Synthesis of (R)-α-benzylmethionine: a novel rearrangement during alkylation of the Seebach (R)-methionine oxazolidinone

Article abstract of DOI:10.1039/b303471m

Alkylation of the enolate of the Seebach (R)-methionine oxazolidinone with benzyl bromide gave the expected benzylated product in low yield. The major product was a novel amine arising from oxazolidinone cleavage, decarboxylation, alkylation and finally hydrolysis. The rearrangement could be suppressed by using a more reactive electrophile or by using the N-Cbz instead of the N-benzoyl protecting group, and the required (R)-α-benzyl-methionine was obtained in 78% yield and in an enantiomeric ratio of 90 : 10.

Full text of DOI:10.1039/b303471m

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis of (R)-ꢀ-benzylmethionine: a novel rearrangement during
alkylation of the Seebach (R)-methionine oxazolidinone
Panayiotis A. Procopiou,* Mahmood Ahmed, Séverine Jeulin and Rossana Perciaccante
GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage, Hertfordshire,
UK SG1 2NY. E-mail: pan.a.procopiou@gsk.com; Fax: 01438-763438; Tel: 01438-768428
Received 27th March 2003, Accepted 25th June 2003
First published as an Advance Article on the web 14th July 2003
Alkylation of the enolate of the Seebach (R)-methionine oxazolidinone with benzyl bromide gave the expected
benzylated product in low yield. The major product was a novel amine arising from oxazolidinone cleavage,
decarboxylation, alkylation and finally hydrolysis. The rearrangement could be suppressed by using a more reactive
electrophile or by using the N-Cbz instead of the N-benzoyl protecting group, and the required (R)-α-benzyl-
methionine was obtained in 78% yield and in an enantiomeric ratio of 90 : 10.
Table 1 Variation of yield of 4 with base used for enolising 3
As a part of a research programme directed toward the design
and synthesis of novel peptidomimetics, we required an efficient
synthesis of (R)-α-benzylmethionine (1). There are no reports
in the literature describing the preparation of 1 apart from a
short communication outlining a chiral TLC method for resol-
ving racemic α-dialkyl amino acids and reporting Rf values of
the respective enantiomers.1
Entry
Base
4 (%)
3 (%)
1
2
3
4
LiNEt2
7
24
9
6
19
LiN(TMS)2
NaN(TMS)2
KN(TMS)2
17 (87 : 13)
17 (72 : 28)
10
Methods for the asymmetric synthesis of homochiral
α-substituted α-amino acids include Schöllkopf’s bis-lactim
ether,2 Seebach’s oxazolidinones,3 Williams’ diphenyloxazin-
ones4 and Corey’s5 and Lygo’s6 asymmetric alkylation of
alanine Schiff’s bases. The most commonly used method, at the
time that this work was initiated, was Seebach’s oxazolidinone
methodology or one of its variants.7
The oxazolidinone 3 was prepared from (R)-methionine (2),
pivalaldehyde and benzoyl chloride in 30% yield according to
the Seebach protocol3 (Scheme 1). The trans (2S,4R) isomer 3
was also obtained (6%). Enolisation of cis-3 with lithium
diethylamide in tetrahydrofuran at Ϫ70 ЊC and reaction with
benzyl bromide gave only 7% of the expected alkylation prod-
uct 4 and 6% recovered starting material 3. Seebach recom-
mended the use of lithium diethylamide as the base having
previously examined a number of bases for the generation of
the corresponding enolate. Fadel subsequently recommended
the use of lithium diethylamide particularly for the enolisation
of methionine and valine oxazolidinones.8 The relationship
between the low yield of the alkylation product 4 and the base
used to form the enolate was not obvious to us and we therefore
examined the enolisation of 3 using three more bases and the
results of its alkylation with benzyl bromide are shown in Table
1. The yield of alkylated product 4 remained low (maximum
24%), and there was always recovered starting material. Inter-
estingly the recovered starting material was not pure cis-3 but a
mixture of cis : trans isomers (87 : 13 in entry 3 and 72 : 28 in
entry 4). This suggested that enolisation does occur with other
bases and that alkylation is slow. The unreacted enolate was
quenched on work-up to give 3 as a mixture of isomers based
on the 2-H, tert-Bu and MeS signals in the 1H NMR spectrum.
More importantly an additional unexpected product (5), which
Scheme 1 Reagents and conditions: i) NaOH, tert-BuCHO, PhCOCl,
CH2Cl2; ii) base, BnBr, THF; iii) silica gel chromatography or aqueous
acid.
was unstable to silica gel and aqueous acid, was observed in
entries 3 and 4. This product was less polar than 4 and was
probably present in entries 1 and 2, however, it was not dis-
tinguished from benzyl bromide by the TLC system used at the
time. Careful chromatography of the unknown product 5 gave a
new more polar product 6 (the most polar component in the
reaction mixture). The polar decomposition product 6 was
identified as the α-amino ketone as shown based on spectro-
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 8 5 3 – 2 8 5 8
2853

View Article Online
scopic evidence and microanalytical data of its hydrochloride
salt.
1H NMR using chiral solvating agent (R)-2,2,2-trifluoro-1-
(9-anthryl)ethanol and also by chiral HPLC and found to be
90 : 10.
Compound 5 was the second least polar component in the
reaction mixture (after benzyl bromide) on TLC. Although 5
was the major component in the reaction mixture (TLC) before
work-up only a small amount of 5 could be isolated, following
quenching of the reaction mixture with 2.5% aqueous ammo-
nium chloride solution (pH 6) and flash chromatography. The
A search of the literature for any reports of oxazolidinones
giving unexpected by-products revealed the following: See-
bach10 described the enolates of oxazolidinones as less stable
and poorer nucleophiles than the corresponding enolates of
imidazolidinones, reacting only with more reactive electro-
philes. No decomposition products were reported however, and
only the alkylation of (2S,4S)-3 with the least hindered and
very reactive methyl iodide was described. Kemp11 reported
that the above enolate failed to react with 2-phthalimidoethyl
bromide or tosylate, however, it reacted with bromoacetonitrile
to give the expected alkylation product in 42% yield.
1
structure of 5 was deduced from the H NMR spectrum. The
presence of two AB quartets due to the benzylic protons was
consistent with two geometrical isomers of the imine 5. The
Seebach group reported the X-ray diffraction structure of oxa-
zolidinone 3 which showed the five-membered ring to be essen-
tially planar with the tert-butyl and acyl groups occupying
opposite sides of the ring plane which was effected by pyramid-
alisation of the amide N-atom.9 The amide carbonyl O-atom
occupied an s-cis-position with the tert-butyl-substituted acetal
C-atom. The enolate of 3 would be expected to retain the five-
membered ring planar, and the carbonyl of the amide group
anti to the C4 double bond (Scheme 2).
Mutter,12 using alanine trans-oxazolidinone and ethyl iodide
as the electrophile, reported unsuccessful alkylation possibly
because of “ring-opening to an N-acyliminium species and
lability of the configuration at C2”, however, no degradation
products were isolated. Abell,13 using phenylalanine and valine
trans-oxazolidinones and benzhydryl or ethyl bromoacetate as
the electrophile, observed “self-addition by-products”.
Although in the current study the formation of 5 and 6 was
suppressed by using a more reactive electrophile, nevertheless,
the enantiomeric ratio of the resulting dialkyl amino acid was
90 : 10. It was therefore of interest to compare some other
oxazolidinone variants for the stability of the corresponding
enolates and for the enantiomeric ratio of the resulting amino
acid. The carbamate carbonyl of N-Cbz oxazolidinones is less
electrophilic than the benzamide carbonyl of N-COPh oxazol-
idinones and hence not expected to give rise to rearrangement
products. The oxazolidinone 12, prepared by the method of
Duffy,14 was treated with lithium bis(trimethylsilyl)amide and
benzyl bromide in tetrahydrofuran and gave the oxazolidinone
13 in 43% yield (Scheme 4).
Scheme 2 Proposed mechanism for the rearrangement of enolate 7 to
5 and 6.
As alkylation with benzyl bromide was slow, the enolate 7
could attack intramolecularly the amide carbonyl to produce
the strained bicyclic intermediate 8, which would rapidly
rearrange to the β-ketoacid 9. This in turn would readily de-
carboxylate to form the new enolate 10, which on reaction with
benzyl bromide would produce imine 5. As the geometry of the
enolate 10 is not known, the configuration of 5 is not known
either and was assumed to be racemic. Hydrolysis of 5 would
produce amine 6. One way of minimising the degradation of
the enolate of 3 and increasing the formation of the alkylation
product 4 would be to use a more reactive electrophile. Thus,
treatment of 3 with lithium bis(trimethylsilyl)amide at Ϫ65 ЊC,
followed by benzyl iodide gave after crystallisation the expected
product 4 in 80% yield. Deprotection of 4 by heating to reflux in
hydrobromic acid gave 1 in 96% yield (Scheme 3).
Scheme 4 Reagents and conditions: i) NaOH, H2O; ii) tert-BuCHO,
CH2Cl2, 4 Å MS; iii) BnOCOCl; iv) LiN(TMS)2, THF, Ϫ78 ЊC;
v) BnBr; vi) KOSiMe3, THF, 80 ЊC; vii) AcCl, Et3N, CH2Cl2.
1
The H NMR spectrum of 13 as in the case of 4 was very
broad, however, when the 1H NMR spectrum of 13 was run in
DMSO-d6 at 80 ЊC the broadening was eliminated, indicating
the presence of rotamers. The broadness of the NMR spectra
of alkylated oxazolidinones was also reported by Karady.7
Attempted cleavage of the oxazolidinone with sodium hydrox-
ide in refluxing methanol over a period of three days was
unsuccessful, however, treatment of 13 with potassium tri-
methylsilanolate in refluxing tetrahydrofuran over 2 h com-
pletely de-protected 13 and provided 1 in 98% yield.15 The
enantiomeric ratio was determined on the crude acetyl
derivative 11 and found to be 91 : 9.
The amino acid 1 was finally converted to the N-acetyl
derivative 11 for further characterisation and enantiomeric
purity determination. The enantiomeric ratio was obtained by
The (R)-methionine derivative 14, prepared by the method
of Davies,16 was found to be very unstable and therefore the
alkylation was not attempted.
Scheme 3 Reagents and conditions: i) conc. HBr, reflux; ii) AcCl, Et3N,
CH2Cl2.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 8 5 3 – 2 8 5 8
2854

View Article Online
at a flow rate of 3 ml minϪ1. The mass spectra were recorded on
a Fisons VG Platform spectrometer using electrospray positive
and negative mode (ESϩve and ESϪve). Analytical HPLC was
conducted on a Supelcosil LCABZϩPLUS column (3.3 cm ×
4.6 mm) eluting with 0.1% formic acid and 0.01 M ammonium
acetate in water (solvent A), and 0.05% formic acid and 5%
water in acetonitrile (solvent B), using the following elution
gradient: 0–2.00 min 0% B, 2.00–22.00 min 100% B, 22.00–
27.00 min 100% B, 27.00–29.00 min 0% B, 29.00–30.00 min 0%
B, at a flow rate of 1 ml minϪ1, injecting 5 µl of solution and
detecting between 215 and 330 nm. Column chromatography
was performed on Merck Kieselgel 60 (art. 9385), or Biotage
pre-packed silica gel cartridges containing KP-Sil run on a flash
12i chromatography module. Optical rotations were measured
with an Optical Activity AA100 digital Polarimeter at 20 ЊC
The oxazolidinone 15 was prepared from 2 using the method
of Karady7 and also from N-Cbz-(R)-methionine using the
modification of Marlowe17 and Jones.18 The cis isomer pre-
dominated (4 : 1), however, the isomers could not be separated
by crystallisation, chromatography or preparative HPLC and
this was not pursued further (Scheme 5). Kapadia19 reported yet
another modification for the preparation of Karady oxazol-
idinones, which we had successfully utilised for the preparation
of other amino acids.15 This method, however, was incom-
patible with 2 as a multitude of decomposition products were
formed.
1
and are given in units of 10Ϫ1 deg cm2 gϪ1. H NMR spectra
were recorded at 400, 250 or 200 MHz and 13C NMR at 100, 75,
62 or 25 MHz. The chemical shifts are expressed in ppm relative
to tetramethylsilane.
(2R,4R)-3-Benzoyl-2-tert-butyl-4-[2-(methylthio)ethyl]-1,3-
oxazolidin-5-one (cis-3)
Was prepared according to the Seebach3 method: white crys-
tals, mp 126–127 ЊC (from methanol), lit.3 mp 126.2–126.6 ЊC;
[α]D20 Ϫ61 (c 1.181 in CHCl3), lit.3 [α]D20 ϩ 62.2 (c 1 in CHCl3)
(ent-3). [Found: C, 63.7; H, 7.3; N, 4.4; S, 9.8. C17H23NO3S
requires C, 63.5; H, 7.2; N, 4.4; S, 10.0%]; νmax (CHBr3)/cmϪ1
1787, 1665 and 750; δH (CDCl3; 250 MHz) 7.55–7.35 (5H, m,
Ph), 6.08 (1H, s, 2-H), 4.18 (1H, dd, J 10 and 3 Hz, 4-H), 2.62–
1.95 (4H, m, CH2CH2S), 1.88 (3H, s, SCH3) and 1.04 (9H, s,
C(CH3)3).
(2S,4R)-3-Benzoyl-2-tert-butyl-4-[2-(methylthio)ethyl]-1,3-
oxazolidin-5-one (trans-3)
Scheme 5 Reagents and conditions: i) NaOH, H2O, EtOH; ii) PhCHO,
cyclohexane, Dean and Stark apparatus; iii) BnOCOCl, CH2Cl2;
iv) PhCH(OMe)2, Et2O, BF3ؒOEt2, Ϫ78 ЊC; v) PhCH(OMe)2, SOCl2,
ZnCl2, THF, 0 ЊC.
20
White needles, mp 154–156 ЊC (from methanol), [α]D Ϫ129
(c 1.027 in CHCl3). [Found: C, 63.45; H, 7.2; N, 4.4; S, 10.0.
C17H23NO3S requires C, 63.5; H, 7.2; N, 4.4; S, 10.0%];
νmax (CHBr3)/cmϪ1 1783, 1648, 1380, 1232 and 1010; δH (CDCl3;
250 MHz) 7.70–7.45 (5H, m, Ph), 6.20 (1H, s, 2-H), 4.50 (1H,
dd, J 5 and 1 Hz, 4-H), 2.30 (2H, t, J 7 Hz, CH2S), 2.10–1.50
(2H, m, SCH2CH2), 1.85 (3H, s, SCH3) and 1.02 (9H, s,
C(CH3)3).
In summary, we have examined a variety of methods for the
formation of cis-oxazolidinones of (R)-methionine. Of the
four different oxazolidinone variants examined the ferro-
cenecarboxaldehyde analogue 14 was unstable, the benzalde-
hyde analogue 15 was obtained as an inseparable mixture of cis
and trans isomers, the oxazolidinone 12 was obtained in pure
form as an oil after chromatography, and finally the oxazol-
idinone 3 was obtained in pure form after crystallisation. Oxa-
zolidinone 3 on enolisation and reaction with benzyl bromide
gave novel rearrangement products 5 and 6. The rearrangement
was suppressed by reacting the derived enolate with benzyl iod-
ide. Alkylation of 12 with benzyl bromide did not give the
rearrangement product. Both alkylation products 4 and 13 gave
Alkylation of cis-3 using sodium bis(trimethylsilyl)amide and
benzyl bromide
A solution of cis-3 (964 mg, 3 mmol) in tetrahydrofuran (6 ml)
was added dropwise at Ϫ70 ЊC to a solution of sodium bis-
(trimethylsilyl)amide in tetrahydrofuran (1 M, 3 ml) under
nitrogen. An orange colouration was obtained immediately on
addition to the base, and the solution was stirred at Ϫ70 ЊC for
40 min before benzyl bromide (0.36 ml, 3 mmol) was added.
The mixture was allowed to warm to 20 ЊC and stirred for 5 h
whereupon the orange colour turned yellow. TLC [silica, ethyl
acetate–hexane (1 : 19)] indicated the presence of a major prod-
uct (5) slightly more polar than benzyl bromide, the alkylation
product 4, and a trace of starting material 3. The reaction mix-
ture was quenched by addition of 5% aqueous ammonium
chloride solution (25 ml) and extracted into diethyl ether. The
organic solution was dried, filtered and concentrated to give the
crude mixture: νmax (Nujol)/cmϪ1 3350, 1790, 1680, 1655, 1600,
1580, 1500, 1258, 1230, 1180, 1050, 930, 840 and 700. The crude
product was dissolved in dichloromethane (10 ml) and stirred
with hydrochloric acid (2 M, 15 ml) until all of the major
component 5 hydrolysed. The phases were separated and the
organic phase was extracted with water (× 4). The combined
aqueous extracts were basified with 2 M sodium hydroxide solu-
tion and extracted with diethyl ether. The ether layer was dried
and concentrated to give 2-amino-2-benzyl-4-(methylthio)-1-
1
very broad H NMR spectra due to restricted rotation. Alkyl-
ation was stereoselective but the minor isomer could not be
completely removed. The alkylated oxazolidinone 4 was cleaved
to the amino acid 1 by heating in concentrated hydrobromic
acid whereas the oxazolidinone 13 was cleaved by potassium
trimethylsilanolate. Both methods gave the amino acid 1 in an
enantiomeric ratio of 90 : 10.
Experimental
Organic solutions were dried over anhydrous MgSO4. TLC was
performed on Merck 0.25 mm Kieselgel 60 F254 plates. Products
were visualised under UV light and/or by staining with aqueous
KMnO4 solution. LCMS analysis was conducted on a Supel-
cosil LCABZϩPLUS column (3.3 cm × 4.6 mm) eluting with
0.1% formic acid and 0.01 M ammonium acetate in water (sol-
vent A), and 0.05% formic acid and 5% water in acetonitrile
(solvent B), using the following elution gradient: 0–0.7 min 0%
B, 0.7–4.2 min 100% B, 4.2–5.3 min 100% B, 5.3–5.5 min 0% B
20
phenylbutan-1-one (6) (289 mg, 32%) as a yellow oil: [α]589 0;
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 8 5 3 – 2 8 5 8
2855

View Article Online
20
[α]546 Ϫ5.6 (c 0.948 in methanol); λ
(EtOH) 243.2 nm
had signals in the NMR spectrum at 3.48 and 3.06 (two d, 1 H
each, J 14 Hz). A trace amount of pivalaldehyde, 1725 cmϪ1
and 9.5 ppm, was present in the IR and NMR spectra of 5
respectively, suggesting facile hydrolysis. Further elution of the
column gave 4 (1.053 g, 88%) as a colourless gum, which was
crystallised twice from methanol to give 4 as a white solid: mp
82–83 ЊC. [Found: C, 70.0; H, 7.1; N, 3.5; S, 7.7. C24H29NO3S
requires C, 70.0; H, 7.1; N, 3.4; S, 7.8%]. Other data as reported
above. The aqueous phase was acidified with 2 M hydrochloric
acid to pH 1 and extracted with dichloromethane. The aqueous
phase was then basified with 2 M sodium hydroxide to pH 13
and extracted with diethyl ether. The organic solution was
dried, filtered and evaporated to give 6 as a yellow gum (8 mg,
1%). Characterising data as above.
max
(ε 9000), inflexion at 261.6 nm (3000); νmax (Nujol)/cmϪ1 3370,
1673, 1595, 1495, 1450, 750 and 700; δH (CDCl3; 250 MHz) 7.80
(2H, d, J 8 Hz, ortho-PhCO), 7.55–7.35 (3H, m, Ph), 7.30–7.20
(3H, m, Ph), 7.17–7.05 (2H, m, Ph), 3.44 (1H, d, J 13 Hz,
CH2Ph), 2.97 (1H, d, J 13 Hz, CH2Ph), 2.65–2.30 (3H, m), 2.12–
1.90 (1H, m), 2.02 (3H, s, SCH3) and 1.53 (2H, m, NH2);
δC (CDCl3; 62 MHz) 206.0 (CO), aromatic C [137.4, 135.6, 131.6,
130.3, 128.5, 128.2, 128.0, 126.8], 66.6 (CNH2), 46.7 (CH2Ph),
40.4 (CH2CH2S), 28.7 (CH2S) and 15.2 (SCH3); FABϩve m/z
300 (M ϩ H)ϩ. The original dichloromethane solution was dried,
concentrated and purified by chromatography on silica gel (37 g)
eluting with ethyl acetate–hexane (1 : 19, 1 : 9, 1 : 6, 1 : 4) to give
(2R,4R)-3-benzoyl-4-benzyl-2-tert-butyl-4-[2-(methylthio)ethyl]-
20
1,3-oxazolidin-5-one (4) (113 mg, 9%) as a colourless gum: [α]D
ϩ23.4 (c 1.047 in CHCl3); νmax (CHBr3)/cmϪ1 1783, 1640, 1376
and 715; δH (CDCl3; 250 MHz) 7.45–7.15 (8H, m, Ph), 7.0–6.5
(2H, br, Ph), 5.42 (1H, s, 2-H), 4.1–3.65 (1H, m), 3.45–3.2 (1H,
m), 3.2–2.7 (3H, m), 2.5–2.25 (1H, m), 2.20 (3H, s, SCH3), 0.67
(9H, s, C(CH3)3); δC(CDCl3; 25 MHz) 173.9 (CO), 171.0 (CO),
aromatic C [136.0, 134.6, 130.5, 130.1, 128.6, 127.9, 127.8, 127.3],
95.2 (2-C), 67.7 (4-C), 39.6 (PhCH2), 38.6 (CHCH2), 38.1
[C(CH3)3], 28.8 (CH2SCH3), 25.4 [C(CH3)3], 15.3 (SCH3); the
starting material 3 (160 mg, 17%) as a mixture of cis and trans
isomers (87 : 13): δH (CDCl3; 250 MHz) 7.70–7.35 (5H, m), 6.2
(0.13H, s), 6.08 (0.87H, s), 4.50 (0.13H, dd, J 5 and 1 Hz), 4.18
(0.87H, dd, J 10 and 3 Hz), 2.62–1.95 (4H, m), 1.88 (2.61H, s),
1.84 (0.39H, s) and 1.04 (9H, s), and finally the amine 6 (33 mg,
4%), same data as above.
(R)-ꢀ-Benzylmethionine (1) by hydrolysis of 4
The oxazolidinone 4 (8 g, 19 mmol) was suspended in concen-
trated hydrobromic acid (9 M, 100 ml) and the mixture was
heated to reflux for 3 h. On cooling to 20 ЊC long white needles
crystallised. Water was added and the mixture was extracted
with dichloromethane. The organic solution was discarded and
the aqueous solution was evaporated to dryness. The residue
was redissolved in water and applied to Dowex 50X8-200 ion-
exchange resin (240 g). The resin was washed with water until
the pH was neutral. The product was eluted with aqueous 1 M
ammonium hydroxide solution and appropriate fractions were
combined and evaporated to dryness under reduced pressure.
The residue was dissolved in water (250 ml) and freeze-dried to
give 1 (4.28 g, 94%) as a white solid: [α]D20 ϩ15 (c 0.63 in water).
[Found: C, 60.6; H, 7.1; N, 6.2. C12H17NO2S requires C, 60.2; H,
7.2; N, 5.85%]; νmax (Nujol)/cmϪ1 3580–2100, 1590, 1450, 1375,
735 and 695; δH (D2O; 250 MHz) 7.45–7.23 (5H, m, Ph), 3.35
(1H, d, J 14 Hz, CH2Ph), 3.02 (1H, d, J 14 Hz, CH2Ph), 2.7–
2.44 (2H, m), 2.4–2.25 (1H, m) and 2.12 (3H, s, SCH3);
FABϩve m/z 240 (M ϩ H)ϩ; FABϪve m/z 238 (M Ϫ H)Ϫ.
2-Amino-2-benzyl-4-(methylthio)-1-phenylbutan-1-one
hydrochloride (6ؒHCl)
The amine 6 (75 mg, 0.25 mmol) in diethyl ether (3 ml) was
treated with hydrogen chloride solution in dioxane (4 M, 62 µl).
The solid was collected by filtration to give the hydrochloride
salt (69 mg, 82%) as a white solid: mp 217–220 ЊC (decomp)
(from ethanol–diethyl ether). [Found: C, 64.25; H, 6.8; N, 4.1; S,
9.3; Cl, 10.5. C18H21NOSؒHCl requires C, 64.4; H, 6.6; N, 4.2; S,
9.55; Cl, 10.6%]; νmax (Nujol)/cmϪ1 3290, 2700–2250, 1670,
1593,1500, 1455, 753 and 700; δH (DMSO-d6; 250 MHz) 8.8–8.4
(3H, br, NH3), 8.17 (2H, d, J 8 Hz, ortho-PhCO), 7.78 (1H, t, J 8
Hz, para-PhCO), 7.65 (2H, t, J 8 Hz, meta-PhCO), 7.34–7.15
(5H, m, Ph), 3.65 (1H, d, J 14 Hz, CH2Ph), 3.56 (1H, d, J 14 Hz,
CH2Ph), 2.92–2.66 (2H, m), 2.5–2.0 (2H, m), 1.9 (3H, s, SCH3).
N-Acetyl-(R)-ꢀ-benzylmethionine (11)
A suspension of the amino acid 1 (2.39 g, 10 mmol) in di-
chloromethane (40 ml) was treated with triethylamine (3.1 ml,
22 mmol) with ice cooling. Acetyl chloride (1.5 ml, 21 mmol)
was added and the mixture was stirred at 20 ЊC for 3 h. The
mixture was evaporated to dryness and the residue was dis-
solved in ethyl acetate. The organic solution was washed with
2 M hydrochloric acid, and sodium bicarbonate solution. The
basic solution was acidified with 6 M hydrochloric acid and
extracted with ethyl acetate. The organic solution was washed
with brine, dried, and concentrated to give 11 (2.37 g, 84%) as a
white solid: mp 187–188 ЊC; [α]D20 ϩ16.7 (c 1.348 in methanol).
[Found: C, 59.7; H, 6.8; N, 4.9. C14H19NO3S requires C, 59.8; H,
6.8; N, 5.0%]; νmax (Nujol)/cmϪ1 3380, 3350, 3500–2100, 1700,
1610, 1525, 1445, 1210, 745 and 705; δH (DMSO-d6, 250 MHz)
7.54 (1H, s, NH ), 7.33–7.18 (3H, m, Ph), 7.10–7.05 (2H, m, Ph),
7.0–6.0 (1H, br, OH), 3.19 (2H, s, PhCH2), 2.39 (2H, m,
SCH2CH2), 2.15–1.8 (2H, m, SCH2CH2), 2.02 (3H, s, CH3CO),
and 1.88 (3H, s, CH3S). The enantiomeric ratio was obtained by
integration of the signal at 2 ppm using the chiral solvating
agent (R)-2,2,2-trifluoro-1-(9-anthryl)ethanol (25 mg) and 11
(5 mg) in CDCl3 and found to be 89 : 11. The enantiomeric ratio
was confirmed by chiral HPLC on a Chiralpak AD-RH column
(15 cm × 0.46 cm) eluting with 50% iso-propanol–0.1% aqueous
phosphoric acid, flow rate 0.5 ml minϪ1, detecting at 215 nm,
tr 11.2 min, 11% (S-enantiomer), and tr 17.8 min, 89%
(R-enantiomer).
Alkylation of cis-3 using lithium bis(trimethylsilyl)amide and
benzyl iodide
A solution of 3 (940 mg, 2.92 mmol) in tetrahydrofuran (5 ml)
was added dropwise at Ϫ65 ЊC to a solution of lithium bis-
(trimethylsilyl)amide in tetrahydrofuran (1 M, 3 ml) under
nitrogen. An orange colouration was obtained immediately on
addition to the base, and the solution was stirred at Ϫ60 ЊC for
30 min before benzyl iodide (654 mg, 3 mmol) was added. The
colour of the reaction mixture turned slowly yellow and after
5 h at 20 ЊC TLC [silica, ethyl acetate–hexane (1 : 19)] indicated
complete disappearance of the starting material, a trace of the
unstable intermediate 5, and mainly the alkylation product 4.
The reaction mixture was quenched by addition of 2.5% aque-
ous ammonium chloride solution and extracted into diethyl
ether. The organic solution was dried, filtered, concentrated
and purified by chromatography on silica gel (35 g) eluting with
ethyl acetate–hexane (1 : 10) to give the unstable intermediate 2-
benzyl-2-{[(1E/Z)-2,2-dimethylpropylidene]amino}-4-(methyl-
thio)-1-phenylbutan-1-one (5) as a colourless gum (16 mg, 1%):
νmax(CHBr3)/cmϪ1 1675, 1655, 1500 and 1450; δH(CDCl3; 200
MHz) 7.90 (2H, d, J 8 Hz, ortho-PhCO), 7.55–7.20 (7H, m, Ph,
Benzyl (2R,4R)-2-tert-butyl-4-[2-(methylthio)ethyl]-5-oxo-
1,3-oxazolidine-3-carboxylate (12)
N᎐CH ), 7.20–7.00 (2H, m, Ph), 3.40 (1H, d, J 14 Hz, PhCH ),
᎐
2
3.25 (1H, d, J 14 Hz, PhCH2), 2.6–2.3 (3H, m), 2.2–1.9 (1H, m),
2.02 (3H, s, CH3S) and 1.0 [9H, s, C(CH3)3]. The minor isomer
Aqueous sodium hydroxide (1 M, 11.2 ml) was added to
(R)-methionine (1.67 g, 11.2 mmol) and the mixture was stirred
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 8 5 3 – 2 8 5 8
2856

View Article Online
for 5 min. The solution was evaporated to dryness under
reduced pressure and the white solid was dried in vacuo at 60 ЊC
for 40 min, before it was suspended in dichloromethane (50 ml).
Molecular sieves (4 Å, 4 g) were added, followed by pivalalde-
hyde (1.83 ml, 16.8 mmol) and the mixture was stirred at 20 ЊC
for 2 h. The molecular sieves were removed by filtration and
washed with methanol. The filtrate and washings were concen-
trated under reduced pressure to give another white solid a
portion of which (1.37 g, 5.7 mmol) was suspended in di-
chloromethane (80 ml) and cooled to Ϫ18 ЊC. A solution of
benzyl chloroformate (1.23 ml, 8.6 mmol) in dichloromethane
(10 ml) was added dropwise and the mixture was stirred vigor-
ously for 10 h and then allowed to warm to 20 ЊC. The reaction
mixture was washed with water, sodium bicarbonate solution,
water and dried. The solution was filtered, concentrated under
reduced pressure and the residue was purified by chromato-
graphy on silica gel eluting with ethyl acetate–hexane (0 : 1, to
1 : 19) to give 1214(1.07 g, 27%) as a colourless oil: νmax (film)/
cmϪ1 1788, 1713, 1400 and 700; δH (CDCl3, 250 MHz) 7.43–7.30
(5H, m, Ph), 5.57 (1H, s, 2-H), 5.18 (2H, s, CH2Ph), 4.53 (1H, t,
J 7 Hz, 4-H), 2.82–2.64 (2H, m), 2.30–2.00 (2H, m), 2.03 (3H, s,
SCH3) and 0.96 (9H, s, C(CH3)3); δC (CDCl3, 75 MHz) 172.4
(CO), 156.0 (CO), aromatic C [135.1, 128.8, 128.7], 96.4 (2-C),
68.5 (CH2Ph), 55.8 (4-C), 37.0 [C(CH3)3], 32.3 (CHCH2), 30.6
(CH2SCH3), 24.9 [C(CH3)3], 15.1 (SCH3); NOE was observed
from 2-H to 4-H.
m, Ph), 3.35 (1H, d, J 14 Hz, CH2Ph), 3.03 (1H, d, J 14 Hz,
CH2Ph), 2.67–2.48 (2H, m), 2.38–2.28 (1H, m), 2.14 (3H, s,
SCH3) and 2.11–2.03 (1H, m). The amino acid 1 was con-
verted as before to the N-acetyl derivative 11: LCMS tr 2.58
min, 100%; ESϩve m/z 282 (M ϩ H)ϩ and ESϪve m/z 280
20
(M Ϫ H)Ϫ; [α]D ϩ17 (c 1.1 in methanol); δH (DMSO-d6, 400
MHz) 12.9 (1H, br, OH ), 7.54 (1H, s, NH ), 7.31–7.17 (3H, m,
Ph), 7.10–7.00 (2H, m, Ph), 3.18 (2H, br s, CH2Ph), 2.42–2.30
(2H, m, CH2CH2S), 2.1–1.8 (2H, m, SCH2CH2), 2.02 (3H,
s, CH3CO), 1.86 (3H, s, SCH3); δC (DMSO-d6, 100 MHz)
173.5 (CO), 169.1 (CO), aromatic C [136.5, 130.0, 127.9, 126.5],
62.4 (NHCCO2H), 38.2 (CH2Ph), 33.8 (SCH2CH2), 27.7
(CH3SCH2), 22.9 (CH3CO), 14.7 (SCH3). The enantio-
meric ratio was determined by chiral HPLC (Chiralpak
AD-RH) tr 11.2 min, 10% (S-enantiomer) and 18.1 min, 90%
(R-enantiomer).
(2R,4R)-2-Ferrocenyl-4-[2-(methylthio)ethyl]-3-pivaloyl-
1,3-oxazolidin-5-one (14)
Was prepared according to the Davies16 method: unstable
orange solid (2.28 g, 97%): νmax (KBr)/cmϪ1 1785 and 1650;
δH (CDCl3, 400 MHz) 7.09 (1H, s, 2-H), 4.72 (1H, dd, J 5,
10 Hz, 4-H), 4.55–4.17 (4H, m, Cp), 4.25 (5H, m, CpЈ), 2.75–
2.68 (2H, m, CH2CH2S), 2.12–2.00 (2H, m, CH2S), 2.08 (3H, s,
SCH3), 1.29 [9H, s, C(CH3)3]; δC (CDCl3, 100 MHz) 176.5 (CO),
171.5 (CO), 87.6 (2-C), 83.5 (quaternary C in Cp), 69.8 (5 × CH
in CpЈ), 68.3, 67.1, 66.7, 64.7 (4 × CH in Cp), 53.4 (4-C), 39.8
[C(CH3)3], 32.3 (CH2CH2S), 28.7 (CH2S), 26.9 [C(CH3)3], 14.7
(CH3S).
Benzyl (2R,4R)-4-benzyl-2-tert-butyl-4-[2-(methylthio)ethyl]-
5-oxo-1,3-oxazolidine-3-carboxylate (13)
A solution of 12 (0.4 g, 1.1 mmol) in tetrahydrofuran (10 ml)
was added to lithium bis(trimethylsilyl)amide solution in tetra-
hydrofuran (1 M, 1.14 ml) at Ϫ78 ЊC under nitrogen. The solu-
tion turned immediately to orange coloured and was stirred
for 30 min before benzyl bromide (0.18 ml, 1.5 mmol) was
added and the resulting mixture was allowed to warm to 20 ЊC.
After stirring for 16 h 5% aqueous ammonium chloride
solution was added and the mixture was extracted with ethyl
acetate. The organic solution was dried, concentrated and puri-
fied by chromatography on silica gel eluting with ethyl
acetate–hexane (1 : 49 to 1 : 2) to give 13 (210 mg, 43%) as a
colourless gum: analytical HPLC (30 min run) tr 17.2 min,
100%; [α]D22 ϩ37 (c 0.65 in CHCl3). [Found: C, 68.1; H, 7.0; N,
3.2. C25H31NO4S requires C, 68.0; H, 7.1; N, 3.2%]; νmax (film)/
cmϪ1 1788, 1713, 1397, 1314, 1180, 1044, 754 and 700;
δH (DMSO-d6, 250 MHz at 80 ЊC) 7.50–7.15 (8H, m, Ph), 7.00–
6.90 (2H, m), 5.30 (1H, d, J 12 Hz, OCH2Ph), 5.17 (1H, d, J 12
Hz, OCH2Ph), 4.65 (1H, s, 2-H), 3.59 (1H, d, J 14 Hz,
CCH2Ph), 3.01 (1H, d, J 14 Hz, CCH2Ph), 2.80 (1H, m), 2.64
(1H, m), 2.46–2.41 (2H, m), 2.06 (3H, s, SCH3) and 0.82 [9H,
s, C(CH3)3] (at ambient temperature the NMR spectrum
was very broad); δC (CDCl3, 100 MHz at 30 ЊC) 173.7 (CO),
154.1 (CO), aromatic C [134.9, 134.0, 129.8, 129.1, 128.8,
128.7, 128.6, 128.2, 128.1, 127.5, 127.4], 95.0 (2-C), 67.9
(OCH2Ph), 67.9 (4-C), 40.0 (very broad PhCH2), 38.0
[C(CH3)3], 37.9 (CH2CH2S), 29.0 (CH2S), 25.7 [C(CH3)3], 15.4
(SCH3); ESϩve m/z 442 (M ϩ H)ϩ, 459 (M ϩ NH4)ϩ, 901 (2M
ϩ NH4)ϩ.
Benzyl (2R,4R)-4-[2-(methylthio)ethyl]-5-oxo-2-phenyl-
1,3-oxazolidine-3-carboxylate (15)
a) From (R)-methionine. Was prepared according to the
Karady7 method: colourless oil (220 mg, 6%); LCMS tr 3.48
min, 18% and 3.54 min, 82%; ESϩve m/z 372 (M ϩ H)ϩ, 389
(M ϩ NH4)ϩ; νmax (KBr)/cmϪ1 1792, 1712, 1595, 1455, 1404,
1350, 1321, 1239, 1017, 731 and 695; δH (CDCl3, 400 MHz)
7.47–7.23 (10H, m, Ph), 6.74 (0.86H, s, 2-H), 6.47 (0.14H, s,
2-H), 5.20 (2H, s, PhCH2O), 4.64 (1H, t, J 7 Hz, 4-H), 2.70–2.50
(2H, m, CH2CH2S), 2.17–1.92 (2H, m, CH2CH2S) and 1.97
(3H, s, CH3S); δC (CDCl3, 100 MHz) 171.7 (CO), 154.0 (CO),
aromatic C [136.9, 135.1, 130.1, 129.6, 128.8, 128.7, 128.3,
126.4, 125.9], 89.9 (2-C, minor), 88.8 (2-C, major), 68.3
(CH2Ph), 54.8 (4-C), 31.8 (CH2CH2S), 29.7 (CH2CH2S), 15.0
(SCH3).
b) From N-Cbz-(R)-methionine. N-Cbz-(R)-Methionine
(566 mg, 2 mmol) and benzaldehyde dimethyl acetal (304 mg,
2 mmol) in diethyl ether (10 ml) was cooled to Ϫ78 ЊC. Boron
trifluoride etherate (1.23 ml, 10 mmol) was added and the mix-
ture was allowed to warm to 20 ЊC and stirred for 4 days. The
reaction mixture was then treated with aqueous sodium bi-
carbonate solution until the pH was neutral, and then extracted
with ethyl acetate. The organic layer was dried, concentrated
and purified by chromatography on Biotage (90 g cartridge)
eluting with ethyl acetate–petroleum ether (0 : 1, 1 : 19, 1 : 9) to
give 15 (485 mg, 65%) as an oil: LCMS tr 3.49 min, 15% and
1
3.53 min, 85%; ESϩve m/z 372 (M ϩ H)ϩ; H NMR data as
Cleavage of 13 with potassium trimethylsilanolate
above.
A solution of 13 (180 mg, 0.4 mmol) in tetrahydrofuran (3 ml)
was treated with potassium trimethylsilanolate (90% pure, 180
mg, 1.26 mmol) and the mixture was heated to 80 ЊC for 2 h.
The mixture was diluted with methanol, and concentrated
under reduced pressure. The concentrate was applied to an
SCX-2 cartridge eluting with methanol and then with 0.2 M
ammonia in methanol. Evaporation of the ammoniacal
fractions gave 1 (94 mg, 98%) as a white solid: LCMS tr 1.99
min, 100%; ESϩve m/z 240 (M ϩ H)ϩ and ESϪve m/z 238
(M Ϫ H)Ϫ; δH (D2O, 400 MHz) 7.45–7.36 (3H, m, Ph), 7.29 (2H,
Acknowledgements
We are grateful to the late Professor Sir Derek H. R. Barton for
helpful discussions about the mechanism of the rearrangement,
to Dr Philip Sidebottom for the determination of the enantio-
meric ratio of 11 by NMR spectroscopy, to Mr Eric Hortense
for the determination of the enantiomeric ratio of 11 by chiral
HPLC and to Dr Keith Biggadike and Mr Charles Baker-Glenn
for proof reading the manuscript.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 8 5 3 – 2 8 5 8
2857

View Article Online
W. Petter, J. Escalante, E. Juaristi, D. Quintana, C. Miravitlles and
E. Molins, Helv. Chim. Acta, 1992, 75, 913–934.
10 A. K. Beck and D. Seebach, Chimia, 1988, 42, 142–144.
11 D. S. Kemp and E. C. Young, Tetrahedron Lett., 1994, 35, 8315–
8318.
12 K. Nebel and M. Mutter, Tetrahedron, 1988, 44, 4793–4796.
13 A. D. Abell, J. M. Taylor and M. D. Oldham, J. Chem. Soc.,
Perkin Trans. 1, 1996, 1299–1304.
References
1 K. Günther, M. Schikedanz, K. Drauz and J. Martens, Fresenius’
Z. Anal. Chem., 1986, 325, 298–299.
2 U. Schöllkopf, W. Hartwig, U. Groth and K. O. Westphalen,
Liebigs Ann. Chem., 1981, 696–708.
3 D. Seebach and A. Fadel, Helv. Chim. Acta, 1985, 68, 1243–
1250.
4 R. M. Williams and M.-N. Im, J. Am. Chem. Soc., 1991, 113, 9276–
9286.
5 E. J. Corey, F. Xu and M. Noe, J. Am. Chem. Soc., 1997, 119, 12414–
12415.
6 B. Lygo, J. Crosby and J. A. Peterson, Tetrahedron Lett., 1999, 40,
8671–8674.
7 S. Karady, J. S. Amato and L. M. Weinstock, Tetrahedron Lett.,
1984, 25, 4337–4340.
8 A. Fadel and J. Salaün, Tetrahedron Lett., 1987, 28, 2243–2246.
9 D. Seebach, B. Lamatsch, R. Amstutz, A. K. Beck, M. Dobler,
M. Egli, R. Fitzi, M. Gautschi, B. Herradon, P. C. Hidber, J. J. Irwin,
R. Locher, M. Maestro, T. Maetzke, A. Mourino, E. Pfammatter,
D. A. Plattner, C. Schickli, W. B. Schweizer, P. Seiler, G. Stucky,
14 K. J. Duffy, L. H. Ridgers, R. L. DesJarlais, T. A. Tomaszek,
M. J. Bossard, S. K. Thompson, R. M. Keenan and D. F. Veber,
Bioorg. Med. Chem. Lett., 1999, 9, 1907–1910.
15 D. M. Coe, R. Perciaccante and P. A. Procopiou, Org. Biomol.
Chem., 2003, 1, 1105–1111.
16 F. Alonso, S. G. Davies, A. S. Elend and J. L. Haggitt, J. Chem. Soc.,
Perkin Trans. 1, 1998, 257–264.
17 W. D. Shrader and C. K. Marlowe, Bioorg. Med. Chem. Lett., 1995,
5, 2207–2210.
18 H. Cheng, P. Keitz and J. B. Jones, J. Org. Chem., 1994, 59, 7671–
7676.
19 S. R. Kapadia, D. M. Spero and M. Eriksson, J. Org. Chem., 2001,
66, 1903–1905.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 8 5 3 – 2 8 5 8
2858