Enzymatic one-pot resolution of two nucleophiles: Alcohol and amine

Article abstract of DOI:10.1016/S0957-4166(00)00069-0

Enzymatic aminolysis of racemic secondary alcohol ester derivatives with racemic amines leads to the resolution, in one reaction, of the alcohol and the amine with very high enantioselectivity. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(00)00069-0

Tetrahedron: Asymmetry 11 (2000) 1459±1463
Enzymatic one-pot resolution of two nucleophiles:
alcohol and amine
Eduardo Garcõa-Urdiales, Francisca Rebolledo and Vicente Gotor*
Â
Departamento de Quõmica Organica e Inorganica, Universidad de Oviedo, 33071, Oviedo, Spain
Â
Â
Received 2 February 2000; accepted 18 February 2000
Abstract
Enzymatic aminolysis of racemic secondary alcohol ester derivatives with racemic amines leads to the
resolution, in one reaction, of the alcohol and the amine with very high enantioselectivity. # 2000 Elsevier
Science Ltd. All rights reserved.
Lipases and esterases accept a wide range of substrates, which are usually converted with high
enantioselectivity. Kinetic resolutions of primary and secondary alcohols in organic solvents are
carried out, in most cases, through lipase-catalyzed transesteri®cation and esteri®cation
reactions.1 3 In order to shift the equilibrium inherent to these kinds of reactions towards product
synthesis, activated esters such as vinyl acetate are routinely employed.4 Taking into account that
lipases do not have amidase activity,5,6,y resolution of alcohols by aminolysis of their corresponding
ester derivatives constitutes an alternative to the use of activated esters. Moreover, this strategy
opens the possibility to the one-pot resolution of both nucleophiles when a racemic amine is used.
Optically active alcohols and amines are important classes of compounds. Thus, amines bearing
an a stereogenic centre have been extensively employed as resolving agents,8 chiral auxiliaries and
intermediates in the synthesis of a wide range of biologically active molecules.9 More recently,
they have also been used for the induction of helicity in polymers.10 On the other hand, enantio-
merically pure secondary alcohols are useful chiral auxiliaries in organic chemistry both for
analytical and synthetic applications.11
Despite the consequent advantages of carrying out two resolutions in a single process, this
possibility has only been attempted using alcohols and a thiol12 or a diol as nucleophiles.13 In the
former case, problems of reversibility led to enantiomeric excesses (ee) lower than for the corre-
sponding single resolutions. However, this strategy has never been applied to amines to the best
* Corresponding author. Tel: +34 985 10 34 51; fax: +34 985 10 34 48; e-mail: vgs@sauron.quimica.uniovi.es
y
The hydrolysis of N-acetylamines catalyzed by Candida antarctica lipase B in phosphate bu€er has been described
but >10 days of reaction were necessary to achieve a high conversion at 50ꢀC.7
0957-4166/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(00)00069-0

Â
E. Garcõa-Urdiales et al. / Tetrahedron: Asymmetry 11 (2000) 1459±1463
1460
of our knowledge. Here we report for the ®rst time the one-pot kinetic resolution of di€erent
secondary alcohols and amines bearing an a stereogenic centre.
All reactions (Table 1) were carried out using lipase B from Candida antarctica (CALB) as
catalyst (500 mg), 1,4-dioxane as solvent (10 mL) at 30ꢀC and 250 r.p.m. Enzyme and reaction
conditions were selected on the basis of previous results obtained in the stereoselective amidation
of racemic amines.14 In all cases the corresponding acetyl derivative of the alcohol and the amine
were in equimolar ratio (3.0 mmol). In order to avoid the competitive enzymatic hydrolysis of the
ester, reagents and solvent were dried prior to use and nitrogen atmosphere was used. Moreover,
4 A molecular sieves were added (350 mg) to ensure a low water activity in the reaction medium.
When the reaction was stopped, the crude of the reaction consisted of a mixture of four
compounds Ð ester, alcohol, amine and amide Ð but the isolation and puri®cation of each one
was easily accomplished by acidic extraction and subsequent column chromatography of the organic
phase.{ Absolute con®gurations for both nucleophiles were determined by comparison of the sign
of the speci®c rotation of either substrate or product with the data published in the literature.
Two conversion values were calculated for each aminolysis reaction: ca obtained from ee values for
Table 1
Compounds 1a and 3a: R1=Ph, R2=Me; 1b and 3b: R1=n-hexyl, R2=Me; 1c and 3c: R1=Ph,
R2=methoxymethyl; 2a and 4a: R3=Ph, R4=Me; 2b and 4b: R3=n-pentyl, R4=Me; 2c and
4c: R3=1-naphthyl, R4=Me; 2d and 4d: R3=(3-tri¯uoromethyl)benzyl, R4=Me; 2e and 4e:
R3=ethyl, R4=Me; 2f and 4f: R3=Ph, R4=n-propyl
{
Once the enzyme had been ®ltered o€ and washed with methylene chloride, the organic solution was extracted with
2N HCl and water. The crude resulting from the evaporation of the organic phase was puri®ed by silica gel chroma-
tography employing a hexane/ethyl acetate gradient. Evaporation of the aqueous phase yielded the remaining amine in
its hydrochloride form. In all cases, ester and alcohol were isolated in >75% yield and amine and amide in >85%
yield. Yields were calculated taking into account conversion values.

Â
E. Garcõa-Urdiales et al. / Tetrahedron: Asymmetry 11 (2000) 1459±1463
1461
substrate (amine) and product (amide) and ce obtained from ee of ester and alcohol.15x Coincidence
in both conversion values was the criterion considered to con®rm that side reactions had not
taken place.{ Results obtained are summarized in Table 1.
First we examined the eciency of the CALB-catalyzed aminolysis of (þ)-1-(phenyl)ethyl
alcohol employing its corresponding acetate (þ)-1a as substrate. Di€erent racemic amines such
as (þ)-1-(phenyl)ethylamine (þ)-2a, (þ)-2-heptylamine (þ)-2b, (þ)-1-[1-(naphthyl)]ethylamine
(þ)-2c and (þ)-[1-(3-tri¯uoromethyl)phenyl]-2-propylamine (þ)-2dk (entries 1±4, respectively)
were used as nucleophiles. In all cases the enzyme reacts faster with the R enantiomer of both
ester and amine. Thus, we obtained the S esters and amines and the R alcohols and amides.
Values of the enantiomeric ratio15 obtained for the resolution of the alcohol with these four
amines (Ee) were very high. On the other hand, the E values obtained for the amines (Ea) were
excellent as well.
From the aminolysis of (þ)-2d with (þ)-1a (entry 4), acetamide (R)-4d is straightforwardly
obtained. This amide 4d constitutes an intermediate in the synthesis of fen¯uramine,16 a com-
pound with biological properties as anorectic agent.17 Due to the fact that both enantiomers of
fen¯uramine exhibit di€erent biological properties, obtaining both R- and S-4d is always desirable.
In our case, the remaining substrate (S)-2d can be transformed into (S)-4d thereby allowing the
preparation of both enantiomers of fen¯uramine easily.
Next, we used (þ)-2-octyl acetate as substrate (þ)-1b and (þ)-1-(phenyl)ethylamine (þ)-2a as
nucleophile (entry 5). Again results show the same enantiopreference of CALB towards R enant-
iomers and high enantiomeric ratios for both alcohol and amine.
These results are in agreement with the general rule for the stereospeci®ty and eciency of
lipases in the resolution of secondary alcohols.18 This rule is a consequence of the common
structure of the active site of lipases.19 This structure shows that the cavity of the active site,
which hosts the leaving group and the nucleophile, is the same one. When nucleophiles are sec-
ondary alcohols, the higher di€erence in size between the substituents at the stereogenic centre,
the greater E values will be obtained. In addition to this, when kinetic resolutions of secondary
alcohols are accomplished with CALB, it has been shown that if the medium sized substituent at
the stereogenic centre is larger than an ethyl group the enantiomeric ratio is signi®cantly
reduced.20 Results obtained in the kinetic resolution of a-disubstituted primary amines with
CALB show that these amines behave similarly to secondary alcohols.12,14
At this point, the question of the independence of both resolutions with regard to the
enantioselectivity arose. The proposed mechanism for lipase catalysis (Scheme 1) shows that
enantiodiscrimination of each substrate takes place in di€erent steps. Thus, this mechanism supports
the hypothesis that E values measured for each nucleophile of the pair do not depend at all on the
nature of its counterpart. In order to shed light on this aspect we decided to carry out the resolution
x
Enantiomeric excesses of compounds 1a, 3a, 1c, 3c, 4a and 4d were determined by means of chiral HPLC using a
Chiralcel OB-H column. Amines 2a and 2d were transformed into their corresponding acetamides (4a and 4d) prior to
ee analysis. The ee of ester 1b was determined by chiral GC using a Cydex-B chiral capillary column. Alcohol 3b was
transformed into 1b in order to determine its ee. The rest of the amines and amides were analyzed through chiral HPLC
with a Chiralcel OD column: compounds 2b, 4b, 2e and 4e had to be transformed into their N-benzoylated derivatives
prior to analysis, amides 4c and 4f were analyzed directly and amines 2c and 2f were previously transformed into their
corresponding acetamides.
{
Due to errors inherent in the measure of reagents and enantiomeric excesses, slight di€erences between ce and ca
kvalues were tolerated.
Amine (þ)-2d was prepared by reductive amination of 3-(tri¯uoromethyl)phenylacetone. Yield: 78%.

Â
E. Garcõa-Urdiales et al. / Tetrahedron: Asymmetry 11 (2000) 1459±1463
1462
Scheme 1. Catalytic mechanism: acylation and deacylation steps
of acetate (þ)-1a employing as nucleophiles (þ)-2-butylamine (þ)-2e and (þ)-1-phenyl-1-butylamine
(þ)-2f (entries 6 and 7). In these cases and according to the aforementioned rule, amines (þ)-2e
and (þ)-2f should be resolved with low Ea values. If the nature of the nucleophile does not a€ect
the resolution of the ester, the Ee value measured for (þ)-1-(phenyl)ethyl acetate (þ)-1a should
remain higher than 200.
Results obtained with 2-butylamine con®rm this hypothesis (entry 6). While the enantiomeric
ratio of the alcohol (Ee) remained higher than 200, the Ea value was 4.** When amine 2f was the
nucleophile (entry 7) the reaction rate dropped so much that the non-enzymatic aminolysis
became signi®cant under reaction conditions.yy Thereby, E values obtained for the corresponding
alcohol and amine underestimate the real ones and, consequently, they cannot be taken into
account to con®rm the independence of both resolutions. Despite this fact, amide (R)-4f was
obtained with ee=97%. The apparent enantiomeric ratio for the amine was 60. This apparent
enantioselectivity is surprisingly greater than the one we expected. If we consider that a medium
sized substituent larger than an ethyl group lowers the enantiomeric ratio of the reaction con-
siderably, the enantiomeric ratio for amine 2f should be similar to that obtained for analogous
secondary alcohols bearing an n-propyl group as the medium sized substituent, such as (þ)-4-
decanol. When this alcohol is resolved by means of CALB-catalyzed transesteri®cation, an E
value of 10 is obtained.20 We are currently investigating this fact with other secondary alcohols
and a-disubstituted primary amines in simple and simultaneous resolutions of both nucleophiles.
Finally, we employed (þ)-2-methoxy-1-(phenyl)ethyl acetate (þ)-1c as substrate, an ester
derived from an alcohol bearing a medium sized substituent larger than an ethyl group, and (þ)-
1-(phenyl)ethylamine (þ)-2a (Table 1, entry 8).{{ The reaction rate was again signi®cantly lower
than when esters derived from a-methyl substituted alcohols were used and the enantiomeric ratio
obtained was moderate (Ee=71). The E value measured for the amine (Ea>200) does not di€er
Because of the low boiling point of amine 2e (63ꢀC) it was dicult to keep an equimolar amine/ester ratio under
**
reaction conditions. This may account for the discrepancy in the conversion values measured for both nucleophiles (ce
and ca). In addition to this, partial solubility of acetamide 4e in both organic and aqueous phases made conventional
extraction inecient. Instead, the organic solution was submitted to continuous extraction over 12 h using methylene
chloride as organic solvent.
yy
Analysis of control reaction showed that the non-enzymatic reaction took place with a contribution to the total
conversion degree lower than 5%.
{{
In this case, the nature of substituents at the stereogenic centre of ester 1c leads to the opposite absolute con®guration
for the same stereochemistry.

Â
E. Garcõa-Urdiales et al. / Tetrahedron: Asymmetry 11 (2000) 1459±1463
1463
from the one obtained when (þ)-1-(phenyl)ethyl acetate (þ)-1a was used as substrate (compare
entries 1 and 8).
In conclusion, we have developed a simple and ecient procedure for the one-pot resolution of
racemic secondary alcohols and analogous primary amines. The simplicity of the isolation pro-
cedure of the four products obtained in each reaction makes the methodology very attractive. The
results obtained imply that this methodology can be applied successfully to other racemic alcohols
and amines which show high E values in simple enzymatic catalyzed kinetic resolutions.
Acknowledgements
We are grateful to the CICYT (Project BIO 98-0770) for ®nancial support and to Novo Nordisk
Co. for the generous gift of lipase B from Candida antarctica. E.G.-U. thanks Ministerio de
Educacion y Ciencia for a predoctoral fellowship.
References
1. Schmid, R. D.; Verger, R. Angew. Chem., Int. Ed. Engl. 1998, 37, 1609±1633.
2. Wong, C. H.; Whitesides, G. M. Enzymes in Synthetic Organic Chemistry; Pergamon Press: Oxford, 1994.
3. Faber, K. Biotransformations in Organic Chemistry, 3rd ed.; Springer: Berlin, 1997.
4. For a review including recent examples, see: Roberts, S. M. J. Chem. Soc., Perkin Trans. 1 1999, 1±21.
5. West, J. B.; Wong, C. H. Tetrahedron Lett. 1987, 28, 1629±1632.
6. Margolin, A. L.; Klibanov, A. M. J. Am. Chem. Soc. 1987, 109, 3802±3804.
7. Smidt, H.; Fischer, A.; Fischer, P.; Schmid, R. D. Biotechnol. Tech. 1996, 10, 335±338.
8. Jacques, J.; Collet, A.; Wilen, S. H. Enantiomers, Racemates, and Resolutions; John Wiley and Sons: New York,
1981.
9. Nogradi, M. Stereoselective Synthesis, 2nd ed.; VCH: Weinheim, Germany, 1995.
10. Yashima, E.; Matsushima, T.; Okamoto, Y. J. Am. Chem. Soc. 1997, 119, 6345±6359.
11. Morrison, J. D. Asymmetric Synthesis; Academic Press: New York, 1984.
12. Oehrner, N.; Orrenius, C.; Mattson, A.; Norin, T.; Hult, K. Enzyme Microb. Technol. 1996, 19, 328±331.
13. Lin, G.; Liu, S.-H.; Chen, S.-J.; Wu, F.-C.; Sun, H.-L. Tetrahedron Lett. 1993, 34, 6057±6058.
14. For a recent review, see: Gotor, V. Bioorg. Med. Chem. 1999, 7, 2189±2197.
15. Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am. Chem. Soc. 1982, 104, 7294±7299.
16. Fogagnolo, M.; Giovannini, P. P.; Guerrini, A.; Medici, A.; Pedrini, P.; Colombi, N. Tetrahedron: Asymmetry
1998, 9, 2317±2327.
17. Goument, B.; Duhamel, L.; Mauge, R. Tetrahedron 1994, 50, 171±188.
18. Kazlauskas, R. J.; Weiss¯och, A. N. E.; Rappaport, A. T.; Cuccia, L. A. J. Org. Chem. 1991, 56, 2656±2665.
19. For CALB structure, see: Uppenberg, J.; Ohrner, N.; Norin, M.; Hult, K.; Kleywegt, G. J.; Patkar, S.; Waagen,
V.; Anthonsen, T.; Jones, T. A. Biochemistry 1995, 34, 16838±16851.
20. Rottici, D.; Hñ€ner, F.; Orrenius, C.; Norin, T.; Hult, K. J. Mol. Catal. B: Enzym. 1998, 5, 267±272.