Designing enzymatic resolution of amines

Article abstract of DOI:10.1039/a808028c

A new strategy, utilizing IR and mass spectrometry, has been developed to design appropriate reagents and reaction conditions for enantioselective enzymatic protection of amines with readily removable protecting groups.

Full text of DOI:10.1039/a808028c

Designing enzymatic resolution of amines
Shuichi Takayama, Shelly T. Lee, Shang-Cheng Hung and Chi-Huey Wong*
Department of Chemistry and the Skaggs Institute for Chemical Biology, The Scripps Research Institute, 10550 North
Torrey Pines Road, La Jolla, CA 92037, USA. E-mail: wong@scripps.edu
Received (in Corvallis, OR, USA) 15th October 1998, Accepted 23rd November 1998
A new strategy, utilizing IR and mass spectrometry, has
been developed to design appropriate reagents and reaction
conditions for enantioselective enzymatic protection of
amines with readily removable protecting groups.
but can be suppressed under special conditions; (c) reagents that
do not react spontaneously with amines. Based on our
experience, reagents in category (a) are not useful for selective
enzymatic protection of amines. Reagents in category (b) may
be useful under conditions that suppress spontaneous reactions
(high dilution and/or use of a reaction-suppressing solvent such
as 3-methylpentan-3-ol4 ). Reagents in category (c) are perhaps
the most useful and can be used under any condition compatible
with the enzyme, so the use of high concentrations of reactants
is not a problem and in many cases is actually beneficial for the
reaction. Since reagents in category (b) do not have good
leaving groups, they are poor reagents for protecting alcohols,
and in fact can be used to selectively protect amines in the
presence of hydroxy groups (Table 1, entry 6). In contrast, some
of the reagents in categories (a) and (b) possess good leaving
groups and are thus useful for the protection of alcohols and
amino alcohols (e.g. 7–9).
Stereoselective enzymatic acylation of alcohols via transester-
i,j
ification in organic solvents has become a useful strategy in
1
enzymatic synthesis. This method has been further improved to
eliminate the problems of reversible reactions and product
inhibition by using irreversible acyl donors,2 and among these
,3
3
reagents enol esters have proven to be the most useful. Much
less developed, however, is the enzymatic protection of
4
amines, as two major problems are often encountered: (i)
amines are much more nucleophilic than alcohols and react non-
enzymatically with the esters commonly used in the enzymatic
acylation of alcohols, and (ii) unlike esters, which can be readily
cleaved under basic conditions, once amines are acylated, harsh
conditions are often required to liberate the free amine.
We have systematically investigated a number of potentially
useful amine protecting reagents and their reactivities and
conditions in order to develop a general strategy for the
selection of appropriate protecting reagents. To tackle the first
problem, the reactivities of a number of amine protecting
reagents were studied by comparing, with TLC, the amount of
spontaneous background reaction with an amine in a non-
To solve the second problem of liberating free amines, we
then selected reagents that will give readily removable amine
protecting groups.6 These exercises allowed us to quickly
4
i
identify several novel (e.g. 6, 8, 11), and some known (e.g. 9,
4e
10 ) enzymatic amine protecting reagents, along with suitable
reaction conditions for their use.
Benzylisopropenyl carbonate 8 is also a mildly activated
reagent that gives benzyl carbamates under spontaneous
reaction suppressing conditions. Dibenzyl carbonate 11 is a less
reactive version of reagent 8 and useful for amine resolution
when used under highly concentrated reaction conditions.
reaction-suppressing (toluene) or reaction-suppressing solvent
d,i,j
(
3-methylpentan-3-ol),4 and by measuring the IR absorption
maxima of the carbonyl groups [n(CNO)]. The n(CNO) values
reflect the CNO bond length and correlate with the reactivity of
carbonyl compounds; the larger the wavenumber, the shorter
Previous attempts to use 11 in amine resolutions have been
unsuccessful,4
conditions.
e,7
perhaps due to poor design of reaction
the CNO bond, and the more reactive the carbonyl group (Fig.
5
1
). The correlation between the IR absorption maxima and the
Allyl pent-4-enoate 6 is a less reactive version of 9 and can be
used under highly concentrated conditions whereas reagent 9
requires use of spontaneous reaction-suppressing conditions.
Representative new examples of using these protecting
reagents/reaction conditions for enzymatic resolution of
amines, including the pharmaceutically important 1-amino-
indane,4j are shown in Table 1. Reagents 6, 10 and 11 are
especially useful, given that they are readily available, that the
reactivity is most accurate for comparing compounds with
similar structures, but can also be useful for estimating the
reactivity of dissimilar esters and carbonates.
As seen in Fig. 1, amine protecting reagents can be
categorized into three categories depending on their reactivity
with amines of interest: (a) reagents that react spontaneously
with amines; (b) reagents that spontaneously react with amines
More reactive
Less reactive
(c)
(a)
(b)
F
F
F
F3C
O
FH2C
O
O
O
O
O
O
O
O
O
F
F
O
O
O
34a (1754)
4 (1749)
54k (1741)
1
(1788)
2 (1765)
6 (1739)
O
CF3
O
O
O
CN
O
O
O
O
O
O
O
94i (1753)
O
O
74d
(1761)
8 (1755)
104e (1748)
11 (1742)
Fig. 1 Reactivity of protecting reagents and their usefulness in enzymatic protection: (a) too reactive, (b) useful under spontaneous reaction suppressing
conditions and (c) useful under reaction promoting conditions. The number in parenthesis is IR absorption maxima for the carbonyl group.
Chem. Commun., 1999, 127–128
127

Table 1 Examples of amine resolution using the newly designed amine
protecting reagents and reaction conditions†
We thank Dr J. Wu and Professor G. Siuzdak for mass
spectral analysis, and the NIH (GM 44154) for financial
support.
Yield Ee
24c
Entry Product
Reagent Conditionsa (%)b (%) [a]
D
Ph
1
O
O
6
A
43
(R)
+98
Notes and references
99% (0.73)
N
H
† Condition A: Amine (0.94 mmol), toluene (230 ml), protecting reagent
(1.69 mmol), molecular sieves 4 Å powder (114 mg) and Candida
antarctica lipase (20 mg) was stirred for 48–70 h at 24 °C. The mixture was
directly chromatographed (SiO ; hexanes–Et O, 3:1) to give the products.
The ees were determined by HPLC (Chiralpak AD or Chiralcel OD-H).
Condition B: (±)-Methylbenzylamine (168 mg, 1.39 mmol), hexane (10
ml), 8 (197 mg, 1.03 mmol), molecular sieves 4 Å powder (920 mg), and
Candida antarctica lipase (240 mg, Sigma) was stirred for 60 h at 24 °C.
2
2
Ph
2
10
A
46
(R)
+48
N
O
H
83% (0.74)
The reaction was filtered through Celite, the filtrate was diluted with Et
washed (dilute HCl; brine), dried (MgSO ) and concentrated in vacuo. The
residue was chromatographed (SiO ; hexanes–Et O, 4:1) to give the product
120 mg, 34%).
2
O,
4
2
2
Ph
Ph
(
3
4
11
8
A
33
34
(R)
+44
‡ Measurement of reaction efficiency using mass spectrometry: (R)-a,4-
Dimethylbenzylamine (35 mg, 0.29 mmol), 6 (80 mg, 0.57 mmol), 11 (138
mg, 0.57 mmol), toluene (3 ml) molecular sieves 4A (300 mg), and Candida
antarctica lipase (20 mg) was stirred for 24 h at room temperature. After
filtering through Celite, a portion of the filtrate (5 ml) was mixed with an
internal standard (10 ml of a 1:1 solution of 14:12), and injected directly into
a PE SCIEX API100 electrospray mass spectrometer in the positive
ionization mode. The relative amount of products formed was determined
NHZ
NHZ
99% (0.59)
B
(R)
+38
86% (0.85)
+
by the peak intensity ratio of the [M + H ] peaks (218/204 = 14.5) of the
HN
O
+
5
6
A
46
(R)
+56
pent-4-enamides and the [M + Na ] peaks (292/278 = 3.14) of the benzyl
99% (0.89)
carbamates (14.5:3.14 = 4.6:1).
1
2
A. M. Klibanov, Acc. Chem. Res., 1990, 23, 114.
Y.-F. Wang, J. J. Lalonde, M. Momongan, D. E. Bergbreiter and C.-H.
Wong, J. Am. Chem. Soc., 1988, 110, 7200; H. M. Sweers, C.-H. Wong,
J. Am. Chem. Soc., 1986, 108, 6421; J. M. Fang and C.-H. Wong, Synlett,
1994, 6, 393; M. Degueil-Castaing, B. DeJeso, S. Drouillard and B.
Maillard, Tetrahedron Lett., 1987, 28, 953; J. B. West, J. Scholten, N. J.
Stolowich, J. L. Hogg, A. I. Scott and C.-H. Wong, J. Am. Chem. Soc.,
OH
6
7
8
11
11
11
A
36
(S,S) +1.8
8
8
5
2% (0.80)
NHZ
1
988, 110, 3709; A. Ghogare and G. S. Kumar, J. Chem. Soc., Chem.
A
A
41
(R)
–2.8
Commun., 1989, 1533; V. Gotor and R. Pulido, J. Chem. Soc., Perkin
Trans. 1, 1991, 491; M. Pozo, R. Puliddo and V. Gortor, Tetrahedron,
1
1% (0.47)
NHZ
992, 48, 6477; G. Kretzsdchmar and M. Schudok, Tetrahedron Lett.,
997, 38, 387.
1
3
4
A. Zaks and D. R. Dodds, Drug Discovery Today, 1997, 2, 513; K. Faber,
Biotransformations in Organic Chemistry, Springer, Heidelberg, 1995;
C.-H. Wong and G. M. Whitesides, Enzymes in Synthetic Organic
Chemistry, Pergamon, Oxford, 1994.
(a) F. Balkenhohl, K. Ditrich, B. Hauer and W. Ladner, J. Prakt. Chem.,
1
Gasset, J. Chem. Soc., Chem. Commun., 1993, 2453; (c) V. Gotor,
Tetrahedron, 1993, 49, 4321; (d) H. Kitaguchi, P. A. Fitspatrick, J. E.
Huber and A. M. Klibanov, J. Am. Chem. Soc., 1989, 111, 3094; (e) B.
Orsat, P. B. Alper, W. J. Moree, C.-P. Mak and C.-H. Wong, J. Am.
Chem. Soc., 1996, 118, 712; (f) M. Pozo and V. Gotor, Tetrahedron,
1993, 49, 10725; (g) M. Pozo, V. Gotor, Tetrahedron, 1993, 49, 4321; (h)
S. Puertas, F. Robelledo, V. Gotor, Tetrahedron, 1995, 51, 1495; (i) S.
Takayama, W. J. Moree and C.-H. Wong, Tetrahedron Lett., 1996, 35,
19
(R)
7% (0.70)
+19
N
Boc
Z
a
A: toluene (high concentration), Candida antarctica lipase (CAL); B:
a
b
21
2
21 c
hexane (dilute), CAL. Isolated yield. In 10 deg cm g . Concentra-
tion (c) in CHCl shown in parenthesis.
997, 339, 381; (b) V. Gotor, E. Menendez, Z. Mouloungui and A.
3
reaction conditions are simple and user friendly (does not even
require a pH meter), and that they give amides and carbamates
that are widely utilized as amine protecting groups.6
Finally, since the efficiency of enzymatic reactions may not
correlate with chemical reactivity, quantitative mass spectrome-
try has been used to compare reagents 6 and 11 for their
6
287; (j) A. L. Gutman, E. Meyer, E. Kalerin, F. Polyak and J. Sterling,
8
Biotechnol. Bioeng., 1992, 40, 760; (k) K.-E. Jaeger, K. Liebeton, A.
Zonta, K. Schimossek and M. T. Reetz, Appl. Microbiol. Biotechnol.,
1996, 46, 99; (l) S. Fernandez, R. Brieva, F. Rebolledo and V. Gotor,
J. Chem. Soc., Chem. Commun., 1992, 2885; (m) E. Schoffers, A.
Golebiowski and C. R. Johnson, Tetrahedron, 1996, 52, 3769; (n) M.
Pozo, R. Pulido and V. Gotor, Tetrahedron, 1992, 48, 6477.
For n values, see: R. M. Silverstein, G. C. Bassler and T. C. Morrill,
Spectrometric Identification of Organic Compounds, Wiley, New York,
1
T. W. Greene and P. G. M. Wuts, Protective Groups in Organic
Synthesis, Wiley, New York, 1991; R. Madsen, C. Roberts and B. Fraser-
Reid, J. Org. Chem., 1995, 60, 7920.
7 A. Maestro, C. Astorga and V. Gotor, Tetrahedron: Asymmetry, 1997, 8,
3153.
8 S. Takayama, R. Martin, J. Wu, K. Laslo, G. Siuzdak and C.-H. Wong,
J. Am. Chem. Soc., 1997, 119, 8146; J. Wu, S. Takayama, C.-H. Wong
and G. Siuzdak, Chem. Biol., 1997, 4, 653.
efficiency in Candida antarctica lipase catalyzed protection of
amines. The amount of protected amine, formed in an
enzymatic reaction containing equimolar amounts of protecting
reagents 6 and 11 but with a limiting amount of amine, was
measured by directly injecting a quenched reaction mixture into
a mass spectrometer and comparing the peak intensities of the
products to those of internal standards.‡ Compound 6 was found
to be approximately five times as efficient as 11 in enzymatic
amine protection. Thus, the chemical reactivity of a reagent
determined by its IR absorption combined with a rapid
assessment of its enzymatic reactivity using mass spectrometry
illustrated in this study provides a new effective strategy for the
development of new protecting reagents and conditions for
efficient enzymatic amine resolutions. We have used 6 and 11
for enantioselective enzymatic transformation of more than 50
amines so far and work is in progress to further expand the
scope of their application.
5
6
991.
Communication 8/08028C
128
Chem. Commun., 1999, 127–128