Enzymatic resolution of (RS)-2-(1-aminoethyl)-3-chloro-5-(substituted) pyridines

Article abstract of DOI:10.1016/j.tetasy.2004.07.043

The enantioselectivity of the lipase-catalyzed acetylation of several (RS)-2-(1-aminoethyl)-3-chloro-5-(substituted)pyridines has been examined. An enantiomeric excess (substrate) of 94% at 55% conversion was obtained for acetylation of the (R)-isomer of

Full text of DOI:10.1016/j.tetasy.2004.07.043

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 2797–2799
Enzymatic resolution of
(RS)-2-(1-aminoethyl)-3-chloro-5-(substituted)pyridines
Amy E. Sigmund and Robert DiCosimo*
DuPont Central Research and Development, Experimental Station, PO Box 80328, Wilmington, DE 19880-0328, USA
Received 1 June 2004; accepted 6July 2004
Available online 11 September 2004
Abstract—The enantioselectivity of the lipase-catalyzed acetylation of several (RS)-2-(1-aminoethyl)-3-chloro-5-(substituted)pyr-
idines has been examined. An enantiomeric excess (substrate) of 94% at 55% conversion was obtained for acetylation of the (R)-iso-
mer of (RS)-2-(1-aminoethyl)-3-chloro-5-bromopyridine by Candida antartica lipase B in ethyl acetate, whereas the (R)-isomer of
(RS)-2-(1-aminoethyl)-3,5-dichloropyridine and (RS)-2-(1-aminoethyl)-3-chloro-5-(difluoromethoxy)pyridine were each acetylated
with significantly lower enantioselectivity. Substitution of methyl propionate, methyl isobutyrate or methyl methoxyacetate for ethyl
acetate resulted in decreased enantioselectivity.
ꢀ 2004 Elsevier Ltd. All rights reserved.
X
Cl
X
Cl
X
Cl
1. Introduction
H
N
CALB
EtOAc
+
NH2
NH2
N
N
N
Numerous examples of the enzyme-catalyzed resolution
of racemic amines by the enantioselective aminolysis of
esters have been reported.1–4 The resolution of ( )-1-(2-
pyridyl)ethylamine was performed using Candida antar-
tica lipase B (CALB) to catalyze the enantioselective
aminolysis of ethyl acetate by the (R)-enantiomer of
the racemic amine.5,6 The resolution of bicyclic hetero-
arylamines using CALB in ethyl acetate has also been
reported to result in high enantioselectivity for the pro-
duction of the (S)-amine.6 This method has now been
examined for the resolution of several 2-(1-amino-
ethyl)-3-chloro-5-(substituted)pyridines to determine if
this particular functional-group substitution of the pyr-
idyl ring affected the enantioselectivity of the reaction.
O
(RS)-1a-c
(S)-1a-c
(R)-2a-c
a: X = Br; b: X = Cl; c: X = HF2CO
Scheme 1.
to the reactions as the hydrochloride salt. Enzyme-cata-
lyzed reaction rates were significantly slower in the ab-
sence of TEA, and only low conversions (<20%) to
amide were obtained. There was no aminolysis of ethyl
acetate by 1a hydrochloride or 1a/TEA in the absence
of added enzyme.
Only CALB catalyzed the production of (S)-1a11 with
high enantioselectivity (Fig. 1, E = 27); a 94% ee at
55% conversion of 0.10M (RS)-1a was obtained at
30ꢁC. Increasing the concentration of (RS)-1a from
0.10 to 0.30M resulted in only a slight decrease in ee
of (S)-1a at 55% conversion (Table 1). In all the reac-
tions, ee (R)-2a13 was 75–76% at ca. 55% conversion,
indicating that (R)-2a may partially racemize as it is pro-
duced. A similar result was previously reported for the
resolution of 2-(1-aminoethyl)pyridine by CALB in neat
ethyl acetate at 60ꢁC (99% ee (S)-1, 75% ee (R)-2 at 56%
conversion).6 Optimal reaction temperature was be-
tween 22 and 30ꢁC, where a slightly lower (S)-1a ee
was observed at 40ꢁC.
2. Results and discussion
Seventeen lipases7 were screened as catalysts for the ami-
nolysis of ethyl acetate by (RS)-2-(1-aminoethyl)-3-
chloro-5-bromopyridine8,9 (RS)-1a (Scheme 1). Triethyl-
amine (TEA, 1equiv) was added to all the reactions,10 as
the halide-substituted pyridylamines 1a–c8,9 were not
stable as the free amine for long periods of time (as
either pure compound, or in solution), and were added
*
Corresponding author. Tel.: +1 302 695 9416; fax: +1 302 695
8114; e-mail: robert.dicosimo@usa.dupont.com
0957-4166/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.07.043

2798
A. E. Sigmund, R. DiCosimo / Tetrahedron: Asymmetry 15 (2004) 2797–2799
acylating agent, and does not result in a yield loss of the
100
80
60
40
20
0
desired chiral amine. The low ee for (R)-2a obtained in
these reactions does not allow for the simultaneous pro-
duction of (R)-1a in high ee [hydrolysis of (R)-2a using
6N HCl produces no decrease in ee of the resulting
(R)-1a], but it may be possible to thermally racemize
(R)-2a to (RS)-2a (as was demonstrated for the N-acetyl
derivative of 2-(1-aminoethyl)pyridine6), and make pos-
sible the complete utilization of the racemate for the
production of (S)-1a.
0
10
20
30
40
50
60
Acknowledgements
(RS)-1a conversion (%)
The authors thank John Daub and Earl Reed (DuPont
Crop Protection Products) for the preparation of 1a–c
and 2a, and Michael Walker, Raphael Shapiro, Albert
Casalnuovo, Peter Bloxham, Kingmo Sun and William
Moberg (DuPont CPP) for helpful discussions, and con-
firmation of the chirality of the reaction products.
Figure 1. Dependence of enantiomeric excess of (S)-1a on conversion
of (RS)-1a (0.10–0.30M) to (R)-2a using CALB (50mg/mL) in ethyl
acetate (j), and calculated E = 27 (fitted curve).
The CALB-catalyzed aminolysis of ethyl acetate by
(RS)-2-(1-aminoethyl)-3,5-dichloropyridine (RS)-1b or
(RS)-2-(1-aminoethyl)-3-chloro-5-(difluoromethoxy) pyri-
dine (RS)-1c was each less enantioselective than when
using (RS)-1a (Table 1). At conversions of less than
50%, reaction rates for aminolysis of ethyl acetate by
(RS)-1b were ca. 2-fold greater than for (RS)-1a,
whereas reaction rates were ca. 2-fold less for (RS)-1c
than for (RS)-1a; these relative reaction rates may be
related to the effect of changes in the steric bulk of the
pyridyl 5-substituent on the reaction rate. Increasing
the steric bulk of the acylating agent employed in the
CALB-catalyzed resolution of (RS)-1a (Table 1, entries
4 and 5) also resulted in a significant decrease in enantio-
selectivity, and only moderately affected the already-low
enantioselectivity for the resolution of (RS)-1b (Table 1,
entries 4 and 5).
References
1. Alfonso, I.; Gotor, V. Chem. Soc. Rev. 2004, 33, 201–209.
2. van Rantwijk, F.; Sheldon, R. A. Tetrahedron 2004, 60,
501–519.
3. Breen, G. Tetrahedron: Asymmetry 2004, 15, 1427–1430.
´
4. Pamies, O.; Ell, A. H.; Samec, J. S. M.; Hermanns, N.;
´
Ba¨ckvall, J.-E. Tetrahedron Lett. 2002, 43, 4699–4702.
´
5. Iglesias, L. E.; Sanchez, V. M.; Rebolledo, F.; Gotor, V.
Tetrahedron: Asymmetry 1997, 8, 2675–2677.
6. Skupinska, K. A.; McEachern, E. J.; Baird, I. R.; Skerlj,
R. T.; Bridger, G. J. J. Org. Chem. 2003, 68, 3546–3551.
7. Lipases (ICR nos 101–117; ICR 110 = CALB lyo.) were
obtained from BioCatalytics, Inc., Pasadena, CA.
8. Foor, S. R. PCT Int. Appl. WO2003080576A2, 2003;
Chem. Abstr. 2003, 139, 276823.
9. Neubert, T. D.; Piotrowski, D. W.; Walker, M. P. U.S.
Patent Appl. 2004044040 A1, 2004; Chem. Abstr. 2002,
136, 263098.
10. In a typical reaction, 55.2mg (0.20mmol) of (RS)-1a
hydrochloride,8,9 28lL (0.20mmol) of triethylamine, and
100mg/mL of CALB in 2.0mL of ethyl acetate were mixed
at 30ꢁC for 24h. To the reaction mixture was then added
2.0mL of 20mM tetradecane (internal standard) in 1:1
acetonitrile/methanol, and the resulting mixture filtered
Although the use of CALB as catalyst for the resolution
of (RS)-1a–c was only successful for (RS)-1a, this reac-
tion can be used to prepare (S)-1a in high ee. The use of
an inexpensive acylating agent (ethyl acetate) as neat
solvent eliminates the need to run the reaction under
scrupulously dry conditions (such as in the presence of
molecular sieves14), as any water present in the reaction
mixture will simply hydrolyze a small percentage of the
Table 1. Aminolysis of alkyl esters by (RS)-1a–c catalyzed by Candida antartica lipase Ba
Entry
Compound
Concentration
(mM)
Time
(h)
Alkyl ester
(RS)-1
(mM)
(RS)-2
(mM)
Conversion
(%)
(S)-1 ee
(%)
(R)-2 ee
(%)
Eb
1
2
(RS)-1a
(RS)-1a
(RS)-1a
(RS)-1a
(RS)-1a
(RS)-1b
(RS)-1b
(RS)-1b
(RS)-1b
(RS)-1cc
102
200
298
102
104
107
102
103
105
052
24
24
24
Ethyl acetate
Ethyl acetate
Ethyl acetate
45
92
53
92
55
54
55
54
94
91
89
75
7625
7619
5
27
3
133
150
52
23
4
6Methyl methoxyacetate
6Methyl isobutyrate
6Ethyl acetate
49
81
ND
ND
5
ND
ND
ND
3.7
ND
75
1
6
59
47
45
54
47
6
7
6Methyl methoxyacetate
6Methyl isobutyrate
6Methyl propionate
56 ND
ND
10
5
8
86ND
77
17
3.5
1
9
10
ND
27
ND
2
25
Ethyl acetate
26
ND
49
69
68
12
a Reactions run at 30ꢁC in neat alkyl ester as solvent, 50mg CALB/mL, and 1equiv TEA per (RS)-1a–c hydrochloride.
b Calculated for production of (S)-1a–c according to Ref. 12.
c 37.5mg CALB/mL reaction.

A. E. Sigmund, R. DiCosimo / Tetrahedron: Asymmetry 15 (2004) 2797–2799
2799
(Gelman 0.2lm GHP syringe filter) and analyzed by gas
chromatography (GC). For HPLC analysis, a 0.100mL
aliqout of the filtrate was mixed with 0.400mL of 150mM
N,N-dimethylbenzamide (HPLC internal standard) in 1:1
acetonitrile/methanol. Chiral GC analysis of (R)-2a–c and
Wiant, C. F.; Jordan, C. F. J. Am. Chem. Soc. 1973, 95,
811–818. The absolute stereochemistry for products 2a–c
were inferred from the configuration of the corresponding
amines, which were acetylated.
12. Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J.
Am. Chem. Soc. 1982, 104, 7294–7298.
(S)-2a–c was performed using
a GTA-40 column
(Advanced Separation Technologies) at 160ꢁC. No
changes in relative amounts of (R)- and (S)-enantiomers
of 2a were produced by the analysis, verified by analyzing
a mixture of 9:1 (S)- and (R)-2a. Chiral HPLC analysis of
(R)-2a–b and (S)-2a–b was also performed using UV
detection at 280nm on a (S,S)-Whelk-01 column (Regis,
25cm · 4.6mm id), using 30% isopropanol/70% hexanes
(isocratic); ee was consistent with that measured by chiral
GC analysis. Quantitation of (RS)-2a was performed on a
DB-1701 GC column (J&W Scientific, 30m, 0.53mm OD,
1-lm film thickness). Chiral HPLC analysis of (R)-1a–c
and (S)-1a–c was performed using UV detection at 280nm
on a Chiralcel OD-H column (Daicel, 25cm · 4.6mm id)
at 30ꢁC and a flow rate of 0.6mL/min of 0.2% diethyl-
amine in 10% 2-propanol/90% hexanes; N,N-dimethyl-
benzamide was used as internal standard.
13. Acetyl chloride (2.31g, 29.4mmol) and (RS)-1a hydro-
chloride8,9 (8.00g, 29.4mmol) were dissolved in dichloro-
methane (53mL) under nitrogen. Triethylamine (6.55g,
64.7mmol) was added dropwise with stirring while main-
taining temperature <30ꢁC. The solvent was removed by
rotary evaporation, the remaining mixture suspended in
dichloromethane (20mL), and triethylamine hydrochlor-
ide removed by filtration. The filtrate was washed with
dichloromethane (5mL), the wash and filtrates combined,
and (RS)-2a recovered as a yellow solid by flash chroma-
tography on silica (300mL) using 9:1 ethyl ether/dichlo-
1
romethane (5.33g, 66% yield): mp 123–125ꢁC; H NMR
(500MHz, CDCl3): d 8.50 (d, J = 2.0Hz, 1H), 7.85 (d,
J = 2.0Hz, 1H), 6.86 (b s, 1H), 5.52 (m, 1H), 2.04 (s, 3H),
1.40 (d, J = 6.7Hz, 3H); 13C NMR (125MHz, CDCl3): d
169.1, 156.8, 148.1, 139.6, 130.0, 118.8, 46.5, 23.4, 20.9;
HRMS calcd for C9H10N2OBrCl (M+1) 276.9743, found
276.9733.
11. Resolution of (RS)-1a and (RS)-1b with (L)-(+)-tartaric
acid [yielding (S)-1a and (S)-1b, respectively], and (D)-(À)-
tartaric acid [yielding (R)-1b] produced standards used to
confirm the enantiospecificity of CALB-catalyzed amino-
lysis reactions: Smith, H. E.; Schaad, L. J.; Banks, R. B.;
14. Sigmund, A. E.; McNulty, K. C.; Nguyen, D.; Silverman,
C. E.; Ma, P.; Pesti, J. A.; DiCosimo, R. Can. J. Chem.
2002, 80, 608–612.