Available via license: CC BY 4.0

Content may be subject to copyright.

*Corresponding author: xuyi@sit.edu.cn

1 INTRODUCTION

Preparation methods of optically active compounds

are classified into two broad categories: the optical

resolution of racemic compounds and the asymmetri-

zation of prochiral compounds. Biocatalysts are wide-

ly used in both cases. When the starting material is a

racemic mixture, the most popular enzymatic ap-

proach to obtaining the optically active compounds is

kinetic resolution. However, the maximum theoretical

yield is limited to 50% and the tedious procedures for

the separation of the recovered starting material and

the product are inevitable and half of the starting ma-

terial (or product) has the wrong absolute configura-

tion for certain purposes.

To overcome these drawbacks, several methods have

been offered, such as the dynamic kinetic resolution.

Another method is the inversion of the stereogenic cen-

tre of the substrate (or product) after a biocatalytic res-

olution. For example, the lipase/Mitsunobu process of

secondary alcohol

[1-11]

or an acid hydrolysis/inversion

of the remaining epoxide in the epoxide hydro-

lase-catalysed enantiomeric hydrolysis of epoxide

[12-18]

.

By these methods, one can obtain the chiral compounds

with high optical purity at 100% theoretical yield. Alt-

hough pioneer works had been made before1 or 10, the

derivations were incomplete. In the following paragraph,

a complete derivation was made. Moreover, the possi-

bility for predicting the time (t

max

) which is needed to

reach the maximum enantiomeric excess of the final

product (ee

f

) was firstly explored.

2 EXPERIMENT

2.1 Generalization

All the chemicals and reagent were commercially

obtained and of analytical grade.

2.2 Enantioselective hydrolysis of 3-(2-nitrophenoxy)

propylene oxide (1a) by Trichosporon loubierii

ECU1040

Lyophilized yeast cells (3 g) were rehydrated in so-

dium phosphate buffer (90 ml, 100 mm, pH 7.0) for 30

min on a shaker (160 rpm, 30

o

C). Then 10 ml DMSO

containing 500 mg of the substrate was added and the

mixture was agitated at 30

o

C. Samples were taken at

different time. The ee value of epoxide was directly

determined by HPLC analysis through using Chiralcel

OD column. The mobile phase was hexane/ isopropa-

nol (90/10, v/v) at a flow rate of 1.0 ml/min and de-

tected at 254 nm.

2.3 Enantioselective hydrolysis of

trans-3-(4-methoxyphenyl)glycidic acid methyl

ester (MPGM) by Serratia sp. lipase

Experiments were performed through using a substrate

concentration of 50 mm in 10 ml toluene solution and

10 ml culture supernatant (the pH value was adjusted

to 7.5 by Tris-HCl buffer). The reactions were carried

out at 30

o

C and 160 rpm in 100 ml flasks equipped

Theoretical Prediction and Experimental Verification of ee

s

Versus Time

in Biocatalytic Resolution and ts Application in a Bioresolution-

inversion Process

Yi Xu* & Dan Zhou

School of Chemical and Environmental Engineering, Shanghai Institute of Technology, Shanghai, China

Jianbo Chen

College of Life and Environment Science, Shanghai Normal University, Shanghai, China

ABSTRACT: A systematic theoretical derivation of bioresolution-inversion process was made. An equation

was derived between the maximum ee value of final product (ee

f(max)

) and enantiomeric ratio (E) of a reaction.

The corresponding equations of conv.

(max)

, ee

p(max)

, ee

s(max

) versus E-value were also derivative and the interrela-

tionships among ee

f(max)

, conv.

(max)

, ee

p(max)

and ee

s(max)

were deduced. Furthermore, a simple equation was de-

veloped to predict the enantiomeric excess of substrate (ee

s

) at any other time of the whole reaction course based

on the ee

s

value which was determined at a certain reaction time. This equation of ee

s

versus time was verified by

three different experiments. Based on the equation of ee

s

versus time, a new equation for predicting the time

(t

(max)

) needed to reach the maximum enantiomeric excess of the final product (ee

f (max)

) after the resolu-

tion-inversion was developed.

Keywords: theoretical prediction; bioresolution-inversion; epoxide hydrolase; lipase

DOI: 10.1051/

C

Owned by the authors, published by EDP Sciences, 2015

/

00 ( 2015)

201conf

Web of Conferences

,

5

MATEC

2

00

atec

m

25

5

2

2

03

0

7KLVLVDQ2SHQ$FFHVVDUWLFOHGLVWULEXWHGXQGHUWKHWHUPVRIWKH&UHDWLYH&RPPRQV$WWULEXWLRQ/LFHQVH ZKLFKSHUPLWV

XQUHVWULFWHGXVHGLVWULEXWLRQDQGUHSURGXFWLRQLQDQ\PHGLXPSURYLGHGWKHRULJLQDOZRUNLVSURSHUO\FLWHG

4

3

Article available at http://www.matec-conferences.org or http://dx.doi.org/10.1051/matecconf/20152502003

MATEC Web of Conferences

with tight plugs. Samples were taken at different time

for the determination of ee value of MPGM. The ee

value was determined by HPLC with a chiral column

(Chiralcel OJ, 25×4.6 cm, Daicel Chemical Industries,

Tokyo, Japan) and elution by hexane/isopropanol (60:

40, v/v; 0.8 ml/min) and detection at 254 nm. The

retention time was respectively 13.5 and 15.7 min for

(2S, 3R)-MPGM and (2R, 3S)-MPGM.

2.4 Enzymatic transesterification of (R,

S)-4-hydroxy-3-methyl-2-(2-propenyl)-2-cyclope

nten-1-one) [HMPC] with vinyl acetate catalyzed

by Lipase PS

50 mm (R, S)-HMPC dissolved in vinyl acetate was

added to the lipase PS, and the reaction was conducted

at 30

o

C, 160 rpm. Samples were taken at different

time for the determination of ee value HMPC. The

enantiomeric excess of substrate (ee

s

) and product (ee

p

)

was determined by GLC using β-DEXTM 120 column

(oven temperature, 150

o

C; injector and detector tem-

perature, 280 ◦C). The retention time was respectively

15.4, 16.1, 20.6 and 21.2 min for (R)-HMPC acetate,

(S)-HMPC acetate, (S)-HMPC and (R)-HMPC.

3 DERIVATION OF EQUATIONS

3.1 Derivation of equations for resolution-inversion

process

If we define the enantiomeric excess of the final

product (after the biocatalytic resolution and inversion)

as ee

f

, the value of ee

f

would be directly dependent

upon the conversion ratio and the enantioselectivity of

biocatalyst, E-value. For a simple irreversible biocat-

alytic kinetic resolution-inversion process, supposing

that no racemization occurred in the whole course, we

can obtain equations 1~5

[19]

:

0

0

ln

ln

B

B

A

A

E

(1)

AB

AB

ee

s

(2)

QP

QP

ee

p

(3)

ps

s

eeee

ee

BA

QP

C

00

(4)

ps

ps

f

eeee

eeee

AQPB

AQPB

ee

2

)(

)()(

(5)

In this equation, A and B refer to the fast- and

slow-reacting enantiomers of the substrate; P and Q

refer to the corresponding enantiomers of the product;

ee

s

and ee

p

are respectively the enatiomeric excess of

the substrate and the product; E is the enantiomeric

ratio.

If we define

x

B

B

0

(6)

Then,

E

x

A

A

0

(7)

By substituting equations (6) and (7) into equations

2~5, we can obtain as follows:

22

1

E

xxC

(8)

E

E

s

xx

xx

ee

(9)

E

E

p

xx

xx

ee

2

(10)

E

f

xxee

(11)

Figure 1. Graphic plots of Conv. versus ee

f

at different

E-values

Figure 2. Graphic plots of ee

s

versus ee

f

at different E-values

02003-p.2

EMME 2015

By combination of equations 8~11, we can get plots

of C versus ee

f

(Figure 1), ee

s

versus ee

f

(Figure 2) and

ee

p

versus ee

f

(Figure 3). From Figures 1~3, we can

see that there is a maximum ee

f

value (ee

f(max)

) for a

fix ‘E’ and the ee

f(max)

varies with the change of

E-value.

Figure 3. Graphic plots of ee

p

versus ee

f

at different E-values

It is obvious that if

0

'

xf

ee

,

That is:

)1(

1

1

E

E

x

One can get the maximal ee

f

value: ee

f

(max) and

the corresponding conv., ee

p

, ee

s

are defined as conv.

(max), ee

p

(max), ee

s

(max). By substituting the x

value into equations 8~11, we can get the following

equations:

)1()1(

1

(max)

11

E

E

E

f

EE

ee

(12)

2

11

1.

)1()1(

1

(max )

E

E

E

EE

Conv

(13)

)1()1(

1

)1()1(

1

(max )

11

2

11

E

E

E

E

E

E

p

EE

EE

ee

(14)

1

1

11

11

)1()1(

1

)1()1(

1

(max )

E

E

EE

EE

ee

E

E

E

E

E

E

s

(15)

From equations 12~15, we can clearly know the

maximum ee

f

and corresponding conv., ee

p

and ee

s

.

For example, if E-value equals 200, we can get the

maximum ee

f

value 96.9% at 51.1% conversion or at

94.9% ee

p

or at 99.0% ee

s

. In practical process, the

chemical inversion (y) is not always 100%, perhaps

90% or others in some cases. Considering the above

mentioned condition, some modifications should be

made for an incompletely chemical inversion. In fact,

only equation 12 should be changed to equation 16

and others are kept unchanged (y is the efficiency of

chemical inversion):

)1()1(

1

(max )

11

E

E

E

f

EE

yee

(16)

Figure 4 shows the curves of

ee

f(max)

, ee

s(max)

,

ee

p(max)

and Conv.

(max)

vs. E. It is interesting to see

that the

ee

f(max

) value is always larger than ee

s(max)

value, but smaller than ee

p(max)

value.

Figure 4. Theoretical plots of ee

f (max)

, ee

s(max)

, ee

p(max)

and

conv

(max)

as a function of E according to equations (12) - (15)

Now, a question arises. How to predict the time for

the reaction to stop at appropriate moment to reach

ee

f(max)

?

3.2 Prediction of the time-dependant changes in

enantiomeric excess of substrate (ee

s

~ t)

According to Chen et al.

[20]

and Lu et al.

[21]

, for a

simple irreversible kinetic resolution, the E-value is

shown as follows:

0

0

ln

ln

B

B

A

A

E

This indicates that the distinction between two

competing enantiomers (A and B) by an enzyme is

equal to a constant E. Equation 1 can be re-written as

follows:

02003-p.3

MATEC Web of Conferences

E

kt

kt

B

B

A

A

0

0

ln

ln

(17)

Then the equation 17 can be derived to:

)exp(

0

ktAA

(18)

)exp(

0

E

kt

BB

(19)

It is at a low initial substrate concentration accord-

ing to Lu’s derivation and Michaelis-Menten equation

(if the substrate concentration is low enough and rela-

tive to Km, the reaction is the first order). Here, A

0

and B

0

are initial concentrations of the fast- and slow-

reacting enantiomers, k is the rate constant for the

fast-reacting enantiomer. For the kinetic resolution of

a racemate (A

0

=B

0

=0.5S

0

), it is known that:

AB

AB

ee

s

By substituting equation 18 and equation 19 into

equation 2, we can write as follows:

s

e

s

e

ee

t

E

k

A

B

1

1

ln)

1

1()ln(

(20)

Considering that both k and E are constants, we can

acquire as follows:

2

2

21

1

1

1

1

ln

1

1

1

ln

1

s

e

s

s

e

s

e

ee

te

ee

t

(21)

ee

s1

and ee

s2

are respectively ee

s

values at t

1

and t

2

.

Equation 21 can be written as follows:

1

1

1

1

1

1

1

2

1

1

1

2

1

1

2

t

t

s

s

t

t

s

s

s

ee

ee

ee

ee

ee

(22)

It can be concluded from equation 22 that if we

know ee

s1

at t

1

, then we can theoretically predict the

ee

s

value at another time (t

2

) in the same reaction

mixture.

By substituting equation 15 into equation 22, we

can get:

1

1

1(max )

1

1

ln

ln

s

e

s

e

ee

E

tt

(23)

If one knows the ee

s

value at t

1

and the E-value,he

can calculate the time which is needed to reach the

maximum ee

f

according to equation 23.

4 EXPERIMENTAL VERIFICATION OF

TIME-DEPENDANT CHANGES IN ENANTI-

OMERIC EXCESS OF SUBSTRATE (EES ~ T)

The equation 22 was verified by three different bio-

catalytic kinetic resolution experiments.

It can be seen from Figure 5 that for the first exam-

ple, , both of the theoretical curves (the curves were

respectively plotted according to equation 22 and ee

s

values at 30 min and 60 min) fit the experimental data

quite well in the resolution of 3-(2-nitrophenoxy)

propylene oxide by epoxide hydrolase of Tricho-

sporon loubierii ECU1040 [22]. This enables one to

stop the reaction at a proper time (e.g. ee

s

> 98%) to

get both high optical purity and high yield of the

epoxide. And this will also simplify the work of

measurement.

Figure 5 Variation of ee

s

in the resolution of racemic 1 by

lyophilized cells of Trichosporn loubierii ECU1040 (100

g/L). Symbols: Ƶ Measured; ũű Calculated with the

ee

s

at t = 30 min (10 mM) and 60 min (10 mM), respectively.

The second example is related to enzymatic resolu-

tion of MPGM. (2R, 3S)-MPGM, a very important

intermediate in the synthesis of Diltiazem Hydrochlo-

ride, can be prepared according to enantioselective

hydrolysis of the racemic MPGM catalyzed by Serra-

tia sp. Lipase

[23]

. So it is necessary to stop the reaction

when the ee

s

value was enough high so that we can get

(2R, 3S)-MPGM at both high yield and optical purity.

It can be seen from Figure 6 that the theoretical curves

fit the experimental data quite well. This enables one

to stop the reaction at a proper time (e.g. ee

s

98%) to

get both high optical purity and high yield of the

MPGM.

02003-p.4

EMME 2015

Figure 6. Variation of ee

s

in the resolution of racemic MPGM

by by Serratia sp. lipase. Symbols: Ƶ Measured; ũ

Calculated with the ee

s

at t = 1.5 h.

Chiral HMPG and its ester are very important agri-

cultural intermediates. Figure 7 showed the time

course of ee

s

value in transesterification of (R,

S)-HMPC with vinyl acetate which is catalyzed by the

Lipase PS. The theoretical curve was plotted based on

equation 22 and the ee

s

at 3h. The theoretical curves

fit the experimental data quite well. The t

(max)

value

can be also calcultated from equation 23. This enables

one to stop the reaction at a suitable time to obtain

high ee value and yield of the substrate ((S)-HMPC)

and product ((R)-HMPC acetate). And the highest

yield and ee value of final product ((R)-HMPC acetate)

can be obtained after the bioresolution/chemical in-

version.

Figure 7. Time course of ee

s

value in transesterification of (R,

S)-HMPC with vinyl acetate catalyzed by Lipase PS. Ʒ

Measured; ũCalculated with the ee

s

at t = 3 h.

5 CONCLUSIONS

A systematic theoretical derivation of bioresolu-

tion-inversion process was made. An equation was

derived between the ee

f(max)

and E-value of a reaction.

The corresponding equations of conv.

(max)

, ee

p(max)

,

ee

s(max)

versus E-value were also derived and the in-

terrelationships among ee

f(max)

, conv.

(max)

, ee

p(max)

and

ee

s(max)

were deduced. Furthermore, a simple equation

was developed to predict the enantiomeric excess of

substrate (ee

s

) at any other time of the whole reaction

course based on the ee

s

value which was determined at

a certain reaction time. This equation of ee

s

versus

time was verified by three different experiments.

Based on the equation of ee

s

versus time, a new equa-

tion for predicting the time (t

(max)

) needed to reach the

maximum enantiomeric excess of the final product

(ee

f(max)

) after the resolution-inversion which was

developed. The current work will be beneficial to the

biocatalytic resolution-inversion study.

ACKNOWLEDGEMENT

This work was supported by the Shanghai Committee

of Science and Technology (No. 13430503400), the

Science Foundation of Shanghai Institute of Technol-

ogy (YJ2010-04, YJ2011-54), the Scientific Research

Foundation for the Returned Overseas Chinese Schol-

ars, State Education Ministry (No. ZX2012-05) and

the Innovation Program of Shanghai Municipal Edu-

cation Commission (No. 11YZ227).

REFERENCES

[1] Woo M H, Kim H S. & Lee E. Y. 2012. Development

and characterization of recombinant whole cells ex-

pressing the soluble epoxide hydrolase of Danio rerio

and its variant for enantioselective resolution of racemic

styrene oxides. J. Ind. Eng. Chem., , 18: 384-391

[2] S. Easwar. & N.P. Argade. 2003. Amano PS-catalyzed

enantioselective acylation of (±)-α-methyl-1,

3-benzodioxole-5-ethanol: an efficient resolution of chi-

ral intermediates of the remarkable antiepileptic drug

candidate, (-)-talampanel. Tetrahedron: Asymmetry

14 :333-337

[3] A. Wallner, H. Mang, S.M. Glueck, A. Steinreiber, S.F.

Mayer. & K. Faber. 2003. Chemo-enzymatic enan-

tio-convergent asymmetric total synthesis of

(S)-(+)-dictyoprolene using a kinetic resolu-

tion-stereoinversion protocol. Tetrahedron: Asymmetry

14 : 2427-2432

[4] F. Compostella, L. Franchini, G.B. Giovenzana,

L.Panza, D. Prosperib. & F. Ronchettia. 2002.

Chemo-enzymatic stereo-convergent synthesis of

3-O-benzoyl-azidosphingosine. Tetrahedron: Asym-

metry, 13 : 867-872

[5] J. Oshida, M. Okamoto. & S. Azuma. 1999. Chemoen-

zymatic synthesis of 1α, 24(R)-dihydroxycholesterol.

Tetrahedron: Asymmetry 10 : 2337-2342

[6] A. Kamal, G.B.R. Khanna, R. Ramu. & T. Krishnaji.

2003. Chemoenzymatic synthesis of duloxetine and its

enantiomer: lipase-catalyzed resolution of

WK

HH

V

5HDFWLRQWLPHK

HH

V

02003-p.5

MATEC Web of Conferences

3-hydroxy-3-(2-thienyl) propanenitrile. Tetrahedron

Lett. 44 :4783-4787

[7] D.R. Boyd, N.D. Sharma. & J. Mol. 2002. Enzymatic

and chemoenzymatic synthesis of arene trans- dihydro-

diols. Catal. B: Enzymatic 19-20:31–42

[8] Steinreiber A, Stadler A, Mayer SF., Faber K. & Kappe

CO. 2001. High-speed microwave-promoted Mitsunobu

inversions. Application toward the deracemization of

sulcatol. Tetrahedron Lett. 42: 6283-6286

[9] P. Virsu, A. Liljeblad, A. Kanerva. & L.T. Kanerva.

2001. Preparation of the enantiomers of

1-phenylethane-1, 2-diol. Regio- and enantioselectivity

of acylase I and Candida antarctica lipases A and B.

Tetrahedron: Asymmetry 12: 2447-2455

[10]Thiago S. F., Marcos R.S., Maria CO, Telma L.G.L.,

Ricardo A.M. & Marcos C. M., 2015, Chemoenzymatic

synthesis of rasagiline mesylate using lipases. Appl.

Catal. A: General, 492: 76-82

[11]Zahia H., Mounia M..-K, Nassima B., Olivier R. &

Louisa A.-Z. 2013. A green route to enantioenriched

(S)-arylalkyl carbinols by deracemization via combined

lipase alkaline-hydrolysis/Mitsunobu esterification. Tet-

rahedron: Asymmetry, 24:290-296

[12]S.P. Moreau, C. Morisseau, J. Baratti, A. Archelas. & R.

Furstoss. 1997. Triplex stability of oligodeoxynucleo-

tides containing substituted quinazoline-2, 4-(1H,

3H)-dione. Tetrahedron, 53 :8457-8478

[13] R.V.A. Orru, S.F. Mayer, W. Kroutil. & K. Faber. 1998.

Chemoenzymic deracemization of (±)-2,2-disubstituted

oxiranes. Tetrahedron, 54 :859-874

[14]E. Vanttinen. & L.T. Kanerva. 1995. Combination of the

lipase-catalyzed resolution with the Mitsunobu esterifi-

cation in one pot. Tetrahedron: Asymmetry,

6 :1779-1786

[15]Wu Y-W., Kong X-D., Zhu Q-Q, Fan L-Q. & Xu J-H.

2015. Chemoenzymatic enantioconvergent hydrolysis of

p-nitrostyrene oxide into (R)-p-nitrophenyl glycol by a

newly cloned epoxide hydrolase VrEH2 from Vigna ra-

diate. Catal. Comm., 58: 16-20

[16] Lilian L. B., Bruna Z. C., Marcelo A.S. T., Clelton A. S.,

Aline C., Marianna T.P. F., André S. S., Juliano S. M.,

Anita J. M. & Anete P. S. 2013. A novel and enantiose-

lective epoxide hydrolase from Aspergillus brasiliensis

CCT 1435: Purification and characterization. Prot. Expr.

Pur. 91 :175-183

[17] R.V.A. Orru, W. Kroutil. & K. Faber. 1997. Deracemi-

zation of (+)-2, 2-disubstituted epoxides via enantiocon-

vergent chemoenzymic hydrolysis using Nocardia EH1

epoxide hydrolase and sulfuric acid. Tetrahedron Lett.

38:1753-1754

[18]C. Morisseau, H. Nellaiah, A. Archelas, R. Furstoss &

J.C. Baratti. 1997. Asymmetric hydrolysis of racemic

para-nitrostyrene oxide using an epoxide hydrolase

preparation from Aspergillus niger. Enzyme Microbiol.

Technol. 20:446-452.

[19]C.S. Chen, Y. Fujimoto, G. Girdaukas, C.J. Sih. & J.

Am. 1982. Quantitative analyses of biochemical kinetic

resolutions of enantiomers. Chem. Soc. 104:7294-7299

[20]C. S. Chen, Y. Fujimoto, G. Girdaukas, C. J. Sih. & J.

Am. 1982. Microbial degradation of the phytosterol side

chain. II. Incorporation of [14C]-NaHCO3 onto the

C-28 position Chem. Soc. 104 : 7294-7298

[21]Y. Lu, X. Zhao. & Z. N. Chen. 1995. A convenient

method for evaluation of the enantiomeric ratio of kinet-

ic resolutions. Tetrahedron: Asymmetry 6 : 1093-1097

[22] Y. Xu, J-H Xu, J. Pan, L. Zhao. & S.L. Zhang. 2004.

Biocatalytic resolution of nitro-substituted phenoxypro-

pylene oxides with Trichosporon loubierii epoxide hy-

drolase and prediction of their enantiopurity variation

with reaction time. Mol. Catal. B: Enzymatic 27:155-159

[23] S Takeji., O.Kenji, Akatsuka H., Kawai E. & Matsumae

H. 2000. Enzymatic resolution of diltiazem intermediate

by Serratia marcescens lipase: molecular mechanism of

lipase secretion and its industrial application J. Mol.

Catal. B: Enzymatic, 10: 141-149

02003-p.6