A 2,3-butanedione protected chiral glycine equivalent--a new building block for the stereoselective synthesis of enantiopure N-protected alpha-amino acids.

Article abstract of DOI:10.1039/b210673f

A new chiral glycine equivalent 7 has been synthesised from glycidol using a chiral memory protocol, and its use in the synthesis of N-Z protected alpha-amino acids was demonstrated in a series of diasteroselective lithium enolate alkylation reactions and subsequent acid hydrolyses.

Full text of DOI:10.1039/b210673f

A 2,3-butanedione protected chiral glycine equivalent—a new building
block for the stereoselective synthesis of enantiopure N-protected
a-amino acids
Darren J. Dixon,a Christopher I. Harding,a Steven V. Ley*a and D. Matthew G. Tilbrookb
a Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge, UK CB2 1EW.
E-mail: svl1000@cam.ac.uk
b Department of Chemistry, University of Western Australia, Nedlands, Perth, Western Australia, Australia
6907
Received (in Cambridge, UK) 30th October 2002, Accepted 22nd November 2002
First published as an Advance Article on the web 20th January 2003
A new chiral glycine equivalent 7 has been synthesised from
glycidol using a chiral memory protocol, and its use in the
synthesis of N-Z protected a-amino acids was demonstrated
in a series of diasteroselective lithium enolate alkylation
reactions and subsequent acid hydrolyses.
the reaction, facilitating complete conversion in 22 h at 255
°C.
In a typical example, a THF solution of 7 was added via a
cannula to 1.1 equivalents of lithium diisopropylamide (LDA)
in THF (0.125 M) at 278 °C, followed by the immediate
addition of 1.1 equivalents of HMPA. After 1 h, 3 equivalents of
the alkyl halide were added via syringe and the mixture warmed
to 255 °C. After a further 21 h, 1.1 equivalents of acetic acid
were added to quench the reaction, followed by 5 ml of diethyl
ether to aid precipitation, and the solution was allowed to warm
to room temperature over 1 h. The precipitous mixture was
filtered through a short plug of silica (2–3 cm), eluting with
diethyl ether. Evaporation afforded the crude product which
was purified by silica gel chromatography. The results for a
range of monosubstitution reactions with alkyl halides are
summarized in Table 1.
In general, the reactions proceeded with good to excellent
yields and usually exhibited high diastereoselectivities. The
relatively low selectivities observed with propargyl bromide
and tert-butyl bromoacetate are difficult to rationalise but may
arise through secondary chelation effects in the transition state.
NOE experiments or single-crystal X-ray diffraction studies on
the major diastereoisomers confirmed the introduced alkyl
group was located in the equatorial position. This suggested
A large variety of routes to synthetically important, substituted
a-amino acids exist and have been reviewed extensively.1 Of
these, the use of chiral glycine anion equivalents is one of the
most commonly employed procedures and is founded on the
wealth of knowledge of metal enolate formation and the
diversity of alkylating electrophiles that are available. The
range of chiral glycine anion equivalents published2 gives an
indication of their importance.
Recently within our group we have developed a new butane-
2,3-diacetal desymmeterised glycolic acid equivalent.3 This a-
hydroxy acid equivalent is readily synthesised on a large scale
using a chiral memory protocol, and undergoes highly selective
alkylation, aldol and Michael reactions and the products are
deprotected to the desired a-hydroxy acid derivatives under
mild conditions. Clearly the extension of this chemistry to the
chiral glycine analogue is of considerable importance. Here we
describe the preparation and alkylation reactions of this glycine
equivalent 7, and demonstrate its use in the synthesis of N-Z
protected mono- 9 and disubstituted a-amino acids 11.
We believed that the enantiomerically and diastereomerically
pure glycine equivalent 7 could be produced using a chiral
memory protocol, employing the chirality of an amino alcohol
building block 4 to establish the chirality of the 2,3-mixed acetal
aminal functionality in 5. It was reasoned that the alkyl halide
product would readily undergo elimination of the halide to give
the exo-methylene enol ether, which after oxidative cleavage
would yield the desired glycine equivalent 7.
Starting from commercially available (S)-glycidol 1, we
prepared the N-benzyloxycarbonyl (Z) protected amino alcohol
4 in three steps and 64% overall yield.4 These reactions were
performed on multigram quantities and only required silica gel
chromatography after the final step to obtain analytically pure
material. Treatment of amino alcohol 4 with 5 equivalents of
2,2,3,3-tetramethoxybutane and 0.1 equivalents of boron tri-
fluoride–tetrahydrofuran complex5 in dichloromethane for 2.5 h
at room temperature afforded the cyclic precursor 5 in 69%
yield. An HBr elimination using 2 equivalents of potassium
bis(trimethylsilyl)amide (0.5 M KHMDS in toluene) in tetra-
hydrofuran (278 °C to room temperature, overnight, 64%) and
subsequent oxidative cleavage with ruthenium trichloride
catalyst and sodium periodate oxidant using the Sharpless
conditions (carbon tetrachloride, water, acetonitrile, 85%)6
yielded the crystalline glycine equivalent 7 in 24% yield from
glycidol 1 (Scheme 1).
Scheme 1 Reagents and conditions: (a) phthalimide, PPh3, di-tert-butyl
azodicarboxylate, THF, rt, overnight; (b) 48% HBr, reflux, 5 h, 77%; (c)
benzyl chloroformate, MeOH then diisopropylethylamine, 0 °C to rt, 2 h, 83
%; (d) 5 eq. 2,2,3,3-tetramethoxybutane, DCM, 0.1 eq. BF3.THF, 30 °C, 2.5
h, 69%; (e) 2 eq. potassium bis(trimethylsilyl)amide, THF, 278 °C to rt, 24
h, 64%; (f) RuCl3, NaIO4, MeCN, H2O, CCl4, rt, 85%; (g) 1.1 eq. lithium
diisopropyl amide (0.125 M in THF), THF, 1.1 eq. HMPA, 278 °C, 1 h,
then 3 eq. RX, 255 °C, 21 h, then 1.1 eq. AcOH followed by Et2O, rt, 1h;
(h) 9+1 TFA–H2O, rt, 30 min. The absolute and relative stereochemistry of
5 and the relative stereochemistry of 7 were confirmed by single-crystal X-
ray diffraction methods. The enantiopurity of 7 was established by chiral
HPLC and was shown to be greater than 99%.7
After a series of trial reactions, it was apparent that the
lithium enolate of 7 was the most selective in alkylation
reactions. The initial reactions were slow, but the addition of 1.1
equivalents of hexamethylphosphoramide (HMPA) accelerated
468
CHEM. COMMUN., 2003, 468–469
This journal is © The Royal Society of Chemistry 2003

Table 1 Results of the monoalkylation reactions of 7
Table 2 Results of cleavage reactions to the N-Z-protected a-amino acids
Reaction yield
(%)
Diastereoisomeric
ratio
Amino Acid
Product
Yield (%)
71
E.e. (%)
> 99a
Alkyl halide
MeI
Major product
9b
8a
8b
8c
8d
8e
73
82
74
92
72
11+1ab
22+1ab
14+1ab
3+1c
9c
63
> 99a
19+1ac
9a
74
64
> 99a
> 99b
8f
81
89
67
14+1cd
10+1c
2+1c
11b
8g
8h
a Enantiomeric excess was measured for the methyl esters by chiral HPLC.
When only one enantiomer was observed, we estimated this to be at least
99% e.e. In all cases the racemic mixtures were found to separate.10 b Direct
determination of the e.e. was not possible and was estimated by analogy
with the previous examples.
a Recrystallization afforded diastereomerically pure products as observed
by 1H NMR spectroscopy. b The relative stereochemistry of the major
diastereoisomer was confirmed by single crystal X-ray diffraction. c The
relative stereochemistry of the major diastereoisomer was confirmed by
NOE experiments. d (3-Furyl)bromomethane was synthesised from 3-fur-
anmethanol.8
Notes and references
1 (a) R. M. Williams, Synthesis of Optically Active a-Amino Acids,
Pergamon, Oxford, 1989; (b) R. O. Duthaler, Tetrahedron, 1994, 50,
1539; (c) L. Hegedus, Acc. Chem. Res., 1995, 28, 299; (d) R. Bloch,
Chem. Rev., 1998, 98, 1407; (e) M. Calmes and J. Daunis, Amino Acids,
1999, 16, 215.
attack of the alkyl halide on the enolate carbon atom was
occurring from the side opposing the 1,3-related axial methoxy
group.
A second alkylation reaction on previously monoalkylated
material also proved very effective, and no loss of reactivity or
selectivity was observed. In the first example studied, 8b was
alkylated as before with benzyl bromide with an 82% yield and
10a was the only observable product by 1H NMR spectroscopy.
In a complimentary study, 8c was alkylated with iodomethane
2 (a) A. G. Myers and J. L. Gleason, Org. Syn., 1998, 76, 57; (b) R.
Moretti, S. Thomi and W. Oppolzer, Tetrahedron Lett., 1989, 30, 6009;
(c) D. A. Evans and A. E. Weber, J. Am. Chem. Soc., 1986, 108, 6757;
(d) D. Pettig and U. Schöllkopf, Synthesis, 1988, 173; (e) N. Hoffmann
and D. Seebach, Eur. J. Org. Chem., 1998, 1337; R. Moretti, S. Thomi
and D. Seebach, Angew. Chem., Int. Ed. Engl., 1996, 35, 2708; (f) S. D.
Bull, S. G. Davies, S. W. Epstein, M. A. Leech and J. V. A. Ouzman, J.
Chem. Soc., Perkin Trans. 1, 1998, 2321; (g) M.-N. Im and R. Williams,
J. Am. Chem. Soc., 1991, 113, 9276; (h) C. Garbay-Jaureguiberry, W.-
Q. Liu and B. P. Roques, Tetrahedron: Asymmetry, 1995, 6, 647; (i) G.
Chassaing, H. Josien and A. Martin, Tetrahedron Lett., 1991, 32, 6547;
(j) R. Chinchilla, L. Falvello, N. Galindo and C. Najera, J. Org. Chem.,
2000, 65, 3034; (k) M. Ahn, K. Fuji, K. Tanaka and Y. Watanabe,
Tetrahedron: Asymmetry, 1996, 7, 1771; P. Blomgren, P. Cheng, J. Z.
Gougoutas, H. H. Gu, E. J. Iwanowicz, M. F. Malley, Y. Y. Pan and K.
Smith, Synlett, 1998, 665.
1
with a 66% yield and only 10b was observed by H NMR
spectroscopy (Scheme 2).
3 (a) D. K. Baeschlin, D. J. Dixon, A. C. Foster, S. J. Ince, S. V Ley and
H. W. M. Priepke, Chem. Rev., 2001, 101, 53; (b) E. Diez, D. J. Dixon
and S. V. Ley, Angew. Chem., Int. Ed., 2001, 40, 2906; (c) D. J. Dixon,
S. V. Ley, A. Polara and T. S. Sheppard, Org. Lett., 2001, 3, 3749; (d)
D. J. Dixon, S. V. Ley and F. Rodriguez, Org. Lett., 2001, 3, 3753; (e)
D. J. Dixon, S. V. Ley and F. Rodriguez, Angew. Chem., Int. Ed., 2001,
40, 4763; (f) D. J. Dixon, A. Guarna, S. V. Ley, A. Polara and F.
Rodriguez, Synthesis, 2002, 1973.
Scheme 2 Dialkylation reactions. Reagents and conditions: (a) (R = Me),
1.1 eq. LDA, THF, 1 eq. HMPA, 278 °C, 1 h then benzyl bromide, 255 °C,
21 h, then 1.1 eq. AcOH, Et2O, rt, 1 h, 82%; (b) (R = Bn), 1.1 eq. LDA,
THF, 1.1 eq. HMPA, 278 °C, 1 h then 3 eq. MeI, 255 °C, 21 h, then AcOH,
Et2O, rt, 1 h, 66%; (c) 2+1 TFA/H2O, rt, 30 min, then 1 M NaOH, MeOH,
rt, 1 h, 82% (R = Me), 66% (R = Bn).
Treatment of certain diastereomerically pure monoalkylated
4 (a) J.-W. Chern, H.-W. Liu, A. Gutcait and K.-C. Wang, Tetrahedron:
Asymmetery, 1996, 7, 1641; (b) D. H. Ball and J. M. Williams Jr, J. Org.
Chem., 1963, 28, 1589; (c) derivatisation of amino alcohol 4 as the (R)-
and (S)-Mosher’s esters showed the e.e. to be > 98% and confirmed the
stereochemistry as S (d) I. Ohtani, H. Kakisawa, Y. Kashman and T.
Kusumi, J. Am. Chem. Soc., 1991, 113, 4092; (e) J. Dale and H. Mosher,
J. Am. Chem. Soc., 1973, 95, 512.
5 (a) A. Hense, S. V. Ley, H. M. I. Osborn, D. R. Owen, J. F. Poisson, S.
L. Warriner and K. E. Wesson, J. Chem. Soc., Perkin Trans. 1, 1997,
2023; (b) U. Berens, D. Leckel and S. C. Oepen, J. Org. Chem., 1995,
60, 8304.
products 8 with aqueous trifluoroacetic acid (9+1 TFA/H2O, 30
min) afforded the desired N-Z monosubsituted -a-amino acids
D
9 in good to excellent yields. The enantiomeric excess of the
products were determined as greater than 99% and therefore
confirmed no loss of stereochemical integrity was occurring in
the hydrolysis step. The disubstituted amino acid 11b was
obtained by a sequential acid then base hydrolysis9 (2+1 TFA/
H2O, 30 min then 1 M NaOH, MeOH, 1 h) in excellent yield
(Table 2).
6 P. H. J. Carlsen, T. Katsuki, V. Martin and K. B. Sharpless, J. Org.
Chem., 1981, 46, 3936.
In summary, the new chiral glycine equivalent 7 readily
undergoes alkylation reactions in good yields and diaster-
eoselectivities to afford, after hydrolytic cleavage, the N-Z
protected mono- and disubstituted a-amino acids.
7 Chiralcel AD column, 95+5 hexane–isopropanol, 1 ml min21
.
8 (a) A. Barba and A. Mateos, J. Org. Chem., 1995, 60, 3580; (b) J. Sandri
and J. Viala, Synth. Commun., 1992, 22, 2945.
We gratefully acknowledge the financial support from the
EPSRC (to D. J. D. and C. I. H.) and the BP endowment and the
Novartis Research Fellowship (to S. V. L.). We also thank Dr
John Davies for the X-ray structure determinations and Dr Peter
Grice and Chris Cowburn for the NOE experiments.
9 J. Amato, S. Karady and M. Weinstock, Tetrahedron Lett., 1984, 25,
4337.
10 Chiralcel OD column, 90+10 hexane–isopropanol, 1 ml min21. These
conditions were developed from information on the Chiralcel website:
http://www.daicel.co.jp/chiral/application/application.html.
CHEM. COMMUN., 2003, 468–469
469