Chemoenzymatic Synthesis of (4S)- and (4R)-4-Methyl-2-oxoglutaric Acids
FULL PAPER
Scheme 2. GOT-catalysed transamination of (4S)- and (4R)-4-methyl-2-oxoglutarates
commercially available 1 solution in THF) was added dropwise.
After stirring for 1 h at Ϫ78°C, 14 mL of methyl iodide (0.225 mol,
3 equiv.) was added and the mixture was allowed to stand for an-
other hour at this temperature before being raised to Ϫ30°C. After
1 h, water and ethyl acetate were introduced at room temperature
and the solution transferred into a separatory funnel. The aqueous
phase was neutralised with 1 HCl and the organic phase was
extracted, washed with a saturated aqueous sodium chloride solu-
tion, dried with magnesium sulfate and concentrated in vacuo to
give 12.8 g (73% yield) of a colourless liquid after column chroma-
The activity of GOT for these close analogues of α-oxog-
lutaric acid (KG) was determined in phosphate buffer at
pH ϭ 7.6 in the presence of aspartic acid (Scheme 2).(4R)-
MKG is the best substrate. Its behaviour is very close to
the natural substrate, with a Km value of 0.16 m (against
0.18 m for KG) and a Vmax value of 72% relative to KG.
Apparently, GOT does not discriminate between its natural
substrate and this analogue. (4S)-MKG is a poorer sub-
strate with a Km value of 9 m and a Vmax value of only
11% relative to KG. The enantioselectivity of the enzyme is tography on silica gel (Rf ϭ 0.38, cyclohexane:ethyl acetate, 7:3).
Ϫ 1H NMR (400 MHz, CDCl3): δ ϭ 1.03 (d, 3 H, J ϭ 8 Hz, CH3),
mainly due to the best affinity of the enzymatic site for
1.74 (dd, 1 H, J ϭ 4 Hz, J ϭ 14 Hz, CH2), 2.30 (dd, 1H, J ϭ 8
(4R)-MKG as indicated by the ratio of the Km values
Hz, J ϭ 14 Hz, CH2), 2.37 (m, 1 H, CH), 3.12 (s, 6 H, 2 CH3
(around 1:45), while the kcat values are in a ratio of only 6:1.
acetal), 3.53, 3.67 (each s, each 3 H, 2CH3 ester). Ϫ 13C NMR
(100 MHz, CDCl3): δ ϭ 18.4 (4-CH3), 34.5 (C4), 37.0 (C3), 49.7,
49.8 (2 CH3 acetal), 51.5, 52.3 (2 CH3 ester), 101.2 (C2), 169.0 (C1),
176.1 (C5).
Experimental Section
General Procedure for the Resolution of Dimethyl (R,S)-2,2-Dimeth-
General: Melting points were determined with a Reichert hot-stage
apparatus and are reported uncorrected. Ϫ IR spectra were re-
corded with a PerkinϪElmer 801 spectrophotometer. Ϫ Flash
chromatography was carried out on 220Ϫ400 mesh silica gel. Ϫ
Thin-layer chromatography was performed on Merck precoated sil-
ica gel 60 F254 plates and spots were visualised with vanillin [vanil-
lin (1 g) is dissolved in MeOH (60 mL) and conc. H2SO4 (0.6 mL)].
oxy-4-methylglutarate (4): Substrate 4 (12.8 g, 54.7 mmol) was dis-
solved in phosphate buffer (650 mL, 0.1 , pH ϭ 7.0). Pseudo-
monas mendocina lipase (5 g) was then added and the pH was main-
tained by addition of 1 NaOH. After addition of 30 mL of
NaOH solution (55% conversion), the mixture was centrifuged and
the supernatant was transferred into a separatory funnel. The aque-
ous phase was extracted with toluene and the organic phase was
washed twice with a 10% solution of NaHCO3, dried with MgSO4
and concentrated in vacuo to afford 5.63 g (44%) of (S)-4; [α]D25 ϭ
ϩ31.9 (c ϭ 2.1, CHCl3), ee > 96% [1H NMR with Eu(hfc)3]. The
combined aqueous extracts were brought to pH ϭ 1Ϫ2 with 1
HCl, saturated with NaCl and extracted three times with ethyl acet-
ate. The combined organic phases were dried (MgSO4) and concen-
trated in vacuo to afford (R)-5. (R)-5 was esterified with SOCl2 in
refluxing MeOH in quantitative yield, and after the usual workup
was hydrolysed again with the enzyme (c ϭ 85%) in order to in-
crease the enantiomeric purity. Thus, another esterification under
the same conditions afforded a 42% overall yield (5.05 g) of (R)-4;
Ϫ
1H- (400 MHz), 13C-NMR (100 MHz) experiments were per-
formed with an AC 400 Bruker spectrometer. Ϫ Enzymes were
from commercial origin: Pseudomonas mendocina lipase from Gen-
encor, all others from Sigma.
Dimethyl 2,2-Dimethoxyglutarate (3): α-Oxoglutaric acid (60 g, 0.41
mol) was dissolved in 500 mL of methanol in which trimethyl or-
thoformate (160 mL, 1.4 mol, 3.5 equiv.) was introduced. Sulfuric
acid (10 mL, 32 ) was added dropwise and the solution was re-
fluxed during 12 h. The solution was concentrated in vacuo and
ethyl acetate (200 mL) was added. The solution was washed with a
saturated aqueous bicarbonate solution, dried with magnesium sul-
fate and concentrated to dryness to give 90.4 g (96%) of a colour-
less liquid. Ϫ 1H NMR (400 MHz, CDCl3): δ ϭ 2.17, 2.26 (each
m, each 2 H, 2 CH2), 3.20 (s, 6 H, 2 CH3 acetal), 3.60, 3.74 (each
s, each 3 H, 2 CH3 ester). Ϫ 13C NMR (100 MHz, CHCl3): δ ϭ
28.1, 28.7 (C3 and C4), 49.9 (2 CH3 acetal), 51.7, 52.6 (2 CH3 ester),
101.5 (C2), 169.0 (C1), 172.8 (C5).
25
[α]D ϭ Ϫ31.7 (c ϭ 2.58, CHCl3), ee > 96% [1H NMR with
Eu(hfc)3].
Dimethyl (R)- and (S)-4-Methyl-2-oxoglutarate (6): To a solution of
5 g (21.3 mmol) of (R)-4 or (S)-4 in 25 mL of dichloromethane and
2.5 mL of water was added dropwise 25 mL of trifluoroacetic acid
at room temperature. After 2 h of stirring, the solution was concen-
trated in vacuo to give a colourless oil.
(R,S)-Dimethyl 2,2-Dimethoxy-4-methylglutarate (4): Under argon
was introduced 16.5 g (0.075 mol) of 3 into 200 mL of anhydrous
THF. After cooling to Ϫ78°C, 1.2 equiv. of LHMDS (90 mL of a
Dimethyl (R)-4-Methyl-2-oxoglutarate: Liquid (3.8 g, 95%);
[α]D25 ϭ ϩ13.0 (c ϭ 1.3, CHCl3). Ϫ 1H NMR (400 MHz, CDCl3):
Eur. J. Org. Chem. 1999, 3403Ϫ3406
3405