Synthesis and evaluation of three structurally related 18F- labeled orvinols of different intrinsic activities: 6-O-[18F] fluoroethyl-diprenorphine ([18F]FDPN), 6-O-[18F] fluoroethyl-buprenorphine ([18F]FBPN), and 6-O-[18F] fluoroethyl-phenethyl-orvinol ([18F]FPEO)

Article abstract of DOI:10.1021/jm500503k

We report the synthesis and biological evaluation of a triplet of 6-O- ¹⁸F-fluoroethylated derivatives of structurally related orvinols that span across the full range of intrinsic activities, the antagonist diprenorphine, the partial agonist buprenorphine, and the full agonist phenethyl-orvinol. [¹⁸F]fluoroethyl-diprenorphine, [ ¹⁸F]fluoroethyl-buprenorphine, and [¹⁸F]fluoroethyl- phenethyl-orvinol were prepared in high yields and quality from their 6-O-desmethyl-precursors. The results indicate suitable properties of the three 6-O-¹⁸F-fluoroethylated derivatives as functional analogues to the native carbon-11 labeled versions with similar pharmacological properties.

Full text of DOI:10.1021/jm500503k

Brief Article
pubs.acs.org/jmc
18
Synthesis and Evaluation of Three Structurally Related F‑Labeled
Orvinols of Different Intrinsic Activities: 6‑O‑[18F]Fluoroethyl-
diprenorphine ([18F]FDPN), 6‑O‑[18F]Fluoroethyl-buprenorphine
([18F]FBPN), and 6‑O‑[18F]Fluoroethyl-phenethyl-orvinol ([18F]FPEO)
Bent W. Schoultz,*,† Trine Hjørnevik,‡,§ Brian J. Reed,†,‡ Janos Marton,∥ Christopher S. Coello,‡
́
Frode Willoch,‡,⊥ and Gjermund Henriksen#,‡
†Department of Chemistry, University of Oslo, P.O. Box 1033, Blindern, N-0315 Oslo, Norway
‡Institute of Basic Medical Sciences, University of Oslo, P.O. Box 1110, Blindern, N-0317 Oslo, Norway
§The Intervention Centre, Oslo University Hospital, Oslo, Norway
∥ABX Advanced Biochemical Compounds, Biomedizinische Forschungsreagenzien GmbH, Heinrich-Glaeser-Strasse 10-14, D-01454
Radeberg, Germany
⊥Aleris AS, Oslo, Frederik Stangs Gate 11/13, N-0264 Oslo, Norway
#Norwegian Medical Cyclotron Centre, P.O. Box 4950, Nydalen, 0424 Oslo, Norway
S
* Supporting Information
ABSTRACT: We report the synthesis and biological evaluation
of a triplet of 6-O-18F-fluoroethylated derivatives of structurally
related orvinols that span across the full range of intrinsic
activities, the antagonist diprenorphine, the partial agonist
buprenorphine, and the full agonist phenethyl-orvinol. [18F]-
fluoroethyl-diprenorphine, [18F]fluoroethyl-buprenorphine, and
[18F]fluoroethyl-phenethyl-orvinol were prepared in high yields
and quality from their 6-O-desmethyl-precursors. The results
indicate suitable properties of the three 6-O-18F-fluoroethylated derivatives as functional analogues to the native carbon-11
labeled versions with similar pharmacological properties.
antagonist properties, the intrinsic activity of the tracer, have so
far not been systematically studied on these parameters.
INTRODUCTION
■
Positron emission tomography (PET) imaging of opioid
receptors (OR) with the nonselective antagonist [11C]-
diprenorphine ([11C]1a) has a long history in measuring OR
and its functions in brain.1 A large variety of PET studies for
detecting OR occupancy with [11C]1a and [18F]fluoroethyl-
diprenorphine ([18F]5a) are reported, e.g., within measure-
ments on conditions of experimental pain and emotional
physiological states like euphoria.2−4 Variable sensitivities have
been reported for the measurements of OR occupancy caused
by the release of endogenous endorphin peptides. On the other
hand, consequently low sensitivity is reported from studies in
detecting changes to the availability of ORs from competition
and OR occupation from exogenous potent opioid.5,6
Current tracer design of opioid receptor tracers for use in
PET aims at optimizing the sensitivity of the tracer for
detecting changes to the availability of receptors from
physiological, physical, or pharmacological stimuli relative to
control conditions. The effects of tracer affinity and subclass
selectivity on the tracer sensitivity toward endogenous and
exogenous opioid receptor binding compounds have been
addressed.5,6 In contrast, the effects of the tracer agonist or
The ORs belong to the super family of G protein-coupled
receptors that exist in two different conformational states,
corresponding to the coupled or high affinity state and
uncoupled or low affinity state. This suggests that agonists
preferentially bind with higher affinity to the receptor when in
the coupled state, while antagonists bind with the same affinity
to both, independent of the coupling status as reported for
dopamine receptors.7 Current available OR PET tracers
possessing different intrinsic activity originate from different
compound classes. Investigations on the effects of intrinsic
activity by these tracers could be difficult because tracers with
different structures are likely to have different binding pattern
toward the receptor pocket which potentially could shade for
apparent binding selectivity toward OR coupling states.
Investigations with structurally related PET tracers possessing
different intrinsic activity could clarify these relations and be
useful in optimizing PET tracers selectivity to obtain more
Received: March 31, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm500503k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Brief Article
specific physiological information about the presence of OR
states by PET imaging.
BPN (1b), and PEO (1c), are available by the original method
of Bentley8,9 and more recently developed procedures.12,17−20
6-O-Desmethyl derivatives DDPN (2a), DBPN (2b), and
DPEO (2c) and the 3-O-trityl protected precursor compounds
TDDPN (3a), TDBPN (3b), and TDPEO (3c) were prepared
according to the method of Luthra et al.11 previously used to
provide precursor compounds for production of the [6-O-
methyl-11C]labeled versions of PEO, BPN, and DPN.
For provision of 6-O-fluoroethylated orvinol reference
standards (5a−c), the 3-O-trityl-protected-6-O-desmethyl
precursors (3a−c) were treated with fluorethyl bromide
(Scheme 1). Selective 6-O-alkylation was accomplished using
2.33 equiv of 2-fluoroethyl-bromide in the presence of 10 equiv
of sodium hydride in DMF at room temperature.
For compounds 3a and 3c, the 6-O-fluoroethylated products
F-TDDPN (4a) and F-TDPEO (4c) were isolated in good
yields (65% and 64%) with no formation of orvinol side
products. Under identical conditions, a 6-O-vinyl byproduct, V-
TDBPN (4d), was formed in 20% yield during the alkylation of
TDBPN (3b) with 2-fluoroethyl-bromide.
Finally, the reference substances 6-O-fluoroethyl-DPN
(FDPN, 5a), 6-O-fluoroethyl-BPN (FBPN, 5b), and 6-O-
fluoroethyl-PEO (FPEO, 5c) were achieved by deprotection of
the corresponding 3-O-trityl derivatives (4a−c) using boiling
aqueous acetic acid.21
The antagonist [11C]1a belongs to the class of orvinols
established by Bentley and his associates at the company
Reckitt and Colman in the 1960s.8−10 Previously, together with
the earlier established partial antagonist [11C]buprenorphine
([11C]1b),11 we completed the spectrum of structurally related
orvinol PET tracers possessing all intrinsic activities by the
provision of the agonist [11C]phenethyl-orvinol ([11C]1c).12
The availability and flexibility to design PET imaging
protocol with this orvinol triplet could benefit from a shift
from the short-lived carbon-11 label (t1/2 = 20.3 min) to the
longer lived fluoride-18 (t1/2 = 109.7 min). Recently, we
reported the production of 6-O-[18F]fluoroethyl-diprenorphine
([18F]5a) based on a new method for fully automated provision
of high quality 2-[18F]fluoroethyl-tosylate.13 With 6-O-[18F]-
fluoroethyl-phenethyl-orvinol ([18F]5c) and [18F]5a previously
reported with different production protocols,14−16 we here
aimed for production of 6-O-[18F]fluoroethyl-buprenorphine
([18F]5b), not previously reported, and to provide the triplet of
orvinols ([11C]1a−c) as fluoride-18 versions ([18F]5a−c) with
a single production protocol (Figure 1).
The radiosynthesis of [18F]5a−c was accomplished by 18F-
fluoralkylation of the 6-O-desmethyl 3-O-trityl protected
orvinol precursors (3a−c) with 2-[18F]fluoroethyl-tosylate
([18F]FETos) in a two-pot, three-step synthesis. In this work,
we used the same method (Scheme 2) and set up for the
automated production of the three 6-O-[18F]fluoroethylated
orvinols as previously described for the production of [18F]5a.13
Scheme 2. Preparation of 18F-Labeled Orvinols by Three-
Step, Two-Pot 18F-Fluoroalkylation
Figure 1. Three 6-O-18F-fluoroethylated orvinol PET tracers with
different intrinsic activity.
RESULTS AND DISCUSSION
The starting materials for the synthesis of the reference
standards and precursor compounds (Scheme 1) DPN (1a),
■
Scheme 1. Synthesis of Orvinol Precursors and Fluorine-18
and Fluorine-19 Labeled Orvinol Drivatives
The set up for the automated synthesis of the 18F-labeled
orvinols comprised the following processes: (1) fixation, wash,
and release of [18F]fluoride from target water on strong-anion
exchange SPE cartridge before (2) drying and formation of the
[K+ ⊂ 2.2.2 ]18F- complex, (3) synthesis of [18F]FETos, (4)
isolation and drying of [18F]FETos, (5) 18F-fluoro-ethylation
and (6) deprotection of the 3-O-trityl moiety from the 18F-
fluorethylated intermediates ([18F]4a−c), and finally (7)
isolation by means of preparative HPLC of [18F]5a−c before
SPE fixation, release with EtOH, and finally formulation in PBS.
Preparative isolated yield of [18F]FETos was >45% with
>99% in radiochemical purity (TLC and HPLC). 18F-
Fluoralkylation of 3a−c provided >90% in radiochemical
yield (based on [18F]FETos) with >99% in yield for the
following deprotection (TLC/HPLC). Starting from 10 to 30
GBq of [18F]fluoride, the decay corrected formulated product
yield was 26% 8% for all three products (n > 90). The total
radiosynthesis was completed in a synthesis time of ∼100 min.
B
dx.doi.org/10.1021/jm500503k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Brief Article
The identity of [18F]5a−c was confirmed by coelution with
the respective authentic reference 5a−c by means of HPLC.
Final quality control of [18F]5a−c revealed a specific activity in
the range 50−300 GBq/μmol (HPLC) depending on the
starting specific activity.
Animal PET. In vivo brain imaging with PET was performed
with [18F]5a−c in rats together with naloxone for blocking of
each tracer. Figure 2 shows the measured uptake kinetics for all
In separate experiments, we investigated the effect of
blocking the ORs with the nonselective antagonist naloxone,
which was injected 15 min prior to the three fluoride-18 labeled
orvinol tracers (Figure 2A−2C). The measurement of activity
in brain revealed the same fast and high initial uptake of all
three tracers as seen without any pretreatment, followed by a
fast washout of activity from all brain regions. The uptake of all
three tracers in basal ganglia approached that of cerebellum,
indicating that the binding of the tracers is specifically to opioid
receptors.
Ratios between basal ganglia and cerebellum uptake (Table
1) revealed the highest standard uptake values (SUV) ratio for
Table 1. Binding Ratios with Test Retest for [18F]5a−c in
a
Rat Brain (Mean SD)
orvinol
binding ratio
test/retest
[18F]5a
[18F]5b
[18F]5c
5.03 0.30
3.01 0.26
2.82 0.15
4.64/4.97
3.19/3.24
2.78/2.83
a
Left column shows binding ratio between basal ganglia and
cerebellum for [18F]5a−c with average SUV and SD from 55 to 85
min post injection (n = 4−8).
[18F]5a, followed by that of [18F]5b and of [18F]5c. Also,
intrasubject variability of SUV is interestingly low (CoV (%):
6.0%, 8.6%, and 5.3% for [18F]5a−c, respectively).
Test retest measurement was performed by injection of [18F]
5a−c to the same animal with minimum 7 days apart for
investigating the reproduction of ratios of measured SUV of
basal ganglia and cerebellum. The results revealed a high degree
of reproducibility within subject for all three tracers.
Characterization. For characterization and biological
evaluation of [18F]5a−c, and to compare physiochemical
properties and structure−activity relations (SAR) caused by
altering the structure of the native 6-O-methylated orvinol by a
6-O-fluoroethylation, [18F]5a−c and 5a−c were investigated in
different in vitro and in vivo assays.
The octanol/PBS partition coefficient (log P octanol/PBS)
of [18F]5a−c was measured to determine the apparent
lipophilicity. log P for [18F]5a−c was measured to be 2.20
0.03, 3.08 0.07, and 2.92 0.15, respectively. The measured
log P for [18F]5c represented an 22% increase in value
compared to the earlier measured value for the native drivative
[11C]1c. The observed increase in lipophilicity for [18F]5c is in
accordance with the effect of an extended alkyl moiety from the
substitution of 6-O-methyl with a 6-O-fluoroethyl group. All
measured log P values are within the range of optimal
lipophilicity for PET tracers (log P optimal range =2.0−3.5)
for crossing the blood−brain barrier which is a prerequisite to
reach OR targets in the CNS.22
Figure 2. Regional time−activity curve of [18F]5a (A), [18F]5b (B),
and [18F]5c (C) with number (n) of rats: n = 4, 8, and 6, respectively.
For Naloxone n = 1.
In Vitro Receptor Binding Assay. The opioid receptor
affinities of all three fluoroethylated orvinols (5a−c) toward μ,
δ, and κ-ORs were determined in radioligand binding assay in
vitro using human cloned receptors stably expressed on CHO
cells (μ- and δ-OR) and HEK-293 (κ-OR) according to
published procedures. The results are given in Table 2 as Ki
values (nM), together with literature values for the native 6-O-
methyl orvinol versions.
The results from the in vitro binding study in Table 2
revealed similar high affinities for human cloned μ-OR to κ-OR
for the native and the 6-O-fluoroethylated orvinol triplet. 5c
was found to possess similar high affinity binding to human
tracers of the μ-OR high density region, the basal ganglia, and
the μ-OR low density region, the cerebellum. All data (mean
SD) are normalized for injected dose and body weight. [18F]
5a−c showed fast brain uptake and specific accumulation in
basal ganglia. The measured uptake kinetics of [18F]5b−c
(Figure 2B,C) shows a plateau after ∼3 min as previously
observed with [11C]1c12 and in accordance with [11C]1b in
baboon.21 In contrast, [18F]5a shows a pronounced washout of
activity from basal ganglia after reaching maximum at about 10
min as previously reported with [11C]1a and [18F]5a in mice.14
C
dx.doi.org/10.1021/jm500503k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Brief Article
Measurements of proteins precipitated with acetonitrile
showed for all time points a stable protein bound radioactivity
fraction of 9 3%, 11 2%, and 11 2% (mean, SD) relative
to radioactivity in serum for [18F]5a−c, respectively. Compared
to the protein filtering method, washing the red blood cell
fraction resulted alone in a 10% increase in recovery of activity
while precipitating the proteins from the serum with
acetonitrile represented a significant increase in work-up of
radioactivity from the full blood samples.
Metabolite Assay. The metabolic fate of the three tracers
was assessed from samples worked up in the in vivo protein
binding assay by use of the above-described method for
washing the red blood cell and the protein fraction. Intact
tracers were determined by radio-TLC measurements analyzed
by exposure of phosphor imaging plate before digital readout.
The fractions of intact tracer in the serum samples measured
by TLC were used to calculate the extent of total intact tracer
in full blood given in Table 4. Measured fractions of intact
Table 2. Binding Affinity (Ki in nM) of 6-O-Methylated and
6-O-Fluoroethylated Orvinol Triplet towards Human
Cloned Opioid Receptors
orvinol
μ-OR (nM)
δ-OR (nM)
κ-OR (nM)
a
1a /5a
0.07/0.24
0.09/0.24
0.18/0.10
0.23/8.00
1.15/2.10
5.1/0.49
0.02/0.2
0.07/0.12
0.12/0.08
b
1b /5b
c
1c /5c
a
b
Values from Raynor et al. 1994.23 Values from Romero et al. 1999.24
Previous study, Marton et al. 2009.12
c
cloned μ-OR and to κ-OR with a lowered affinity for δ-OR. The
measured affinities to μ- and κ-OR are close to affinities
previously reported for 1c. The affinity to δ-OR was increased
for 5c over that of 1c by a factor of 10. In contrast, both 5a and
5b revealed a slight decrease in affinity toward δ-OR relative to
their native versions by a factor of 35 and 2, respectively.
Determination of pharmacological potency of 5a−c was
performed in a separate experiment by agonist stimulation of
[35S]GTPγS. The measured agonist stimulation of [35S]GTPγS
toward the different OR subtypes are given in Table 3 for 5a−c
Table 4. Intact [18F]5a−c from Blood (n = 3, Mean SD)
[18F]5a (%)
[18F]5b (%)
[18F]5c (%)
Table 3. [35S]GTPγS Stimulation by 5a−c at 1 nM,
5 min
12.5 7.0
6.9 1.8
4.8 3.2
39.3 3.5
27.4 2.8
27.5 1.7
26.8 8.3
17.4 5.6
14.3 2.8
20 min
40 min
Pharmacological Potency (% Relative Agonist Response)
orvinol
μ-OR (%)
δ-OR (%)
κ-OR (%)
5a
5b
5c
12
55
97
22
9
70
43
95
tracer was compensated for activity bound to the red blood cell
fraction and proteins to achieve the total fraction of available
tracer assuming that bound activity is either a metabolite of the
tracer in question or irreversibly bound intact tracer.
92
From the results in Table 4 and measured values for log P,
there is a correlation between higher available tracer content in
the serum samples for tracers possessing higher lipophilicity
(log P: [18F]5b > [18F]5c > [18F]5a).
as % agonist response relative to the maximum stimulation by
standards: the full μ-agonist DAMGO, the δ-agonist DPDPE,
and the full κ-agonist U69593 at 1 nM concentration of the
ligand in question.
The integrity of [18F]5a−c in brain was investigated after
collecting whole rat brains at 5, 20, and 40 min after iv
injections. Brains were frozen and homogenized before
extracting the activity with acetonitrile. Measured average
extraction yield remained stable for all time point with 94 3,
The assay revealed high potency for 5c to all three receptor
subtypes as previously reported for the native version 1c.12 Less
agonist response was observed for 5a and 5b toward all OR
subtypes as expected from properties of the native versions.
Protein Binding Assay. Protein binding properties of [18F]
5a−c were investigated by an in vitro assay in full blood from
rats (n = 3). After separating the red blood cell fractions, only
14, 16, and 6% of the total radioactivity was measured in the
serum fraction for [18F]5a−c, respectively. After filtering off the
proteins, 68, 96, and 94% of the activity measured in serum
remained in the protein fractions.
The observed high binding of radioactivity to the red blood
cell and the protein fraction in the in vitro assay limited the
method relevance for preparation of serum samples to be
analyzed for tracer stability and integrity toward metabolism.
Therefore, we aimed for a method to increase the total activity
amount in serum after removing the proteins. The method was
changed by adding a washing step for the red blood cell fraction
by use of PBS followed by the use of acetonitrile to precipitate
proteins from serum. The changed method was applied in an
assay with in vivo incubation. Blood samples were collected 5,
20, and 40 min (n = 3) after iv injection of [18F]5a−c in male
Sprague−Dawley rats. The combined serum and PBS fraction
revealed a content 48 13%, 60 8%, and 48 6% of the
total activity after 5 min in vivo incubation of [18F]5a−c,
respectively. In samples, 40 min post injection showed a slight
reduction of activity in serum to 29 14%, 50 1%, and 46
7% (mean % value, SD).
92
3, and 93
1% extracted activity for [18F]5a−c,
respectively. Radio TLC on the extracts revealed constantly
high content of intact tracer with 94.6 1.3, 98.0 0.7, and
96.9 1.8% (mean SD) combined for all time points for
[18F]5a−c, respectively, showing that all three orvinols possess
a high integrity toward metabolism in brain. This also indicates
that uptake of radioactive metabolites are low and that
metabolic species have no accumulation in brain. The measured
high tracers integrity in brain is in accordance with earlier
report with 80% intact [18F]5a after 30 min in mice.
CONCLUSION
■
The structurally matched orvinol triplet 1a−c is now available
as the fluoride-18 version [18F]5a−c from a fully automated
production based on a SPE method by use of a single module
with performance and quality suitable for the production of
potent OR PET tracers.
PET imaging of [18F]5a−c in rat brain showed fast brain
uptake and selective accumulation in μ-rich regions such as the
basal ganglia. Different clearance kinetics from the brain was
observed, with a plateau after 3 min for [18F]5b and [18F]5c,
while [18F]5a levels decreased constantly after a maximum at 10
min. The in vitro assays revealed that the agonist potency and
affinity toward the OR for the fluorine-18 labeled version of
D
dx.doi.org/10.1021/jm500503k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Brief Article
PEO was conserved except for a slightly higher affinity toward
the δ-OR. This first report on the 6-O-18F-fluoroethylated
version (5b) of buprenorphine (1b) showed no significant
change in subtype selectivity or affinity compared to the native
structure.
ASSOCIATED CONTENT
* Supporting Information
New compound characterization and procedures for compound
synthesis and biological assays. This material is available free of
charge via the Internet at http://pubs.acs.org.
■
S
The results open for continued investigations in character-
izing the dependency of agonist and antagonist properties on
sensitivity of imaging of OR in different functional states with
now available [18F]5a−c.
AUTHOR INFORMATION
Corresponding Author
*E-mail: bentws@kjemi.uio.no.
■
Notes
EXPERIMENTAL SECTION
The authors declare no competing financial interest.
■
Chemistry. The purity of the synthesized compounds was
determined by analytical HPLC to be >95%. 3a and 3b were
produced by ABX Biochemical, Dresden, Germany. Compound 3c
was prepared as reported previously.12 The syntheses of reference
standards 5a and 5c are previously reported.14,15 The method of
preparation of 5b is described in the Supporting Information.
Radiosynthesis. The general method for 18F-fluorethylation of
TDDPN (3a), TDBPN (3b), and TDPEO (3c), with subsequent
removal of the 3-O-trityl-group and preparative HPLC purification of
[18F]5a−c, was performed with modifications to a previously reported
method14 in accordance with the method described earlier.13 All
labeling steps were performed in an automated fashion in a two-pot,
three-step reaction on the same synthesis module (Hot-box III,
Scintomics GmbH, Fuerstenfeldbruck, Germany).
ACKNOWLEDGMENTS
■
We thank Einar Mantor Iversen and Anne Toril Aalvik at the
Norwegian Medical Cyclotron Centre for their skilled technical
assistance in providing high-quality batches of [18F]fluoride. We
will also thank Hong Qu for technical assistance in animal
experiments.
ABBREVIATIONS USED
■
BPN, buprenorphine; CPM, cyclopropylmethyl group; DPN,
diprenorphine; DBPN, 6-O-desmethyl-buprenorphine; DDPN,
6-O-desmethyl-diprenorphine; DPEO, 6-O-desmethyl-pheneth-
yl-orvinol; FBPN, 6-O-(2-fluoroethyl)-buprenorphine; FDPN,
6-O-(2-fluoroethyl)-diprenorphine; F-TDBPN, 6-O-(2-fluo-
roethyl)-3-O-trityl-6-O-desmethyl-buprenorphine; F-TDDPN,
6-O-(2-fluoroethyl)-3-O-trityl-6-O-desmethyl-diprenorphine;
F-TDPEO, 6-O-(2-fluoroethyl)-3-O-trityl-6-O-desmethyl-phe-
nethyl-orvinol; FPEO, 6-O-(2-fluoroethyl)-phenethyl-orvinol;
PEO, (20R)-phenethyl-orvinol; TDDPN, 3-O-trityl-6-O-des-
methyl-diprenorphine; TDBPN, 3-O-trityl-6-O-desmethyl-bu-
prenorphine; TDPEO, 3-O-trityl-6-O-desmethyl-phenethyl-or-
vinol; tBu, tert-butyl group; Tr, trityl group = triphenylmethyl
group
Cyclotron produced [18F]fluoride in water was trapped and washed
on a QMA-cartridge Sep Pac Light (Waters) before forming the [K
+
⊂ 2.2.2]18F complex. Then 7.5 mg of ethylene-ditosylate in 1 mL of
acetonitrile was added for reaction in 5 min at 75 °C. Unreacted
ethylene-ditosylate precursor was precipitated from the reaction
mixture at room temperature by the addition of 3 mL of 0.007 M
acetic acid. The precipitate was removed online by means of a 0.45 μm
polypropylene filter disk while transferring the mixture to two stacked
Sep-Pac Plus C-18 cartridges (Waters). Immobilized [18F]FETos was
eluted with 35% (v/v) of methanol/water in the interval 7−17 mL.
Subsequent immobilization followed by gas-jet drying on a 30 mg
Strata-X column (Phenomenex) and finally eluting [18F]FETos with
200 μL of anhydrous DMF and trapped in a second vial. The 18F-
fluoroalkylation of 3-O-protected orvinol precursor in question was
conducted as follows: 2.0 mg of orvinol precursor, 3a−c, was dissolved
in 200 μL of anhydrous DMF and activated with 5 mg of NaH for 5
min. Excess NaH was removed before transfer of activated orvinols
precursor to vial containing [18F]FEtos. The reaction was allowed to
proceed for 10 min reaction at 100 °C. After cooling to 40 °C, 0.5 mL
of 2.0 M HCl in EtOH was added and the mixture allowed reacting for
5 min in order to remove the trityl protecting group. Before product
isolation HPLC and work-up, the reaction mixture was brought to pH
∼8 by adding of 0.53 mL of 2.0 M ammonium hydroxide. The product
peak was collected from the HPLC (LC, Scintomics GmbH) mobile
phase before product immobilization on a Sep-Pak Light C18
cartridge, with subsequent product formulation in physiological
phosphate buffer saline and sterile filtering. The purity and specific
activity (SA) was determined by HPLC system (Shimadzu)
comprising a μ-Bondapak C-18 HPLC column 300 mm length × 4
mm internal diameter (CS Chromatographie GmbH, Germany) fitted
with a diode array UV detector (Shimadzu), which was set at 220 nm,
coupled in-line with a GABI radioactivity detector (Ray Test,
Straubenhardt, Germany). Samples were eluted with 1/1 (v/v)
acetonitrile and 0.1 M ammonium formate at 1.5 mL/min. The
identity of [18F]5a−c was confirmed by means of analytical radio-
HPLC after coinjection of authentic reference samples.
REFERENCES
■
(1) Henriksen, G.; Willoch, F. Imaging of opioid receptors in the
central nervous system. Brain 2008, 131, 1171−1196.
(2) Sprenger, T.; Willoch, F.; Miederer, M.; Schindler, F.; Valet, M.;
Berthele, A.; Spilker, M. E.; Forderreuther, S.; Straube, A.; Stangier, I.;
̈
Wester, H. J.; Tolle, T. R. Opiodergic changes in the pineal gland and
̈
hypothalamus in cluster headache: a ligand PET study. Neurology
2006, 66, 1108−1110.
(3) Sprenger, T.; Valet, M.; Boecker, H.; Henriksen, G.; Spilker, M.
E.; Willoch, F.; Wagner, K. J.; Wester, H. J.; Tolle, T. R. Opiodergic
̈
activation in the medinal pain system after heat pain. Pain 2006, 122,
63−67.
(4) Boecker, H.; Sprenger, T.; Spilker, M. E.; Henriksen, G.;
Koppenhoefer, M.; Wagner, K. J.; Valet, M.; Berthele, A.; Tolle, T. R.
Runner’s high: opiodergic mechanisms in the human brain. Cereb.
Cortex 2008, 18, 2523−2531.
(5) Melichar, J. K.; Hume, S. P.; Williams, T. M.; Daglish, M. R.;
Taylor, L. G.; Ahmad, R.; Malizia, A. L.; Brooks, D. J.; Myles, J. S.;
Lingford-Hughes, A.; Nutt, D. J. Using [11C]diprenorphine to image
opioid receptor occupancy by methadone in opioid addiction: clinical
and preclinical studies. J. Pharmacol. Exp. Ther. 2005, 312, 309−315.
(6) Hume, S. P.; Lingford-Hughes, A. R.; Nataf, V.; Hirani, E.;
Ahmad, R.; Davies, A. N.; Nutt, D. J. Low sensitivity of positron
emission tomography ligand [11C]diprenorphine to agonist opiates. J.
Pharmacol. Exp. Ther. 2007, 322, 661−667.
(7) Narendran, R.; Hwang, D.-R.; Slifstein, M.; Talbot, P. S.; Erritzoe,
D.; Huang, Y.; Cooper, T. B.; Martinez, D.; Kegeles, L. S.; Abi-
Dargham, A.; Laruelle, M. In vivo vulnerability to competition by
endogenous dopamine: comparison of the D2 receptor agonoist
Yields and physical data of the prepared compounds are collected in
the Supporting Information.
E
dx.doi.org/10.1021/jm500503k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Brief Article
radiotracer (−)-N-[11C]propyl-norapomorphine ([11C]NPA) with the
D2 receptor antagonist radiotracer [11C]-raclopride. Synapse 2004, 52,
188−208.
(8) Bentley, K. W.Thebaine and oripavine derivatives. (Reckitt and
Sons Ltd.) British Patent 1,136,214 1968; Chem. Abstr. 1969, 70,
78218s.
12. Buprenorphine is a potent and selective μ/κ antagonist in the
[35S]-GTP-γ-S functional binding assay. Synapse 1999, 34, 83−94.
(9) Bentley, K. W.; Hardy, D. G. Novel analgesics and molecular
rearrangements in the morphine-thebaine group. III. Alcohols of the
6,14-endo-ethenotetrahydrooripavine series and derived analogs of N-
allylnormorphine and -norcodeine. J. Am. Chem. Soc. 1967, 89, 3281−
3292.
(10) Husbands, S. M. Buprenorphine and related orvinols. In
Research and Development of Opioid-Related Ligands; ACS Symposium
Series 1131; Ko, M-C, Husbands, S. M., Eds.; American Chemical
Society: Washington, DC, 2013; pp 127−144.
(11) Luthra, S. K.; Brady, F.; Turton, D. R.; Brown, D. J.; Dowsett,
K.; Waters, S. L.; Jones, A. K. P.; Matthews, R. W.; Corwder, J. C.
Automated radiosyntheses of [6-O-methyl-11C]diprenorphine and [6-
O-methyl-11C]buprenorphine from 3-O-trityl protected precursors.
Appl. Radiat. Isot. 1994, 45, 857−873.
(12) Marton, J.; Schoultz, B. W.; Hjørnevik, T.; Drzezga, A.; Yousefi,
B. H.; Wester, H.-J.; Willoch, F.; Henriksen, G. Synthesis and
evaluation of a full agonist orvinol for PET-imaging of opioid
receptors: [11C]PEO. J. Med. Chem. 2009, 52, 5586−5589.
(13) Schoultz, B. W.; Reed, B. J.; Marton, J.; Willoch, F.; Henriksen,
G. A Fully Automated radiosynthesis of [18F]fluoroethyl-diprenor-
phine on a single module by use of SPE cartridges for preparation of
high quality 2-[18F]fluoroethyl tosylate. Molecules 2013, 18, 7271−
7278.
(14) Wester, H.-J.; Willoch, F.; Tolle, T. R.; Munz, F.; Herz, M.; Øye,
̈
I.; Scadrack, J.; Schwaiger, M.; Bartenstein, P. 6-O-(2-[18F]-
Fluoroethyl)-6-O-desmethyldiprenorphine ([18F]FDPN): synthesis,
biologic evaluation, and comparison with [11C]DPN in humans. J.
Nucl. Med. 2000, 41, 1279−1286.
(15) Marton, J.; Henriksen, G. Design and Synthesis of an 18F-
Labeled version of phenylethyl orvinol ([18F]FE-PEO) for PET-
imaging of opioid receptors. Molecules 2012, 17, 11554−11569.
(16) Riss, P. J.; Hong, Y. T.; Marton, J.; Caprioli, D.; Williamson, D.
J.; Ferrari, V.; Saigal, N.; Roth, B. L.; Henriksen, G.; Fryer, T. D.;
Dalley, J. W.; Aigbirhio, F. I. Synthesis and evaluation of 18F-FE-PEO
in rodents: an 18F-labeled full agonist for opioid receptor imaging. J.
Nucl. Med. 2013, 54, 299−305.
(17) Marton, J.; Szabo,
Ethenomorphinan-Derivaten. Liebigs Ann. Chem. 1993, 915−919.
(18) Marton, J.; Hosztafi, S.; Berenyi, S.; Simon, Cs.; Makleit, S.
́
Z.; Hosztafi, S. Herstellung von 6,14-
́
Herstellung von 6,14-Ethenomorphinan-Derivaten. Monatsh. Chem.
1994, 125, 1229−1239.
(19) Werner, L.; Machara, A.; Adams, D. R.; Cox, D. P.; Hudlicky, T.
Synthesis of buprenorphine from oripavine via N-demethylation of
oripavine quaternary salts. J. Org. Chem. 2011, 76, 4628−4634.
(20) Machara, A.; Werner, L.; Endoma-Arias, M. A.; Cox, D. P.;
Hudlicky, T. Improved synthesis of buprenorphine from thebaine and/
or oripavine via palladium-catalyzed N-demethylation/acylation and/
or concomitant O-demethylation. Adv. Synth. Catal. 2012, 354, 613−
626.
(21) Galynker, I.; Schlyer, D. J.; Dewey, S. L.; Fowler, J. S.; Logan, J.;
Galley, S. J.; MacGregor, R. R.; Ferrieri, R. A.; Holland, M. J.; Brodie,
J.; Simon, E.; Wolf, A. P. Opioid receptor imaging and displacement
studies with [6-O-[11C]methyl]buprenorphine in Baboon Brain. Nucl.
Med. Biol. 1996, 23, 325−331.
(22) Waterhouse, R. N. Determination of lipophilicity and its use as a
predictor of blood−brain barrier penetration of molecular imaging
agents. Mol. Imaging Biol. 2003, 5, 376−389.
(23) Raynor, K.; Kong, H.; Chen, Y.; Yasuda, K.; Yu, L.; Bell, G. I.;
Reisine, T. Pharmacological characterization of the cloned κ-, δ-, and
μ-opioid receptors. Mol. Pharmacol. 1994, 45, 330−334.
(24) Romero, D. V.; Partilla, J. S.; Zheng, Q.-X.; Heylinger, S. O.; Ni,
Q.; Rice, K. C.; Lai, J.; Rothman, R. B. Opioid peptide receptor studies.
F
dx.doi.org/10.1021/jm500503k | J. Med. Chem. XXXX, XXX, XXX−XXX