Chemo-enzymatic dynamic kinetic resolution of amino acid thioesters

Article abstract of DOI:10.1002/adsc.200700050

The L-forms of racemic-N-protected-β,γa,aunsaturated a-amino acid thioesters were found to be substrates for the subtilisin-catalysed hydrolysis to the corresponding acids. The D-enantiomer was continuously racemised in the presence of an organic base. The combined reactions in a biphasic system allowed the deracemisation of the amino acid derivatives based on a dynamic kinetic resolution. Excellent yields and enantioselectivities were achieved.

Full text of DOI:10.1002/adsc.200700050

COMMUNICATIONS
DOI: 10.1002/adsc.200700050
Chemo-Enzymatic Dynamic Kinetic Resolution ofAmino Acid
Thioesters
Dario Arosio,a Antonio Caligiuri,a Paola D’Arrigo,a
Giuseppe Pedrocchi-Fantoni,a,b Cristina Rossi,a Caterina Saraceno,a
Stefano Servi,a,b,* and Davide Tessaroa
a
Dipartimento di Chimica, Materiali, Ingegneria Chimica “G. Natta”, Politecnico di Milano, via Mancinelli 7,
20131 Milano, Italy
Phone: (+39)-02-2399-3047; fax: (+39)-02-2399-3080; e-mail: stefano.servi@polimi.it
ICRM, CNR, Via Mario Bianco 9, 20131 Milano, Italy
b
Received: January 24, 2007
Abstract: The l-forms of racemic-N-protected-b,g-
unsaturated a-amino acid thioesters were found to
be substrates for the subtilisin-catalysed hydrolysis
to the corresponding acids. The d-enantiomer was
continuously racemised in the presence of an organ-
ic base. The combined reactions in a biphasic
system allowed the deracemisation of the amino
acid derivatives based on a dynamic kinetic resolu-
tion. Excellent yields and enantioselectivities were
achieved.
uct, and a rate of racemisation of the substrate higher
than the rate of conversion to product. These condi-
tions require the design of suitable substrates. In thio-
esters, the acidity of the hydrogen in the a-position is
higher in comparison to the corresponding oxo esters,
amides or acids. The enzymatic transformation of a
thioester into the corresponding carboxylate with a
higher pKa of the a-proton is the basis for a successful
DKR, provided that the enzymatic systems resist the
basic conditions required for substrate racemisation.
This concept has been applied in the DKR of a-alkyl
thioesters.[4] Our interest in the synthesis and dera-
cemisation of non-natural amino acids[5] prompted us
to design racemic compounds suitable for this novel
application. From previous data[6] aryllycines and ana-
logues were considered as candidates. In fact not only
p-hydroxyphenylglycine esters have been shown to be
the substrate for subtilisin, but also phenylglycine
Keywords: amino acids; biotransformations; dy-
namic kinetic resolution; enzyme catalysis; racemi-
zation
Hydrolytic enzymes are the most readily available (Phg) has been partly deracemised in the presence of
biocatalysts for synthetic applications, particularly in chiral copper catalysts indicating a suitable pKa value
the kinetic resolution of racemic mixtures.[1] Like in for the a-proton.[7]
any resolution method, a maximum of 50% yield of
We initially checked the ability of subtilisin Carls-
enantiomerically pure product can be obtained, based berg to transform amino acid thioesters 1–6
on racemic starting material. When kinetic resolution (Scheme 1) and found a good activity on these N-Boc
is coupled with an in situ racemisation of the substrate amino acid derivatives with reaction rates comparable
either chemical or enzymatic, the yield limitation can although inferior to the one observed with the corre-
be overcome leading to a much more efficient pro- sponding oxoesters.[8]
cess: a deracemisation based on a dynamic kinetic
Racemisation conditions were studied by observing
resolution (DKR).[2] A well known example in amino the exchange rate of the a-proton of the thioester
1
acid chemistry is the industrial process d-hydantoi- with deuterium from CD3OD by H NMR. In a typi-
nase-carbamoylase which allows the large scale prepa- cal experiment to 20 mg of substrate dissolved in
ration of relevant d-amino acids.[3] In this process a 500 mL of DMSO-d6 and 100 mL of CD3OD, 0.5
microbial catalyst containing the two enzymatic activ- equivs. of trioctylamine were added at 318C. The
ities transforms the R-starting material into the d- 50% exchange rate varied from a few minutes (for
amino acids, while the remaining hydantoin is contin- compound 6) to several hours for the 2-substituted
uously racemised at slightly basic pH. Requisites for a phenylglycines. In all cases, no exchange was observed
successful DKR are an enzyme selective for one of in the absence of trioctylamine. Control experiments
the enantiomers, a racemising system (chemical or en- with the corresponding oxo esters and acids showed
zymatic) acting on the substrate but not on the prod- no measureable proton exchange under similar condi-
Adv. Synth. Catal. 2007, 349, 1345 – 1348
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1345

Dario Arosio et al.
COMMUNICATIONS
Scheme 1.
tions. This indicated that the thioester, but not the N-Boc-Phg-OH was recovered by extraction of the
oxo ester, was suitable for the in situ racemisation water phase with ethyl acetate (3”50 mL) at pH 3
method and that the product acid was configuration- after separation from the organic layer containing the
ally stable under thioester-racemising conditions. Fur- trioctylamine and further extraction at pH 8. Drying
ther experiments showed that thioesters were not and evaporation of the solvent gave a quantitative
prone to spontaneous hydrolysis. The deuterium ex- amount of l-N-Boc-Phg-OH as a yellow oil. The en-
change rate was highest for 6 (t1/2 of 10 min) and antiomeric excess evaluated with chiral HPLC analy-
lowest for the 2-substituted arylglycines 1–5 (t1/2 of sis was >99%.
several h).
l-N-Boc-Phg-OH was deprotected with dioxane sa-
The resolution of 1 was initially studied under non- turated with HCl at 608C to give 1.106 g of l-Phg hy-
racemising conditions: 2 g of 1 in 150 mL of a 2:1 drochloride as a white solid (87% yield from 1, ee>
mixture of water:methyl tert-butyl ether (approxi- 99%), [a]2D5: +147.4 (c 0.8 on free amino acids ; HCl
mately 50 mM) at pH 8.0 and 378C were treated with 1N). A procedure similar to the one described was
500 U of subtilisin and the pH controlled in a pHstat applied to thioesters 2–6 and the results are summar-
apparatus. The base addition stopped sharply at 50% ised in Table 1.
conversion (24 h). The hydrolysis was restored by ad-
The efficiency of the deracemisation process de-
dition of trioctylamine (1.5 mL, 3.38 mmol, 0.5 pends on the rate of enzymatic hydrolysis and the
equivs.) and stoichiometric base addition was com- proton extraction rate. The latter can be modulated
plete after 48 h. The profile of the reaction as regis- by the trioctylamine concentration which influences
tered on a pHstat before and after trioctylamine addi- the racemisation rate with pseudo-first order kinetics.
tion is reported in Figure 1.
Figure 2 shows the correlation between TOA equiva-
Due to the hydrophobic nature of the amino acids lent and time course (arbitrary units) as measured on
1
and to the presence of the N-protecting group, the l- compound 1 in DMSO-d6 from H NMR studies.
Figure 1. Time course of the DKR of compound 1 before and after trioctylamine addition.
1346
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 1345 – 1348

Chemo-Enzymatic Dynamic Kinetic Resolution of Amino Acid Thioesters
COMMUNICATIONS
Table 1. Subtilisin-catalysed DKR of compounds 1–6 in the
presence of trioctylamine.
thioesters, and not limited to b,g-unsaturated a-amino
or amino acid derivatives. Moreover, the opposite
enantiomers are expected from hyrolysis catalysed by
enzymes of opposite stereochemical preferences.
Computational studies are in progress in order to
define the influence of the groups affecting the acidity
of the a-proton thus identifying new substrates for
the DKR.
Substrate
Conversion Base/Sub-
Time
[h]
ee AA
[%]
strate
1 (R=Ph)
2 (R=4-Cl-
C6H4)
>95%
>95%
0.5
0.5
48
36
>99%
>99%
3 (R=4-F-
C6H4)
4 (R=2-Cl-
C6H4)
5 (R=2-F-
C6H4)
6 (R=Th)
>95%
>95%
>95%
>95%
0.5
1
20
96
12
5
>99%
>99%
>99%
>99%
Experimental Section
1
Subtilisin Carlsberg from B. subtilis from Fluka and Alcalase
2.5L DX from Novozymes were used in all experiments and
gave similar results.
0.5
The data reported in Table 1 reflect the combined
effect of these two parameters. Due to the complete
and well known specificity of subtilisin for l-amino
acid esters, the enantiomeric excesses of all products
were higher than 99%.
General Procedure for d,l-N-Boc-amino Acids SEt
Synthesis
d,l-N-Boc-Phg-SEt: To a solution of d,l-N-Boc-Phg-OH
(3 g, 11,9 mmol) in 100 mL of dichloromethane, DCC
(2.71 g, 13.1 mmol, 1.1 equivs.) and DMAP (150 mg,
1.2 mmol, 0.1 equiv.) were added and then ethanethiol
(1.1 mL, 14.3 mmol, 1.2 equiv.) was added at room tempera-
ture. After a short induction period, the temperature rises to
about 30–358C and a crystalline white precipitate was
formed. The reaction was brought to completion by further
stirring overnight at room temperature. At this time, a
yellow suspension was obtained and the reaction was finish-
ed. The reaction mixture was filtered to eliminate a portion
of white solid (N,N’-dyclohexylurea) and the filtrate was
evaporated under reduced pressure using a rotary evapora-
tor. The crude product was purified by column chromatogra-
phy on silica gel using hexane/ethyl acetate (95:5) as eluent
to give a white solid; yield: 2.9 g (82%). Rf: 0.33 (hexane-
ethyl acetate, 9/1). 1H NMR (DMSO-d6): d=1.13 (t, 3H),
1.41 (s, 9H), 2.80 (2H), 5.27 (br s, 1H), 7.35 (m, 3H), 7.42
(m, 2H), 7.99 (br s, 1H); 13C NMR (CDCl3): d=198.65
Conditions for the effective chemo-enzymatic dy-
namic kinetic resolution of racemic compounds are
being continuously sought after.[9] The present
method of DKR of N-protected b,g-unsaturated a-
amino acid thioesters is highly efficient allowing the
complete deracemisation of the amino acid derivative
in high yield and selectivity. The enzyme selectivity
parallels the one observed with the oxoesters[5c,d] and
assures the enantioselectivity of the hydrolytic pro-
cess. However, the presence of the sulphur atom is es-
sential in providing a useful racemisation rate. The
latter can also be dependent on the nature of the thio-
ester, the R group and the nitrogen protecting group.
The method should therefore be of general applicabil-
ity to acyclic and cyclic a-substituted carboxylic acid
Figure 2. Rate of exchange of a-H in compound 1 as a function of base concentration.
Adv. Synth. Catal. 2007, 349, 1345 – 1348
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1347

Dario Arosio et al.
COMMUNICATIONS
(CHCOS), 154.65 (NHCOO), 136.82 (C ar.), 128.89 (C ar),
[3] a) A. S. Bommarius, M. Kottenhahn, H. Klenk, K.
Drauz, A direct route from hydantoins to d-amino acids
employing a resting cell biocatalyst with d-hydantoinase
and d-carbamoylase activity, in: Microbial reagents in
organic synthesis, (Ed. S. Servi), Kluwer Academic Pub-
lisher, Nato ASI series, 1992, vol. 381, pp 161–175;
b) O. May, S. Verseck, A. Bommarius, K. Drauz, Org.
Proc. Res. Dev. 2002, 6, 452–457.
128.50 (C ar), 127.44 (C ar), 80.23 [C
ACHTREUNG
(NHCHCOS), 28.22 [C
(SCH2CH3).
ACHTREUNG
General Procedure for the Dynamic Kinetic
Resolution on d,l-N-Boc-amino Acids SEt
Dynamic Kinetic Resolution of d,l-N-Boc-Phg-SEt: To a
solution of d,l-N-Boc-Phg-SEt (2 g, 6.67 mmol) in 50 mL of
MTBE, water (100 mL) and trioctylamine (1.5 mL,
3.38 mmol, 0.5 equivs.) were added. The pH was adjusted to
8.0 with NaOH (0.5M) and 510 U of subtilisin was added.
The reaction was kept at pH 8.0 by the automatic addition
of NaOH (0.5M), at 378C, under vigorous mechanical stir-
rer. After 2 days, the consumption of NaOH 0.5 N reached
100%. The water phase was separated from organic layer
and extracted (at pH 8) with 30 mL of diethyl ether. l-N-
Boc-Phg-OH was recovered by extraction of the water
phase with ethyl acetate (3”50 mL) at pH 3. Drying and
evaporation of the solvent gave a quantitative amount of l-
N-Boc-Phg-OH as a yellow oil. The enantiomeric excess an-
alysed by chiral HPLC was >99%.
[4] a) P.-J. Um, D. G. Drueckhammer J. Am. Chem. Soc.
1998, 120, 5605–5610; b) S. D. Tan, M. M. Günter,
D. G. Drueckhammer J. Am. Chem. Soc. 1995, 117,
9093–9094; c) C.-N. Lin, S.-W. Tsai, Biotechnol.
Bioeng. 2000, 69, 31–38.
[5] a) A. Caligiuri, P. D’Arrigo, E. Rosini, D. Tessaro, G.
Molla, S. Servi, L. Pollegioni, Adv. Synth. Catal. 2006,
348, 2183–2190; b) A. Caligiuri, P. D’Arrigo, T. Gef-
flaut, G. Molla, L. Pollegioni, E. Rosini, C. Rossi, S.
Servi, Biocat. Biotrans. 2006, 24, 409–413; c) E.
Agosta, A. Caligiuri, P. D’Arrigo, S. Servi, D. Tessaro,
R. Canevotti, Tetrahedron: Asymmetry 2006, 17, 1995–
1999; d) M. Cambi›, P. D’Arrigo, E. Fasoli, S. Servi, D.
Tessaro, R. Canevotti, L. Del Corona, Tetrahedron:
Asymmetry 2003, 14, 3189–3196.
[6] R. M. Williams, J. A. Hendrix, Chem. Rev. 1992, 92,
889–917.
[7] Y. Z. Belokon, I. E. Zeltzer, V. I. Bakhmutov, M. B.
Saporovskaya, M. G. Ryzhov, A. I. Yanovsky, Y. T.
Struchkov, V. M. Belikov, J. Am. Chem. Soc. 1983,105,
2010–2017.
[8] The ratio of the specific activities of the protease cata-
lyzed hydrolysis d,l-N-Boc-Phg-OEt/d,l-N-Boc-Phg-
SEt=2.52.
l-N-Boc-Phg-OH was deprotected with dioxane saturated
with HCl at 608C to give L-Phg-OH·HCl as a white solid;
yield: 1.106 g (87%; ee>99%); [a]2D5: +149.4 (c 0.8 1N HCl,
lit[10]: +156 (c1, 1 N HCl).
References
[1] a) K. M. Koelle, C.-H. Wong, Nature 2001, 409, 232–
240; b) R. J. Kazlauskas, U. T. Bornscheuer, Biotrans-
formations with lipases, in: Biotechnology, (Eds.: H.-J.
Reem, G. Reed), Wiley-VCH, Weinheim, 1998, vol. 8a,
pp 37–191; c) J. M. J. Williams, R. J. Parker, C. Neri,
Enzymatic kinetic resolution, in: Enzyme Catalysis in
Organic Synthesis, (Eds.: K. Drauz, H. Waldmann),
Wiley-VCH, Weinheim 2002, vol. 1, pp 287–310.
[2] a) H. Pellissier, Tetrahedron 2003, 59, 8291–8293C;
b) N. J. Turner, Curr. Opin. Chem. Biol. 2004, 8, 114–
119; c) C. Gruber, I. Lavandera, K. Faber, W. Kroutil,
Adv. Synth. Catal. 2006, 348, 1789–1805.
[9] a) J. Paetzold, J. E. Bꢁckvall, J. Am. Chem. Soc. 2005,
127, 17620–17621; b) B. Marti N-Matute, M. Edin, K.
Bogꢂr, J. E. Bꢁckvall, Angew. Chem. Int. Ed. 2004, 43,
6535–6539; c) O. Pamies, J. E. Bꢁckvall, Trends. Bio-
technol. 2004, 22,130–135; d) O. Pamies, J. E. Bꢁckvall,
Chem. Rev. 2003, 103, 3247–3261; e) M.-J. Kim, Y.
Ahn, J. Park, Curr. Opin. Biotechnol. 2002, 13, 578–
587.
[10] Fluka catalogue No. 78560.
1348
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 1345 – 1348