinities towards CB2 and CB1.
With a subnanomolar affinity (K
over CB1, RoSMA-18-d exhibited outstanding in vitro performance characteristics and was
radiofluorinated with an average radiochemical yield of 10.6 ± 3.8% (n = 16) and molar activities
i
for CB2) of 0.8 nM and a remarkable selectivity factor of > 12’000
6
1
8
ranging from 52 - 65 GBq/µmol (radiochemical purity > 99%). [ F]RoSMA-18-d
6
showed exceptional
CB2 attributes as demonstrated by in vitro autoradiography, ex vivo biodistribution and positron
18
emission tomography (PET). Further, [ F]RoSMA-18-d was used to detect CB2 upregulation on post-
6
1
8
mortem human ALS spinal cord tissues. Overall, these results suggest that [ F]RoSMA-18-d
6
is a
promising CB2 PET radioligand for clinical translation.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 57
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
Introduction
Although the effects of cannabis sativa were known for centuries, the main pharmacologically active
1
component, Δ9-tetrahydrocannabinol (THC), was first isolated and characterized in 1964 . Notably, the
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
isolation of THC has prompted calls for extensive structure activity-relationship studies, ultimately
leading to the discovery of the cannabinoid receptors type 1 (CB1) and type 2 (CB2) in the early nineties
2, 3. Characterized by seven transmembrane domains, CB1 and CB2 belong to the rhodopsin-family of
G-protein coupled receptors (GPCRs) that mainly transduce extracellular signals via inhibition of
adenylyl cyclase (AC), thereby suppressing the cyclic adenosine monophosphate (cAMP) pathway 4.
Given the complexity of cannabinoid downstream signaling and the various interactions with other
signaling pathways, it is not surprising that cannabinoid receptor activation may result in diverse cellular
responses such as cell migration, proliferation, survival and apoptosis 5-7. Initial drug discovery activities
focused on the development of non-subtype selective agonists, however, their use was plagued by severe
psychotropic adverse events including memory loss, altered locomotor activity and impaired cognition
8-10.
These adverse events were primarily attributed to the high abundancy of CB1 in the central nervous
11
system (CNS) . The identification of CB2 as a key player in (neuro)inflammatory processes along with
its relatively low abundancy in the healthy mammalian CNS have channeled the development of CB2-
selective ligands 12, 13. Particularly, CB2-targeted agonists were anticipated to exert pharmacological
effects while being well tolerated 14-17. Further, CB2-selective positron emission tomography (PET)
imaging probes may allow the subclinical diagnosis of inflammatory conditions, thereby enabling
18
precision medicine . Notwithstanding the diversity of CB2-targeted probes in preclinical development,
no suitable CB2 PET radioligand is available in current clinical practice. Although the previously
11
reported CB2 PET radioligand [ C]NE40 has been tested in healthy volunteers and Alzheimer’s Disease
1
1
(
AD) patients, it was not possible to detect CB2 upregulation with [ C]NE40 in AD patients,
conceivably owing to off-target binding to CB1 19, 20. Other reported CB2 PET radioligands were deemed
unsuitable due to inappropriate pharmacokinetics, the presence of radiometabolites in the brain and high
lipophilicity as well as non-specific binding 21-28
.
2
ACS Paragon Plus Environment

Page 3 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
As part of our efforts to develop a suitable CB2 PET radioligand, we recently reported on a pyridine-
18
based probe, [ F]3 (Figure 1), that exhibited a CB2 affinity of 6 nM and a selectivity factor of nearly
29
18
7
00 over CB1 . In spite of the single digit nanomolar CB2 affinity, [ F]3 displayed poor in vivo
2
9
18
specificity and a fast washout from CB2-rich rat spleen . To overcome the shortcomings of [ F]3, and
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
based on knowledge gained from our previous study, we designed and synthesized 15 new derivatives
1
8
of [ F]3, which were obtained by structural modifications at position 6 of the pyridine core. These
1
8
structure activity-relationship (SAR) studies led to the identification of [ F]RoSMA-18-d
6
(Figure 1).
1
8
This new compound exhibited an 8-fold improved CB2 binding affinity of 0.8 nM compared to [ F]3
1
8
and a remarkable selectivity factor of > 12’000 over CB1. Preclinical evaluation of [ F]RoSMA-18-d
6
revealed a substantially improved CB2 specificity in vivo and a strikingly prolonged tracer retention in
the CB2-rich rat spleen, underlining the outstanding performance characteristics of this new radioligand
for CB2-targeted in vivo imaging.
1
8
18
Figure 1: Structures of [ F]3 (previous work) and [ F]RoSMA-18-d
6
(this work) with sites of modification indicated in
boxes at positions 2 (blue) and 6 (red).
Results
Structure-activity relationship (SAR)
Structural modifications of lead compound 3 were performed according to the simplified retrosynthetic
analysis depicted in Scheme 1. While methoxy-azetidine was introduced by Pd-catalyzed cross-
coupling to a bromo-pyridine precursor (retrosynthetic step 1), ester formation was accomplished via
the reaction of free carboxylic acids with the respective alkyl halides (retrosynthetic step 2). Carboxylic
acids were obtained by saponification of the respective ethyl esters (retrosynthetic step 3). Importantly,
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 57
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
the ethoxy moiety served as a protection group during amide formation that was typically achieved with
an activated ester using the coupling reagent TBTU (retrosynthetic step 4). Concurrently, the use of an
ethoxy protection group during amide formation obviated the necessity for retrosynthetic steps 2 and 3
for target compounds (1-8 and 15, Table 1) bearing an ethyl ester functionality. Alkyl fluoride groups
were introduced either by transforming an OH functionality into a sulfonate leaving group, followed by
nucleophilic substitution with fluoride or by direct treatment of the carboxylic acids with commercially
available fluorinated alkylating reagents.
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 1: General retrosynthetic scheme for the design of 2,5,6-trisubstituted pyridines depicted in Table 1.
From previous studies, we have established that a 3-methoxy-azetidine substituent at position 5, along
with an amide at position 2 of the pyridine core were key features that led to high CB2 binding affinities
29
for this class of compounds . Additionally, the adjacent diethyl moiety was important for CB2 affinity
and selectivity over CB1. The ester functionality was well tolerated and allowed fast derivatization of
29
the alkoxy side chain . Accordingly, the main focus of the present SAR study was to screen different
substituents at position 6 of the pyridine core in combination with either an ethoxy- or fluoroalkoxy side
chain at the ester functionality. One important aspect for compound design was the introduction of a
18
fluorine atom in one of these two positions to allow for F-labeling.
Cis and trans isomers (racemic, relative stereochemistry) of disubstituted cyclopropanes were
envisioned for substituents at position 6 of the pyridine core. Accordingly, disubstituted cyclopropanes
were prepared from commercially available diethyl cis-cyclopropane-1,2-dicarboxylate and diethyl
trans-cyclopropane-1,2-dicarboxylate, respectively (Supplemental Figure 1).
Starting from picolinic acid VI, target compounds 1-7 were synthesized either in five steps (compounds
1
, 2, 5 and 7) or in six steps (compounds 3, 4 and 6) as shown in Scheme 2. Initially, ethers III, VII and
4
ACS Paragon Plus Environment

Page 5 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
XII were obtained via nucleophilic aromatic substitution of the chlorine atom in VI with the appropriate
alcoholate. Subsequent amide couplings were accomplished via the activated ester using TBTU as
coupling reagent in chemical yields ranging from 44-78%. Buchwald-Hartwig cross-coupling conditions
were used to introduce 3-methoxy-azetidine, providing amides I, X and XV in chemical yields of 85%,
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
87% and 87%, respectively. Alcohol deprotections were performed either by treatment of benzyl-
bearing intermediates X and XV with Pd/C or under acidic conditions for the TBS-protected
intermediate I. Esterification of alcohols II, XI and XVI with fluoromethyl and fluoroethyl iodide,
respectively, gave final compounds 1, 2, 5 and 7 in chemical yields ranging from 10-37%. Fluorination
was carried out via the activated mesylate precursors V, IX and XIV, which were synthesized from the
respective alcohols in quantitative yields using excess mesyl chloride. Treatment of the mesylate
precursors with TBAF afforded target compounds 3, 4 and 6 in variable yields as depicted in Scheme
2
. Enantiopure key intermediates XVII and XIX were obtained via chiral resolution of XVI by HPLC
and were used for the synthesis of enantiopure target compounds 8-15 (Scheme 3). While the synthesis
of target structures 8 and 15 was achieved by direct fluoromethylation of the respective alcoholate, target
compounds 9-14 were obtained in two steps via ethyl ester hydrolysis and subsequent introduction of
fluoroalkoxy side chains of variable chain lengths (n = 1-3) as shown in Scheme 3. The precursor for
radiofluorination was obtained by reacting carboxylic acid XVIII with propyl tosylate. Further, a de
novo synthesis of enantiopure precursor was developed via the previously reported procedure of
30
diastereoselective cyclopropanation
(Supplemental Figures 2 and 3). The product of this
diastereoselective synthesis was used to confirm the absolute configuration of target compounds
depicted in Scheme 3 by comparison of retention times on chiral HPLC and light polarimetry
experiments (Supplemental information).
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 57
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
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
Scheme 2: Synthesis of 2,5,6-trisubstituted pyridines 1-7. Comp. 1-3, cis/racemic; Comp. 6 and 7, trans/racemic.
6
ACS Paragon Plus Environment

Page 7 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Binding affinities towards hCB2 and selectivities over hCB1 were determined for all final compounds
and are summarized in Table 1. In order to estimate blood-brain barrier penetration and non-specific
binding to brain tissue, logD7.4 and LIMBA logDBPL values were measured. Initially, we investigated
the influence of cyclopropyl functionalization either via fluoromethylation or by introduction of
fluoroalkoxy groups of variable chain lengths. These modifications afforded derivatives 1-3, 6-8 and 15
with high CB2 affinities in the range from 0.3-13 nM (Table 1). In sharp contrast, replacement of the
cyclopropyl substituent with a geminal dimethyl moiety led to a remarkable reduction in potency and
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
selectivity values, as evidenced by target compounds 4 (K
i
hCB2 > 10 μM) and 5 (K hCB2 = 143 nM).
i
Scheme 3: Synthesis of enantiopure 2,5,6-trisubstituted pyridines 8-15.
Notably, the stereochemistry of disubstituted cyclopropyl moieties significantly affected the CB2
binding affinity, with trans-racemic target compounds 6 and 7 outperforming their cis-racemic
stereoisomers 3 and 2 (Table 1, 6 vs. 3; K
i
hCB2 = 0.3 nM vs. 1 nM and 7 vs. 2; K
i
hCB2 = 4 nM vs.
13 nM). Despite exhibiting the highest CB2 binding affinity, target compound 6 displayed a relatively
low selectivity factor of 272 over CB1. Given the high abundancy of CB1 in the CNS, this compound
was not selected for radiolabeling. Bioisosteric substitution of the fluorine atom with a hydroxyl group
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 57
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
and concurrent introduction of fluoroalkoxy side chains of variable lengths via esterification led to a
remarkable increase in selectivity over CB1 while retaining the subnanomolar affinity for CB2, as
evidenced by the K values of derivatives 9, 10 and 11 (Table 1). Further, the results support the concept
i
that longer fluoroalkoxy side chains favor higher selectivity over CB1. Notably, while the S,S-
enantiomers 9-11 exhibited subnanomolar CB2 affinities, their respective R,R-enantiomers 12-14
displayed a significant reduction in CB2 affinity and selectivity over CB1.
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
i
Table 1: Binding affinity (K ), selectivity index and Lipophilicity (logD7.4 and LIMBA logD BPL) of 2,5,6-trisubstituted pyridines 1-15.
LIMBA
logDBPL
K
[µM]
i
hCB2
K
[µM]
i
hCB1
K
i
Cmp.
R
1
R
2
logD7.4
CB1/CB2
1
2
3.42
4.08
1.34
2.23
0.005
0.013
> 10
> 2173
2.981
0.590
229
675
3
4
5
6
3.06
2.45
2.64
0.001
> 10
*
*
n.d.
> 10
2.437
0.094
n.d.
*
*
n.d.
n.d.
0.143
0.0003
17
3.14
2.46
272
7
8
4.19
2.89
2.04
<0.1
0.004
0.002
2.585
0.806
703
4031
9
3.54
3.60
0.93
0.91
1.70
1.00
0.91
0.0004
0.0003
0.0007
0.010
0.547
1.601
> 10
1503
5337
1
1
0
1
3.90
> 14’286
167
(
RoSMA-18)
1
1
2
3
3.49
3.62
1.622
3.497
0.011
307
8
ACS Paragon Plus Environment

Page 9 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
4
3.80
3.59
1.16
1.88
0.019
0.008
> 10
> 532
224
5
1.828
*
not determined, Comp. 1-3, cis/racemic; Comp. 6 and 7, trans/racemic
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
Overall, the highest CB2 affinities were achieved for hydroxymethylated and fluoromethylated
cyclopropyl derivatives as evidenced by the hCB2 K -values of compounds 6 (0.3 nM), 10 (0.3 nM), 9
0.4 nM) and 11 (0.7 nM). Among all the newly designed compounds, 11 exhibited the highest
i
(
selectivity factor of > 12’000. With this remarkable affinity and selectivity, 11 provided the most
appropriate in vitro properties for evaluation as a potential CB2 PET radioligand. Hence, this structure
was selected for advanced profiling, radiolabeling and biological assessment.
Advanced profiling of target compound 11
The inhibitory constant (K
i
) and half maximal effective concentration (EC50) towards the CB2 receptor
were assessed for different species. Competitive binding assays were performed as previously
2
9
reported , and revealed single K
i
value estimations of 2.0 nM (mouse), 0.7 nM (rat) and 0.1 nM
(
cynomologus monkey), respectively, indicating that compound 11 maintains an excellent binding
affinity across species. Similarly, functional activity was determined for the CB1 and CB2 receptor
using Chinese hamster ovary (CHO) cells stably expressing CB2 in a validated cAMP assay29.
Compound 11 was found to be a potent inverse agonist for human CB2 as well as for rodent CB2 (data
not shown), while exhibiting virtually no activity on human CB1. To identify potential off-targets,
compound 11 was screened at high concentration (1 µM) against a customized panel of representative
31
proteins as previously described , and was found to be highly selective for the CB2 receptor (selectivity
screening included the following targets: CB1, adrenergic receptors α , β , adenosine A and A
receptor, angiotensin II receptor type 1, GABA receptor, cholecystokinin A receptor, Dopamine
receptor D , glycine receptor, muscarinic acetylcholine receptors M and M , μ- and κ-opioid receptor,
PCP receptor, serotonin 1A receptor, diltiazem-inducible calcium channel, monoamine oxidase A,
xanthine oxidase, HIV-1 protease, norepinephrine transporter, acetylcholinesterase, 5-HT2A/B and 5-HT
1
, α2A, β
1
2
1
3
A
1
1
2
3
receptor, serotonin transporter, glucocorticoid receptor, gelatinase B, peroxisome proliferator-activated
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 57
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
receptor γ, zeta-chain-associated protein kinase 70, prostaglandin F receptor, glycogen synthase kinase-
3
α and β, cyclin-dependent kinase 2, nicotinic acetylcholine receptors, tyrosine-protein kinase ABL1,
angiotensin-converting enzyme, phosphodiesterase 3 and 4D2, estrogen receptor α and cyclooxygenase
). In spite of its ester group, inverse agonist 11 is chemically stable in aqueous buffers at different pH
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
values (1, 4, 6.5, 8, 10) over a period of 2 h at 37 °C showing only slight decomposition at pH 1 (%
recovery after 2 h at pH: 1, 87%; 4, no degradation detected; 6.5, no degradation detected; 8, 97%; 10,
93%). The molecule is highly soluble (high throughput solubility = 86 µg/mL) and even more
importantly is capable of efficiently permeating through membranes as illustrated by a high PAMPA
permeation coefficient Peff of 5.02 cm/s*10E-6 (55.3% donor / 32.3% membrane / 12.2% acceptor
compartment) and favorable P-glycoprotein (P-gp) assay data (Papp 217 nm/sec). In the mouse P-gp
assay compound 11 exhibits an efflux ratio of 4.6, which suggests together with its high permeability
that crossing the blood brain barrier (BBB) in rodents is feasible. In humans, the P-gp efflux ratio is only
2.7, therefore it is expected that PET tracer candidate 11 would efficiently cross the BBB in humans.
An efflux ratio < 4 is considered favorable for brain penetration. In summary, PET tracer candidate 11
possesses all physicochemical and pharmacological characteristics that allow for radiolabelling and
assessment in vivo.
Docking Studies
In order to assess potential interactions within the active site of the CB2 receptor, inverse agonist 11
was docked into the CB2 receptor pocket, based on the X-ray structure of inactive state CB2 receptor
3
2
with antagonist AM10257 (PDB: 5ZTY) . The docking pose illustrates an excellent fit of the ligand
within the binding cavity and is supported by detailed structure-activity relationship information. In
addition to the previously reported pyridine-based PET radioligand [ F]3 29 the extra primary hydroxyl
group in position 2 undergoes a critical hydrogen bond interaction with Thr114 (Figure 2) which might
contribute to the subnanomolar binding affinity of target compound 11.
18
1
0
ACS Paragon Plus Environment

Page 11 of 57
Journal of Medicinal Chemistry
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
Figure 2: CB2 inverse agonist 11 docked into the binding cavity of inactive state CB2 X-ray structure (PDB: 5ZTY). A. Front
view, showing binding pose of docked compound 11. B. Side view, highlighting the (1S,2S)-hydroxy-cyclopropyl
configuration, which allows for critical hydrogen bond interaction with the hydroxy group of Thr114. In the foreground single
‘
‘toggle switch,’’ Trp6.48 (W2586.48) (orange sticks) is depicted.
Tritium-Labeling of Compound 11
Encouraged by the outstanding in vitro profile and docking findings of 11, and with the aim to assess
3
3
the performance characteristics of tritium-labeled [ H]11, codenamed [ H]RoSMA-18, by in vitro
autoradiography experiments, compound 11 was radiolabeled with tritium after condensing bis-vinyl
amino ester XXV and picolinic acid XXVII. The bis-vinyl group was inserted over two steps starting
form diphenylmethyleneglycine ethyl ester XXI. Two phenyl sulfoxide ethylene groups were introduced
via double Michael-Addition with KOtBu resulting in XXII. The bis-vinyl amino ester XXIII was
obtained by thermolytic elimination of the two phenyl sulfoxide groups. From there, amino ester XXV
was accessed via saponification followed by esterification with 3-fluoropropanol. Picolinic acid XXVII
was obtained by decomposition of carboxylic acid XVIII over 2 steps via Curtius rearrangement and
3
hydrolysis. Remarkably, attempts for a direct hydrolysis of amide XVIII failed. [ H]RoSMA-18-
precursor was then successfully synthesized via amide bond formation of picolinic acid XXVII and bis-
vinyl amino ester XXV.
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 57
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
3
Scheme 4: Synthesis of [ H]RoSMA-18-precursor via bis-vinyl amino ester XXIII and picolinic acid XXVII.
3
The tritiated probe [ H]RoSMA-18 was obtained by tritium enriched hydrogenation of the bis-olefin
3
precursor in DMF (Scheme 5). An amount of 4.59 GBq of the [ H]RoSMA-18 was obtained with a
radiochemical purity of 98.5% and a molar activity of 4.17 GBq/µmol, as determined by MS
spectrometry.
3
Scheme 5: Radiosynthesis of [ H]RoSMA-18.
Fluorine-18 Radiochemistry
The tosylate precursor was synthesized via esterification of XVIII as outlined in Scheme 3. The
18
radiofluorinated probe, codenamed [ F]RoSMA-18, was obtained by nucleophilic substitution using
[18
the radiochemical purity was > 99%. The decay-corrected radiochemical yield was 10.6 ± 3.8% (n =
6). The crude product was purified by semipreparative HPLC, trapped on a C-18 Sep-Pak cartridge,
F]KF/Kryptofix222 in acetonitrile (Scheme 6). Molar activities ranged from 52 - 65 GBq/µmol and
1
and eluted with ethanol. The final formulation for biological evaluations was prepared with 5% ethanol
in water for injection (WFI). The identity of the final product was confirmed by HPLC co-injection of
the respective non-radioactive reference compounds.
1
2
ACS Paragon Plus Environment

Page 13 of 57
Journal of Medicinal Chemistry
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
Scheme 6: Radiosynthesis of [ F]RoSMA-18.
LogD7.4 determination
18
A mean LogD7.4 value of 1.7 ± 0.4 (n = 4) was determined for [ F]RoSMA-18 using the shake-flask
3
3
method . The partition coefficient for non-radioactive reference 11 was assessed according to the
recently reported carrier mediated distribution system (CAMDIS) method and amounted to 3.85 33, 34
.
In vitro autoradiography
18
CB2 specificity of [ F]RoSMA-18 was assessed by in vitro autoradiography using CB2-positive rat
spleen tissues. In accordance with reported CB2 expression patterns in splenic lymphocytes, a high
spleen binding was observed under baseline conditions, which was significantly reduced by 71 ± 2%
(n = 4) following the addition of commercially available CB2 ligand GW-405,833 (10 µM), thus
18
confirming the specificity of [ F]RoSMA-18 towards CB2 (Figure 3A). The CB2 specificity was
3
further confirmed on rat and mouse spleen tissue with the tritium-labeled analogue, [ H]RoSMA-18
(
Figure 3B). Considering the relatively high CB2 Bmax of 697±30 fmol/mg in the rodent spleen, a Bmax/Ki
18
ratio of 996 can be calculated for [ F]RoSMA-18, while a ratio >10 is considered appropriate for
35
3
specific receptor imaging . Notably, when [ H]RoSMA-18 was employed on spleen tissue originating
from CB2 knockout (ko) mice, tissue binding virtually disappeared. These results underline the high
3
specificity of [ H]RoSMA-18 for CB2 receptors. In contrast to the spleen autoradiograms, CB2-negative
1
8
brain slices of healthy Wistar rats did not reveal any specific tissue binding of [ F]RoSMA-18 (Figure
4). The latter findings were expected given that CB2 is hardly detectable in the healthy brain (Bmax/Ki
<< 10).
To further elucidate the clinical relevance, post-mortem sections of human spinal cord tissue were
evaluated for their CB2 protein expression levels by in vitro autoradiography using post-mortem human
spinal cord tissues (Figure 5).
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 57
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
Figure 3: Representative in vitro autoradiography of the rodent spleen. A. Autoradiograms after incubation with either
[18
F]RoSMA-18 only or in combination with excess CB2 ligand GW-405,833 (10 µM). Quantification revealed a specificity
3
of 71±2% (n = 4). B. Autoradiograms after incubation with [ H]RoSMA-18 alone or in combination with excess CB2 ligand
CP55,940 (10 µM) confirmed the high CB2 specificity in rat (80% signal reduction) and mouse (92% signal reduction). The
specificity of the signal was confirmed by a 92% reduced binding to CB2 knockout mouse spleen tissue. wt, wild type. ko,
knockout.
Figure 4: Representative in vitro autoradiography of the Wistar rat brain. A. Autoradiograms after incubation with either
[18
F]RoSMA-18 only or in combination with excess CB2 ligand GW-405,833 (10 µM). B. Autoradiographic data quantification
revealed no evident specific tissue binding in the healthy rat brain (n = 4).
1
4
ACS Paragon Plus Environment

Page 15 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
18
An increased binding of [ F]RoSMA-18 was observed in the diseased tissue, providing evidence for
CB2 upregulation in neurodegenerative ALS patient’s spinal cord. Addition of GW-405,833 (10 µM)
led to a reduction of tracer binding confirming CB2 specificity. These findings are in accordance with
3
6
previous reports on activated CB2-overexpressing microglial cells in patients suffering from ALS .
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
18
Figure 5: Representative in vitro autoradiograms of [ F]RoSMA-18 on post-mortem human spinal cord sections (amyotrophic
lateral sclerosis - ALS - patient vs. healthy control). For blockade conditions, CB2 ligand GW-405,833 was used in a
1
8
concentration of 10 µM. Higher [ F]RoSMA-18 binding was found in the diseased tissue compared to the healthy control,
which was reduced under blockade conditions.
Ex vivo biodistribution
1
8
Encouraged by the remarkable CB2-specificity observed in the autoradiograms, [ F]RoSMA-18 was
administered to Wistar rats and organ biodistribution was determined 30 min post injection. Highest
tracer accumulation was detected in the intestine, urine and spleen. Intestinal and urinary uptake was
not reduced under blockade (GW-405,833, 2 mg/kg) conditions, suggesting that the clearance of
[18
F]RoSMA-18 and its radiometabolites is mainly via the hepatobiliary and renal pathways. In contrast,
radioactivity in the spleen was reduced by an average of 86 ± 2% (n = 4) under blockade conditions,
demonstrating specific target engagement in vivo (Figure 6). In accordance with autoradiographic
findings, there was no specific binding in the healthy Wistar rat brain in vivo. Specific binding was
further observed in the blood (60 ± 9%), the lung (50 ± 9%) and the femur (49 ± 18%). The latter findings
3
7
are in line with previous reports on CB2 expression in circulating leukocytes , bone marrow derived
B-lymphocytes 38 and lung-resident macrophages 39
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 57
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
18
Figure 6: Ex vivo biodistribution of [ F]RoSMA-18 in Wistar rats (n = 4). Animals were euthanized at 30 min post injection.
A specificity of 86 ± 2% (n = 4) was observed for the spleen (highlighted by red arrow). Results are presented as % normalized
injected dose per gram organ tissue. No specific binding was found in the brain. High unspecific uptake in urine and intestine
suggest a combination of renal as well as hepatobiliary elimination of radioactivity.
PET and in vivo displacement
1
8
Upon tail-vein injection of 12 - 17 MBq (0.14 - 0.48 nmol) [ F]RoSMA-18 into Wistar rats, time-
activity curves were determined for the spleen - presented as standardized uptake values (SUV) from 0-
60 min post injection (Figure 7). Radioactivity in the spleen reached a maximum SUV of 3.4 ± 0.3 (n
=
3) at 15 min post injection and remained constant for additional 30 min, followed by a slow washout.
To assess whether CB2 binding in the spleen is reversible, the CB2-specific partial agonist GW-405,833
1.5 mg/kg) was administered as a bolus 10 min after tracer administration. Significant tracer
(
displacement was observed consistently for all three animals treated with GW-405,833. At the last time
point measured, baseline animals revealed an averaged SUV of 2.9 ± 0.6 while the displacement group
exhibited a significantly reduced averaged SUV of 1.1 ± 0.5 (P = 0.016). These results show that
[18
F]RoSMA-18 binds specifically and reversibly to CB2 receptors in vivo.
1
6
ACS Paragon Plus Environment

Page 17 of 57
Journal of Medicinal Chemistry
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: Time-activity curves of [ F]RoSMA-18 are presented as standardized uptake values (SUV) ± standard error
(
STDERR) for the rat spleen under baseline and displacement conditions. GW-405,833 (1.5 mg/kg) was administered 10 min
after tracer injection (both intravenous). Significant displacement was observed for all tested animals (baseline n = 3,
displacement n = 3).
Metabolite studies
Ex vivo metabolite studies were performed with four male Wistar rats by intravenous injection of the
radiotracer (394 - 426 MBq) to determine intact tracer fraction in the plasma over time, and to assess
the presence or absence of radiometabolites in the brain. Plasma samples were harvested at the time
points of 5, 15, 30 and 45 min post injection. Fractions of intact parent radioligand in the plasma are
depicted in Table 2. Two polar radiometabolites were identified in the plasma, however, only intact
tracer was present in the brain samples, which were harvested at 5 and 45 min post injection. These
findings suggest that radiometabolites are formed outside the CNS and do not enter the brain.
18
Table 2: Percentage of intact [ F]RoSMA-18 in plasma and brain samples of Wistar rats at 0, 5, 15, 30 and 45 min p.i.
Plasma
Brain
[18F]RoSMA-18
Radio-
metabolites
1
8
Time
[ F]RoSMA-18
(%)
Radio-
(
min)
(%)
metabolites (%)
(
%)
0
5
100
69
0
not determined
31
54
69
74
100
-
-
15
46
not determined
not determined
3
4
0
5
31
26
100
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 57
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
Although RoSMA-18 was chemically stable in solutions of varying pH and at elevated temperatures,
18
the fluoropropyl in [ F]RoSMA-18 can potentially be cleaved by the action of enzymes, thereby
18
releasing 3-[ F]fluoropropanol or oxidized derivatives. These small radioactive fragments are
susceptible to defluorination, which can lead to an accumulation of radioactive fluoride in the bone
mineral matrix. In particular, the latter is critical for neuroimaging due to the close proximity of the skull
to adjacent brain regions such as the cortex or cerebellum. Radioactivity spill-over from the skull to
adjacent brain regions complicates data analysis. To assess potential defluorination, we investigated the
metabolic stability using microsomes and identified the carboxylic acid metabolite depicted in Figure
8 by LC-MS/MS. Notably, despite steric hinderance, the fluorine-bearing side chain was efficiently
cleaved by rat and human liver enzymes in vitro (Figure 8), suggesting that compound RoSMA-18 and
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
its radioactive analogue, [ F]RoSMA-18, may indeed be susceptible to in vivo defluorination.
Figure 8: Extracted mass chromatograms of RoSMA-18 and its dealkylated metabolite formed after 30 min of incubation
with (A) rat and (B) human liver microsomes.
In accordance with the metabolite identification results described above, we found substantial
radioactivity accumulation in ventral as well as dorsal skull segments during PET studies of
1
8
[ F]RoSMA-18 in Wistar rats (Figure 9A). It has been previously reported that terminal fluoropropoxy
moieties can be subjected to enzyme-mediated defluorination 40, 41. Initial studies indicated an oxidative
1
8
ACS Paragon Plus Environment

Page 19 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mechanism leading to a dealkylation of the ester. To overcome this limitation, we replaced all the
hydrogen atoms of the fluoropropyl side chain with deuterium atoms. The radiosynthesis of
[18
18
F]RoSMA-18-d
6
was performed under identical conditions as with non-deuterated [ F]RoSMA-18
(
Scheme 7). Indeed, the deuteration strategy led to a remarkable reduction of radioactivity uptake in
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
ventral and dorsal skull regions, confirming the hypothesis that a deuterated fluoropropyl side chain
reduces enzyme-mediated defluorination compared to its hydrogen-bearing analogue (Figure 9B).
1
8
6
Scheme 7: Radiosynthesis of [ F]RoSMA-18-d .
Due to the low abundancy of CB2 receptors in the healthy rat brain, however, the time-activity curve
18
revealed a fast washout of [ F]Ro-SMA-18-d (Supplemental Figure 5).
6
Figure 9: Coronal PET images of the head region averaged from 0-60 min post injection are presented after tail-vein injection
1
8
18
of either [ F]RoSMA-18 (panel A) or [ F]RoSMA-18-d
6
(panel B) into Wistar rats. Results are presented as standardized
uptake values (SUV). Lateral and dorsal skull regions are highlighted by arrows. Skull radioactivity is significantly reduced in
18
6
the animal scanned with [ F]RoSMA-18-d , indicating diminished defluorination of the deuterated tracer in vivo.
In vitro metabolite identification studies of RoSMA-18-d revealed a remarkably reduced dealkylation
6
by rat (Figure 10A) and human liver microsomes (Figure 10B), as compared to RoSMA-18.
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 57
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
Figure 10: Extracted mass chromatograms of RoSMA-18-d
6
and its dealkylated metabolite formed after 30 min of incubation
with (A) rat and (B) human liver microsomes.
Although isotopic exchange should not affect pharmacodynamic properties, we determined the binding
affinity of RoSMA-18-d
conditions as with RoSMA-18, an inhibitory constant (K
was found for the deuterated form, RoSMA-18-d , which is identical to the K
non-deuterated form. In addition, a K value of > 10 µM towards human CB1 was obtained for RoSMA-
8-d , thus confirming the high selectivity over CB1. For other species such as mouse, rat and
cynomologus monkey, the K values obtained for RoSMA-18-d were 1.2±0.3 nM, 0.3±0.3 nM and
.6±0.3 nM, respectively. Moreover, we extended the autoradiographic findings by showing that
6
for CB2 and its selectivity over CB1. Applying identical experimental
) of 0.8±0.4 nM (n = 4) towards human CB2
value of 0.7 nM for the
i
6
i
i
1
6
i
6
0
[18
F]RoSMA-18 and its deuterated analogue, [ F]RoSMA-18-d , provide essentially identical in vitro
18
6
autoradiograms. This was the case for the rat spleen and the rat brain, with substantial CB2 specificity
detected in the spleen but not in the brain (Figures 11 and 12, respectively).
2
0
ACS Paragon Plus Environment

Page 21 of 57
Journal of Medicinal Chemistry
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
18
Figure 11: Representative in vitro autoradiography of [ F]RoSMA-18-d
6
using rodent spleen. Autoradiograms after incubation
1
8
with either [ F]RoSMA-18-d
6
only or in combination with GW-405,833 (10 µM). Quantification of the data confirmed the
high CB2 specificity (n = 4).
1
8
Figure 12: Representative in vitro autoradiography of [ F]RoSMA-18-d
6
using rodent brain. A. Autoradiograms after
only or in combination with GW-405,833 (10 µM). B. Quantification of the data
confirmed the lack of CB2-specific binding in the healthy rat brain (n = 4).
1
8
incubation with either [ F]RoSMA-18-d
6
18
To further show that [ F]Ro-SMA-18-d can detect CB2 upregulation in the human spinal cord, we
6
conducted in vitro autoradiography experiments, thereby performing data quantification and statistical
analysis (n = 4). It was found that CB2 expression was significantly higher in post-mortem ALS spinal
cord tissues compared to the control tissues (Figure 13, quantification in the grey matter). This signal
was significantly reduced by blockade with either GW-405,833 or the previously reported selective CB2
ligand RSR-056 with the same structural scaffold (K
i
hCB2: 2.5 nM) 42. There was no significant
difference between the blockade by GW-405,833 and RSR-056, supporting the lack of off-target binding
1
8
by [ F]RoSMA-18-d .
6
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 57
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
6
Figure 13: Representative in vitro autoradiography of [ F]RoSMA-18-d using post-mortem human ALS and control spinal
18
6
cord tissue. Autoradiograms are presented after incubation with either [ F]RoSMA-18-d only or in combination with GW-
405,833 (10 µM) or RSR-056 (10 µM), respectively. Quantification of the data confirmed CB2 upregulation in the grey matter
of ALS patient tissues compared to healthy control sections (n = 4).
Discussion
Structure-activity relationship studies led to the discovery of RoSMA-18, a highly potent (K
i
for hCB2
= 0.7 nM) fluorinated CB2 ligand that displayed an outstanding selectivity factor of > 12’000 over CB1,
1
8
thereby substantially outperforming the previously reported lead compound [ F]3 (selectivity factor
29
<
700 over CB1) . The high selectivity over CB1 is particularly relevant with respect to CB2-targeted
brain imaging, since CB1 is known to be highly expressed in the CNS 43. Due to the promising
properties, RoSMA-18 was selected for radiofluorination and the respective tosylate precursor was
synthesized. Of note, our initial concerns about the diminishing role of a non-protected hydroxyl-group
of the tosylate precursor on radiosynthesis outcome proved to be unfounded. A robust radiosynthetic
procedure was established that yielded the designated radioligand in excellent pharmaceutical quality
for biological evaluation. Significant differences in the lipophilicity values of RoSMA-18 and the
18
respective radiofluorinated analogue, [ F]RoSMA-18, were observed. We attribute this discrepancy to
the different experimental methods used for the determination of the logD7.4 values. Whereas the logD7.4
value of RoSMA-18 was determined by the recently reported carrier mediated distribution system
CAMDIS) 33, 34, the value for [ F]RoSMA-18 was obtained by the widely used shake-flask method .
18
44
(
18
An identical average logD7.4 value of 1.7 was obtained by the shake-flask method for [ F]RoSMA-18
2
2
ACS Paragon Plus Environment

Page 23 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
8
and its deuterated analogue, [ F]RoSMA-18-d . This value compares well to previously reported brain
6
penetrating probes and lies within the recommended range of 1 – 3 suggested by Waring et al. 45.
[18
F]RoSMA-18 exhibited outstanding specificity for CB2-positive spleen by in vitro autoradiography,
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
in vivo PET and ex vivo biodistribution. Compared to the previously reported lead compound [ F]3,
[18
F]RoSMA-18 displayed a substantial increase in specific spleen binding by in vitro autoradiography
1
8
18
29
(
[ F]3 vs. [ F]RoSMA-18, 31% vs. 71%) as well as by in vivo PET displacement studies . Indeed,
1
8
18
spleen uptake was approximately 3-fold higher for [ F]RoSMA-18 than [ F]3 from 15-60 min after
1
8
radioligand injection into Wistar rats primarily because [ F]3 exhibited a rapid washout from the spleen,
1
8
while [ F]RoSMA-18 spleen binding remained at a high level during the whole scanning period of 60
2
9
min . In accordance with previous reports, we did not detect any CB2 in the healthy rodent brain,
46
neither in vitro nor in vivo . With regard to the clinical relevance of CB2-imaging, we were able to
18
show that [ F]RoSMA-18 can be used to image CB2 receptor upregulation in ALS using
autoradiography on post-mortem human spinal cord sample tissues. ALS is a fatal neurodegenerative
disorder affecting motor neurons and with no current effective treatment 47. Its etiology is not yet
understood, and early disease detection is hampered by a challenging differential diagnosis often leading
1
8
to false positive results. PET experiments with [ F]RoSMA-18 has the potential to become a valuable
tool for non-invasive patient diagnosis and further help to understand ALS disease etiology.
18
To study the in vivo stability of [ F]RoSMA-18, we analyzed plasma and brain samples at predefined
time points after radioligand injection into Wistar rats. Two polar radiometabolites were detected in the
18
plasma samples and only intact [ F]RoSMA-18 was found in the brain samples. We therefore conclude
that the radiometabolites are formed outside the CNS and do not cross the blood-brain barrier. This is
particularly relevant as radiometabolites complicate image analysis and thus are not desired in CNS-
based PET. Despite initial promising chemical stability results, a relatively high skull radioactivity
uptake was observed, possibly owing to in vivo enzymatic defluorination and subsequent accumulation
of radioactive fluoride in the bone mineral matrix. Consequently, a considerable spill-over effect was
observed, which was most pronounced in the cerebellar region. A strategy of deuteration was proposed
to overcome the limitation, and hence, all the hydrogen atoms in the fluoropropyl side chain were
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 57
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
replaced with deuterium atoms. This strategy abolished radioactivity uptake in lateral and dorsal skull
regions as confirmed by the respective PET images and supports the hypothesis that a deuterated
fluoropropyl side chain prevents enzyme-mediated defluorination. Metabolite identification studies
confirmed the improved stability of RoSMA-18-d towards enzymatic degradation. Further, the
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
subnanomolar affinity for CB2 and the outstanding selectivity over CB1 were confirmed for the
deuterated version, paving the way for the assessment of inflammatory CB2 upregulation in the
mammalian CNS.
Conclusion
In the present study, we performed a structure-activity relationship of fluorinated pyridine-based CB2
18
ligands starting from the previously reported lead compound [ F]3. RoSMA-18 emerged as the most
promising candidate, exhibiting a subnanomolar affinity (K
i
for hCB2 = 0.7 nM) and a remarkable
1
8
selectivity factor of > 12’000 over CB1, which substantially outperformed [ F]3 (K
i
for hCB2 = 6 nM
29
18
and selectivity factor of 698) . [ F]RoSMA-18 proved to be highly specific/selective in autoradiograms
1
8
of CB2-positive rat spleen. We further showed that [ F]RoSMA-18 can be used to image CB2 receptor
upregulation in ALS patients by autoradiography on post-mortem human spinal cord samples. In vivo
18
evaluation of the [ F]RoSMA-18 in the CB2-positive spleen revealed a specificity of 86 ± 2%. In vivo
18
defluorination of [ F]RoSMA-18 was circumvented by the replacement of all the hydrogen atoms in
18
the fluoropropyl side chain with deuterium atoms to afford [ F]RoSMA-18-d
6
. Overall, these results
18
suggest that [ F]RoSMA-18-d
6
is a promising CB2 PET radioligand for clinical translation.
Experimental section
Wistar rats were purchased from Charles River (Sulzfeld, Germany) and kept under standard conditions.
All animal studies were carried out in accordance with the Swiss Animal Protection Law as well as the
ARRIVE guidelines and further approved by the local veterinary office in Zurich, Switzerland. Post-
mortem human spinal cord tissue of healthy controls and ALS patients were kindly provided by the
Cantonal Hospital St. Gallen (Switzerland) with a written consent of the donors.
2
4
ACS Paragon Plus Environment

Page 25 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Reagents and solvents were purchased from Sigma Aldrich Chemie GmbH (Germany), Acros Organics
(
Belgium), ABCR GmbH (Germany), Merck (Germany) or Fluka (Switzerland) and were used without
1
further purification. A Bruker Avance FT-NMR spectrometer (300 or 600 MHz) was used to record H-
1
3
and C-NMR spectra and the chemical shifts (δ) are given in ppm referenced to tetramethylsilane (TMS,
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
0
ppm). Coupling constants (J) are presented in Hz and multiplicities in the H NMR spectra are reported
as either s = singlet, d = doublet, t = triplet, m = multiplet or br = broad peak. Mass spectrometry (MS)
was carried out on a Bruker's maXis (ESI-Qq-ToF-MS) spectrometer (Bruker Daltonik GmbH,
Germany) and MS results are reported in m/z. Purity of target compound was assessed by HPLC and
1
8
18
was at least > 95%. [ F]RoSMA-18 and [ F]RoSMA-18-d
Merck-Hitachi L2130 system), equipped with a radiation detector VRM 202 (Comecer, Netherlands)
in combination with an ACE 5 C-18-300 (250 x 10.0 mm, 5 µm) column and a gradient solvent system.
The following system was used for semipreparative purifications: 0.1% H PO in H O (solvent A),
MeCN (solvent B); 0.0-8.0 min, 20% B; 8.1-30.0 min, 20-90% B; 30.1-35.0 min, 90% B; 35.1-37.0 min,
0-20% B; 37.1-43.0 min, 20% B. A flow rate of was 4 mL/min was used. The UV signal detection was
6
were purified by semi-preparative HPLC
(
3
4
2
9
conducted at 230 nm. Analytical quality control was performed on an Agilent 1100 series HPLC system,
equipped with UV detector and a GabiStar radiodetector (Raytest). A reverse phase ACE C18-AR
column (50 x 4.6 mm, 3 mm) was used for analytical quality control, in combination with the following
separation conditions: 0.1% TFA in H O (solvent A), MeCN (solvent B); 0.0-2.0 min, 20% B; 2.1-12.0
2
min, 20-90% B; 12.1-14.0 min, 90% B; 14.1-15.0 min, 90-20% B; 15.1-20.0 min, 20% B. The flow rate
was 1 mL/min and the UV signal was recorded at 320 nm. Molar activities were calculated by comparing
the UV intensity of the formulated products with calibration curves of the respective standards.
Chemistry
General procedure for the ether formation reactions yielding compounds 1, 2, 5, 7, 8, 15:
To a solution of the respective alcohol in DMF (62-133 mM) was added NaH (2-5 equiv.) at 0 °C. The
resulting reaction mixture was stirred at ambient temperature for 30 min before the corresponding halide
(
5 equiv.) was added. The reaction mixture was stirred at RT for 0.5-28 h. The solvent was evaporated
under reduced pressure and the residue diluted with EtOAc. The organic phase was extracted with brine,
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 57
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
dried over Na
2
SO
4
and concentrated in vacuo. Column chromatography with variable eluent systems
consisting of heptane/EtOAc was performed to obtain the title compounds generally in low yields as
depicted in Schemes 1-2.
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
General procedure for fluorination reactions yielding compounds 3, 4 and 6:
A solution of the respective mesylate precursor in acetonitrile (12-37 mM) was combined with tetra
butyl ammonium fluoride in tetrahydrofuran (5 equiv.) under argon atmosphere. The resulting mixture
was stirred at 80 °C for 1-16 h. The reaction mixture was diluted with ethyl acetate and extracted 3x
with aqueous HCl (1M) and 1x with brine. Column chromatography with variable eluent systems
consisting of heptane/EtOAc was carried out to obtain the title compounds variable yields (Scheme 2).
General procedure for esterifications to give compounds 9, 10, 11, 12, 13, 14:
A solution of the appropriate carboxylic acid in DMF (40 mM) was combined with K
2
CO (3 equiv.),
3
followed by the addition of the respective halide (3-5 equiv.). The resulting mixture was stirred at room
temperature for 2-4 h. Volatiles were removed in vacuo and the resulting crude was diluted with EtOAc.
The organic layer was extracted with brine, dried over Na
2
SO and concentrated in vacuo. Silica column
4
chromatography with eluent systems of variable heptane/EtOAc combinations gave the title compounds
in yields that are shown in Scheme 3.
General procedure for esterifications yielding the precursor and precursor-d :
6
To a solution of the respective carboxylic acid in DMF (12-24 mM) was added potassium carbonate (3
equiv.) and the appropriate ditosylate (2-3 equiv.). The reaction mixture was stirred for 16-17 h at
ambient temperature, subsequently diluted with water and extracted 3x with ethyl acetate. The combined
organic layers were extracted with brine, dried and evaporated under reduced pressure. Silica column
chromatography with eluent systems of variable hexane/EtOAc combinations gave the title compounds.
General procedure for Buchwald-Hartwig aminations to give compounds I, X, XV:
To a solution of the appropriate brominated precursor in toluene (31-72 mM) was added the
corresponding amine (1.5 equiv.), Cs
2
CO
3
(3 equiv.), rac-BINAP (0.2 equiv.) and [Pd(OAc) ] (0.2
2
2
6
ACS Paragon Plus Environment

Page 27 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
equiv.) and the reaction mixture was stirred at 110 °C for 1-3 h. The reaction mixture was diluted with
EtOAc and filtered through Celite. Subsequently, the organic phase was washed 3x with aqueous HCl
(
1 M) and 1x with brine, then dried over Na
2
SO . Column chromatography was performed with gradient
4
elution using a system of heptane/EtOAc to give the title compounds in yields very good yields ranging
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
from 85-87% (Scheme 2).
General procedure for the formation of methoxy-pyridines III, VII, XII:
To a solution of the respective alcohol in DMF (53-141 mM) was added NaH (3 equiv.) at 0 °C. The
resulting reaction mixture was allowed to stir at ambient temperature for 30 min before 1.4-1.5 equiv.
of the respective alcohol was added and the reaction mixture was stirred at room temperature (or 80 °C)
for 3-16 h according to Scheme 2. The volatiles were removed under reduced pressure and the residue
diluted with EtOAc. The organic phase was extracted with brine, dried over Na
2
SO and concentrated
4
under reduced pressure. Column chromatography with variable eluent systems consisting of
heptane/EtOAc was performed to obtain the title compounds.
General procedure for the amide coupling reactions yielding compounds IV, VIII, XIII:
To the appropriate picolinic acid in DMF (16-228 mM) was added DIPEA (3-5 equiv.) and TBTU (1.0
– 1.2 equiv.), followed by the addition of ethyl 2-amino-2-ethylbutanoate (1.2-2.0 equiv.). Subsequently,
the reaction mixture was stirred at ambient temperature for 1-28 h, depending on starting material
consumption monitored by TLC. The volatiles were removed under reduced pressure and the resulting
crude was subsequently diluted with EtOAc. The organic layer was extracted with brine, dried over
Na
2
SO and concentrated under reduced pressure. Column chromatography using variable combinations
4
of heptane and EtOAc gave the title compounds as depicted in Scheme 2.
General procedure for saponification reactions yielding compounds XVIII, XX:
A solution of the appropriate ester in THF/MeOH/H O 1:1:1 (42-78 mM) was combined with KOH (5
2
equiv.) and the resulting mixture was stirred at 90 °C for 18 h. The organic solvents were removed under
reduced pressure. Subsequently, the aqueous phase was acidified (pH = 2) and extracted 3x with ethyl
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 57
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
acetate. The combined organic layers were extracted with brine and dried over Na
2
SO . The organic
4
phase was evaporated in vacuo and the crude residue was purified by column chromatography using
varying combinations of hexane/ethyl acetate to give the title compounds with excellent yields as shown
in Scheme 3.
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
General procedure for mesylation reactions yielding compounds V, IX, XIV:
A solution of the respective alcohol precursor in DCM (97-111 mM) was combined with triethylamine
(3 equiv.) and mesyl chloride (2 equiv.) at 0 °C. The reaction mixture was stirred at ambient temperature
for 0.5-2 h and subsequently diluted with ethyl acetate. The organic layers were washed 3x with
saturated sodium hydrogen carbonate and 1x with brine. The combined organic layers were dried over
Na
2
SO and evaporated in vacuo. The crude residue was purified by column chromatography using
4
combinations of hexane/ethyl acetate to give the title compounds in quantitative yields as shown in
Scheme 2.
General procedures for alcohol deprotection reactions yielding II, XI, XVI:
Compound II: Silyl ether I was dissolved in a mixture of acetic acid, water and tetrahydrofuran (3:1:1,
71 mM) and the mixture was stirred at ambient temperature for 1 h. The organic solvents were removed
under reduced pressure and the crude residue was diluted with ethyl acetate. The organic layers were
extracted 3x with saturated aqueous NaHCO
3
and 1x with brine, dried over Na
2
SO and concentrated in
4
vacuo. The product was used without further purification for the subsequent step (purity by analytical
HPLC: 100%).
Compounds XI and XVI: To a solution of the respective ether in ethyl acetate/methanol (10:1, 37-61
mM) was added palladium on carbon (Pd-C, 4-5 equiv.) and the suspension was stirred under H
2
atmosphere for 48 h. The reaction mixture was filtered through Celite and the solvents were removed
under reduced pressure. The product was used without further purification for the subsequent step
(purity by analytical HPLC: 95-100%).
2
8
ACS Paragon Plus Environment

Page 29 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Ethyl 2-ethyl-2-(6-(((cis)-2-((fluoromethoxy)methyl)cyclopropyl)methoxy)-5-(3-methoxyazetidin-
1
-yl)picolinamido)butanoate (1)
1H NMR (600 MHz, Chloroform-d) δ 8.63 (s, 1H), 7.64 (d, J = 7.9 Hz, 1H), 6.60 (d, J = 7.9 Hz, 1H),
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
3
0
.32 (ddd, J = 56.8, 9.0, 1.9 Hz, 2H), 4.68 – 4.52 (m, 1H), 4.37 – 4.20 (m, 6H), 4.01 – 3.71 (m, 4H),
.34 (s, 3H), 2.70 – 2.58 – 2.50 (m, 2H), 1.94 – 1.80 (m, 2H), 1.38 – 1.25 (m, 5H), 1.01 – 0.93 (m, 1H),
+
.82 – 0.72 (m, 6H), 0.55 – 0.42 (m, 1H). LC-MS (ESI): 482.2657 [M+H ]; calculated for C24
H36FN O
3 6
+
+
H : 482.2666.
Ethyl 2-ethyl-2-(6-(((cis)-2-((2-fluoroethoxy)methyl)cyclopropyl)methoxy)-5-(3-methoxyazetidin-
1-yl)picolinamido)butanoate (2)
1H NMR (600 MHz, Chloroform-d) δ 8.63 (s, 1H), 7.64 (d, J = 7.9 Hz, 1H), 6.60 (d, J = 7.9 Hz, 1H),
4
3
0
.55 (dt, J = 47.7, 4.1 Hz, 2H), 4.40 – 4.19 (m, 7H), 3.68 (dt, J = 29.6, 4.2 Hz, 2H), 3.59 – 3.50 (m, 4H),
.34 (s, 3H), 2.70 – 2.54 (m, 2H), 1.94 – 1.80 (m, 2H), 1.53 – 1.39 (m, 1H), 1.38 – 1.25 (m, 4H), 0.98 –
.90 (m, 1H), 0.76 (t, J = 7.3 Hz, 6H), 0.50 – 0.41 (m, 1H). LC-MS (ESI): 496.3 [M+H+].
Ethyl 2-ethyl-2-(6-(((cis)-2-(fluoromethyl)cyclopropyl)methoxy)-5-(3-methoxyazetidin-1-
yl)picolinamido)butanoate (3)
1H NMR (600 MHz, Chloroform-d) δ 8.62 (s, 1H), 7.64 (d, J = 7.9 Hz, 1H), 6.60 (d, J = 7.9 Hz, 1H),
4.75 – 4.45 (m, 3H), 4.32 – 4.21 (m, 6H), 3.92 – 3.82 (m, 2H), 3.34 (s, 3H), 2.64 – 2.57 (m, 2H), 1.91 –
1.84 (m, 2H), 1.63 – 1.59 (m, 1H), 1.49 – 1.43 (m, 1H), 1.32 (t, J = 7.3 Hz, 3H), 1.02 – 0.95 (m, 1H),
+
0
.79 – 0.74 (m, 6H), 0.57 – 0.52 (m, 1H). LC-MS (ESI): 452.2567 [M+H ]; calculated for C23
H34FN O
3 5
+
+
H : 452.2561.
Ethyl 2-ethyl-2-(6-(3-fluoro-2,2-dimethylpropoxy)-5-(3-methoxyazetidin-1-
yl)picolinamido)butanoate (4)
1H NMR (600 MHz, Chloroform-d) δ 8.63 (s, 1H), 7.64 (d, J = 7.9 Hz, 1H), 6.60 (d, J = 7.9 Hz, 1H),
.52 – 4.09 (m, 9H), 3.96 – 3.81 (m, 2H), 3.34 (s, 3H), 2.58 (dt, J = 14.7, 7.5 Hz, 1H), 1.88 (dq, J =
4
2
9
ACS Paragon Plus Environment