close

Вход

Забыли?

вход по аккаунту

?

General Aspects and Optimization of Enantioselective Biocatalysis in Organic Solvents The Use of Lipases [New Synthetic Methods (76)].

код для вставкиСкачать
General Aspects and Optimization of Enantioselective Biocatalysis
in Organic Solvents: The Use of Lipases
By Ching-Shih Chen” and Charles J. Sih*
New Synthetic
Methods (76)
Enantioselective biocatalysis in nonaqueous media is becoming increasingly important in
preparative synthetic chemistry. This article discusses (1) the general catalytic properties of
enzymes in nonaqueous environments, ( 2 ) the basic principles that govern lipase-catalyzed
enantioselective esterification and transesterification reactions in organic media for the preparation of optically active acids and alcohols, (3) the determination of kinetic and thennodynamic parameters, and (4) the quantitative analysis of published data.
Table 1. Representative enzyme catalysis in organic media
1. Introduction
Enzymatic reactions are generally conducted in aqueous
media because of the general notion that an aqueous environment is optimal for maintaining the catalytically active
conformations of the enzyme protein for binding and catalysis. In an aqueous medium, the folding of a soluble enzyme
protein is driven by the tendency of hydrophobic amino acid
residues to be excluded from water, so that they are buried
in the interior of the molecule, whereas the charged and
hydrophilic residues are on the surface in contact with the
aqueous solvent. When water is replaced with an apolar solvent, the interior hydrophobic residues tend to disperse, resulting in a reorganization of the enzyme tertiary structure.
This conjecture implies that the conformation of the enzyme
protein is drastically altered and, in the extreme case, the
enzyme pocket may be viewed as “turned inside
in
organic media. Consequently, the catalytic behavior of the
enzyme is expected to be less optimal than in water. This
perception has perhaps impeded the investigation of the
stereoselective properties of enzymes in nonaqueous media
in the past. In more recent years, however, it has become
increasingly apparent that many biocatalytic reactions in
nonaqueous media can proceed with high degrees of enantioselectivity.
It has been known for some time that a variety of enzymes
are catalytically active in apolar media (Table I), and some
of these enzymes have been used for the transformation of
water-insoluble substrates[’] (such as the transesterification
of fats) and for peptide synthesis1261via reversal of the hydrolytic process. However, the enantioselective properties of
the lipases[2-’01 (triglycerol acylhydrolases EC 3.1.1.3), in
particular, have attracted the attention of synthetic organic
chemists because of their synthetic utility. Lipases are
uniquely stable in nonpolar organic solvents and have the
remarkable ability of assuming a variety of conformations to
accomodate substrates of varying sizes and stereochemical
complexities. Moreover, compounds that are unstable in water (anhydrides, halogenated compounds, etc.) could be used
as substrates in nonaqueous media. But more importantly,
lipase-catalyzed esterification reactions in organic solvents
are often more enantioselective than the corresponding hy[*] Prof. C . S. Chen
College of Pharmacy, University of Rhode Island
Kingston, R102881-0809 (USA)
Prof C . J. Sih
School of Pharmacy, University of Wisconsin
Madison, W153706 (USA)
Angew. Chem. lnr. Ed. Engl. 28 (1989) 695-707
Biocatalyst
Solvent
Catalyzed reaction
Ref.
Lipases
(crude powder)
apolar solvents
ester synthesis/
interchange.
peptide synthesis.
macrocyclic lactone
formation
peptide synthesis
[2-201
Proteases
(thermolysin,
subtilisin,
chymotrypsin)
Immobilized
carboxyesterase
Immobilized alcohol
dehydrogenase
Immobilized
polyphenol oxidase
Immobilized
mandelonitrile
lyase
Cholesterol
oxidase
ethyl acetate,
amyl alcohol,
etc.
methyl
propionate
isopropyl
alcohol
chloroform
ethyl acetate
heptane, CCI, ,
butyl acetate
enantioselective acylation
of racemic alcohols
asymmetric reduction of
ketonic substrates
oxidative conversion of
phenols to o-quinones
synthesis of optically
active R-cyanohydrins
[21, 221
[23-25]
t261
[2, 91
[27]
1281
[29]
oxidation of 3P-hydroxy- [30]
steroids
drolytic reactions in water. For these reasons, nonaqueous
biocatalytic technology is becoming increasingly important
as a method for the preparation of homochiral (optically
active) compounds. This article is intended to bring this
methodology to the attention of synthetic chemists by discussing the basic principles and the more recent studies, but
limiting the treatment to the enantioselective properties of
lipases in nonaqueous media. The recent development of
enzyme catalysis in reverse micelle~[~and supercritical flui d ~ [ ~ is
’ ] not covered, and the enzymology and kinetics of
lipases have been discussed e 1 ~ e w h e r e . f ~ ~ ~
2. Enzyme Catalysis in Organic Solvents - General
Considerations
Three types of solvent systems are used in biocatalytic
conversions: water-miscible organic cosolvent systems, anhydrous organic solvents ( < 1 % water), and biphasic systems consisting of water and a water-immiscible organic solvent. The potential advantages of conducting biocatalytic
reactions in these solvent media have been summarized by
many w ~ r k e r ~(1): efficient
~ ~ ~ -catalysis
~ ~ ~ may be achieved
with substrates poorly soluble in water; (2) reactions catalyzed by hydrolytic enzymes may be shifted towards the synthetic direction; (3) undesirable side reactions in organic media as well as substrate and product inhibition may be
0 VCH VerlagsgesellschufimbH, 0-6940 Weinheim, 1989
0570-0833/89j0606-0695S 02.SOjO
695
reduced; (4) recovery of product and biocatalyst from systems may be easier; ( 5 ) enzymes may be more stable in anhydrous organic media.
Water-miscible organic cosolvents such as acetone, methanol, ethanol, dimethylformamide (DMF), and dimethyl sulfoxide (DMSO) are often added to the aqueous reaction media to increase the solubility of substrates that are poorly
soluble in water.[381 Moreover, they have been frequently
used in peptide synthesis to shift the ionic equilibrium. By
lowering the dielectric constant of the medium, the organic
cosolvent diminishes the hydration of ionic groups, thereby
reducing the acidity of the carboxyl group and the basicity of
the a-amino group.[39]However, while lower concentrations
of these polar solvents may not have serious adverse effects
(concentrations of DMSO as high as 20% have been used in
lipase-catalyzed enantioselective hydrolyses of
higher solvent concentrations d o progressively lead to
protein denaturation and the attendant loss of stereochemical specificity. The use of suitable quantities of polar cosolvents may enhance the reaction rate and, in some cases, may
also improve the stereoselectivity of biocatalytic syst e m ~ . [ -431
~ ’ However, this article will concentrate only on
biocatalysis in anhydrous organic solvents o r in biphasic
solvent systems consisting of water and a water-immiscible
organic solvent. Of special interest is the lipase-catalyzed
esterification and transesterification reactions in organic solvents, because the observed enantioselectivity of condensation reactions in organic media is often higher than that of
’‘I
the corresponding hydrolytic reactions in
It is now well known that lipases can accept a broad range
of artificial substrates of varying sizes in both aqueous and
apoIar media, which suggests that the protein backbone of
lipases is flexible and can adopt a variety of conformations
to accommodate substrate molecules. Thus, lipases must
possess low energetic barriers for conformational transitions, a behavior reminiscent of “induced-fit”1441enzymes.
These features make it rather difficult for one to predict the
stereochemical interaction of the enzyme with substrates.
That is, in some cases the enantioselectivity of hydrolysis of
esters in water is higher than the corresponding esterification
reaction o r vice versa. However, when hydrolytic reactions
in aqueous media proceed with poor enantioselectivity, the
corresponding esterification mode in organic solvent should
be examined. In contrast, for “non-induced fit” enzymes,
which possess more rigid conformations (high energetic barriers for the disruption of noncovalent forces), one envisages
that they need not undergo major conformational changes to
assume the catalytically active conformation in nonaqueous
environments. Thus, this class of enzymes is expected to lose
only a modicum of catalytic activity and enantiosekctivity
when water is replaced with apolar media, because the catalytic cavity is retained in nonaqueous media. However, these
enzymes tend to have a more rigid substrate specificity and
enhancement of enantioselectivity in apolar media is not
expected.
The thermodynamic changes during formation of an enzyme-substrate (EnzSu) complex in water and in organic
media have been considered by Kieh~om.[~’]
In aqueous solutions, the polar groups in the catalytic site of the enzyme
and the substrate molecules form hydrogen bonds with water. In the process of enzyme-substrate complexation, some
696
A
I
z
,* (Enz + Prlorg.
‘
L
-
G
(Enz + SU),,
I
IEnzSuI
IEnzPrl
~
Reaction coordinate
Fig. 1. Comparison of free energy profiles in aqueous solutions and organic
solvents for an enzyme reaction in which substrate and product form hydrogen
bonds with solvent water. Enz = enzyme, Su = substrate. Pr = product,
H-bond = hydrogen bond.
of the hydrogen bonds are disrupted and the energy required
for this disruption also contributes to the free energy change
of complex formation (AGaq) (Fig. 1). On the other hand, in
a non-hydrogen-bonding solvent, the EnzSu-forming process does not involve the disruption of hydrogen bonds and
the free energy of EnzSu complex formation[451
is
given by Equation (I),
where nEnzand n,, denote the number of hydrogen bonds
disrupted between water and, respectively, the enzyme active
site and the substrate. Thus, the relationship of the apparent
association constants of the EnzSu complex in water (K,,,.,)
and organic solvent (Ka(,,rg,,)can be expressed by Equation (2).
It is noteworthy that although the energy of hydrogen
bonding is relatively weak (4.5 kcal mol-I), a transition
from an aqueous to an apolar environment may result in an
extremely large apparent association constant for either enzyme-substrate or enzyme-product complex for highly polar
substrate molecules (e.g., polyhydroxy compounds o r their
phosphate derivatives) (Table 2). As a consequence, severe
substrate and/or product inhibition will occur.
Table 2. Effect of hydrogen bonding energy on enzyme-substrate cotnplexation in apolar solvents. Calculated For non-hydrogen-bonding solvents such as
isooctane and hexane.
2000
3.9 x 106
1.7 109
i s x 10’3
3. Lipases and Interfacial Catalysis
There are now about 20 different commercially available
lipases from microbial, plant, and animal sources (Table 3).
Many of these lipases have been purified to homogeneity and
obtained in crystalline form. However, only the crystals of
Angeu.. Chem. Inr. Ed. Engi.2X (1989) 695- 707
Table 3. Source of commercial lipase preparacions.
Source
Suppliers
Alculigmrs sp.
Achromohuctrr sp.
Asprrgi1iu.s niger
5uc;11u.\ .suhtriis
Cundidu cvlindrucru
Amano
Meito Sangyo
Amano
Towa Koso
Sigma. Amano. Meito Sangyo,
Boehringer-Mannheim
Amdno
Sigma. Toyo Jozo
Sigma, Amano
Amano
Amano, NOVO
Rhbne-Poulenc
Tdkeda Ydkuhin
Sigma. Amdno, Boehringer-Mannheim
Amano
Amano
Sigma, Boehringer-Mannheim
Sigma, Boehringer-Mannheim
Sigma, Amano, Tanabe seiyaku
Amano, Nagdse Sangyo, Osaka Saiken Lab.
Am m o
Amano, Serva
Sigma
Cundidu lipoljticu
Cliromohucierium siscosum
Georviclruni cundidum
Humii~nlulunuginosu
M u w r mi(,Jiei
Prnri~illiumcumemherti
Phwriniwes niiens
Porcine pancreas
P.scudomonus urruginosu
P.srudomonus ,f?uorescrns
Psrudomonus sp.
Rhi~opiiau r r h i x s
Rhizopu" ddcmur
Rhixpii.s juponicus
Rhizopws o r w u
Rhrzopus sp.
Wheat gcrm
the lipase of Geotrichurn candidurn have been investigated by
X-ray analysis, which, at 6-A1461and 2.8-Ac4'] resolution,
revealed the presence of nine a-helices and four p-strands.
The helical content was calculated to be about 20 %. The
molecule is ellipsoidal with dimensions of 50 x 50 x 70 A.
Near the molecular center, there exists a large cleft, which is
assumed to be the active site of the enzyme.
Very few studies have been made on the structure and
function of lipases. For pancreatic lipase, serine and histidine residues and a carboxylate group were found to be
essential for activity,f4'] but in the case of C. candidurn, there
seems to be no involvement of a reactive serine residue. On
the other hand, the lipase of Humicola lanuginosa requires
histidine and tyrosine residues and a carboxylate group for
Lipases are characterized by a common catalytic feature the involvement of a lipid-water interface in their catalytic
process!331 This unique feature of interfacial catalysis provides the lipases with an inherent affinity for hydrophobic
environments and distinguishes them from other hydrolytic
enzymes.
The general principles of interfacial enzyme catalysis were
put forward by Verger et a1.1501and may be envisaged by
Scheme 1, where Enz is enzyme in solution, Enz* is enzyme
"penetrated" in the interface. Su represents lipid substrate
molecules at the interface, and Pr represents reaction products.
Scheme 1
Inherent in this model is the presence of two successive
equilibria. The first step describes the reversible binding of
the enzyme to the interface, which is designated as the penetration step and is assumed to confer on the biocatalyst a
Angen. Chon. Inf. Ed. En$ 28 (1989) 695 - 707
new conformation, Enz*. The second step is the reversible
binding of the "penetrated" enzyme, Enz*, with the substrate, Su, to form Enz*Su, which undergoes catalysis to
regenerate Enz* with liberation of products, Pr.
Using this concept, the action of lipolysis of two insoluble
competing e n a n t i ~ m e r s , ' ~A~and
'
B, may be envisioned as
shown in Scheme 2.
Enz*R
Enz*S
Enz*
+P
L' E n P + Q
+
Scheme 2.
This model makes the assumption that the catalytic step is
irreversible, as is the case during initial stages of the reaction.
for two
It differs from the conventional enzyme model
competing substrates only by the presence of the additional
Enz + Enz* interconversion step and is similar to the "C
induced-fit'' model described by F e r ~ h t . [Since
~ ~ ] the surfaceactive properties of the enantiomers are identical, the ratio
k,/k, is a constant and independent of the relative concentration of the enantiomers in the interface. As a consequence,
the penetration step does not participate in the discriminaand the enantioselectivity of irreversible
tion
lipase-catalyzed reactions in water-immiscible organic solvents or in anhydrous organic media also obeys the classical
homocompetitive equation [Eq. (311. k,,,/K, is the apparent
second-order rate constant for the reaction of the enzyme
and the substrate at infinitely low substrate concentrations
to give product(s). k,,, and K, denote catalytic and Michaeiis
constants, respectively, and E is the enantiomeric ratio.f5z1
(3)
This relationship has indeed been independently confirmed
by steady-state kinetics derivation. Moreover, kinetic data of
p h o s p h o l i p a ~ eand
~ ~ l~i ~p a ~ e [ ~appear
']
to obey this model
reasonably well.
4. Stereochemical Specificity of Lipase Catalysis
in Solvents
In organic solvents, lipases catalyze the transfer of acyl
groups from suitable donors to a wide array of acceptors
other than water. Depending on the type of acyl donors and
acceptors, lipase-catalyzed reactions include esterification,
transesterification, amidation, peptide synthesis, and macrocyclic lactone formation. Of these, enantioselective ester synthesisjinterchange is of interest because it provides synthetic
chemists with a facile method for the preparation of optically
active acids and alcohols. Mechanistically, enzymatic ester
synthesis/interchange has some features in common with the
corresponding hydrolytic reaction, but the underlying principles governing the stereochemical specificity are somewhat
different and thus warrant consideration.
697
4.1. Resolution of Racemic Acids via Enantioselective
Ester Synthesis
4
_--
The kinetics of lipase-catalyzed acyl transfer in organic
media may be represented by a ping-pong reaction mechan i ~ m , [as
~ ~depicted
]
in Scheme 3. This synthetic mechanism
is analogous to that of a hydrolytic reaction when RZ = H.
However, ester synthesis/interchange is further complicated
by the reversibility of enzyme catalysis with the participation
of an acyl acceptor (R'OH), whose concentration and species may alter the thermodynamic and/or kinetic functions of
the catalytic system."'] Nevertheless, this synthetic mode of
catalysis in organic media may often yield a higher degree of
enantioselectivity than the corresponding hydrolytic reaction in water. The rationale may be envisaged as follows:
+RCO R l k
- RC0,R' k,
+---
+R'OHk,
k,
RCO-Enz .-R'oH~,
RCO-Enz . R'OH
RCO,R'k ,
+--5 Enz
& ENZ . RC02R2+RCO,R'k,,
k,o
& RCO-Enz . R'OH
Enz +
Enz . RC0,R'
+RCO,R'K,
-
RCO,R'K,
-R'OHk5
+---
* RCO-Enz
+ R10HK6
k,o
Enz . R'C0,R2
+ R'OHk,
-R2OHVs
kA
RCO-Enz . R'OH
RCO,R'K,,
-
+----
Jn
RCO-Enz.R20H
Enz.RC02R2
-
Enz+RC02R2
Reaction coordinate
Fig. 2. Reaction-coordinate diagram illustrating a lipase-catalyzed esteritica
tion reaction.
These equations clearly indicate that the discrimination
between two competing substrates depends on the entire reaction sequence.
For hydrolytic reactions, there will be a partial loss of
specificity due to the high concentration of water solvent
( 5 5 . 5 ~ ) .As a consequence, the second terms in Equations (6) and (77, contributed mainly by the steps following
the first irreversible step, can be neglected. Thus, Equations (8) and (9) are obtained.
+R'OHk,
+
RCO-Enz
RCO-Enz . R'OH
Enz _1_) Enz . RC0,R'
- R'OH k ,
L"
+ R'C0,R2k;,
Enz
Scheme 3
During the initial stages of the reaction, the complex catalytic sequence shown in Scheme 1 may be reduced to a minimal reaction mechanism consisting of the presence of two
bond-breaking and two bond-forming steps (Scheme 4),
Enz + RCO,R'
RCO-Enz
A
Enz + RC0,R'
B
-
-R'OHLB RCO-Enz
+R'OHkb
+R*OHkb
Enz + RC0,R2
P
Enz + RC0,R2
Q
Scheme 4.
where k,, k,, Ka, and Kb are net rate constants.r551Kinetically, k, and k', are equivalent to k,,,lK, for A and B, respectively; k, and K, represent the gross binding and catalytic performance of each diastereomeric acyl-enzyme complex toward
the achiral nucleophile (R'OH). The implication of these net
rate constants can be further illustrated by the free energy
diagram in Figure2, where k, and k , are given by Equations4 and 5, respectively. T, and T, refer to transition
states.
k, = exp(-AG,,/RT)
(4)
kb = exp(-AG,,/RT)
(5)
If the reaction rate of substrate A is uA and that of the
competing substrate B is us, then application of the net rate
constant method of CZeZar~d[~~'
yields Equations (6) and (7),
respectively.
Combining Equations (8) and (9) affords the familiar
Equation (10),1521
which indicates that in hydrolytic reac-
tions, the enantioselectivity is solely contributed by the catalytic sequence leading to and including the first irreversible
step. This difference explains why, in some instances, the
enantiomeric purity can often be enhanced by conducting
biocatalytic ester synthesis/interchange in organic solvents
in lieu of the corresponding hydrolytic reaction in water,I1O.18. 571
One should keep in mind that in enzyme-catalyzed condensation reactions, the gradual accumulation of products
(R'OH and RC0,R2) during the reaction will give rise to
reverse catalysis. Thus, the equilibrium constant ( K ) plays a
pivotal role not only in the expression of enzymatic enantioselectivity, but also in the maximal obtainable chemical
conversion [c = 1/(1 + K ) ] . To envisage the progressive
changes of the stereochemical behavior of enzymes, the reaction scheme is further simplified to Equations (1 1 ) and (12),
where k , , k , , k , , and k , are the rate constants for the forward and backward reactions. In an achiral environment,
the equilibrium constants for the pair of enantiomers should
be equal [Eq. (1 3)].
K = k2/k, = k,/k3
698
(13)
Angew. Chem. Inf. Ed. Engl. 28
(1989) 695-707
This relationship reveals that the sense of preferred chirality for the forward and reverse reactions is retained; i.e., if
k, > k , , then k , > k,.
For reversible biocatalytic systems, assuming that the acyl
acceptor is kept a t saturating levels, a new expression, which
incorporates the thermodynamic parameter K , is required
for the calculation of the enantiomeric ratio, E [Eq. (14)],
where E = k , / k , .
Enz +R'C02R
Reaction coordinate
When k, and k , = 0, Equation (14) is reduced to the homocompetitive Equation (3) for the irreversible case.1521Alternatively, Equation (14) may be transformed into Equations (15 ) and (16) to relate the extent of conversion (c) and
the enantiomeric excess of substrate (ees)and product (ee,).
+
In[l - ( 1
In11 - (1
Fig. 4. Reaction-coordinate diagram illustrating an enzymatic resolution o f a
racemic alcohol via enantiospecific esterification. ----, enantiomer 1 ; --~,
enantiomer 2.
k'
In[t - ( t + K ) ( c + e e , { t - c } ) ]
= E
In11 - (t K ) ( c - ee,{l - c})]
k
- R'C0,R k ,
- t---
& Enz . R ' C 0 , R
+ R'C0,R
k4
R'COIRZ
t.--
--+
+ R'OH k',
R'CO-Enz
k'
These equations show that the enantioselectivity of enzyme-catalyzed synthesis depends on the complex interaction of both kinetic ( E ) and thermodynamic ( K ) functions.
The computer-generated graphstLo1of Equations (15 ) and
(16) (Fig. 3) show that the optical yield of both fractions is
-
-
40
60
80 100
0
20
40
60
80
100
Yield [%I
YieLd 1%1
Fig. 3 . Expression of the percentage enantiomeric excess of product (A) and
remaining substrate fractions (B) as a function of the percentage conversion for
E = 100 and K = 0 (a), 0.1 (b), 0.5(c), 1.0 (d). and 5.0 (e) [lo]. These curves
were computer generated from Equations (13) and (14). [Reprinted with permission from J. Am. Chem. Soc., 109 (1987) 2812. Copyright 1987 American
Chemical Society.]
0
20
inversely related to the magnitude of K. In addition, high
optical purity of the remaining substrate (ees 2 99 YO)
cannot
be attained by extending the conversion much beyond 50 YO
even with a very small K value. These graphic representations allow one to predict precisely when to stop the kinetic
resolution t o maximize chemical and optical yields after the
values of E and K have been defined.
' Enz
k,
R'OH
w
Enz
ROH k,
~
R'CO-Enz . ROH
+ K ) c ( 1 + eep)]= E
+ K) c( l - ee,)]
-
R'CO-Enz . R O H
- R'OH
k
i&
k;
- R'C0,R'k;
Enz . R ' C 0 , R
+---
' Enz
+ R L C O 2 RV6
k,
Scheme 5
tiomer may interact differently with the acyl-enzyme (R'COEnz) to give different transition states (Fig. 4), and the size of
this activation energy difference (AAG,) will dictate the
enantioselectivity of biocatalysis. Of the factors that influence the stereochemical properties of the biocatalytic system,
the acyl donor is of special importance, and its role can be
summarized as follows: (1) Because the kinetic differentiation between two enantiomers is independent of the transfer
of the acyl group to the enzyme, the type of acyl donor, either
acid or ester, does not affect discrimination or the E value.
(2) Since the acyl group of the acyl-enzyme exerts steric
and/or stereoelectronic effects on the deacylation process, it
is possible, in principle, to achieve high enantioselective esterification by the selection of a suitable acyl donor. ( 3 ) Since
the reaction is reversible, the concentration and the type of
acyl donor may alter the equilibrium constant, which in turn
influences the biocatalytic stereochemical behavior.
To facilitate the kinetic treatment, the mechanism outlined
in Scheme 5 may be simplified to Scheme 6.
Enz
Enz
+ R'C02R2
RzH11
+
+ R'C0,R
P
k,
k,
R'CO-Enz
Enz
+ R'C0,R
Q
4.2. Resolution of Racemic Alcohols via Enantioselective
Ester Synthesis
Scheme 6.
In Scheme 5, ROH (A) and R'OH (B) represent two competing enantiomers. The nucleophilic group of each enan-
The rate of disappearance of two enantiomers A and B can
be described by Equations (1 7) and (1 8).
Angen. C'hcm. Inr. Ed. EngI.28 (1989) 695-707
699
uA = k , [A] [RCO-Enz] - k , [PI [Enz]
oB = k , [B] [RCO-Enz] - k , [Q] [Enz]
These equations show that the relative concentration of
acyl-enzyme and free enzyme affects both reaction rates. At
saturating levels of acyl donor, the ratio of the acyl-enzyme
to free enzyme remains constant. The combination of Equations (17) and (18) and the integration of the resulting equation lead to the same format as Equations (15) and (16).
However, at low concentrations of the acyl donor, the
amounts of acyl-enzyme and free enzyme vary as the reaction proceeds.'561 As a result, the stereochemical behavior of
the system becomes more complicated and cannot be predicted by Equations ( 3 5) and (16). Practically speaking, it
would be desirable to keep all the free enzyme in the acylated
form so that the reverse reaction is negligible.
5. Experimental Considerations
5.1. Selection of Organic Solvent
The type of organic solvent will have profound effects on
the reaction kinetics and stability of the biocatalyst. In general, the catalytic efficiency of enzymes diminishes as the
polarity of the solvent
Hydrophilic solvents
may denature enzymes by penetrating into the hydrophobic
core of proteins, thus disrupting their delicate functional
s t r u c t ~ r e sor
, ~by
~~
~
stripping
off the essential water from the
enzyme.[591Recently, some general empirical rules have been
formulated for the optimization of biocatalytic activity in
different organic solvents. The parameter P, the partition
coefficient of the solvent between octanol and water,r601was
introduced as a quantitative measure of solvent polarity[6t.621(Table 4). Generally, the catalytic activity is low in
polar solvents having a log P < 2, is moderate in solvents
having a log P between 2 and 4, and is high in apolar solvents
having a log P > 4. However, the instability of biocatalysts
Table 4. The log P values of some commonly used solvents [a]
Solvent
log P
dodecane
octane
heptane
hexane
cyclohexane
"'1
benzene
chloroform
butyl acetate
diethyl ether
butanol
pyridine
ethyl acetate
tetrahydrofuran
acetone
ethanol
acetonitrile
methanol
diemethylformamide
dioxane
dimethyl sulfoxide
700
Comments
6.6\
3.5
3.2
Apolar solvents. suitable for reactions in which dry enzyme powders
are used
'
2.0
2.0
1.7
0.85
0.80
0.71
0.68
0.49
-0.23
-0.24
-0.33
-0.76
- 1.0
-1.1
-1.3 I
in polar solvents can often be partially overcome by immobilizing the enzymes on hydrophilic supports.I6'I (A detailed
account of this "medium-engineering" of biocatalytic systems is given in Ref. [61].)
Another consideration is the solubility of substrate and
product in the media.[63] Enzymes are catalysts and only
serve to accelerate the attainment of chemical equilibrium.
Therefore, the physical properties of the apolar environments may affect the position of equilibrium and thus the
chemical yield of the enzyme reaction. It is conceivable that
for lipase catalysis, owing to the unique interfacial catalytic
property, one would choose a solvent system in which the
substrate can form micelles or emulsions whereas the product is fully dispersed in solution. As a rule, apolar solvents
such as hexane, isooctane, toluene, and cyclohexane are
commonly used for lipase-mediated reactions. Also, halogenated
such as trichlorotrifluoroethane,
and tert-butyl and diisopropyl ethers
have been successfully employed. However, when the substrates are highly
hydrophilic, polar solvents such as DMF, pyridine, and butanol may be used at the expense of enzyme stability or
immobilized enzymes may be employed.[611
5.2. Water Content of Organic Media
Esterification and transesterification reactions may be
carried out in either monophasic, nearly anhydrous organic
media (water content < 1 %) or in aqueous-organic biphasic
systems. In the latter, the enzyme resides in the aqueous
phase, surrounded by an organic phase. Since enzymes are
insoluble in hydrophobic organic solvents, they form suspensions. Hence, unless the reaction mixture is stirred or shaken,
the enzyme settles. Although published experimental procedures give no clear indication as to the optimal amount of
exogenous water to be added to the organic solvent, there is
a consensus that some water is absolutely needed for the
catalytic function of the enzyme.[661This is because water is
involved in various types of noncovalent bonding interactions for the maintenance of catalytically active conformations of enzymes. For some esterification reactions, the immiscible organic solvent is saturated with water, because the
rate of reaction appears to increase as the water content
On the other hand, anhydrous organic media
(no addition of exogenous water) are preferred for lactoniza(intramolecular esterification). In this case, the reaction rate is relatively slow and higher temperature (e.g.+
65 "C) is used.
5.3. Choice of Acyl Donors and Acyl Acceptors
Moderately to highly polar solvents; may reduce enzyme's catalytic efficiency; enzyme immobilization is recommended for
optimal activity.
Another consideration is the choice of substrates; i.e., either chiral racemic acids or the corresponding chirai racemic
esters could be used as substrates. Mechanistically, these two
types of reactions proceed via different transition states for
all steps leading to and including the formation of the acylenzyme intermediate (Scheme 1). Hence, despite the apparent similarity, the E and Kvalues for esterification and ester
interchange may be different. Also, the physical state of the
substrate has a pronounced effect on the enzymatic rate. For
Angcw. Chem. lnr. Ed. Engl. 28 (1989) 695- 707
instance, in an organic medium, the ester substrate is fully
dispersed and is less susceptible to enzymatic lipolysis due to
a lack of lipid-water interface. On the other hand, the acid
substrate forms micelles or emulsions, which are readily attacked by the lipase. However, if the acid is insoluble in the
organic medium, the reaction rate is markedly reduced due
to the sluggish enzymatic action on a solid substrate, but the
physical separation of the substrate from the product is often
much easier for the acid substrate than for the ester substrate.
To achieve high enantioselective esterification, it is imperative to minimize the extent of the reverse reaction. Several
strategies may be used to circumvent the thermodynamic
restraint imposed by equilibrium. These include: the use of
enolester~["~(R'OH = aldehyde or acetone) or anhydride"'] as acyl donors so that no alcohol or water molecules are formed; the selection of a suitable acyl acceptor
( R 2 0 H = secondary alcohol) to prevent the backward reaction;["' the removal of R'OH to shift the equilibrium position (e.g., by adding molecular sieve to remove water in
esterification reactions); and the addition of a large excess of
acyl acceptor to the medium to reduce the relative concentration of the competing nucleophile, R'OH['O] (Scheme 3).
Table 5. Reversible enantioselective esterification of chiral racemic acids 1 cdtalyzed by C. cylindracru lipase with 1-butanol as acyl acceptor in hexane (containing 0.1 % H,O) [7]. e? = enantiomeric excess, E = enantioselectivity.
Br
Br
+ nBuOH
The E value can then be readily calculated using either
Equation (20) or (21).[521
E=
E=
In[(l - c)(l - ees)]
In[(1 - c)(l ee,)]
+
In[1
-
In11
-
c ( l + eep)]
c ( l - eep)]
Moreover, by inserting the values of E, c, ee,, and eep
determined at a conversion greater than 40% into Equations (I 5) and (I 6), the value of K can also be determined.
The need for a quantitative treatment of biocatalytic resolution data in organic media may be best illustrated by an
analysis of some published results (Tables 5 and 6). In the
A n g m . Clirm. In!. Ed. Ennl. 28 (1989) 695-707
RACO,H
R
(*)-la
Substrate
Conversion
f
[hl
WI
19
48
40
168
30
67
45
65
R
l a n-C,H,
I b p-CIC,H,
S
ee ["h]
Product
Substrate
( Rester)
( S acid)
99
300
3
17
4
-
62
~
79
-
65
~
E
[dl
[a] Calculated using either Equation (20) or (21).
Table 6. Reversible enantioselective esterification of chiral racernic alcohol 2
catalyzed by C. cylindrucra lipase with dodecanolc acid as acyl donor in hexane
or heptane [61[a].
5.4. Determination of E and K Values
The enantioselectivity of biocatalytic reactions is normally
expressed as the E value,[521a biochemical constant that is
independent of substrate concentration and the extent of
conversion. Lipase-catalyzed esterification or transesterification reactions in aqueous-organic biphasic or anhydrous
organic media also obey the conventional homocompetitive
equation provided that the reaction is irreversible (saturating
amounts of achiral acyl donor and acceptors are present). At
the initial stages of the reaction, little product has accumulated. Normally, reversibility of the reaction becomes a significant problem only at conversions around 40% or greater
(Fig. 3 ) . Consequently, if the enantiomeric excess of the substrate (re,) and the product (eep)is determined at low conversions (< 20%), one can calculate the extent of conversion (c)
using Equation (19).
+
C0,nBu
lipase
OH
>
+ CH,(CH,),,CO,H
["/.I
T
["Cl
I
Conversion
[hl
["/I
Product
(ester)
Substrate
(alcohol)
45
40
8
88
45
57
95
69
88
EC
E
-
92
17
[a] Ref. 161 does not indicate which solvent was used.
lipase-catalyzed enantioselective esterification of ( f)-2-bromohexanoic acid 1 a and (_+)-2-p-chlorophenoxy)propionic
acid 1b,"] I-butanol was used as the acyl acceptor. It is
noteworthy that even though the reaction was conducted in
hexane with only a very small amount of exogenous water
(0.1 %), the reactions are clearly reversible. Using Equations (20) and (21), the E values were calculated from the
data of Table 5. As can be seen, the E value decreased from
300 (30% conversion) to 3 (67% conversion) for 1 a and
from 17 (45% conversion) to 4 (65% conversion) for 1 b. If
the reactions were irreversible, the E values should remain
constant throughout the reaction. A similar observation was
noted during the enantioselective esterification of ( L- )-menthol 1 using dodecanoic acid as the acyl donor.[61Again, the
E value decreased from a value of 92 (45 YOconversion) to 17
(57 % conversion). Therefore, when the reactions are reversible, Equations (15) and (16) should be used instead of Equations (20) and (21) for the calculation of the E value to take
into account the contribution of the equilibrium constant, K.
701
5.5. Determination of the Kinetic Parameters V,,, and K,,,
Frequently, it is of interest to know which of the kinetic
parameters, the Michaelis constant, K,, o r the catalytic constant, V,,, is altered when a substituent is introduced onto
the substrate molecule. As mentioned earlier, lipase-catalyzed reactions occur at the oil-water or micelle-water interface.[331Since the lipase is reversibly adsorbed at the interFace, the experimentally determined “apparent K,” may represent the dissociation constant of the interfacial enzymesubstrate complex, termed interfacial K,*. Also, the “apparent Y,,” measured at saturating amounts of lipid substrate
is also an apparent quantity, namely, the real V,,, multiplied
by the factor Su/(K,* Su). This factor tells us what fraction
of the enzyme molecules are in the Enz*Su form.
To overcome the complexity of this interfacial surface
problem, Hirohuru et a1.[671selected experimental conditions
under which all enzyme molecules are adsorbed to the interface by increasing the area of interface with vigorous stirring
(I 000 rpm) and increasing the amount of substrate under
these conditions. They assumed that the concentration of the
enzyme in solution is that at the interface. Lineweaver-Burk
plots with substrate concentration expressed in mole Linstead of mz L- gave straight lines to allow the calculation
of K , and V,,,. In any event, accurate noncompetitive measurement of K , and Vmaxis a difficult problem, especially the
K, for the more slowly reacting enantiomer.
In theory, the kinetic parameters, K , and V,,, for both
enantiomers, A and B, can also be calculated directly from
competitive measurements using Equations (22) and (23),[681
+
where [A], and [B], denote initial concentrations, E is
the enantiomeric ratio, x = ([B],/[A]A/E), y = ([A],/[B]E),
t = time, and (Km)A,(K&, ( Vmax)Ar
and ( Vmax)Rare the Michaelis constants and maximal velocities for A and B, respectively. By inserting the E value and at least three sets of
experimental data [ ( t l ,[A],), ( t z ,[A],), ( t 3 ,[AI3) and
( t , , P I , ) , (f,,P I 2 ) , ( t 3 , lB13)1 into Equations (22) and (23)respectively, the values of Vm,,/i”C,, K,? and V,, for each
enantiomer A and B may be readily calculated.
6. Asymmetric Catalysis in Organic Solvents
Because lipase-catalyzed reactions are generally conducted in the hydrolytic mode in
it is not too surprising
to find that there are relatively few published examples of
enantioselective and enantiotopically selective esterification
reactions in nonaqueous media. Nevertheless, the data presented in Tables 7-11 should provide chemists with a
glimpse of the usefulness of this unconventional methodology. Although Equations (15) and (16) should be employed
for the calculation of the E values, the equilibrium constants
for these systems were unfortunately not determined. Therefore, the Evalues listed in Tables 7,9, and 11 were calculated
702
using Equations (20) and (21), and they represent apparent
minimal values rather than intrinsic E values.
6.1. Optically Active Carboxylic Acids and Esters
Most of the biocatalytic kinetic resolutions of chiral racemic acids were conducted in the hydrolytic mode in water,1691
because the hydrolytic rate in water is generally Faster than
the corresponding rate of esterification in organic media.
That is, if the hydrolytic rate in water is slow, the corresponding rate of esterification in organic media in most cases
is expected to proceed at an even slower rate. Hence, more
enzyme is required for esterification in nonaqueous media.
The catalytic activity of enzymes is strongly influenced by
both electronic and steric effects. For example, acids that
possess electron-donating or bulky substituents (e.g., isopropyl or larger) adjacent to the carboxyl function are attacked
slowly or not at all by lipases and the rate-determining segment of the reaction is the formation of the acyl-enzyme
complex. In these cases, transesterification may be used as a
means of improving the catalytic rate; the chiral racemic acid
is first converted nonenzymatically into an activated ester
(e.g., chloroethyl ester), which is more susceptible to nucleophilic attack by the biocatalyst. Using the lipase of Cundidu
cylindracea, Klibanoi)et al.[’] observed good catalytic activity
and enantioselectivity with substrates that possess an electron-withdrawing group adjacent to the carboxyl function,
such as 1 a and 1b, whereas low catalytic activity was noted
with other substrates devoid of an electron-withdrawing substituent.
Table 7 compares the enantioselectivity of esterification
versus hydrolysis of 2-(p-~hlorophenoxy)propionates.In
both sets of experiments, the observed enantioselectivity ( E
value) of esterification in organic media is four- to fivefold
higher than the corresponding hydrolytic reaction in water.
Similarly, the data in Table 8 show that the enantiomeric
excess (ee) of the product, (S)-3-methylglutarate monomethyl ester, is higher when the product is prepared by enzymatic acylation by the corresponding anhydride in organic
solvent[” than by enzymatic hydrolysis of the diester in
water.
As far as the influence of the acyl acceptor on enantioselectivity is concerned, no prediction may be made with any
degree of confidence at this stage. Although one might surmise that the enantioselectivity may perhaps be improved by
the use of a chiral alcohol, in many cases the observed enantioselectivity using achiral alcohols as acyl acceptors is superior to that of chiral alcohols.[771However, the matching of
appropriate racemic acids with racemic alcohols in a double
kinetic resolution experiment is an exciting area that warrants systematic exploration in the future.
The existence of reverse hydrolysis during the lipase-mediated enantioselective formation of 2-(p-chlorophenoxy)propionic esters”’] 1 c, d is illustrated in Figure 5 a . The Cundidu lipase preferentially attacked the ( R ) enantiomer in the
( R )- 1 C, R = ~ B u
= cyclohexyl
( R )- 1 d, R
Angert.. Chem. Inl. Ed. Engl. 28 (1989) 695-707
'"1
Table 7. Lipase-catalyzed enantioselective esterification of the acid ( i ) - lb in
organicsolvent versusenantioselective hydrolysis in water. Conditions: 24 h for
butanol and 105 h for cyclohexanol. 25° C isooctane as solvent.
I
Condidu-
+ ROH
D o A C o z H
lipase
+ H,O
JJ,O"CO,R
!
a1
IRI-lc
60-
ee
I %1
CI
CI
/.
20Synthesis
Stereopreference
R
R
R
nBu
cyclohexyl
lc
Id
~~~
Conversion
€
["/.I
ecp
[%]
[Oh]
[a1
0.12
0.25
0.81
0.97
13
20
0
84
ees
eep
I"/.]
[%I
Conversion
[Oh]
E
[a1
0.24
0.94
0.20
0.73
55
56
2
22
res
0
1
I
0
20
40
Yield
0
20
#
60
80
-
100
60
-
100
I/'[
~
Hydrolysis Srereopreference
R
lc
Id
R
R
nBu
cyclohexyl
[a] Calculated using Equations (20) and (21).
Table 8. Lipase-catalyzed enantiotopically selectiveacylation by the anhydride
3 in organic solvent versus hydrolysis of the diester 4 in water. Conditions: 6 h.
25 "C. diisopropyl ether as solvent [18a]
),,
lipase from
Pseudomonas-
species
+ ROH
>
x1
(
HO,C
3
lipase from
RO,C
Pseudomonas-
species
+H20
CO,R
'
RO,C
Acylation
R
Stereopreference
ee
[%]
Conversion
CH,
nBu
87
91
92
74
pro-R
pro-R
[%]
80
CO,H
S
4
[%I
Fig. 5. Enantiospecific synthesis of the 2-(p-chlorophenoxy)propionic esters 1 c
and I d from the acids ( I ) - l b in isooctane catalyzed by Cundidu lipase [lo].
+ = experimentally derived values. The curves depicting the relationship between enantiomeric excess and conversion were computer generated from
Equations (15) and (16) with use of the apparent constants E and K . a) Formation of the butyl ester l c, E = 12, K = 0.15; b) formation of the cyclohexylester
Id, E = 84, K = 0.02. [Reprinted with permission from J. Am. Chem. Sac. 109
(1987) 2812. Copyright 1987 American Chemical Society.]
CO,R
R
A
do
Yield
Hydrolysis
Stereoee
preference [Oh]
[%]
pro-R
50
-
74
-
Conversion
-
presence of an excess amount of 1-butanol as the acyl acceptor. The E value for this system was moderate ( E = 12) and
the K value was found to be 0.15. As shown in Figure 5 a, the
characteristic feature of such a reversible system is that the
optical purity of the remaining substrate (ees) decreases as
the conversion is extended beyond 50%. This is in marked
contrast to the irreversible case where ee, increases as the
conversion is extended. Figure 5 b illustrates the advantage
of using cyclohexanol as the acyl acceptor, because esters of
secondary alcohols are more resistant to enzymatic hydrolysis than those of primary alcohols. The K value for this
system is reduced to 0.02.
6.2. Optically Active Alcohols
With the exception of certain sterically hindered alcohols,
lipases catalyze the asymmetric acylation of a wide range
Angew. Chem. Int. Ed. Engl. 28 (1989) 695- 707
of cyclic and acyclic substrates with moderate to high
enantioselectivity (Tables 9 and 10). To achieve highly enantioselective acylation, the selection of a suitable biocatalyst and the optimization of reaction conditions are important.
The acyl donor can have a marked effect on the stereochemical behavior of the enzyme system. This is illustrated by the examples of enantioselective acylation of
racemic 1-phenethanol (i-)-9 mediated by Candida cyfindracea Iipa~eI'~]
(Table 11).
While the physical states of the acyl donor, either as an
ester in a fully dispersed form or as an acid in a micelle form,
had no effect on the enantioselectivity, different catalytic
rates were observed. For example, the rate of the reaction
using tributyrin (tributyrylglycerin) as the acyl donor was
only about one-fifth of that of butyric acid. On the other
hand, increasing the chain length of the acyl group from C ,
to C,, led to higher values of enantiomeric ratio E, but no
significant difference in enantioselectivity was noted for acyl
donors with chain lengths between C,, and C16. Similar
observations were reported for the Mucor miehei lipase,
which was used for the resolution of (f)-2-octanol via enantioselective acylation.r'61 In this case, the E value increased
to > 50 with chain lengths of alkanoic acid up to C,; it
declined beyond C, but only to rise again to > 50 for the C,,
acid. Therefore, it is noteworthy that the optimal chain
703
Table 9. Lipase-catalyzed enantioselective acylation of alcohols 5- 17 and 2
I
Table 10. Lipase-catalyzed enantiotopicdlly selective acylation of alcohols
18-22.
HO
5
OH
2
7
6
18a, R = CH(CH )
18b, R = (CH,),dd=CH,
18c, R = C,H
18d, R = cyclo%exyl
18e, R = OCH,C,H,
OH
HO
20
19
8
10
9
OH
21
/-l
0
\ /
22a,X=O
0
f'l0
22b,X=O
\ /
22c, X = H,>,Cl
11
12
Substrate
13a, R = C,H,
13b, R = CH,
Stereopreference
ee
["/.]
Ref.
PI
PI
-
65
90
92
58
96
isopropenyl acetate 53
96
[lY]
48
84
95
[14]
I201
57
74
79
26
68
84
[14]
'[I41
1141
Acyl donor
Conversion
WI
P1 fraction from porcine pancreatic lipase
18a
18 b
18c
14
15
Substrate Stereo[a1
preference
Acyl donor
16
Conee,
version
["/.I
17
rh]
E
[h]
eep
[Oh]
Ref.
18d
19
pro-R
pro-R
pro-R
pro-R
pro-R
methyl acetate
methyl acetate
methyl acetate
methyl acetate
methyl acetate
91
70
98
90
191
I91
PI
PseudornonaJ sp. lipase
18e
pro-S
Crude porcine pancreatic lipase
Cundidu cylindruceu lipase
(+)-5
(+)-6
(+)-7
(*)-2
(k1-8
(kj-9
1R.2R
1R.2S
1R,2R
1R,2S,5R
R
R
dodecanoic acid
dodecanoic acid
dodecanoic acid
dodecanoic acid
dodecanoic acid
dodecanoic acid
44
34
30
45
31
35
octanoic acid
octanoic acid
octanoic acid
octanoic acid
80
32
50
98
-100
88
95
72
93
>lo0
>lo0
23
92
8
45
[6]
[6]
161
[6]
[lZI
[67]
51
49
44
45
87
83
67
71
83
87
86
87
30
38
27
31
[16]
1161
[16]
[16]
41
51
30
65
295
66
295
90
88
78
>lo0
22
[17]
[17]
[15]
52
>97
91
>lo0
[13]
-
>95
-
>95
>95
-100
>lo0
-
[78]
[78]
1791
-
20
21
pro-S
pro-S
22 a
22 b
22 c
pro-S
pro-S
-
methyl acetate
trichloroethyl
acetate
methyl acetate
methyl acetate
methyl acetate
-
Mucor miehei lipase
(+)-lo
(+)-I1
(+)-I2
(kj-9
R
R
R
R
Crude porcine pancreatic lipase (PPL)
(+)-13a
(+)-I4
R
R
R
(?)-I4
R
(*)-I5
S
R
R
(f)-13b
(+)-I6
(*)-I7
ethyl acetate
ethyl acetate
trichloroethyl
hutyrate
trifluoroethyl
dodecanoate
ethyl acetate
ethyl acetate
trichloroethyl
butyrate
49
&3
-
>lo0
[a] The enantiomers shown react preferentially. [h] Calculated using Equations (20) and (21).
length of the acyl donor required for maximal enantioselection varies with the enzyme and the substrate.
The presence of competing enzymes in the crude lipase
preparations should also be considered. In general, lipases from microbial sources are produced extracellularly and
are virtually homogeneous in terms of lipolytic
In contrast, crude mammalian and plant lipase preparations
contain several other interfering enzymes, including proteases and esterases, which may possess opposite or poor stereo704
Table 11. Enantioselective acylation of (+)-l-phenethanol(+)-9 catalyzed by,
C . qlmdruceu lipase.
0
$
+
Cundidu-
lipase
RCO,R'
OAR
a-+
<
ROH
(k)-9
Acyl donor
I
Conversion
[hl
[%I
butyric acid
66
tributyrin (trihutyrylglycerinj 100
octanoic acid
66
tricapyrin (trioctanoylglycerin) 100
dodecanoic acid
66
hexadecanoic acid
66
5-phenylvaleric acid
100
39
10
20
< 5
19
14
< 5
ee,
[%I
ee,
[%I
E
[a]
54
10
24
84
89
94
20
19
41
-
-
-
22
15
95
95
46
45
-
-
-
[a1 Calculated using Equations (20) and (21).
selectivity as compared with the lipase. As a consequence,
low optical yields or unreproducible results are often encountered in using these crude lipases, depending on the state
of the enzyme preparations. Therefore, to enhance the optical yield, a number of measures can be taken. These include:
Angen. Chem. Inl. Ed. Engl. 28 (1989) 695-707
treatment of the crude enzyme preparation with chemical
reagents'"] (e.g., serine protease inhibitors) to inactivate interfering enzymes; treatment of the crude enzyme preparation by physical means (e.g., partial protein purification;
lyophilization) to remove or selectively inactivate competing
enzymes; and selection of a suitable acyl donor which serves
as a poorer substrate for interfering enzymes and/or as a
more efficient substrate for the target enzyme. These strategies of optical purity enhancement are illustrated by a number of recent examples where crude porcine pancreatic lipase
(PPL) was used.
PPL exhibited a moderate enantioselectivity in the esterification of ( i)-sulcatol ( k)-14."31
When the commercial Sigma crude PPL preparation was dehydrated to constant
weight under vacuum, a threefold increase in the E value was
noted. Although a general increase in the chain length of the
acyl moiety had little effect on the enzyme specificity, the use
of an activated ester, trifluoroethyl dodecanoate resulted in
a fourfold increase in the E value. A combination of enzyme
dehydration and ester selection gave rise to a nearly tenfold
increase in E value.
0
Some of the asymmetric esterification reactions in organic
solvents were actually mediated by the competing enzymes in
the crude lipase preparation. A recent study by Ramos Tomho et al.17'] has shown that an esterase fraction from crude
PPL catalyzed highly enantioselective acylation of 2-substituted 1.3-propanediol, whereas pure PPL was virtually inactive. In addition, several hydrolytic enzymes, including
cholesterol esterase and chymotrypsin, are present in crude
PPL preparations. For example, the claim[2z1that crude
porcine pancreatic lipase (PPL) catalyzes the formation of
peptide bonds may well be due to the contaminating chymotrypsin present in the preparation. Therefore, partial purification of enzyme preparations may resht in more enantioselective biocatalysts. Finally, it is interesting to note the
existence of pancreatic colipase whose main function appears to be to restore the activity of the bile-salt-inhibited
lipase. However, it is unclear as to whether colipase can alter
the Vmaxand K, of the pancreatic lipase.[801
6.3. Macrocyclic Lactones
Lipases have been used for the preparative syntheses of
macrocyclie lactones via the lactonization of hydroxy acids
and esters in organic solvents.[23*241 However, more recent
show that the product profile varies with the
length of the hydroxy acid and the lipase and is considerably
more complex than those reported previ~usly.''~,241 For example. when 10-hydroxydecanoic acid was exposed to a variety of lipases in anhydrous isooctane, no decanolide was
detected; instead, a complex mixture of di-, tri-, tetra-, and
A n g w . Chem Ini. Ed. Engl 28 ( 1 9 x 9 ~695- 707
pentalactones was formed in different ratios depending on
the lipase used. A similar product profile was obtained using
methyl 10-hydroxydecanoate as substrate. On the other
hand, hexadecanolide was found to be the dominant product
after exposure of 16-hydroxyhexadecanoic acid to the lipases, except the lipase of Mucor rnkhei which formed hexadecanodiolide predominantly. Chiral (a-1)-hydroxy acids (C,
and C,,) were also lactonized by the lipases to yield a mixture mainly of di- and trilactones, but no monolactones were
detected (Table 12). The enantioselectivity of the intramolecular esterification was examined using a series of acyclic
diastereomeric dimers, prepared from (R)- and (8-7-hydroxyoctanoic
The relative rates of diolide formation
were found to follow the order R , R > R , S > S , S , and no
cyclization was noted with the (S,R)diastereomer. These results indicate that the stereochemistry of the hydroxyl at both
chiral centers appears to influence the hctonization reacti~n.'~~]
Porcine pancreatic lipase (PPL) in anhydrous organic
solvents such as dry ether catalyzed the lactonization of a
number of esters of y-hydroxy acids 25 in nearly quantitative
yields.[7s1 This highly enantioselective process was used
for the synthesis of (S)-(
- )-y-methylbutyrolactone 26
(R' = CH,; R 2 = H; E = > IOO), (R)-(t)-methylbutyrolactone 26 (R'= H, RZ = CH,), and optically active y-phenylbutyrolactone (the presence of the ( R )or (8form was not
established).
25
26
It was found that a-substituted y-hydroxy esters were very
poor substrates for the enzyme. The initial rates of lactonization of a-bromo-y-hydroxybutyrate and of a-methyl-y-hydroxybutyrate were, respectively, 200 and 50 times lower
than that of y-hydroxybutyrate, It is interesting to note that
the PPL-catalyzed hydrolyses of y-, 6-, and s-chiral racemic
lactones in
proceeded with only modest enantioselectivity; E values ranged from 3 to 20.
Lipases also catalyzed the formation of macrocyclic lactones via the condensation of diacids and diols in anhydrous
nonpolar organic solvents such as i s o o ~ t a n e The
. ~ ~major
~~
products of this reaction were mono- and dilactones accompanied by linear oligomeric esters and a small quantity of
trilactones in some systems. The yield of the Iactone products
varied not only with different substrates, but also with reaction conditions. Optimal temperature for lactonization was
between 55 and 75 " C ; at lower temperatures (< 45 "C), oligomeric esters were the dominant products. Lipases accommodated acyclic diacids 27 (m = 2- 12) and diols 28 (n = 516) of various sizes. However, the optimum yield of lactones
was achieved when the size of the ring is in the range of
24-28.
Cyclic diacids and diols as well as acyclic diacids and diols
bearing heteroatoms'771also served as lactonization substrates for the lipases. However, the enantioselectivity of this
unique biocatalytic reaction has yet to be systematically studied.
70 5
Table 12. Lipase-catalyzed lactonization of w-hydroxy carboxylic acids and their esters
0
23, mono
R'
n
R2
H
H
H
8
H
R'
14
A
0
0
0
0
0
66
62
46
19
A
A
B
B
porcine pancreas
Cundidu c.ylindruceu
Mucor miehei
A
B
Pseudomonas sp.
P.wudomonus sp.
Pseudomonus sp.
Candidu cjlindracea
Yield [%] [b]
23
mono
A
A
Origin of the lipase
5
10
CH,
CH,
Condrtions
[a1
Pseudomonus sp.
Pseudomonas sp.
porcine pancreas
Cundidu cylindrucea
Mucor miehei
porcine pancreas
Cundidu cilindruceu
Mucor miehei
n
R2
H
H
Pseudomonus sp.
Pseudomonus sp.
porcine pancreas
Cundidu cylindrucea
Mucor miehei
14
H
CH,
Origin of the lipase
8
H
CH,
24, di
A
A
A
A
A
B
A
Conditions
[a1
C
D
D
A
24
di
tri
tetra
53
33
57
2
2
16
14
20
24
22
4
15
6
13
18
12
9
15
0
0
55
0
15
24
27
46
9
13
20
54
0
0
5
24
26
6
19
15
*
23, mono
Yield [%I [b]
24, di
RS
S,S
3
0
0
0
0
5
13
4
7
*
7
2
*
*
*
26
30
21
15
43
0
penta
*
4
6
15
*
*
+ R,R
12
30
20
21
tri-1
tri-2
tetra
4
11
17
13
3
9
20
2
12
*
*
*
[a] Reaction conditions: A 65"C, 48 h; B 65 "C. 96 h: C 25 'C, 144 h; D 50"C, 96 h. [b] An asterisk indicates that a trace quantity was detected by thin layer
chromatography.
HO,C(CH,)_CO,H
21
+
HO(CH,)"OH
lipases
28
organic solvent, the biocatalytic reaction may often be more
enantioselective than the corresponding hydrolytic reaction
in water. As such, it provides the synthetic chemist with an
additional methodology for the preparation of optically active compounds.
It is important to recognize the potential broad range of
biocatalysts that may be used for biocatalytic resolution in
organic media. We are therefore in the midst of an exciting
stage in the development of this field. We hope this article
will stimulate interest and greater activity in this relatively
new area.
C.S.C. is grateful to the Petroleum Research Fund of the
American Chemical Society for financial support.
7. Epilogue
Enantioselective biocatalysis in organic media is a rapidly
developing field but it is still at an early stage, because there
is still much to learn about enzyme protein conformations in
water and in organic media. Nevertheless, the usefulness of
this nonaqueous methodology is clearly indicated by a number of published reports. Thus far, only the enantioselective
properties of lipase-catalyzed esterification and transesterifcation reactions have been exploited to a large extent. By
choosing a biocatalyst and an acceptor or donor in a suitable
706
Received: November 8, 1988;
supplemented: January 23, 1989 [A 721 IE]
German version: Angew. Chem. IOf (1989) 71 1
[l] A. A. Klyosov. N. Van Viet, I. V. Berezin, Eur. J Biochem. 59 (1975) 3.
121 A. R. Macrae, 1 Am. Oil Chem. Soc. 60 (1983) 291.
[3] B. Cambou. K. Klibanov, J. Am. Chem. Sac. 106 (1984) 2687.
[4] B. Cambou. K. Klibanov, Biotechnol. Bioeng. 26 (1984) 1449.
[S] S. Koshiro, K. Sonomoto, A. Tanaka, S. Fukui, J. Biofechnol. 2(1985) 47.
[61 G. Langrand, M. Secchi. G. Buono, J. Baratti. C. Trrantaphylides. Tefruhedron Left. 26 (1985) 1857.
[7] G. Kirchner, M P. Scollar, A. M. Klibanov, J Am. Chrm. Soc. 107(1985)
7072.
Angew. Chem. Int. Ed. Engl. 28 (1989) 695-707
[8] C Langrand, J. Baratti, G. Buono. C . Triantaphylides, Tc.trahdron Lett.
27 (1986) 29.
191 G. M. Ramos Tombo. H.-P. Schaer, X. Fernandez I Busquets, 0. Ghisalha. Tetruhedron Leti. 27 (1986) 5707.
[lo] C.-S. Chen. S.-H. Wu, G. Girdaukas, C. J. Sih, J. Am. Chem. Soc. 109
(1987) 2812.
[ l l ] A . L. Margolin, J.-Y. Creme, A. M. Klibanov, Tetrahedron Lett. 28(1987)
1607.
[12] G . Gil, E. Ferre, A. Meou, J. Le Petit, C . Triantaphylides, Tetrahedron
Lc,it. 28 (1987) 1647.
[I31 T. M. Stokes. A. C . Oehlschldger, Tetrahedron Lerr. 28 (1987) 2091.
[14] H. Hemmerle, H.-J. Gais, Tetrahedron Lett. 28 (1987) 3471.
[15] A. Belan. J. Bolte, A. Fauve, J. G. Gourcy. H. Veschambre, J. Org. Chem.
5-7 (1987) 256.
[16] P. E Sonnet, J. Org. Cham. 52 (1987) 3477.
I
1171 F. Francalanci, P. Cesti, W. Cabri, D. Bianchi, T. Martinengo, M. Foa, .
Org. Chem. 52 (1987) 5079.
[18] a) K. Yamamoto, T. Nishioka, J. Oda. Y. Yamamoto, Tetrahedron Lett. 217
(1988) 1717; b) D. Bianchi, P. Cesti, E. Battistel, J. Org. Chem. 53 (1988)
5531.
[19] a ) Y.-F. Wang, C:H. Wong, J1 Org. Chem. 53 (1988) 3127: b) Y. F. Wang,
J. J. Lalonde, M. Momongan, D. E. Bergbreiter, C. H . Wong. J. Am. Chem.
So(.. f f 0(1988) 7200; c) Y. Terao, M. Murata, K . Achiwa, T. Nishio, M.
Akamtsu. M. Kamimura, Tetrahedron Lett. 29 (1988) 5173.
[20] F. Theil, S. Ballschuh, H . Schick. M. Haupt. B. Hafner, S. Schwarz, Sym
~ h e s b1988, 540.
[21] J. B. West. C.-H. Wong, Tetrahedron Lelt. 28 (1987) 1629.
A. L. Margolin, A. M I Klibanov, J. Am. Chem. Soc. 109 (1987) 3802.
1. L. Gatfield, Ann. N . X Acad. Sci. 434 (1984) 569.
A. Makita. T. Nihira. Y. Yamada, Tetrahedron Leti. 28 (1987) 805.
Z.-W. Guo. C. J. Sih, J. Am. Chem. Soc. I 1 0 (1988) 1999.
W. Kullman: Enzymatic Peptide Synthesis, C R C Press, Boca Raton, F L
1987.
J. Grundwald. B. Wirtz. M. P. Scollar, A. M. Klibanov. J. Am. Chem. Soc.
fOH (1986) 6732.
R. 2.Kazandjian, A. M. Klibanov, J. Am. Chem. So<. 107 (1985) 5448.
F. Effenberger, T. Ziegler, S. Forster, Angew. Chem. 99 (1987) 491 ; Angew.
C h m . Int. Ed. Enyl. 26 (1987) 458.
A. M . Snijder-Lambers, H. J. Doddema, H. J. Grande. P. H . van Lelyveld,
Biu~~atul.
Org. Media, Proc. In!. Symp. 1986, 87, see Stud. Org. Chem.
i Amcterdrrm) 29 (1987).
P. L. Luisi, Anyew. Chem. 97 (1985) 449; Angew. Chem. I n ! . Ed. Engl. 24
(1985) 439.
T. W. Randolph. H. W. Blanch, J. M. Praunitz. C. R. Wilke, Biotechnol.
L P I I .7 (1985) 325.
H. L Bruckman in B. Borgstrom, H. L. Brockman (Eds): Lipases. Elsevier. Amsterdam 1984. p. 3.
L. E. S Brink, J. Tramper, Biotechnol. Bioeng. 27 (1985) 1258.
C. Ldane, S. Boeren, K . Vos. C. Veeger, Biotechnoi. Bioeng. 30 (1987) 81.
G. Carrea, Trend.7 Biotechnol. 2 (1984) 102.
L. G. Butler, Entymr Microb. Techno/. f (1979) 253.
M. W. Empie, A. Gross, Annu. Rep. Med. Chem. 23 (1988) 305.
G. A. Homandberg, J. A. Mattis, M . Ldskovski, J r , Biochemistry 17
(1978) 5220.
R. Dernoncour, R. Azerad, Terrahedrun Leir. 28 (1987) 4661.
F. Bjorkling, J. Boutelje, J. Gatenbeck, S. Hult. T. Norin, Tetrahedron Let!.
26 (1985) 4951.
L. K. P Lam. R. A. H. F. Hui, J. B. Jones, J. Org. Chem. 51 (1986) 2047.
F. Bjorkling, J. Boutelje, M. Hjalmarsson, K . Hult, T. Norin, J. Chem. Soc.
Chrm. Commun. 1987, 1041.
Angew. Cli1.m. Int. Ed. Engl. 28 (1989) 695 - 707
[44] D. Herschlag, Bioorg. Chem. I0 (1987) 62
[45] A. P. G. Kieboom, R e d Truv. Chim. Pays-Bas 107 (1988) 347.
1461 Y. Hata, Y. Matsuura, N. Tanaka, M. Kakudo, A. Sugihdra, M. Iwai. Y.
Tsujisaka, J. Biochem. 86 (1979) 1821.
[47] Y. Hata, N. Tanakd, N. Kakudo. A. Sugihard, M. Iwai, Y. Tsujisdka, Aria
Crystollogr. Sect. A37 (1981) C38.
[48] M. SCnieriva, P. Desnuelle, Adv. Enzymoi. Re1ut. Areas Moi. Biol. 48
(1 979) 360.
[49] W. H . Liu, T. Beppd, K . Arima, Agric. Biol. Chem. 41 (1977) 131
[SO] R. Verger, M. C. F. Mieras, G. H. DeHaas. J. Biol. Chem. 248 (1973)
4023.
[51] J. Lavayre. J. Verrier, J. Baratti, Biutechnol. Bioeng. 24 (1982) 2175.
I521 C. S. Chen, Y. Fujimoto, G. Girdaukas, C. J. Sih, J1 Am. Chem. Sue. 104
(1982) 7294.
[53] A. R. Fersht: Enzyme Structure and Mechunism. 2nd ed., Freeman, New
York 1985, Chapter 12.
[54] A. Zdks, A. M. Klibanov, Proc. Natl. Acad. Sci. USA 82 (1985) 3192.
[ S S ] W. W. Cleland. Biochrmi.strv 14 (1975) 3220.
[56] G. Langrdnd. J. Baratti, G. Buono, C. Triantaphylides. Biocata1jsi.s f
(1988) 231.
[57] P. Aldercreutz, B. Mattiasson, Biocaialysis 1 (1987) 99.
[58] M. V. Rodionova, A. B. Belovd. V. V. Mozhaev, K . Martinek. I. V. Berezin, Dokl. Akud. Nauk USSR 292 (1987) 913; Chem. Ahsrr. 106 (1987)
171823.
[59] A. Zaks, A. M. Klibanov, L Biol. Chem. 263 (1988) 3194.
[60] A. Leo, C. Hansch, D. Elkins, Chem. Rev. 7 1 (1971) 525.
[61] C . Ldane, S. Boeren, R. Hilhorst, C. Veeger. Biucatul. Org. Media. Proc.
Inr. Symp. 1986. 65, see Stud. Org. Chem. (Amsterdam) 29 (1987).
[62] C. Laane, BiocOta/ysi5 f (1987) 17.
I631 K . Martinek, A. N. Semenov, I V. Beredn. Btochim. Bt0phy.s. Aria 658
(1981) 76.
[64] C. Gancet, C. Gaignard, P. Fourmentreaux, Eur. Pat. Appl. 239470.Al
(1987); Chem. Abstr. 108 (1988) 73744d.
1651 M. Reslow. P. Adlercreutz. M. Mattiasson. Biocatal. Org. Media, Proc.
Int. Svmp., 1986. 349. see Stud, Org. Chem. (Amsierdam) 29 (1987).
[661 A. M. Klibanov, CHEMTECH f986, 354.
H. Hirohara, S. Mitsuda, E. Ando, R. Komdki, Biaca/ul. Org. Synth.,
Proc. Int. Symp. f98S. 119, see Stud. Org. Chem. (Amslwdam) 22
(1985).
C. S. Chen, C. J. Sih, unpublished data.
C. J. Sih, S. H. Wu, Top. Stereochem. f 9 (1989). in press.
C . 4 . Chen, unpublished data.
D. V. Patel. E VanMiddlesworth. J. Dondubauer, P. Gannett, C. J. Sih. J.
Am. Chem. Soc. 108 (1986) 4603.
G. M. Ramos Tombo, H.-P. Schaer, X. Fernandez I Busquets, 0. Ghisdiba: Biocatrrl. Org. Media, 1986, 43, see Stud, Org. Chem. (Amsterdam) 29
(1987).
Z. W. Guo, T. K. Ngooi, A. Scilimati, G. Fulling. C. J. Sih. Tetrahedron
Leu. 29 (1988) 4927.
A. Scilimati. T. K. Ngooi, C. J. Sih, Tetrahedron Lett. 29 (1988) 5583.
A. L. Gutman, K. Zuobi, A. Boltansky, Tetrahedron Let!. 28 (1987)
3861.
L. Bianco, E. Buige-Jampel, G. Rousseau. Tetruhedron Lett. 29 (1988)
1915.
Z . W. Guo, C. J. Sih, unpublished data.
V. Gotor, R. Brieva, F. Rebolledo, J. Chem. Sue. Chem. Commun. 1988,
957.
P. D. Theisen. C. H. Hedthcock, J. Org. Chem. 53 (1988) 2374.
B. Borgstrom, C. E. Albertsson in B. Borgstrom, H. L. Brockman (Hrsg.):
Lipmes. Elsevier. Amsterdam 1984, p. 151.
707
Документ
Категория
Без категории
Просмотров
6
Размер файла
1 274 Кб
Теги
synthetic, biocatalysts, aspects, general, organiz, method, solvents, enantioselectivity, lipase, optimization, use, new
1/--страниц
Пожаловаться на содержимое документа