A simple enantioconvergent and chemoenzymatic synthesis of optically active α-substituted amides

Article abstract of DOI:10.1002/anie.201105164

Simple and efficient: The combination of an enzymatic, enantioinverting reaction with simple follow-up processes allows the transformation of readily available racemic compounds into versatile chiral α-substituted amides (see picture; Ms=methanesulfonyl).

Full text of DOI:10.1002/anie.201105164

Communications
DOI: 10.1002/anie.201105164
Enantioconvergent Processes
A Simple Enantioconvergent and Chemoenzymatic Synthesis of
Optically Active a-Substituted Amides**
´
Wiktor Szymanski, Alja Westerbeek, Dick B. Janssen,* and Ben L. Feringa*
Enantiopure amides 1, substituted in the a position with an
azide or a simple N, O, or S substituent (Scheme 1), form an
important class of compounds and are key intermediates in
SR) include lipid-lowering agent eflucimibe,[10] TNFa con-
vertase inhibitors,[11] and renin inhibitors.[12]
Considering the importance of the broad family of
enantiopure a-heterosubstituted amides 1, we decided to
investigate the possibility of designing an enantioconvergent
and chemoenzymatic process for their preparation. Herein,
we present a time- and cost-efficient synthetic route that is
characterized by high enantioselectivity and product versa-
tility, and takes advantage of the intrinsically compatible
product reactivity of haloalkane dehalogenase catalyzed
reactions.
Enantioconvergent processes are transformations in
which both enantiomers of a chiral substrate are converted
into the same enantiomer of the product through different
pathways.[13] Such processes allow the highly efficient trans-
formation of cheap racemic starting materials into valuable
enantiopure products by increasing the yield beyond the
theoretical value of 50%, which is the yield that can be
obtained from the common transformation of only one
enantiomer, for example, by kinetic resolution. Chemoenzy-
matic enantioconvergent syntheses can be carried out in three
general ways (Figure 1).
Scheme 1. General structure of 1 and examples of bioactive com-
pounds derived from 1.
medicinal chemistry. The a-azidoamide motif (1, Nu = N3) is
found in bioactive molecules, such as the antibiotic azidocillin
(Scheme 1). The azide group is considered to be an isostere of
the amino group, however, it has increased bulkiness and is
not positively charged at physiological pH values.[1] The
introduction of an azide function into the a position of
peptides is also aimed at the synthesis of enzyme inhibitors
that, after binding to their targets and subsequent Staudinger-
type ligation with biotin–phosphine, are subjected to strepta-
vidin pull-down.[2] Furthermore, enantiopure a-azidoamides
are important reagents for “click” chemistry[3] and valuable
precursors in the synthesis of amino acids[4] and peptides.[5,6]
The chiral scaffold of an amide with a simple O substitu-
ent (1, Nu = OR) in the a position can be found in numerous
bioactive compounds, such as solimastat (Scheme 1), which is
an inhibitor of matrix metalloproteinase,[7] and other anti-
cancer[8] and anticonvulsant[9] agents. Bioactive molecules
that bear a chiral amide motif with an S substituent (1, Nu =
´
[*] Dr. W. Szymanski, Prof. Dr. B. L. Feringa
Center for Systems Chemistry
Stratingh Institute for Chemistry, University of Groningen
Nijenborgh 4, 9747 AG, Groningen (The Netherlands)
E-mail: b.l.feringa@rug.nl
Figure 1. Strategies used in chemoenzymatic enantioconvergent
processes.
A. Westerbeek, Prof. Dr. D. B. Janssen
Department of Biochemistry, University of Groningen
Nijenborgh 4, 9747 AG, Groningen (The Netherlands)
E-mail: d.b.janssen@rug.nl
The first approach (Figure 1a) employs two enzymes that
express opposite enantiopreferences. One enzyme catalyzes
the transformation of substrate A to product C with retention
of configuration, while the other one inverts the stereochem-
istry in the conversion of ent-A to C, usually by an SN2-type
process. This approach has been applied mainly for the
hydrolysis of epoxides, by using two epoxide hydrolases[14,15]
[**] We thank Theodora Tiemersma-Wegman and Monique Smith for
their technical assistance with HPLC and GC analyses.
Supporting Information for this article (including experimental
details and analytical data for the products) is available on the
WWW under http://dx.doi.org/10.1002/anie.201105164.
10712
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 10712 –10715

or one enzyme that catalyzes the convergent transformations
of both enantiomers with distinct regioselectivity.[16–18]
The second approach (Figure 1b) uses one enzyme to
perform a kinetic resolution of racemic substrate A. In a
subsequent set of reactions, enantioenriched compounds ent-
A and B are transformed into a single product C; one of these
transformations proceeds with retention and the other with
inversion of configuration. Such deracemizations can be
performed, for example, by combining the lipase-catalyzed
kinetic resolution of racemic esters with a subsequent stereo-
inversion of the obtained enantioenriched alcohols via
sulfonyl esters[4,19,20] or nitrate esters,[19] or by following a
Mitsunobu protocol.[21]
Scheme 2. Chemoenzymatic enantioconvergent preparation of (R)-1
(compounds 5–8) from rac-2. Ms=methanesulfonyl, TEA=triethyl-
amine, Tris=tris(hydroxymethyl)aminomethane.
The third approach (Figure 1c) relies on the use of a single
enantioinverting enzyme for the conversion of one of the
enantiomers of substrate A into product B. In the follow-up
reaction, this product and, optionally, the unreacted ent-A,
are transformed into the final compound C with retention of
configuration. Most recent examples of such syntheses
include the use of inverting alkyl sulfatases for enantioselec-
tive sulfate ester hydrolysis, which produces an alcohol and an
unreacted enantiomer of the sulfate ester, both with the same
absolute configuration.[22–24] The follow-up chemical hydro-
lysis of the enantioenriched sulfate ester proceeds with
retention of configuration and gives the same enantiomer of
the alcohol as the one produced in the enzymatic inverting
step.
Despite its apparent efficiency and elegance, the use of
the first approach (Figure 1a) is severely limited by low
availability of enzyme pairs that show complementary
enantioselectivity. The efficiency of the latter approaches
(Figure 1b,c) strongly depends on the high enantioselectivity
of the enzymatic reaction and high atom economy of the
chemical steps (which is especially low in case of the
Mitsunobu protocol).[21] Furthermore, it would be advanta-
geous if no separation of intermediates was needed and the
purification steps were reduced to a minimum.[22]
haloalkane dehalogenases as the inverting enzymes in the
biocatalytic step, and sulfonyl esters for the follow-up
chemical transformations.
We have recently established that easily prepared racemic
a-bromoamides (rac-2) can be converted by haloalkane
dehalogenases with high enantioselectivity (E values
> 200).[26] Upon extraction of the aqueous enzymatic reaction
mixture, both products (S)-2 and (S)-3 can be obtained with
high overall yields (> 80%). We expected that no separation
of intermediates would be needed prior to the conversion of
alcohols (S)-3 into methanesulfonyl esters (S)-4, and that this
reaction mixture would also not have to be processed before
the final step (Scheme 2). These simplifications to the
procedure would fulfill the requirements of limiting the
intermediate separation and purification steps in the process-
es described in Figure 1c. In the last reaction, in which the a-
bromoamide (S)-2 and the a-mesyloxyamide (S)-4 are both
converted in an SN2-type process into final products (R)-1, we
employed azide and benzylamine as N nucleophiles, ethane-
thiol as an S nucleophile, and phenol as an O nucleophile.
Racemic a-bromoamide 2a (Scheme 3), which is a
precursor of compound 4a, undergoes kinetic resolution
catalyzed by a haloalkane dehalogenase from Bradyrhi-
zobium japonicum USDA110 (DbjA) with excellent enantio-
selectivity (E > 200).[26] To establish the best conditions for
the reactions of model nucleophiles with mixtures of a-
bromoamides (S)-2 and a-methanesulfonyloxyamides (S)-4,
we performed a study in which the racemic model compound
4a was reacted to racemic products 5a–8a under various
conditions (Table 1).
Haloalkane dehalogenases are enzymes that catalyze the
À
hydrolytic cleavage of carbon halide bonds by employing a
catalytic aspartate ion for the nucleophilic displacement of
the halide in an SN2-type reaction.[25] This process gives a
covalent intermediate in which the product is bound to the
enzyme by an ester bond. This bond is subsequently hydro-
lyzed to form the product and to recover the nucleophilic
aspartate ion. Overall, the reaction proceeds with inversion of
configuration, therefore the enzyme can be efficiently used in
the process described in Figure 1c.
Additional possibilities stem from the fact that this
reaction gives two enantiopure compounds, a haloalkane
and an alcohol, both of which possess high intrinsic reactivity
and outstanding potential for functionalization. In particular,
both the halide (directly) and the hydroxide (upon simple
activation) can be regarded as efficient leaving groups, which
sets the stage for a number of nucleophilic displacements
leading to diverse classes of products.
Taking advantage of this reactivity, we designed a chemo-
enzymatic enantioconvergent process (Scheme 2) for the
synthesis of enantiopure amides 1 (Scheme 1) by using the
general approach described in Figure 1c. This process uses
Scheme 3. Racemic a-bromoamides.
The reaction of compound 4a with sodium azide in DMF
as the solvent resulted in the full conversion of the substrate in
2 h, and product 5a was isolated in almost quantitative yield
(Table 1, entry 1). When similar conditions were employed
for the introduction of the N nucleophile benzylamine in the
presence of potassium carbonate as the base, the reaction
took 2 days to reach full conversion and the product was
Angew. Chem. Int. Ed. 2011, 50, 10712 –10715
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10713

Communications
Table 1: Model transformations of racemic compound 4a.
Table 2: Enantioconvergent chemoenzymatic preparation of compounds
5–8.
Nucleophile
Substrate
rac-2b[b]
rac-2a[a]
rac-2c[c]
NaN3
BnNH2
PhOH
EtSH
5a, 81%, 98% ee 5b, 73%, 97% ee 5c, 80%, 95% ee
6a, 81%, 97% ee 6b, 96%, 91% ee 6c, 85%, 90% ee
7a, 70%, 95% ee 7b, 57%, 93% ee 7c, 66%, 96% ee
8a, 82%, 28% ee 8b, 76%, 92% ee 8c, 63%, 73% ee
Entry Nucleophile
Solvent Amount
of K2CO3
t
Product Yield[a]
1
2
3
4
5
6
NaN3 (5 equiv)
BnNH2 (2 equiv) DMF
BnNH2 (2 equiv) MeCN 2 equiv
PhOH (2 equiv) MeCN 2 equiv
EtSH (2 equiv)
EtSH (2 equiv)
DMF
–
2 h
5a
6a
6a
7a
8a
8a
96%
<50%
71%
73%
71%
2 equiv
46 h
46 h
20 h
46 h
20 h
[a] DbjA (1.7 wt%, 42 h). [b] DbjA (1.1 wt%, 20 h). [c] LinB (2.3 wt%,
44 h).
MeCN 2 equiv
DMF 2 equiv
70%
which compounds (S)-3 were mesylated to give intermediates
(S)-4 (Scheme 2). We observed that these reactions pro-
ceeded cleanly and did not affect the a-bromoamides (S)-2.
Crude reaction mixtures obtained in the second step were
concentrated in vacuo, and no purification or separation of
intermediates (S)-2 and (S)-4 was required. This step was
followed by the introduction of nucleophiles by using the
optimized conditions for SN2 reactions (Table 1, entries 1, 3, 4,
and 6).
The reactions leading to a-azidoamides (R)-5a–c gave
these products in 73–81% yield (based on rac-2a–c), thus
providing the proof-of-principle for the proposed enantio-
convergent synthetic route by showing that both products of
the enzymatic reaction can be efficiently transformed into a
single product (Table 2). The high enantiomeric excess (ee >
95%) of the obtained products renders our methodology
suitable for the preparation of chiral a-azidoamides, which
are important precursors for peptide synthesis.
Transformations in which benzylamine was used as a
nucleophile led to a-(benzylamino)amides (R)-6a–c (Table 2)
with even better yields than in the case of the azide
nucleophile. The high enantiomeric excess of the products
(up to 97% for (R)-6a) further confirms the potential of the
presented methodology for the preparation of peptide
analogues.
The yields obtained in the syntheses of a-phenoxyamides
(R)-7a–c were generally lower (Table 2) than for the reac-
tions with sodium azide and benzylamine, yet were still
significantly superior to the maximum theoretical yield of
50% that could be obtained if only one product of the
enzyme-catalyzed kinetic resolution was used for further
transformations. Considering the high enantiomeric excess of
the products (93–96% ee, Table 2), we conclude that the
presented methodology can be efficiently used for the
stereoselective preparation of chiral amides substituted with
O nucleophiles.
[a] Yield of isolated product. Bn=benzyl.
isolated with low yield because of the formation of side
products (Table 1, entry 2). The use of acetonitrile instead of
DMF as the solvent (Table 1, entry 3) resulted in a much
cleaner reaction, and product 6a was isolated in good yield
after 46 h. When these conditions were used for the reaction
of compound 4a with phenol (Table 1, entry 4), full con-
version of substrate was already achieved after 20 h. Under
the same conditions, 4a could be converted to the a-ethyl-
thioamide 8a in 71% yield within 46 h (Table 1, entry 5).
Virtually the same yield was obtained with DMF as solvent,
but the reaction time was shorter (Table 1, entry 6). On the
basis of these results, the latter conditions were chosen for the
planned enantioconvergent chemoenzymatic synthesis of
compound 8a and its analogues.
To study the feasibility of the transformations presented in
Scheme 2, we focused on three model substrates (Scheme 3).
Racemic compounds 2a–c are readily accessible in more than
90% yield in one step from simple starting materials.[26]
Furthermore, compound 2b, which includes a moiety derived
from glycine, the most simple amino acid, was chosen to show
that our methodology has potential in the preparation of
peptides. Moreover, compound 2c features an ethyl substitu-
ent on the a position, which enabled us to assay the influence
of steric hindrance on the efficiency of the final step of the
transformation presented in Scheme 2.
Subsequently, we conducted a series of reactions aiming at
the enantioconvergent chemoenzymatic preparation of com-
pounds 5–8 (Scheme 2, Table 2). The racemic precursors rac-2
were initially subjected to a haloalkane dehalogenase medi-
ated kinetic resolution (Scheme 2), thus providing a selective
transformation to a-hydroxyamides (S)-3 by using DbjA and
haloalkane dehalogenase from Sphingomonas paucimobilis
UT26 (LinB). In the enzymatic reactions, conversions of
50%, which are crucial for the efficient enantioconvergent
process,[27,28] were observed after 20–44 h. The enantiomeric
excess of the products was found to be higher than 97% in all
cases. Notably, only a small amount of enzyme, typically
corresponding to 0.01 mol% of biocatalyst, was used for these
conversions (Table 2). The reaction mixtures obtained from
the biotransformations were extracted with ethyl acetate and
no further product separation or purification was required.
After evaporation of the solvent, the crude mixtures of
enantiomerically enriched a-bromoamides (S)-2 and a-
hydroxyamides (S)-3 were subjected to the reactions in
In the transformations aimed at a-ethylthioamides (R)-
8a–c, the products were obtained in good yields (Table 2),
albeit with low ee values. In the least satisfying case,
compound (R)-8a was obtained with only 28% ee. We
suspect that this enhanced racemization may result from the
ethylthio substituent, which stabilizes the intermediate carb-
anion and thus increases the acidity of the proton attached to
the a position of compound 8.[29,30] The deprotonation and
subsequent reprotonation (enolization) of compound 8 under
the basic reaction conditions (two equivalents of potassium
carbonate) would lead to its racemization. We found that the
10714 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 10712 –10715

[8] S. Hanessian, L. Auzzas, A. Larsson, J. B. Zhang, G. Giannini, G.
Gallo, A. Ciacci, W. Cabri, ACS Med. Chem. Lett. 2010, 1, 70 –
74.
[9] S. V. Andurkar, J. P. Stables, H. Kohn, Bioorg. Med. Chem. 1999,
7, 2381 – 2389.
[10] A. J. Lopez-Farre, D. Sacristan, J. J. Zamorano-Leon, N. San-
Martin, C. Macaya, Cardiovasc. Ther. 2008, 26, 65 – 74.
[11] B. Barlaam, T. G. Bird, C. Lambert-van der Brempt, D. Camp-
bell, S. J. Foster, R. Maciewicz, J. Med. Chem. 1999, 42, 4890 –
4908.
[12] J. T. Repine, J. S. Kaltenbronn, A. M. Doherty, J. M. Hamby,
R. J. Himmelsbach, B. E. Kornberg, M. D. Taylor, E. A. Lunney,
C. Humblet, S. T. Rapundalo, B. L. Batley, M. J. Ryan, C. A.
Painchaud, J. Med. Chem. 1992, 35, 1032 – 1042.
ee value of 8c decreased from the initial value of 73% to 69%
after two weeks of storage at 48C, thus confirming our
hypothesis. Subsequently, we incubated the compound for 5 h
under the nucleophilic displacement reaction conditions (two
equivalents of potassium carbonate, DMF) and found that the
ee value further decreased to 65%. These data support our
notion on the enhanced razemisation tendency of products 8.
We have designed an efficient enantioconvergent process
for the synthesis of various enantiopure a-substituted amides,
which form an important class of chiral building blocks in
medicinal chemistry. The process combines the high enantio-
selectivity of the haloalkane dehalogenase catalysed resolu-
tion with the efficient and convergent transformation of both
the starting material and the biocatalysis product into a single
enantiomer of the final compound 1. The products are
obtained in high yield (up to 96% after three steps) and
with high enantiomeric excess (up to 98%) in the case of
configurationally stable compounds 5–7. In the whole process,
the purification and separation steps are eliminated, thus
rendering it a practical and time-efficient methodology. We
are currently directing our efforts toward using directed
evolution methodologies for the broadening of the substrate
scope of haloalkane dehalogenases, and as a consequence
fully enhancing the versatility of the new process.
[13] Y. Simeo, W. Kroutil, K. Faber in Enzymes in Action: Green
Solutions for Chemical Problems (Eds.: B. Zwanenburg, M.
´
Mikołajczyk, P. Kiełbasinski), Kluwer Academic Publishers,
Dordrecht, 2000, pp. 27 – 51.
[14] L. Cao, J. T. Lee, W. Chen, T. K. Wood, Biotechnol. Bioeng. 2006,
94, 522 – 529.
[15] K. M. Manoj, A. Archelas, J. Baratti, R. Furstoss, Tetrahedron
2001, 57, 695 – 701.
[16] E. Y. Lee, M. L. Shuler, Biotechnol. Bioeng. 2007, 98, 318 – 327.
[17] S. Hwang, C. Y. Choi, E. Y. Lee, Biotechnol. Lett. 2008, 30, 1219 –
1225.
[18] M. I. Monterde, M. Lombard, A. Archelas, A. Cronin, M.
Arand, R. Furstoss, Tetrahedron: Asymmetry 2004, 15, 2801 –
2805.
[19] H. Danda, T. Nagatomi, A. Maehara, T. Umemura, Tetrahedron
1991, 47, 8701 – 8716.
[20] K. Lemke, S. Ballschuh, A. Kunath, F. Theil, Tetrahedron:
Asymmetry 1997, 8, 2051 – 2055.
Received: July 22, 2011
Published online: September 16, 2011
Keywords: amides · biocatalysis · chiral resolution ·
enantioselectivity · haloalkane dehalogenases
[21] S. Takano, M. Suzuki, K. Ogasawara, Tetrahedron: Asymmetry
1993, 4, 1043 – 1046.
[22] S. R. Wallner, M. Pogorevc, H. Trauthwein, K. Faber, Eng. Life
Sci. 2004, 4, 512 – 516.
[23] P. Gadler, T. C. Reiter, K. Hoelsch, D. Weuster-Botz, K. Faber,
Tetrahedron: Asymmetry 2009, 20, 115 – 118.
[24] M. Schober, P. Gadler, T. Knaus, H. Kayer, R. Birner-Grꢀn-
berger, C. Gꢀlly, P. Macheroux, U. Wagner, K. Faber, Org. Lett.
2011, 13, 4296 – 4299.
.
[1] K. P. Kokko, M. K. Hadden, K. S. Orwig, J. Mazella, T. A. Dix,
J. Med. Chem. 2003, 46, 4141 – 4148.
[2] W. A. van der Linden, P. P. Geurink, C. Oskam, G. A. van der
Marel, B. I. Florea, H. S. Overkleeft, Org. Biomol. Chem. 2010,
8, 1885 – 1893.
[3] V. P. Mocharla, B. Colasson, L. V. Lee, S. Rçper, K. B. Sharpless,
C.-H. Wong, H. C. Kolb, Angew. Chem. 2005, 117, 118; Angew.
Chem. Int. Ed. 2005, 44, 116 – 120.
[4] W. Szymanski, R. Ostaszewski, Tetrahedron: Asymmetry 2006,
17, 2667 – 2671.
[5] W. Szymanski, M. Zwolinska, R. Ostaszewski, Tetrahedron 2007,
63, 7647 – 7653.
[6] W. Szymanski, R. Ostaszewski, Tetrahedron 2008, 64, 3197 –
3203.
[7] J. F. Fisher, S. Mobashery, Cancer Metastasis Rev. 2006, 25, 115 –
136.
[25] D. B. Janssen, Curr. Opin. Chem. Biol. 2004, 8, 150 – 159.
´
[26] A. Westerbeek, W. Szymanski, H. J. Wijma, S. J. Marrink, B. L.
Feringa, D. B. Janssen, Adv. Synth. Catal. 2011, 353, 931 – 944.
[27] E. Vꢁnttinen, L. Kanerva, Tetrahedron: Asymmetry 1995, 6,
1779 – 1786.
[28] S. Pedragosa-Moreau, C. Morisseau, J. Baratti, J. Zylber, A.
Archelas, R. Furstoss, Tetrahedron 1997, 53, 9707 – 9714.
[29] D. J. Cram, Fundamentals of Carbanion Chemistry, Academic
Press, New York, 1965, pp. 71 – 84.
[30] C. F. Bernasconi, K. W. Kittredge, J. Org. Chem. 1998, 63, 1944 –
1953.
Angew. Chem. Int. Ed. 2011, 50, 10712 –10715
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10715