The preparation of [pentane-5,5,5-3H3]-abnormal- cannabidiol

Article abstract of DOI:10.1002/jlcr.1840

A bromoalkane precursor was synthesized in six steps, and its copper catalysed coupling with methylmagnesium chloride to provide unlabelled abnormal-cannabidiol (1a) was optimized. The methodology was used for an analogous coupling using [³H

Full text of DOI:10.1002/jlcr.1840

Research Article
Received 8 May 2010,
Revised 5 August 2010,
Accepted 19 September 2010 Published online 28 November 2010 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.1840
The preparation of [pentane-5,5,5-3H3]-
abnormal-cannabidiol
Stephen J. Byard,a Steven A. Carr,b Jose F. DeFaria,b Crist N. Filer,b
Ã
John M. Herbert,c and Tatiana Jaramillo Forcadac
A bromoalkane precursor was synthesized in six steps, and its copper catalysed coupling with methylmagnesium chloride
to provide unlabelled abnormal-cannabidiol (1a) was optimized. The methodology was used for an analogous coupling
using [3H3]-methylmagnesium iodide to provide [pentane-2H3]-abnormal-cannabidiol (1b).
Keywords: abnormal-cannabidiol; copper-catalysed coupling; [3H3]-methylmagnesium iodide
obtained commercially; in particular, (4R)-1-methyl-4-(1-methyl-
vinyl)cyclohex-2-en-1-ol (2) was obtained from Sai Life Sciences
Introduction
Limited, and [3H3]iodomethane was produced by Perkin-Elmer.
A number of reports have appeared in the last decade
concerning the biological activity of abnormal-cannabidiol
(1a).1,2 This synthetic isomer of cannabidiol fails to elicit a
response from either CB1 or CB2 receptors and lacks psycho-
tropic activity. However, 1a induces endothelium-dependent
vasodilation via a CB1/CB2/NO-independent mechanism, and
shows hypotensive activity that is not antagonized by canna-
bidiol or SR141716A.2 Given the potential importance of this
ligand as a research tool, and in order to facilitate the further
investigation of the putative CB3 receptor with which it is
associated, a requirement for a tritiated form of 1 containing at
least two tritium nuclei per molecule was identified. Work
towards the tritiation of 1, and the preparation of the labelled
ligand 1b are described in this publication.
Column chromatography was carried out using silica gel (Merck
silica 60). Autoradiography was performed at 01C after spraying
with PPO and exposing plates to X-ray film. TLC plates were also
scanned for applied radioactivity. Preparative and analytical
HPLC were performed on a PerkinElmer instrument and peak
detection was done simultaneously by UV and an IN/US Systems
Beta RAM Model 3 radioactivity detector. Solution assays were
performed with a Perkin Elmer Tri-Carb 3100TR instrument. Mass
spectra for tritiated product were obtained on a Kratos Model
MS25 RF instrument with direct injection.
4-(3,5-Dimethoxyphenyl)-1-butene (5).3
A solution of 3,5-
dimethoxybenzyl bromide (3.0 g, 13 mmol) in diethyl ether
(17 mL) was added dropwise during 10 min under a nitrogen
atmosphere to allylmagnesium bromide in diethyl ether (1 M;
25 mL) with ice cooling. The cooling bath was removed, and the
reaction was stirred at RT for 23 h. After this time, the mixture
was cooled to 01C and quenched with saturated aqueous
ammonium chloride (20 mL). Additional water (15 mL) was
added, the phases were separated, and the aqueous layer was
re-extracted with diethyl ether (3 Â 20 mL). The combined
organic phases were dried (Na2SO4) and solvent was removed
under reduced pressure. The residue was purified by chromato-
graphy in ethyl acetate–hexane (1:9) to give 4-(3,5-dimethoxy-
phenyl)-1-butene (2.313 g, 84%). dH(CDCl3) 2.32–2.44 (2H, m),
R
R
R
HO
OH
1a (R = H), 1b (R = 3H)
Experimental
GC-MS data were obtained using an Agilent 6890 gas
chromatograph fitted with mass-selective detector
(5970MSD) operating in CI mode, with methane as the ionisation
gas. The injector temperature was 2501C and the oven
temperature was increased, after an initial 2 min delay, from
70 to 2301C at 101C per minute. LC-MS data were obtained cIsotope Chemistry and Metabolite Synthesis Department, Sanofi-Aventis,
using an Agilent 1200 HPLC system connected to a Bruker
micro-TOF Focus mass spectrometer operating in esi mode and
with sodium formate as internal calibrant. 1H NMR spectra were
a
aAnalytical Sciences Department, Sanofi-Aventis, Willowburn Avenue, Alnwick,
Northumberland NE66 2JH, UK
bPerkin-Elmer Health Sciences Inc., 940 Winter St., Waltham, MA 02451, USA
Willowburn Avenue, Alnwick, Northumberland NE66 2JH, UK
*Correspondence to: John M. Herbert, Isotope Chemistry Department, Covance
Laboratories Ltd., Willowburn Avenue, Alnwick, Northumberland, NE66 2JH.
recorded using a Bruker DRX-500 instrument. Reagents were E-mail: john.herbert@covance.com
J. Label Compd. Radiopharm 2011, 54 180–184
Copyright r 2010 John Wiley & Sons, Ltd.

S. J. Byard et al.
2.66 (2H, dd, J 8.3, 7.5 Hz), 3.79 (6H, s), 4.99 (1H, dm, Jcis 10.3 Hz), chromatography in ethyl acetate–hexane, using a gradient from
5.07 (1H, dm, Jtrans 16.7 Hz), 5.83–5.91 (1H, m), 6.32 (1H, t, 1:9 to 1:1, to give:
J 2.3 Hz), 6.37 (2H, d, J 2.3 Hz); m/z 193 (100%, MH1), 151.
4-(3,5-Dimethoxyphenyl)-1-butanol (6).4 9-Borabicyclo[3.3.1]- (i) 5-(4-bromobutyl)-2-[(1R,6R)-3-methyl-6-(1-methylethenyl)-2-
nonane (0.4 M in hexanes; 19.2 mL; 7.7 mmol) was added slowly cyclohexen-1-yl]-1,3-benzenediol (B; 199 mg, 17%) dH(CDCl3)
at room temperature to a stirred solution of 4-(3,5-dimethoxy- 1.48 (3H, s), 1.57–1.68 (2H, m), 1.70 (3H, br s), 1.74–1.83 (4H,
phenyl)-1-butene in toluene (19 mL). The mixture was warmed m), 2.04–2.12 (1H, m), 2.18–2.27 (2H, m), 2.38–2.48 (1H, m), 2.61
to 401C for 4.5 h, after which time all starting material had been (1H, ddd, J 14.5, 9.1, 5.4 Hz), 3.41 (2H, t, J 6.8 Hz), 3.48 (1H, br d,
consumed (GC-FID). Aqueous sodium hydroxide (2 M, 4.2 mL) Japp 9.4 Hz), 4.50 (1H, s), 4.61 (1H, t, J 1.8 Hz), 5.48 (1H, s), 5.82
was added, followed by hydrogen peroxide (30 wt%, 2.0 mL) and (1H, br s, OH), 6.19 (2H, s); m/z 379 (MH1, 100%).
the mixture was stirred overnight. The mixture was diluted with (ii) 5-(4-bromobutyl)-4-[(1S,6R)-3-methyl-6-(1-methylethenyl)-2-
ethyl acetate (24 mL) and brine (25 mL), the phases were cyclohexen-1-yl]-1,3-benzenediol (A; 389 mg, 33%)dH(CDCl3)
separated, and the aqueous phase was re-extracted with ethyl 1.66 (3H, s), 1.67–1.74 (2H, m), 1.79 (3H, br s), 1.77–1.88 (4H,
acetate (3 Â 25 mL). The combined organic phases were dried m), 2.06–2.13 (1H, m), 2.19–2.28 (1H, m), 2.39 (1H, ddd, J 11.7,
(Na2SO4) and solvent was removed under reduced pressure. The 10.7, 3.5 Hz), 2.48 (2H, t, J 7.5 Hz), 3.40 (2H, t, J 6.7 Hz), 3.85 (1H,
residue was purified by chromatography in ethyl acetate–hex- br d, Japp 9.2 Hz), 4.55 (1H, s), 4.66 (1H, t, J 1.8 Hz), 5.56 (1H, s),
ane (2:3, followed by 1:1) to give 4-(3,5-dimethoxyphenyl)-1- 6.00 (1H, br s, OH), 6.17 (1H, br s), 6.26 (1H, br s); m/z 379 (MH1,
butanol (983 mg, 74%). dH(CDCl3) 1.30 (br s, 1H, OH), 1.59–1.65 100%).
(2H, m), 1.66–1.73 (2H, m), 2.60 (2H, t, J 7.5 Hz), 3.67 (2H, t, (iii) 5-(4-bromobutyl)-4-[(1R,6R)-3-methyl-6-(1-methylethenyl)-2-
J 5.4 Hz), 3.79 (6H, s), 6.31 (1H, t, J 2.2 Hz), 6.36 (2H, d, J 2.2 Hz); cyclohexen-1-yl]-1,3-benzenediol (9; 592 mg, 50%). dH(CDCl3)
1.52 (3H, s, 3-CH3), 1.57–1.68 (2H, m, 5-H), 1.80 (3H, br s,
m/z 211 (100%, MH1), 193.
4-(3,5-Dimethoxyphenyl)-1-bromobutane (7). Triphenylpho- H2C==C–CH3), 1.76–1.91 (4H, m), 2.06–2.13 (1H, m), 2.18–2.25
sphine (2.086 g, 7.9 mmol) was added in one portion to a (1H, m), 2.28 (1H, ddd, J 7.2, 4.9, 3.6 Hz, 4-Hequatorial), 2.47 (1H,
stirred, ice-cooled solution of 4-(3,5-dimethoxyphenyl)-1-buta- ddd, J 12.2, 10.5, 3.6 Hz, 6-H), 2.66 (1H, ddd, J 14.7, 10.5, 5.8 Hz,
nol (972 mg, 4.6 mmol) and carbon tetrabromide (3.094 g, 4-Haxial), 3.41 (2H, t, J 6.7 Hz, CH2Br), 3.51 (1H, br d, Japp 9.4 Hz, 1-H),
9.2 mmol) in acetonitrile. A bright yellow colour formed, which 4.47 (1H, s, HC(H)==C–CH3), 4.66 (1H, t, J 1.8 Hz, HC(H)==C–CH3),
faded after 10 min, at which stage the ice bath was removed and ca. 4.67 (1H, br s, OH), 5.52 (1H, s, 2-H), 6.06 (1H, br s, OH), 6.18
the reaction was stirred at room temperature for 1 h. (1H, d, J 2.6 Hz), 6.22 (1H, d, J 2.6 Hz); m/z 379 (MH1, 100%).
5-(4-Bromobutyl)-3,5-bis(tert-butyldimethylsilyloxy)-2-(6-isopro-
penyl-3-methyl-2-cyclohexenyl)benzene (10). tert-Butyldimethylsilyl
trifluoromethanesulfonate (0.90 mL; 3.8 mmol) and 2,6-lutidine
(452 mL; 3.8 mmol) were added to a solution of 9 (583 mg,
1.54 mmol) in dichloromethane (5 mL) at room temperature. The
resulting pale yellow solution was stirred overnight after which
time the reaction was complete. Water (5 mL) was added, the
phases were separated, and the aqueous phase was re-extracted
with ethyl acetate (2 Â 5 mL). The combined organic layers were
dried (Na2SO4) and evaporated, and the crude product was
purified by chromatography on silica gel in hexane to give 5-(4-
bromobutyl)-3,5-bis(tert-butyldimethylsilyloxy)-2-(6-isopropenyl-
3-methyl-2-cyclohexenyl)benzene (829 mg, 89%). dH(CDCl3) 0.16
(12H, s), 0.96 (9H, s), 0.98 (9H, s), 1.41 and 1.55 (3H, 2s), 1.57–1.75
(1H, m), 1.63 and 1.70 (3H, 2s), 1.76–1.83 (2H, m), 1.84–1.90 (2H,
m), 1.97–2.08 (1H, m), 2.10–2.21 (1H, m), 2.30 (1H, ddd, J 8.1, 6.5,
2.7 Hz), 2.42 (1H, ddd, J 12.0, 12.0, 4.2 Hz), 2.55 (1H, ddd, J 11.8,
8.2, 5.4 Hz), 2.68–2.85 (2H, m), 2.97 (1H, ddd, J 15.0, 10.8, 4.8 Hz),
3.37–3.43 (2H, m), 4.11 (1H, br d, J 9.4 Hz), 4.40 and 4.47 (1H, 2s),
4.53 and 4.56 (1H, 2s), 5.24 and 5.29 (1H, 2s), 6.13 and 6.16 (1H,
2d, both J 2.8 Hz), 6.18 and 6.28 (1H, 2d, both J 2.8 Hz); m/z 607,
607 (ca. 1:1, MH1), 527 ([MH-HBr]1, 100%).
4-[(1R,6R)-3-Methyl-6-(1-methylethenyl)-2-cyclohexen-1-yl]-5-pentyl-
1,3-benzenediol (abnormal-Cannabidiol,1a). Lithium tetrachloro-
cuprate(II) (0.1 M in THF; 25 mL) and methylmagnesium iodide
(3.0 M in THF; 100 mL, 0.3 mmol) were added sequentially to a
solution of 5-(4-bromobutyl)-3,5-bis(tert-butyldimethylsilyloxy)-2-(6-
isopropenyl-3-methyl-2-cyclohexenyl)benzene (20 mg; 33 mmol) in
THF (1.0 mL), forming a fine suspension, which was stirred at room
temperature. Additional methylmagnesium iodide solution (150 mL,
0.45 mmol) was added after 1 h, and the mixture was stirred for a
further 2 h, after which time the reaction was complete (TLC).
Water (5 mL) and ethyl acetate (5 mL) were added, the phases were
separated, and the aqueous layer was re-extracted with ethyl
Triethylamine (1.3 mL) and methanol (1.3 mL) were added, and
the resulting precipitate was removed by filtration. The filtrate
was evaporated and the residue was purified by column
chromatography in ethyl acetate–hexane (5:95 followed by
8:92) to give 4-(3,5-dimethoxyphenyl)-1-bromobutane (1.206 g,
96%). dH(CDCl3) 1.74–1.81 (2H, m), 1.86–1.93 (2H, m), 2.59 (2H, t,
J 7.4 Hz), 3.42 (2H, t, J 6.7 Hz), 3.78 (6H, s), 6.31 (1H, t, J 2.2 Hz),
6.34 (2H, d, J 2.2 Hz); m/z 273, 275 (1:1, MH1), 193 (100%).
4-(3,5-Dihydroxyphenyl)-1-bromobutane (8).5
A solution of
4-(3,5-dimethoxyphenyl)-1-bromobutane (500 mg, 1.8 mmol)
and boron tribromide (1.0 M in dichloromethane; 4.4 mL) in
1,2-dichloroethane (60 mL) was stirred at reflux for 1.5 h.
After this time, TLC indicated complete reaction, and the
mixture was cooled to room temperature. Water (60 mL) was
added, the phases were separated, and the aqueous phase was
re-extracted with ethyl acetate (2 Â 50 mL). The combined
organic phases were washed with brine (50 mL), dried (Na2SO4)
and evaporated to give 4-(3,5-dihydroxyphenyl)-1-bromobutane
(719 mg, 98% allowing for 39% ethyl acetate present by NMR).3
dH(CDCl3) 1.70–1.77 (2H, m), 1.84–1.91 (2H, m), 2.53 (2H, t,
J 7.5 Hz), 3.41 (2H, t, J 6.7 Hz), 4.95 (2H, br s, OH), 6.19 (1H, t,
J 2.1 Hz), 6.24 (2H, d, J 2.1 Hz); m/z 245, 247 (1:1, MH1), 165
([MH-HBr]1, 100%).
5-(4-Bromobutyl)-4-[(1R,6R)-3-methyl-6-(1-methylethenyl)-2-cyclo-
hexen-1-yl]-1,3-benzenediol (9).6 Boron trifluoride diethyl etherate
(40 mL, 0.32 mmol) was added all at once under nitrogen at
À401C to a vigorously stirred suspension of anhydrous magne-
sium sulfate (780mg 6.4 mmol) in dichloromethane (39 mL)
containing
(4R)-1-methyl-4-(1-methylvinyl)cyclohex-2-en-1-ol
(949 mg; 6.2 mmol) and 4-(3,5-dihydroxyphenyl)-1-bromobutane
(1.507 g, 3.1mmol). After 1 h, the reaction was complete (TLC)
and sodium hydrogencarbonate (203 mg, 2.4 mmol) was added.
The mixture was stirred for a further 30 min, and then filtered. The
filtrate was evaporated and the residue was purified by column
J. Label Compd. Radiopharm 2011, 54 180–184
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

S. J. Byard et al.
acetate (3 Â 5 mL). The combined organic phases were dried 495 mCi of crude tritiated intermediate that contained approxi-
(Na2SO4) and solvent was removed under reduced pressure to mately 25% of the desired intermediate by TLC analysis (hexane
leave crude 1,5-bis(tert-butyldimethylsilyloxy)-2-(6-isopropenyl-3- on silica gel). This intermediate was initially purified by passing
methyl-2-cyclohexenyl)-3-pentylbenzene. dH(CDCl3) 0.17 (12H, s), through a silica gel Sep-Pak eluted with hexane, providing
0.90 and 0.91 (3H, 2t, J 6.8 Hz), 0.97 (9H, s), 0.99 (9H, s), 1.26–1.36 95 mCi of crude intermediate. Approximately 40 mCi of this
(2H, m), 1.56 (3H, br s), 1.50–1.57 (1H, m), 1.63 and 1.68 (3H, 2s), intermediate was then evaporated to dryness and dissolved in
1.74–1.88 (2H, m), 1.99–2.09 (1H, m), 2.10–2.22 (1H, m), 2.24–2.32 anhydrous THF (1 mL). Tetrabutylammonium fluoride in THF
(1H, m), 2.43–2.55 (2H, m), 2.63–2.82 (2H, m), 2.97 (1H, ddd, J 15.1, (1 M; 0.05 mL) was added to this solution under argon and the
11.1, 3.8 Hz), 3.40–3.46 (1H, m), 4.09–4.14 (1H, m), 4.42 and 4.47 mixture was stirred at ambient temperature for 30 min. Silica gel
(1H, 2s), 4.53 and 4.55 (1H, 2s), 5.27 and 5.30 (1H, 2s), 6.12 and 6.15 TLC analysis (9:1 hexane–ethyl acetate on silica gel) of the
(1H, 2d, both J 2.2 Hz), 6.21 and 6.29 (1H, 2d, both J 2.2 Hz); m/z 543 mixture showed good conversion to desired product and so
(100%, MH1), 527, 485. This material was redissolved in THF (2 mL) saturated aqueous sodium hydrogencarbonate (5 mL) was
and tetrabutylammonium fluoride (1.0 M in THF; 110 mL) was added. The aqueous phase was extracted with ethyl acetate
added. The resulting dark yellow solution was stirred for a further (3 Â 5 mL) and the combined organic phases were dried
30 min and, the reaction being complete by TLC, saturated (MgSO4). Purification of the final product was accomplished by
aqueous sodium hydrogencarbonate (5 mL) and ethyl acetate reverse phase HPLC eluted with a gradient of water–acetonitrile
(5 mL) were added. The phases were separated, the aqueous phase (1:1) to pure acetonitrile over the course of 50 min with fraction
was re-extracted with ethyl acetate (2 Â 5 mL), and the combined collection. Pooling of appropriate fractions, solvent evaporation
organic phases were dried (Na2SO4) and concentrated under under reduced pressure and reconstitution in ethanol afforded
reduced pressure to leave 2-(6-isopropenyl-3-methyl-2-cyclohexe- 1b (9 mCi), with better than 99% radiochemical purity by reverse
nyl)-3-pentylresorcinol (abnormal-Cannabidiol, 1a) (8.8mg, 85%). phase HPLC (Waters XTerra MS column eluted with the same
dH(CDCl3) 0.90 (3H, t, J 7.1 Hz), 1.28–1.35 (2H, m), 1.43–1.51 (2H, m), gradient as above). The specific activity of the final product was
1.54 (3H, s), 1.72–1.87 (2H, m), 1.79 (3H, br s), 2.06–2.09 (1H, m), measured as 74.3 Ci/mmol by mass spectrometry. m/z 321
2.10–2.13 (1H, m), 2.18–2.30 (2H, m), 2.48 (1H, ddd, J 12.9, 9.5, (100%, MH1 for 1b), 319 (20.7%, [3H2]-1), 317 (9.16%, [3H]-1), 315
3.2 Hz), 2.59 (1H, ddd, J 14.2, 9.1, 7.6 Hz), 3.53 (1H, br d, J 9.5 Hz), (6.05%, 1a).
4.46 (1H, d), 4.64 (1H, t, J 1.5 Hz), ca. 4.80 (1H, br s, OH), 5.52 (1H, s),
6.04 (1H, s, OH), 6.20 (1H, d, J 2.8 Hz), 6.21 (1H, d, J 2.82 Hz); m/z 315 Results and discussion
(100%, MH1), 181. HPLC was carried out using a Waters XTerra
Phenyl; 3.5 mm; 4.6 Â 150 mm with eluant A as acetonitrile–water
(Scheme 1) in the presence of a protic or Lewis acid to give
(19:1) and eluant B as acetonitrile–water (1:1) using a flow rate of
The coupling of trans-(1)-menthadienol (2) and olivetol
cannabidiol (3) commonly gives 1a, along with isomeric
tetrahydrocannabinols as side-products (Scheme 1).6,8 The yield
and the product ratio are dependent on the reaction conditions,
and particularly on the acid used, the best-quoted yield of 1a
being 47%.6 For our purposes, tritiation of the final product by
1 mL/min and a gradient from 100% B at 1 min to 100% A at
36min and UV detection at 240 and 254nm. In this system, 1a had
a RT of 13.97min.
[pentane-5,5,5-3H3]-2-(6-Isopropenyl-3-methyl-2-cyclohexenyl)-
3-pentylresorcinol (abnormal-Cannabidiol, 1b). All the following
exchange was not expected to be viable, since the presence of
two double bonds precludes the use of hydrogen, and tritium
exchanged into the arene from tritiated water is likely to be
labile. Either precursor in Scheme 1 could be prepared in
tritiated form but, with the preparation of a suitable enone (4)
precursor to 2 requiring several steps,9 the possibility that 1,4-
addition of C3H3MgI to 4 could compete with the desired 1,2-
addition and the certainty that 4 prepared in this way would be
racemic, labelling in the olivetol unit would be preferred.
Indeed, one approach to the preparation of [3Hn]-1 would be to
prepare an unsaturated form of olivetol and to carry out
tritiation prior to the coupling; the corresponding reaction using
deuterium has been reported by previous workers.10 Never-
theless, given the complex nature of mixtures expected from the
coupling step, we preferred to modify the synthetic approach to
manipulations were conducted on a vacuum line. A solution of
[3H3]-methylmagnesium iodide (1 mmol at 74.3 Ci/mmol) in
diethyl ether (2 mL) was prepared using the procedure of
Prasad and Franklin7; to this solution at ambient temperature, a
solution of 9 (20 mg, 0.033 mmol) in anhydrous THF (1 mL) was
added with stirring. The reaction was allowed to stir at ambient
temperature for another 10 min and then lithium tetrachlor-
ocuprate in THF (0.1 M; 0.025 mL) was added. The reaction was
stirred for an additional 4 h at ambient temperature and then
quenched by addition of water (2 mL), followed by ethyl acetate
(5 mL). The phases were separated and the aqueous layer was
extracted with ethyl acetate (3 Â 5 mL). The organic phases were
combined and labile tritium was removed by evaporation with
several portions of methanol under vacuum. The mixture was
then dissolved in ethyl acetate–ethanol (9:1, 10 mL), affording
OH
O
OH
+
+
OH
HO
OH
OH
OH
2
olivetol
4
3
1a
Scheme 1.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 180–184

S. J. Byard et al.
permit the introduction of tritium labels following the coupling upon hydroboration of 5 with 9-BBN,4 and oxidation of the
step.
intermediate borane with alkaline hydrogen peroxide, although
The preparation of olivetol, deuterated at the terminus of the the overall yield was actually poorer. Bromination of the alcohol
pentyl chain, and its coupling with 2 to form deuterated thus obtained proceeded efficiently using triphenylphosphine–
tetrahydrocannabinol has also been reported.11 A modification carbon tetrabromide to give the primary alkyl bromide 7, which
of this approach, in which the labelled methyl group was underwent O-demethylation in excellent yield upon treat-
introduced after the coupling step, rather than before, ment with boron tribromide, as described previously for the
would permit the number of labelled steps to be minimized preparation of labelled olivetol.11 These latter conditions were
(Scheme 2). Owing to the presence of phenolic groups in the found to be very much superior to iodotrimethylsilane, as used
final product, protection of the substrate for this last C–C by earlier workers.3 Attempts were also made to convert the
coupling would be essential in order to avoid the generation of alcohol 6 directly into 8 using hydrobromic acid, either alone or
excessive quantities of radioactive waste, and so a single-step in combination with boron tribromide. Indeed, treatment of 6
labelling method was unlikely to be feasible.
with hydrogen bromide in acetic acid did provide an 81% yield
The sequence therefore used for the preparation of 1a and 1b of 8 in a single step. However, purification of the product
is shown in Scheme 3. Grignard allylation of 3,5-dimethoxyben- proved to be difficult, and the process was not improved by
zyl bromide, as described by previous workers,3 provided the microwave heating at 1501C. In addition, the yield using the
terminal alkene 5. Hydroboration of this intermediate has been two-step process was better than that obtained in a single step.
reported previously using borane–dimethyl sulfide complex.3
The olivetol synthon 8 was coupled with commercially
Nevertheless, we were disappointed that, despite an overall obtained trans-(1)-menthadienol, in the presence of a catalytic
yield of approximately 95%, the product from this method was a amount of boron trifluoride etherate, and keeping the
mixture of primary and secondary alcohol products, the temperature at À401C to minimize cyclodehydration of the
separation of which was difficult. Better selectivity was observed initial adducts. Reducing the temperature below À401C did not
result in any advantage and, at À701C the reaction was sluggish
at best. Separation of the mixture of regioisomers and
Br
diastereoisomers produced required extensive chromatography
but did provide pure trans-diastereomer 9 in an acceptable
yield. The cis diastereomer A and the regioisomeric product B
were isolated and identified also, the identity of the last being
immediately clear from the single aromatic resonance observed
in its 1H NMR spectrum. The NMR spectra of the cis diastereomer
A and the desired product 9 are similar, although the aromatic
resonances in the spectrum of A are significantly broadened;
this is likely to be the result of restricted rotation, due in turn to
steric interactions between the aromatic and isopropenyl
CT3
1. CT MgBr
3
2. aq. acid
or TBAF
OSiR3
OH
OSiR3
Scheme 2.
OH
Br
1. 9-BBN
toluene
45oC
MgBr
1. CBr4, PPh3
MeCN
OH
OMe
Br
OMe
Et2O, RT
(84%)
2. NaOH, H2O2
Bu4NHSO4
(74%)
2. Et3N, MeOH
(96%)
MeO
OMe
MeO
OMe
MeO
MeO
5
7
6
BBr3, DCE
reflux (98%)
3
4
5
OH
Br
2
OH
6
1
Br
Br
Br
cat. BF3•Et2O, MgSO4
DCM, -40oC (35%)
HO
OH
HO
OH
HO
HO
OH
9
B
8
A
TBDMSOTf, lutidine
DCM, RT (89%)
TBAF
THF, RT
MeMgCl
Br
LiCuCl4
(85%)
THF, RT
TBSO
OTBS
TBSO
OTBS
HO
OH
11a
10
1a
Scheme 3.
J. Label Compd. Radiopharm 2011, 54 180–184
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

S. J. Byard et al.
TBAF
THF, RT
CT3MgI
T
T
T
Br
LiCuCl4
THF, RT
T
T
T
TBSO
OTBS
TBSO
OTBS
HO
OH
11b
10
1b
Scheme 4.
groups. 1H NMR resonances in the terpenoid unit are shifted
downfield also in the spectrum of A, with the resonance due to
CH3 of the isopropenyl group in particular appearing at d1.66,
compared with d1.52 in the spectrum of 9. Simultaneously, that
due to H1 (geminal to the aromatic ring) is shifted from d3.51 in
9 to d3.85 in A. The assignment of the individual 1H NMR signals
in the spectrum of 9 was established from 2D-COSY spectra
with, in particular, both H1 and H6 correlating to the CH3 of
the isopropenyl group, and key assignments are given in the
experimental section. An unequivocal assignment of the
stereochemistry of 9 was made on the basis of the coupling
between H1 and H6, (J1,6 10.5 Hz), which clearly shows that
these protons are trans to each other. This assignment was
further supported by nOe data. As expected, irradiation at the
frequency corresponding to H1 of the terpene unit (d3.51)
results in enhancement of the signals due to the upfield proton
of the isopropenyl group (d4.47), to the olefinic H2 (d5.52) and
to H6 (d2.66). However, irradiation at the frequency correspond-
ing to H6 results in enhancements of the signals corresponding
to the upfield proton of the isopropenyl group, to H1 and to the
two protons H5, and also in a diminution of the signal due to
H4axial. The conclusion from these experiments is that both H1
and H6 are axial, and that the substituents at these centres must
be equatorial and therefore trans.
Completion of the synthesis in unlabelled form required
protection of the phenolic groups of 9, with tert-butyldimethyl-
silyl proving to be a suitable choice of protecting group. The
protected form 10 underwent copper-catalysed Grignard
coupling with methylmagnesium chloride, providing protected
unlabelled abnormal-cannabidiol (11a), which was not purified,
but rather was desilylated directly to give 1a in an 85% yield
over two steps. The copper-catalysed coupling reaction
described above proved to be general for methylmagnesium
halides, and the procedure using [3H3]-methylmagnesium iodide
was used to prepare high specific activity 1b (Scheme 4).
Following deprotection as above and purification by HPLC, 1b
was obtained with radiochemical purity above 99%, and a
specific activity of 74.3 Ci/mmol.
Conclusion
The preparation of cannabidiol analogue 9 has been effectively
optimized, and the coupling of protected form 10 with [3H3]-
methylmagnesium iodide followed by deprotection provides an
effective means for the preparation of 1b, high-specific tritiated
forms of abnormal-cannabidiol 1 being accessed only with
considerable difficulty by other means.
Acknowledgement
The authors are pleased to thank Drs Francis Barth and Murielle
Cassagne (Sanofi-Aventis, Montpellier) for helpful discussions
and for their consent to publish this work.
References
[1] a) E. Stern, D. M. Lambert, Chem. Biodiversity 2006, 4, 1707;
b) L. Walter, A. Franklin, A. Witting, C. Wade, Y. Xie, G. Kunos,
K. Mackie, N. Stella, J. Neurosci. 2003, 23, 1398.
´
[2] Z. Jarai, J. A. Wagner, K. Varga, K. D. Lake, D. R. Compton, B. R. Martin,
A. M. Zimmer, T. I. Bonner, N. E. Buckley, E. Mezey, R. K. Razdan,
A. Zimmer, G. Kunos, Proc. Natl. Acad. Sci. USA 1999, 96, 14136.
[3] M. Girard, D. B. Moir, J. W. ApSimon, Can. J. Chem. 1987, 65, 189.
[4] E. R. Beerkhardt, K. Matos, Chem. Rev. 2006, 106, 2617–2650
[5] E. Alonso, D. J. Ramon, M. Yus, J. Org. Chem.,1997, 62, 417.
[6] R. K. Razdan, H.C. Dalzell, G. R. Handrick, J. Am. Chem. Soc. 1974,
96, 5860.
[7] V. V. K. Prasad, S. O. Franklin, J. Label. Compd. Radiopharm. 1985,
22, 353.
[8] a) T. Petrzilka, W. Haefliger, C. Sikemeier, Helv. Chim. Acta 1969,
52, 1102; b) B. Cardillo, L. Merlini, S. Servi, Tetrahedron Lett. 1972,
945; c) S.-H. Baek, M. Srebnik, R. Mechoulam, Tetrahedron Lett.
1985, 26, 1083; d) N. Lander, Z. Ben-Zvi, R. Mechoulam, B. Martin,
M. Nordqvist, S. Agureli, J. Chem. Soc. Perkin 1976, 1, 8.
[9] a) R. V. Stevens, K. F. Albizati, J. Org. Chem. 1985, 50, 632;
b) M. Kato, M. Watanabe, Y. Tooyama, B. Vogler, A. Yoshikoshi,
Synthesis 1992, 1055.
[10] E. W. Gill, G. Jones, J. Label. Compd. Radiopharm. 1972, 8, 237.
[11] S. P. Nikas, G. A. Thakur, A. Makriyannis, J. Label. Compd.
Radiopharm. 2002, 45, 1065.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 180–184