First synthesis of (8-2H3)-(all-rac)-δ-tocopherol

Article abstract of DOI:10.1002/ejoc.200300138

Trideuterated tocopherols were needed to be used as internal standards for our study about simultaneous quantitative determination of tocopherols in foodstuffs by mass spectrometry. Whilst procedures for trideuterated α-tocopherol have been recently optimized, no method is reported as regards δ-tocopherol. Different synthetic approaches are discussed, as well as the first procedure for the synthesis of (8-²H₃)-(all-rac)-δ-tocopherol. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejoc.200300138

FULL PAPER
First Synthesis of (8-2H3)-(all-rac)-δ-Tocopherol
Francesco Mazzini,[a] Emanuele Alpi,[a] Piero Salvadori,*[a] and Thomas Netscher[b]
Keywords: Natural products / Vitamins / Deuterium / Isotopic labeling
Trideuterated tocopherols were needed to be used as internal
δ-tocopherol. Different synthetic approaches are discussed,
standards for our study about simultaneous quantitative as well as the first procedure for the synthesis of (8-2H3)-(all-
determination of tocopherols in foodstuffs by mass spectro-
metry. Whilst procedures for trideuterated α-tocopherol have
been recently optimized, no method is reported as regards
rac)-δ-tocopherol.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
Introduction
ternal standards. In fact, on theoretical grounds, the molec-
ular weight of the internal standard has to be increased,
preferably by at least three mass units, to avoid the inter-
ference of natural isotopes of the analyte with the m/z value
of the labeled compound.[8]
Several synthetic procedures for the preparation of poly-
deuterated tocopherols are known. The deuterium is usually
introduced by the use of labeled formaldehyde and reducing
reagents.[9] For the synthesis of trideuterated (all-rac)-α-
and -β-tocopherol we used[10] a recently optimized pro-
cedure,[11] in which the corresponding (2H2)-morpholino-
methyl tocopherol is prepared by a Mannich reaction and
then reduced with NaCNBD3.
In our studies on the development of quantitative mass
spectrometric determinations of compounds showing rel-
evant nutritional properties in foodstuffs we focused our at-
tention on tocopherols, the main components of the pool
of compounds called vitamin E.
Vitamin E is protective against many diseases, such as
cancer, cardiovascular diseases,[1] cell membrane and DNA
damage by free radicals, oxidation of low-density lipo-
proteins[2] and disorders of other lipid-rich body constitu-
ents.[3] Tocopherols are present in oil seeds, leaves and other
green parts of higher plants. Since vitamin E is only synthe-
sized by plants, it is a very important dietary nutrient for
humans and animals.[4] The tocopherol content in foods is
also important to protect food lipids against autoxidation
and thereby to increase their shelf life and value as whole-
some foods.
The preparation of (all-rac)-δ-tocopherol containing lab-
els in the aromatic methyl group required a totally different
approach that we will describe in this paper as the first syn-
thesis of (8-2H3)-(all-rac)-δ-tocopherol.
The preferred separation methods for the determination
of tocopherols are reversed and normal-phase HPLC[5] with
UV and fluorimetric detection, the latter one being the
most sensitive and accurate for quantifying trace amounts.
In addition, there are other methods like GC[6] and capil-
lary electrophoresis (CE).[7] Generally, the total tocopherol
content is determined, even through there is an increasing
interest in the development and improvement of methods
that allow α-, β-, γ- and δ-tocopherol separation and hence
their determination as single components.
Results and Discussion
In order to avoid cross-talk phenomena and to follow the
most abundant ion in quantitative analyses by tandem mass
spectrometry, the labeling position was chosen on the basis
of the tocopherol fragmentation pattern. The fragment re-
sulting from loss of the aliphatic chain and breaking of the
dihydropyran ring is the most abundant in this kind of
analyses[12] (Figure 1).
To obtain accurate determinations by MS/MS analyses
trideuterated α-, β- and δ-tocopherol were needed as in-
[a]
`
Dipartimento di Chimica e Chimica Industriale, Universita di
Pisa
Via Risorgimento 35, 56126 Pisa, Italy
Fax: (internat.) ϩ39-(0)50/918-260
E-mail: psalva@dcci.unipi.it
[b]
Research and Development, Roche Vitamins Ltd
4070 Basel, Switzerland
Fax: (internat.) ϩ41-(0)61/687-2201
E-mail: thomas.netscher@roche.com
Figure 1. Fragments from tocopherols
2840
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejoc.200300138
Eur. J. Org. Chem. 2003, 2840Ϫ2844

First Synthesis of (8-2H3)-(all-rac)-δ-Tocopherol
FULL PAPER
Hence we considered the introduction of convenient lab- (Scheme 2, LG ϭ leaving group). All such compounds pro-
els into the aromatic methyl group to be the most useful for ved to be very unstable. Neither isolation nor conducting
our purposes.
the whole sequence including the reduction reaction in one
As, to the best of our knowledge, no method is known pot were successful. Minami and Kijima have reported a
for the introduciton of a CH3 (CD3) group selectively in the simple reduction of 3,4,5-trimethoxysalicylic acid to 2,3,4-
8-position of tocol, we envisaged the preparation of a suit- trimethoxyphenol using NaBH4,[17] but preliminary experi-
able trideuterated aromatic building block and its sub- ments gave poor yields. These results prompted us to follow
sequent acid-catalyzed condensation reaction with isophy- the monobenzoate route.[18]
tol, in analogy to the preparation of (all-rac)-α-tocoph-
erol.[13] It should be noted, however, that the adaptation of
such methods to the synthesis of the lower homologue(s) is
not trivial, and known procedures for the preparation of
(all-rac)-δ-tocopherol on a laboratory scale are low-yield-
ing.[14] Protection of the phenolic group in position 4 of
the starting hydroquinone is generally required to avoid the
formation of 5- and 7-monomethyl tocol regioisomers and
possible double-alkylation products in the condensation
step (Scheme 1).
Scheme 2. Key: i) LiAlH4, THF, 78%
The retrosynthetic scheme for the preparation of (α,α,α-
2H3)-methylhydroquinone (1) by introduction of deuterium
labels suggests two possible reaction pathways (Scheme 3):
directed metalation[19] (route a) is a very powerful and easy
synthetic technique, so it was the first to be investigated.
Metalation reactions were performed using N,N-diisopro-
Scheme 1. Key: i) selective protection; ii) acid (cat.), isophytol; iii)
deprotection
Due to the small steric effect of the methyl group, selec- pylcarbamoyl, methyl,[20] and methoxymethyl (MOM) as
tive protection of methylhydroquinone is not an easy task. directing metalation groups (DMG) with the use of nBuLi
TIPS and TBDMS have been employed as O-protective as metalating agent and CH3I as electrophile. Very good
groups for this purpose. Good yields (70% using TBDMS conversion was observed with the MOM and methyl groups
chloride;[15] 79% with TIPS triflate/triethylamine/Et2O/ (up to 92%; nBuLi 1.1 equiv., CH3I 5 equiv., room temp.),
Ϫ75Ǟ20 °C) of the monoprotected derivatives were ob- but we were not able to separate the product from about
tained starting from methylhydroquinone but the silyl 8Ϫ10% of the starting hydroquinone derivative. Therefore,
groups were found to be removable under the acidic con- for a clean synthesis of the desired 1, we followed route b.
ditions used in the next step. Attempts to introduce the tri- Commercially available 2,5-dihydroxybenzoic acid was con-
tyl group (Ph3CCl/DMAP/Et3N) gave low selectivity, which verted into the corresponding methyl ester 2 (Scheme 4),
was even lower when the benzylic group (BnBr/K2CO3/ace- and the two phenolic groups in 2 were protected with
tone) was used.
methyl iodide to avoid unnecessary consumption of LiAlD4
Considering those difficulties in direct selective mono- and to increase yields. Bis-ether 3 was then reduced to the
protection of methylhydroquinone, we looked for an benzylic alcohol 4 with LiAlD4. Treatment of the latter with
alternative approach. Successful examples of the transform- PBr3 gave benzylic bromide[21] (5), which was reduced again
ation of the least hindered phenolic group into a benzyl with LiAlD4 (Ǟ 6) and deprotected[22] with BBr3, giving
ether moiety in 2,5-dihydroxybenzoic acid or esters are the pure trideuterated methylhydroquinone 1 in a 50% yield
known in the literature.[16] The O-benzyl group would rep- over six steps, with 98.5% isotopic purity (GC/MS).
resent a good protective group because of its stability dur-
It is noteworthy that protection of the phenolic OH
ing condensation reactions and the mild conditions that can groups as MOM ethers (which would be easier to cleave
be used for its subsequent removal. On a theoretical basis, than methyl ether groups) was not suitable for the synthesis
the aromatic ester moiety can be reduced to a methyl group of the bromide from the benzylic alcohol. We also tried the
through a multi-step reduction sequence with the corre- conversion into the tosylate, mesylate, and chloride under
sponding benzylic alcohol as an intermediate. Unfortu- various conditions, but we always obtained polymeric mate-
nately, we were unable to synthesize the corresponding mes- rial. Considering the successful reaction performed on 5, it
ylate, tosylate, bromide, or chloride for the last step is plausible to suppose an anchimeric assistance of the
Eur. J. Org. Chem. 2003, 2840Ϫ2844
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2841

F. Mazzini, E. Alpi, P. Salvadori, T. Netscher
FULL PAPER
now for the synthesis of 4-benzoyloxy-2-trideuteromethyl-
phenol.
Scheme 6. Key: i) PhCOCl/Py, cryst.; ii) flash chromatography sep-
aration; iii) 1. ZnCl2 (1.5 equiv.), iBuOAc, aq. HCl (cat.), isophytol,
2. KOH/MeOH, room temp.
For the following condensation step the conditions used
for the preparation of (all-rac)-α-tocopherol[23] were ap-
plied. Although the ZnCl2-catalyzed condensation of
monobenzoate 8 with isophytol ran less smoothly, pure (8-
2H3)-(all-rac)-δ-tocopherol (11) was obtained after saponifi-
cation in 60% yield with a 98.5% deuteration of CD3. No
D2 or D1 analogs were detected by mass spectrometry
analysis. It is assumed that the lower yield compared to the
one obtained in α-tocopherol synthesis by the same pro-
cedure is due to partial ester hydrolysis and to the reduced
reactivity of the mono-ester than the hydroquinone.
Scheme 3
Conclusion
Unlike its α- and β-analogs, trideuterated δ-tocopherol
labeled at the aromatic methyl group at the 8-position re-
quires a complex multi-step synthesis. Optimization of each
step was necessary to increase the overall yield. Hence dif-
ferent approaches were followed to obtain (α,α,α-2H3)-
methylhydroquinone derivatives protected at the 4-position
as key building blocks. Benzoylation gave the best results
for selective protection, allowing the successful synthesis of
pure (8-2H3)-(all-rac)-δ-tocopherol with a high percentage
of deuteration (Ͼ98%). Introduction of a benzyl group at
the 4-position is highly attractive too, but this proved to be
difficult, as shown by the various procedures examined for
the synthesis of the desired 4-benzyloxy-2-methylphenol.
Further investigations following the Minami method, as
well as other routes, are in progress.
Scheme 4. Key: i) MeOH, H2SO4, 96%; ii) 1. NaH/THF 2. CH3I,
92%; iii) LiAlD4/THF, 92%; iv) PBr3/CH2Cl2, sublimation, 82%;
v) LiAlD4/THF, 91%; vi) BBr3/CH2Cl2, 82%
MOM group with the leaving group X, promoting the for-
mation of a highly reactive carbocation 7 (Scheme 5) that
gives rise to the polymeric material isolated.
The preparation of (8-2H3)-(all-rac)-δ-tocopherol is, to
the best of our knowledge, the first synthesis ever reported
concerning a deuterated δ-tocopherol. It allows accurate
MS/MS quantitative determinations in different foodstuff
matrices and can be used as a tool to develop metabolic
studies of the non-vitaminic activity of δ-tocopherol.
Scheme 5
Methylhydroquinone (1) prepared according to Scheme 4
was then converted into a mixture of monoprotected benzo-
ates 8 and 9 (ca. 65:35), and dibenzoate 10 (Scheme 6). For
lab-scale isolation of monobenzoate 8, the original pro-
cedure[18] was optimized (by using unlabeled material).
After treatment with benzoyl chloride in pyridine, the 4-
benzoyloxy derivative 8 was separated from 9 and 10 by
fractional crystallization and careful flash chromatography
in reasonable yield (45% on a 800-mg scale). Though this
Experimental Section
General: All reactions were performed under an inert atmosphere
(argon or N2). Melting points were measured on a Reichert-Jung
Thermovar apparatus. 1H and 13C NMR spectra were recorded on
a Varian Gemini 200 spectrometer (200 MHz and 50.3 MHz,
respectively), in CDCl3 solution with Me4Si or CHCl3 as internal
2
standards. H NMR spectra were recorded on a Varian VXR 300
result is not exceptional, this procedure is the best up to spectrometer (46 MHz for 2H) in CHCl3 with CDCl3 as internal
2842
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 2840Ϫ2844

First Synthesis of (8-2H3)-(all-rac)-δ-Tocopherol
FULL PAPER
standard. Chemical shifts are expressed on the δ scale (ppm). GLC
analyses were performed on a PerkinϪElmer 8600 Model instru-
firmed by TLC (Hex/EtOAc, 5:1). After cooling to 0 °C and drop-
wise addition of water (50 mL), the reaction mixture was stirred
ment, equipped with a DB1 column (30 m ϫ 0.52 mm, 5 µm film for an additional 30 min and then filtered. The white aluminate
thickness), and on a PerkinϪElmer AutoSystem XL Model instru- salts were washed with diethyl ether, and the organic solvent was
ment with a BP-1 capillary column (12 m ϫ 0.25 mm, 0.25 µm film evaporated. Usual workup in diethyl ether of the resulting aqueous
1
thickness), both with He as carrier gas and with a flame ionization
detector. The atmospheric pressure chemical ionization mass spec-
tra were acquired on a PerkinϪElmer Sciex API III Plus mass spec-
phase gave pure 4 (3.15 g, 92% yield) as a yellowish oil. H NMR
(CDCl3/TMS): δ ϭ 2.74 (br. s, 1 H, 1-OH) 3.72 (s, 3 H, 5-CH3),
3.80 (s, 3 H, 2-CH3), 6.55Ϫ6.95 (m, 3 H, Harom) ppm; no detectable
trometer. GC/MS analyses were recorded on a Saturn 2000 GC- benzyl proton signal at δ ϭ 4.64 ppm. 2H NMR (CHCl3): δ ϭ
MS/MS Varian Chromatography System mass spectrometer con-
nected to a 3800 Varian gas chromatograph equipped with a DB-
5 capillary column (30 m ϫ 0.25 mm, 0.25 µm film thickness).
Commercial reagents and solvents were purchased from Aldrich,
Fluka, or Merck, and purified by standard methods when neces-
sary. Tetrahydrofuran was distilled from over Na/K alloy before
use. Dichloromethane was distilled from over P2O5. Pyridine was
distilled from over potassium hydroxide and stored over KOH. Li-
AlD4 (96 atom% D) and CD3I (Ն 99.5 atom% D) were acquired
from Aldrich. Deuterated paraformaldehyde (CD2O)n (Ն 99.5
atom% D) was purchased from C/D/N Isotopes (Canada). All
other commercial reagents were used without further purification.
Column chromatography was performed on silica gel 60 (70Ϫ230
mesh and 230Ϫ400 mesh). TLC was performed on silica gel
MachereyϪNagel Alugram Sil G/UV254 (0.20 mm). Hex means n-
hexane. Usual workup involved three extractions into the solvent
specified. The organic extracts were combined, washed with water
until the aqueous phase was neutral, dried over Na2SO4, filtered
and then concentrated on a rotary evaporator under vacuum. The
residue was further dried to constant weight under high vacuum.
All yields given refer to isolated yields.
130.0 (C-1), 151.4 (C-2), 114.4 (C-3), 112.7 (C-4), 153.6 (C-5), 110.9
(C-6), 55.6 (5-CH3), 55.7 (2-CH3), 60.1 (m, CD2OH) ppm.
(α,α-2H2)-2,5-Dimethoxybenzyl Bromide (5): A magnetically stirred
solution of 4 (2.77 g, 16.29 mmol) in dry CH2Cl2 (25 mL) was co-
oled to 0 °C, and a solution of PBr3 (610 µL, 6.49 mmol) in dry
CH2Cl2 (15 mL) was added dropwise over 20 min. The reaction
mixture was allowed to warm to room temp. and stirred overnight,
after which time the end of the reaction was confirmed by TLC
(Hex/EtOAc, 3:1). After usual workup in CH2Cl2 the crude product
was purified by sublimation (90Ϫ100 °C, 0.1 Torr), giving pure 5
(3.11 g, 82% yield) as a white solid. M.p. 74Ϫ76 °C. 1H NMR
(CDCl3/TMS): δ ϭ 3.76 (s, 3 H, 5-CH3), 3.86 (s, 3 H, 2-CH3),
6.70Ϫ6.93 (m, 3 H, aromatic H) ppm; no detectable benzyl proton
2
signal at δ ϭ 4.55 ppm. H NMR (CHCl3): δ ϭ 4.48 (s, benzylic
D) ppm. 13C NMR (CDCl3): δ ϭ 26.8 (C-1), 153.4 (C-2), 114.9 (C-
3), 112.0 (C-4), 151.7 (C-5), 116.3 (C-6), 55.7 (5-CH3), 56.2 (2-
CH3), 29.2 (m, CD2Br) ppm.
(α,α,α-2H3)-1,4-Dimethoxy-2-methylbenzene (6):
A magnetically
stirred suspension of LiAlD4 (630 mg, 15 mmol) in dry THF
(15 mL) was cooled to 0 °C, and a solution of bromide 5 (3.07 g,
13.17 mmol) in dry THF (20 mL) was added dropwise. The reac-
tion mixture was allowed to warm to room temp. and stirred over-
night; the end of the reaction was confirmed by TLC (Hex/EtOAc,
5:1). After cooling to 0 °C and dropwise addition of water (50 mL),
the reaction mixture was filtered by suction filtration. The white
aluminate salts were washed with diethyl ether, and the organic
solvent was evaporated. Usual workup in diethyl ether of the re-
sulting aqueous phase and bulb-to-bulb distillation (50Ϫ60 °C,
Methyl 2,5-Dihydroxybenzoate (2): 2,5-Dihydroxybenzoic acid
(6.73 g, 43.70 mmol) was dissolved in MeOH (45 mL, HPLC grade,
1.11 mol, 25 equiv.) and mixed with 95% H2SO4 (3 mL,
53.22 mmol, 1.2 equiv.). The solution was refluxed overnight,[24]
after which time no starting acid could be detected by TLC (Hex/
EtOAc, 4:1). The solvent was then evaporated and water added.
Usual workup in diethyl ether gave pure 2 (7.05 g, 96% yield) as a
white solid. M.p. 84Ϫ86 °C; NMR (CDCl3/TMS): δ ϭ 3.91 (s, 3
H, 1-CH3), 5.49 (s, 1 H, 5-OH), 6.87 (d, J ϭ 8.6 Hz, 1 H, H-3),
7.02 (dd, J1 ϭ 3.0 Hz, J2 ϭ 8.5 Hz, 1 H, H-4), 7.28 (d, J ϭ 3.6 Hz,
1 H, H-6), 10.39 (s, 1 H, 2-OH) ppm.
1
1 Torr) gave pure 6 (1.86 g, 91% yield) as a yellowish oil. H NMR
(CDCl3/TMS): δ ϭ 3.76 (s, 3 H, 5-CH3), 3.80 (s, 3 H, 2-CH3)
6.65Ϫ6.85 (m, 3 H, aromatic H) ppm; no detectable aromatic
methyl proton signal at δ ϭ 2.24 ppm. 2H NMR (CHCl3): δ ϭ 2.23
(s, CD3) ppm.
Methyl 2,5-Dimethoxybenzoate (3): A magnetically stirred suspen-
sion of dry NaH (1.84 g, 75 mmol) in dry THF (15 mL) was cooled
to 0 °C. A solution of ester 2 (4.2 g, 25 mmol) in dry THF (15 mL)
was added dropwise over 20 min. The reaction mixture was allowed
to warm to room temp. and stirred for an additional hour. The
mixture was cooled again to 0 °C, and a solution of CH3I (6.23 mL,
100 mmol) in dry THF (15 mL) was added dropwise. The reaction
mixture was allowed to warm to room temp. and stirred overnight.
The end of the reaction was confirmed by TLC (Hex/EtOAc, 5:1).
Water (50 mL) was added, and the THF evaporated. After usual
workup in diethyl ether and column chromatography (Hex/EtOAc,
(α,α,α-2H3)-Methylhydroquinone (1): A magnetically stirred solu-
tion of 6 (1.3 g, 8.39 mmol) in dry CH2Cl2 (4 mL) was cooled to
Ϫ78 °C, and BBr3 (27.7 mL of a 1  dichloromethane solution,
27.7 mmol) was added dropwise. The reaction mixture was allowed
to warm to room temp. and stirred for 15 h; TLC (Hex/EtOAc, 3:1)
confirmed the end of the reaction. Next sat. aq. NaCl solution was
carefully added to the mixture. Usual workup in diethyl ether and
column chromatography (CH2Cl2/EtOAc, 6:1) afforded 1 (870 mg,
82% yield) as a white solid. M.p. 126Ϫ127.5 °C. 98.5% deuteration
1
1
of CD3 (MS). H NMR (CDCl3/TMS, 45 °C): δ ϭ 4.34 (s, 2 H, 2
4:1) pure 3 (4.55 g, 92% yield) was obtained as a yellowish oil. H
OH), 6.55 (dd, J1 ϭ 3.5 Hz, J2 ϭ 2.1 Hz, 1 H, H-4), 6.62 (d, J ϭ
2.1 Hz, 1 H, H-3), 7.28 (d, J ϭ 3.6 Hz, 1 H, H-6) ppm; no detect-
able aromatic methyl proton signal at δ ϭ 2.12 ppm. 2H NMR
(CHCl3): δ ϭ 2.11 (s, CD3) ppm. GC/MS (DB-5, 30 m, 70Ǟ300
°C, 20 °C/min, 10 min a 300 °C): tR(1) ϭ 3.9 min. ESI-MS: m/z ϭ
127 (100), 98 (35).
NMR (CDCl3/TMS): δ ϭ 3.65 (s, 3 H, 1-CH3), 3.82 (s, 3 H, 5-
CH3), 3.87 (s, 3 H, 2-CH3) 6.87 (d, J ϭ 8.6 Hz, 1 H, H-3), 7.01
(dd, J1 ϭ 3.0 Hz, J2 ϭ 8.5 Hz, 1 H, H-4), 7.32 (d, J ϭ 3.6 Hz, 1 H,
H-6) ppm.
(α,α-2H2)-2,5-Dimethoxybenzyl Alcohol (4): A magnetically stirred
suspension of LiAlD4 (1.0 g, 23.82 mmol) in dry THF (15 mL) was
cooled to 0 °C. A solution of ester 3 (3.95 g, 20.15 mmol) in dry
(α,α,α-2H3)-4-Benzoyloxy-2-methylphenol (8): A solution of benzoyl
THF (15 mL) was added dropwise over 20 min, and the reaction chloride (0.76 mL, 6.49 mmol, 1.03 mol equiv.) in dry pyridine
mixture was stirred overnight. The end of the reaction was con- (15 mL) was added dropwise at Ϫ10 °C to a stirred solution of
Eur. J. Org. Chem. 2003, 2840Ϫ2844
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2843

F. Mazzini, E. Alpi, P. Salvadori, T. Netscher
FULL PAPER
[5f]
Soc. 1985, 62, 898Ϫ962.
F. Gutierrez-Rosales, J. Garrido-
Fernandez, L. Gallardo-Guerrero, B. Gandul-Rojas, M. I.
Minguez-Mosquera, J. Am. Oil Chem. Soc. 1992, 69, 866Ϫ871.
hydroquinone 1 (800 mg, 6.3 mmol) in dry pyridine (15 mL). The
reaction mixture was stirred for 3 h at Ϫ10 °C, allowed to warm to
room temp., and then acidified with 3  HCl. After usual workup
in diethyl ether, the crude product was dissolved in MeOH (5 mL)
and kept at Ϫ20 °C overnight. The crystals formed (dibenzoate
derivative 10) were filtered and washed with cold MeOH (5Ϫ7 mL).
The filtrate and combined washings were evaporated, and the resi-
due (GC, 8:9 65%:35%) was purified by careful flash chromatogra-
phy (90 g SiO2, Hex/EtOAc, 4:1). From the first eluting zone
370 mg of 8 (GC, 92%) was obtained, and this was recrystallized
from 4 mL of Hex/EtOAc (4:1), giving 330 mg of 8 (purity 98.2%,
GC). From the second eluting zone 500 mg (GC, 8:9 70:30) was
obtained that gave 320 mg of 8 (purity 99%, GC) after a second
flash chromatography. Overall yield 45%. M.p. 112Ϫ114 °C. 1H
NMR (CDCl3/TMS): 5.15 (br. s, OH, 1 H), 6.7 (d, J ϭ 8.6 Hz, 1
H), 6.86 (dd, J1 ϭ 2.6, J2 ϭ 8.6 Hz, 1 H), 6.96 (d, J ϭ 2.6 Hz, 1
H), 7.47Ϫ7.7 (m, 3 H), 8.19 (m, 2 H) ppm; no detectable aromatic
methyl proton signal at δ ϭ 2.23 ppm. 2H NMR (46 MHz, CHCl3):
δ ϭ 2.19 (s) ppm. 13C NMR (CDCl3): δ ϭ 15.9 (m, CD3), 115.9,
120.2, 124.5, 125.8, 128.9, 129.8, 130.6, 134.1, 144.6, 152.2, 166.2
ppm. GC on BP-1, 12 m, 70 Ǟ300 °C, 20 °C/min, 10 min at 300
°C: tR(8) ϭ 8.15 min, tR(9) ϭ 7.95 min. APCI-MS (in MeOH):
m/z ϭ 232 [M ϩ Hϩ].
[5g]
R. Peto, R. Doll, J. D. Buckley, M. B. Sporn, Nature 1981,
290, 201Ϫ208. [5h] R. B. Shekelle, M. Lepper, S. Liu, P. Oglesby,
[5i]
A. M. Shryock, J. Stamler, Lancet 1981, 2, 1185Ϫ1190.
G.
W.
[5j]
W. Burton, K. U. Ingold, Science 1984, 224, 569Ϫ573.
Schüep, R. Rettenmaier, Methods Enzymol. 1994, 234,
[5k]
294Ϫ302.
C. Lenfant, F. C. Thyrion, Ol., Corps Gras,
[5l]
Lipides 1996, 3, 220Ϫ226.
S. L. Abidi, J. Chromatogr. A
2000, 881, 197Ϫ216.
[6] [6a]
A. J. Sheppard, J. A. T. Pennington, J. L. Weihrauch, in
Vitamin E in Health and Disease (Eds.: L. Packer, J. Fuchs),
[6b]
Marcel Dekker, Inc., New York, 1993, pp. 9Ϫ31.
Netscher, Chimia 1996, 50, 563Ϫ567.
T.
[7] [7a]
S. Fanali, P. Catarcini, M. G. Quaglia, E. Camera, M. Rin-
aldi, M. Picardo, J. Pharm. Biomed. Anal. 2002, 29, 973Ϫ979.
[7b]
S. L. Abidi, K. A. Rennick, J. Chromatogr. A 2001, 913,
379Ϫ386. We thank one of the referees for bringing this to
our attention.
[8]
A. P. De Leenheer, M. F. Lefevere, W. E. Lambert, E. S.
Colinet, Adv. Clin. Chem. 1985, 24, 111Ϫ161, and references
cited therein.
L. Hughes, M. Slaby, G. W. Burton, K. U. Ingold, J. Labelled
Compd. Radiopharm. 1990, 28, 1049Ϫ1057.
[9]
[10]
E. Alpi, F. Mazzini, Th. Netscher, P. Salvadori, 4th Italian-
Spanish Symposium on Organic Chemistry (ISSOC-4), August
31-September 3, 2002 Perugia, Italy, contribution PO-44.
Th. Netscher, R. K. Müller, J. Schneider, H. Schneider, P.
Bohrer, R. Jestin, 12th European Symposium on Organic
Chemistry, 13Ϫ18 July 2001, Groningen, The Netherlands,
contribution P2Ϫ81; Fifth Electronic Conference on Synthetic
Organic Chemistry, 1Ϫ30 Sept. 2001, http://www.mdpi.org/
ecsoc-5.htm, Molecular Diversity Preservation International,
Basel, Switzerland, contribution c0007.
S. E. Scheppele, R. K. Mitchum, Ch. J. Rudolph Jr., K. F.
Kinneberg, G. V. Odell, Lipids 1972, 7, 297Ϫ304.
K. U. Baldenius, L. von dem Bussche-Hünnefeld, E. Hilgem-
ann, P. Hoppe, R. Stürmer, in Ullmann’s Encyclopedia of Indus-
trial Chemistry, VCH Verlagsgesellschaft, Weinheim, 1996, Vol.
A27, Chap. 4(Vitamin E), pp. 478.
(8-2H3)-(all-rac)-δ-Tocopherol (11):
A solution of isophytol
(0.57 mL, 1.62 mmol) in iBuOAc (5 mL) was added dropwise very
slowly (over 3 h) at 50 °C to a stirred mixture of monobenzoate 8
(250 mg, 1.08 mmol), 0.5 mL of aq. HCl Ͼ36.5% and anhydrous
ZnCl2 (225 mg, 1.62 mmol) in iBuOAc (5 mL). The reaction mix-
ture was stirred for a further 3 h at 50 °C and then allowed to cool
to room temp. After workup by extraction with toluene/water, the
resulting crude brownish mixture was saponified by stirring for
4.5 h at room temp. with methanolic 1  KOH (12 mL). The mix-
ture was then acidified with 2  HCl, and the MeOH was evapo-
rated. Usual workup and flash chromatography (Hex/EtOAc, 9:1)
gave pure 11 (260 mg, 60% yield) as a brown-yellow oil. 98.5% de-
[11]
[12]
[13]
1
uteration of CD3 (MS). H NMR (CDCl3/TMS): δ ϭ 0.7Ϫ1.9 (m,
[14] [14a]
H2-C(3), H3-C(2a) and C16H33 chain), 2.65 (t, J ϭ 7 Hz, H2-C(4),
2 H), 4.4 (br. s, OH, 1 H), 6.38 [d, J ϭ 2.6 Hz, H-C(5), 1 H], 6.47 [s,
1 H, H-C(7)] ppm; no detectable aromatic 8-methyl proton signal at
M. H. Stern, Ch. D. Robeson, L. Weisler, J. G. Baxter, J.
[14b]
Am. Chem. Soc. 1947, 69, 869Ϫ874.
S. Marcinkiewicz, D.
McHale, P. Mamalis, J. Green, J. Chem. Soc. 1959, 3377Ϫ3378.
[14c]
2
J. Green, D. McHale, P. Mamalis, S. Marcinkiewicz, J.
δ ϭ 2.09 ppm. H NMR (46 MHz, CHCl3): δ ϭ 2.08 [s, D3C-C(8)]
ppm. 13C NMR (CDCl3): δ ϭ 147.8, 146, 127.1, 121.3, 115.8, 112.5,
75.5, 40.1, 39.4, 37.3Ϫ37.6, 32.8, 32.7, 31.4, 28, 24.8, 24.4, 24, 22.7,
22.6, 22.5, 21, 19.5Ϫ19.7 ppm. APCI-MS (in MeOH): m/z ϭ 406
[M ϩ Hϩ], 140 (fragmentation see Figure 1). C27H43D3O2 (405.2):
calcd. C 79.94, H 12.17; found C 79.91, H 12.19.
Chem. Soc. 1959, 3374Ϫ3376.
[15]
J. L. Stanton, E. Cahill, R. Dotson, J. Tan, H. C. Tomaselli, J.
M. Wasvary, Z. F. Stephan, R. E. Steele, Bioorg. Med. Chem.
Lett. 2000, 10, 1661Ϫ1664.
[16]
[17]
U. Schmidt, H. Bokens, A. Lieberknecht, H. Griesser, Tetra-
hedron Lett. 1981, 22, 4949Ϫ4952.
N. Minami, S. Kijima, Chem. Pharm. Bull. 1980, 28,
1648Ϫ1650. We thank one of the referees for bringing this to
our attention.
Acknowledgments
We are very grateful to Mrs. Marion Burglin, Mr. Adrian Friedli,
and Mr. Christian Schöni for their technical assistance.
[18]
P. Mamalis, J. Green, S. Marcinkiewicz, D. McHale, J. Chem.
Soc. 1959, 3350Ϫ3357.
[19]
[20]
V. Snieckus, Chem. Rev. 1990, 90, 879Ϫ933.
´
M. C. Carren˜o, J. L. Garcıa Ruano, M. A. Toledo, A. Urbano,
Tetrahedron: Asymmetry 1997, 8, 913Ϫ921.
S. J. Zweig, N. Castagnoli Jr., J. Med. Chem. 1977, 20,
414Ϫ421.
E. H. Vickery, L. F. Pahler, E. J. Eisenbraun, J. Org. Chem.
1979, 44, 4444Ϫ4446.
U. S. Patent 5,663,376 (2 Sept. 1997), N. Hirose, H. Inoue, T.
Matsunami, T. Yoshimura, K. Morita, Y. Horikawa, N. Iwata,
N. Minami, K. Hayashi, Ch. Seki (Eisai Co., Japan).
B. S. Furniss, A. J. Hannaford, P. W. G. Smith, A. R. Tatchell,
Vogel’s Textbook of Practical Organic Chemistry 5th ed., Long-
man Scientifc and Technical 1989, NY, 1078.
Received March 4, 2003
[21]
[22]
[23]
[1]
K. F. Gey, P. Puska, Ann. N. Y. Acad. Sci. 1989, 570, 268Ϫ282.
[2]
˚
A. Kamal-Eldin, L. A. Appelqvist, Lipids 1996, 31, 671Ϫ701.
[3]
L. S. Ching, S. Mohamed, J. Agric. Food Chem. 2001, 49,
3101Ϫ3105, and references cited therein.
[4]
G. Pongracz, H. Weiser, D. Matzinger, Fat Sci. Technol. 1995,
97, 90Ϫ104.
[5] [5a]
A. Ranalli, Ital. J. Food Sci. 1992, 4, 53Ϫ57. [5b] M. Vitigli-
[5c]
[24]
ano, E. Turri, Olearia 1958, 12, 145Ϫ150.
M. Rahmani, A.
[5d]
Saari Csallany, J. Am. Oil Chem. Soc. 1991, 68, 672Ϫ674.
Y. Endo, R. Usuki, T. Kaneda, J. Am. Oil Chem. Soc. 1984,
61, 781Ϫ784. [5e] A. Kiritsakis, L. R. Dugan, J. Am. Oil Chem.
2844
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 2840Ϫ2844