Note: Descriptions are shown in the official language in which they were submitted.
CA 02332638 2000-12-28
WO 00/01842 PCTIUS99/15400
Visualization of Enzyme-catalyzed Reactions Using pH Indicators: Rapid
Screening of
Hydrolase Libraries for Enahtioselective Enzymes
RELATED APPLICATIONS
This application is a claims priority to U.S. Ser.,No. 60/091,880 filed July
7, 1998 and
U.S. Ser. No. 60/125,708 filed March 2~, 1999, both of which are hereby
incorporated by
reference in their entirety.
FIELD OF THE INVENTIt7~N
This invention relates to pH-dependent assays for detection of enzymatic
activity.
BACKGROUND OF THE I1VVENTION
There is an inereasingl;y important need to develop new biocatalysis processes
rapidly
and inexpensively, especially for the development of novel pharmaceuticals
where time is
extremely valuable. The use of a powerful, analytical screening strategy is
often the key to
speeding up development time; at several different levels of the process. In
the discovery of
novel enzymes, screening plays an important role in identifying which subset
of candidates
contain an enzyme of interest from a collection of organisms, clone banks, or
enzyme
libraries. Directed evolution approaches to engineer custom biocatalysts
require powerful
screening strategies to sift thxough large mutant pools to find enzymes with
properties that
have often been only slightly altered against a high activity background.
Finally, process
optimization and development can often take an excessive amount of time,
especially to
perform a comprehensive analysis of different reaction conditions including
temperature, pH,
cosolvent, reaction time; and .other parameters, both individually and in
combination. This
type of analysis requires the implementation of a rapid, high-throughput assay
which is
amenable to automation and use in a hierarchical screening strategy.
The synthesis of enantiomerically pure compounds (EPC) with one or several
chiral
centers is one of the most challenging tasks in modern organic chemistry.
Enzymes are able
to contribute signif cantly to this challenge and have been increasingly
considered as a useful
class of catalysts for organic synthesis. (.Davies, G. et al.
Biotransformations in Preparative
Organic Chemistry; Academic Press: London, 1989; Wong, et al. Enzymes in
Synthetic
Organic Chemistry; Pergamon: Oxford, 1994; Faber, K. Biotransformations in
Preparative
Organic Chemistry , Springer-Verlag: Berlin-Heidelberg, I995; Drauz, K;
Waldmann, H Eds.
Enzyme Catalysis in Organic Synthesis Vol 1 ~ 2VCH: Weinheim, 1995). Among
these
CA 02332638 2000-12-28
WO 00/01842 PCT/US99/15400
biocatalysts, hydrolases are well established as valuable tools for the food,
pharmaceuticals
and fine chemicals industry. (Gerhartz W. Ed. Enzymes in Industry VCH:
Weinheim, 1990).
The irxiportance of biocatalysis has led to the search of novel enzymes with
singular
activities. Recently, extremophilic microorganisms have been investigated as a
source of
these novel activities. (Kristjansson, J. K. TIBTECH 1989, 7, 349; Adams,et
al.
BiolTechnology , 1995, 13, 6Ei2; Govardhan, et al. Chem. Ind. 1995, 17, 689-
93; Newell, J.,
Chemistry in Britain 1995, 3i'; Vieille, et al. TIBTECH ,1996,14, 183).
Scientists at ThermoGen, Inc. (Chicago, IL} have developed a set of tools to
obtain
libraries of thetxnophilic enzymes by genetic engineering (see, for example,
U.S. Pat. Ser. No.
08/694,078). ThermoCat~ consists of a set of twenty stable esterases, capable
of working
well either at room or high temperature and also in organic solvents. It is of
high interest to
study the selectivity of each enzyme in the library, especially their
enantiodiscrimination
when exposed to racemic substrates. Time is the limiting factor in carrying
out the work
when screening a library of enzymatic activities against an array of
substrates for either
enzyme discovery, enzyme engineering (such as directed evolution) or process
optimization
experiments. The analytical methods typically employed for this purpose
include high-
pressure liquid chromatography (HPLC), thin-layer chromatography (TLC), and
gas
chromatography (GC), which .are often not amenable to high-throughput assays.
The need in the industry for new methods for the identification of new
biocatalysts
requires rapid screening assays combined with hierarchical screening
strategies. The
approach works by eliminating the weakest candidates as one of the earliest
steps in the
bioprocess development timeline, renderiing a streamlined process-viability
study. (See, for
example, Demirjian, et al. Top. Curr. Chem., 1998, 200.)
One.af the most convenient ways to assay an enzyme is through a method that
allows
the development of color a~zd thus can be used in qualitative as well as
quantitative
measurements. A number of c;olorimetric methods to measure enzymatic activity
have been
described. (Michal, et al. in rwfethods of Enzymatic Analysis; Bergmeyer, H.
U., Ed,; Verlag
Chemie: Weinheim, 1983; Vol. I; pp 197; Demirjian, et al. Top. Curr. Chem.,
1998, 200.)
Hydrolytic enzymes can be rapidly screened with chromogenic (nitraphenyl),
fluorogenic (4-
methylurnbelliferone) or indil;ogenic (indoxyl) substrates that yield colored
products upon
hydrolysis. The main limitation of this approach is the presence of the latent
colorimetric
functionality within the substrate, whose introduction is at least time-
consuming and yields a
structure for analysis essentially different from the actual target (usually a
methyl or ethyl
ester).
2
CA 02332638 2000-12-28
WO 00!01842 PCT/US99/15400
Several types of pH-dependent assays have been described including enzyme-
catalyzed processes with hexokinase. (Wajzer, J. Hebd. Seances Acad. Sci.,
1949, 229 1270;
Darrow, et al. Methods in unzymology, 1962; Yol. Y, 226; Crane, et al. Methods
in
Enzymology, 1960; Vol. I, 27'7) and cholinesterase (Lowry, et al. J. Biol.
Chem. 1954, 207,
19) and in enzyme-free studies of carbon dioxide hydration. {Gibbons, et al.
J. Biol. Chem.
1963, 238, 3502). Since the 1970's, such strategies have been used in kinetic
analysis of
enzyme reactions. Examples of this include human carbonic anhydrase
(Khallifah, R. G. J.
Biol. Chem., 1971, 24b, 2561',1, amino acid decarboxylases (Rosenberg, et al.
Anal. Biochem.
1989, 18i, 59) and serine proteases (Whittaker, et al. Anal. Biochem. 1994,
220, 238). The
progress of the hydrolysis is monitored by visual inspection of the solution
color after the
enzyme has been added or by using a microplate reader to get a quantitative
reading.
Whittaker et al. (Anal. Biochem. 1994, 220, 238-243) measured the esterase
activity
of proteases in 96-well microplates using a pH-dependent assay. However, the
Whittaker
assay requires additional experiments beyond those required in practicing the
methodologies
provided herein in that Whittaker does not use an indicator-buffer pair with
the same pKa
values and does not reliably measure the true rates of enzyme-catalyzed
hydrolysis.
Recently, the use of p>=I indicators has been extended to monitor the directed
evolution
of an esterase an a plate assay using a whole cell system, rather than the
isolated enzyme.
(Bornscheuer, et al. Biotechnol. Bioeng. 1L998, 58, 554). However, a pH-
dependent assay has
not been utilized to determine; the enatioselectivity of an enzyme. Kazlauskas
has partially
solved the problem of competition by using a reference non-chiral additive in
classical
chromogenic substrate assays. (Japes, et al. J. Org. Chem., 1997, 62, 4560.)
The same
author developed a quantitative method .for the evaluation of the
enantioselectivity (without
considering the competition factor) for actual substrates based on a pH
indicator/buffer
system (p-nitrophenol / BES) with equal pKa so the linearity of the color
transition allows the
quantitation of the enantiosele~;.tivity. (Danes, et al. Chem. Eur. J. ,1998,
4, 2317).
Despite its impressive accuracy and sensitivity, the method of Kazlauskas
requires
special instrumentation (i.e., a microplate reader) since the color transition
cannot be
visualized and involves data management that could be avoided by using a
suitable indicator
that effectively turns color so the monitoring could be simplified.
Furthermore, in certain
cases, the linearity of the assay (and consequently its accuracy) is
compromised by the
difficulty of choosing a pair buffer / indicator with same pKa and yielding
color change.
3
CA 02332638 2000-12-28
WO 00/41$42 PCT/US99115400
As a solution to the deficiencies in the currently available art, provided
herein is a
simple, colorimetric, pH responsive method for the rapid screening of enzyme
libraries is
provided herein. This is based on the pH change {typically to a lower pH) that
occurs as the
reaction proceeds and the carboxylic acid is released. This drop can be
monitored by visibly
detectable change in color of a solution as determined by a pH indicator,
where the color
profile of the indicator falls into the pH range of the enzymatic activity.
Using this
methodology, the reactivity {detected by a color change) of pairs of
enantiomers
corresponding to the same racemic mixture can be detected and an estimate of
the
enantioselectivity of the enzy~rne cari be made. A large turnover difference
between isomers
indicates a high probability of succesful kinetic resolution if the racemic
mixture is subjected
to the enzyme displaying such time difference. The method involves the use of
single
isomers, so the kinetics obtained in this way do not reflect the competition
that exists when
hydrolyzing the racemic mixture, and therefore E value is approximate, but
very effective,
especially in the case high E, which are the important ones when screening
libraries of
enzymes and substrates.
S1UMMAR'Y OF THE INVENTION
Provided herein is a method for the visualization of hydrolase-catalyzed
reactions
using pH indicators by comibining an enzyme to be tested, a substrate to be
tested, an
appropriate buffer, and a pH-sensitive indicator in a reaction mixture where
the indicator is
selected based on the predicted chemical reaction of the enzyme and the
substrate. In one
embodiment, the indicator will change the color of the reaction mixture in
relation to progress
of a chemical reaction between said enzyme and said substrate. The assay is
useful for
determining qualitatively the enantioselectivity or stereoselectivity of an
enzyme against pairs
of enantiomers or stereoiso~mers. In a preferred embodiment, the assay is
useful for
determining the enantioselec~ivity of an enzyme. In another embodiment, the
assay is useful
for the identification of DNA molecules encoding ester-hydrolyzing enzymes. In
yet another
embodiment, the assay is use;fui for detection a mutation to a DNA molecule
encoding an
ester-hydrolyzing enzyme thavt alters the activity of the enzyme.
BRIEIF DESCRIPTION OF THE FIGURES
Figure 1. pH-dependent .assay using a) bromothymol blue; and, b) phenol red.
Figure 2. Hierarchical screening of a library of thermophilic esterases.
4
CA 02332638 2000-12-28
WO 00/01842 PCTIUS99/15400
DETAILED DESCRIPTION
While twenty enzyme;s were until recently considered a large number to screen
in the
development of a new biocatalytic process, it is only a modest number by
today's standards.
Increasingly, enzyme libraries are becoming larger and larger, and an
effective tool for
screening and cornparzson is becoming a necessity. Provided herein is a pH-
dependent
methodology for quickly one( efficiently optimizing reaction parameters (such
as conditions
for the stereoselective tnans:Formation of a molecule) for identification of
enzymes have
enantiomeric selectivity. In this manner, only a small subset of reactions
need to be analyzed
in more detail, such as by HP'LC. This will significantly increase the number
of samples and
conditions that can be analyzed in a short period of time.
In one embodiment, tl~e assay provided herein is useful for qualtitatively
determining
the selectivity of an enzyme, including but not limited to the
enantioseIectivity or
stereoselectivity of an enzyme, against pairs of enantiomers or stereoisomers.
The assay is
useful to identify enzymes having activity towards compounds with more than
one chiral
center, such as diastereomers. In addition, the assay is also useful for the
study of
enantiomeric mixtures where chirality is determined by symmetry axis or
rotational barriers.
Reagents and methodologies for the use of pH indicators to monitor hydrolase-
catalyzed reactions are providLed herein. In one embodiment, the formation of
acid following
an enzyme-mediated hydrolysis causes a drop in the pH that may be visualized
by a change in
the color of the indicator-coni:aining solution. Preferred indicators are
those showing a color
transition within the operational pH range of the hydrolases, including but
not limited to
bromothymol blue and phenol red. The enantioselectivity of lipases and
esterases may be
estimated using single isomers under identical conditions and comparing the
color turnover
for each one. In one ernbo~diment, this method may be utilized to quickly
evaluate the
enantioselectivity of a lipase vtowards a set of ester substrates and applied
to the hierarchical
screening of a library of thertr~ophilic esterases.
Several advantages are; provided by the methods provided herein. For example,
as the
signal being monitored (ie, color change) does not originate from the
substrate but from the
indicator, the actual substrate can be used for screening. Currently, custom-
made fluorogenic
or chromogenic substrates arf; utilized but the reactions may vary because the
substrate has
been altered. Another advantage is that the enantioselectivity of the enzyme
can be evaluated
if both enantiomers of the ra.cemic mixture are tested separately, typically
suboptimal but
adequate for the qualitative screen demonstrated herein. Also, if both
enantiomers axe not
CA 02332638 2000-12-28
WO 00/01842 PCT/US99/15400
available, the comparison of the results obtained using the racemic mixture
compared to the
results obtained using the available enantiomer will provide useful data. In
that case, the
available enantiomer has to be the slow-reacting isomer of the racemic
mixture. Yet another
advantage is that the present assay does not require complicated hardware
devices, and is
based only on visual observation of the color change of the reaction mixture.
.And, since the
assay is very sensitive, very small amounts of substrate relative to enzyme
are required to
detect the color change, opening the use of microplate wells and the
subsequent automation
of liquid handling systems.
Any suitable buffer miay be utilized in practicing the present invention. Many
such
buffers are known in the art. For instance, the use of phosphate buffer at a
pH equal to its
pKa (7.20) (Beynon, et al. in Buffer Solutions: The Basics, IR.L Press,
Oxford, 1996) would
be preferred because it is compatible with the function of the majority of
lipases and esterases
and their stability profile. Thn phosphate; buffer also provides mild
conditions (neutral pH} if
sensitive substrates are to be hydrolyzed. It is also standard to utilize a pH
slightly above or
below neutral if the enzyme selected shows more desireable activity under
these conditions.
Suitable buffers may include., but are not limited to inorganic buffers such
as potassium
phosphate (pH 6-8), sodium Irhosphate (pH 6-8), or sodium carbonate (pH 8-10);
or organic
buffers such as BES (N,N-bi;~[2-hydroxyethylJ-2-aminomethanesulfonic acid, pH
6-8), Tris
(pH 7-9), citric acid (pH 4-6), and HEPES (N-[2-hydroxyethyl]piperazine-N'-[2-
ethanesulfonic acid; pH 6-8}.
The use of an indicator dye with which to follow the reaction forms another
part of
the present invention. Exemplary pH indicator dyes suitable for following the
hydrolytic
reaction by the color change in the range typically used for this class of
biotransformations is
shown in Table 1. Other indicators are known in the art, such as those
described in Banyai,
et al. (Indicators; Bishop, E;., Ed.;Pergarnon; Oxford, 1971, pp 65-176.} It
should be
understood that many such indicator dyes are available, and that any of those
available dyes
not listed in Table 1 may also gibe useful in practicing the present
invention.
6
CA 02332638 2000-12-28
WO 00!01842 PCT/US99/15400
Table 1
Exemplary pI~' I~adicators :.,~,.
Name - pKa pH range' color change
Bromothymol Blue 7.30 6.0-7.6 yellow-blue
Neutral Red 7.40 6.8-8.0 red-amber
Phenol Red 8.00 6.6-8.0 yellow-red
Brilliant Yellow - 6.6-7.8 yellow-red
Cresol Red 8.46 7.0-8.8 yellow-red
Turmaric (curcumin) - 7.4-8.6 yellow-red
Metacresol Purple 8.3 7.4-9.0 ~T yellow-purple
2-(2,4-dinitrophenylazo)-1-naphthol-3,6-- 6.0-7.0 yellow-blue
disulphonic acid, disodiurn salt
6,8-dinitro-2,4-(1H)quinazolinedione- 6.4-8.0 colorless-yellow
For example, the bromothymol blue / potassium phosphate system falls within
0.1
units of pKa (making it suitable for quantitation) and provides a useful blue-
yellow color
transition in a buffer commonly included in hydrolase-catalyzed
biotransformations. In a
preferred embodiment, Bromothymol Blue (BTB) or Phenol Red are utilized in
combination
with phosphate buffer, which has a similar pKa value. In this manner, the
color transitions
show high contrast as opposed! to using the phosphate buffer in combination
with a dye such
as Neutral Red, which has a similar pKa but displays poor distinction between
red and amber.
It is to be understood by the skilled artisan that various operating pH
conditions may be
utilized based on the individual. enzyme characteristics or process
advantages, and that a
different indicator may be utilized that corresponds to the operating pH
conditions. For
instance, in one embodiment, phenol red may be the optimal indicator where the
reaction is
carried out at slightly basic pH (i.e., pH 8.0). In a preferred embodiment,
preliminary
experiments are performed to determine the concentration of indicator dye in
the reaction that
has no effect on the reaction rate, suggesting that the indicator is does not
inhibit enzyme
activity. In yet another embodiment, control experiments inculding a protein
source such as
BSA, for example, may be performed to demonstrate that the presence of protein
does not
alter the indicator color. In this manner, it can be confirmed that the pH
changes in the
solution were the result of enzyme catalyzed hydrolysis. Further tests of
reaction solutions
containing enzymes and indicators without substrates (control assays) may also
be performed
7
CA 02332638 2000-12-28
WO 00101842 PCT/US99/15400
to establish that the color changes in solution do not result from the
presence of either the
buffer salts or the enzymes without regard to their activity.
In one embodiment, the enantioselectivity of a biocatalyst may be monitored
using an
array of enantiomeric substrates described in the literature to be efficiently
resolved by well-
known enzymes. One such lipase biocatalysts is from Pseudomonas cepacia (PSL;
available
from Amano Pharmaceutical, 3apan). PSL has been used both for enantioselective
hydrolysis
as well as transesterif cations in organic solvents. Exemplary substrates that
display
enantioselectivities in the desired range (from high to very high) are shown
in Example 2.
These allow for the development of a successful biocatalytic resolution in a
cost-effective
manner. Example 2 illustrates substrates ready to use in the hydrolytic assay.
Certain
substrates will be available only as alcohols (both isomers); such substrates
may be acetylated
prior to the assay.
Enantioselectivity (E) is related to the time it takes each enantiomer to
change the
color. The larger the difference between the times to color change for each
enantiomer, the
higher the enantioselectivity of the enzyme for the fast-reacting enantiomer.
The resultant
measurement is an estimated or relative enatioselectivity. Preferred enzymes
demonstrate
large enantioselectivity value: (ie, fast time to color change), and would be
understood to
have the most potential for practical application.
In one embodiment, a reaction is set up in a reaction vessel, such as a 96-
well
microplate, and the reaction solution contains an enzyme showing hydrolytic
activity such as
a hydrolase (such as an esterase, lipases, and proteases), a substrate
solution, a buffer
solution, and an indicator dye.. In a preferred embodiment, the enzyme is a
lipase such as
PSL and the buffer solution h<~s a pH of '7.2. In a more preferred embodiment,
the indicator
dye is present at a concentration by volume of 0.1 %, 0.01 % or 0.001 %. The
amount of
substrate will vary but may be in the range of 1 to 20 mg/ml final
concentration in the
reaction mixture, and is preferably about 10 mg/ml. This low concentration
allows for the
use of small amounts of pure enantiomer, which is often available in limited
quantities. To
optimize the assay, an amount of enzyme is selected that allows for a
convenient time-
window of preferably less than 48 hours for the turnover of one of the
isomers. If the reaction
is too slow with the selected amount of enzyme, the skilled artisan may choose
to increase the
amount of enzyme utilized or to raise the temperature of the reaction such
that evaporation
resulting in a dried reaction does not occur.
The concentration of the buffer to be employed may be determined empirically,
but
should be weak enough to be saturated. by the carboxylic acid being produced
and drive the
8
CA 02332638 2000-12-28
WO 00/01842 PCT1US99I1S400
pH to the turnover point of the indicator. In a preferred embodiment, the
concentration of the
buffer is between 5 mM and 10 mM KPi buffers, which turn color faster, since
their buffer
strength is lower. In a more preferred embodiment, the concentration of the
buffer is 20 mM.
In a another preferred embodiment, the buffer is 20 mM potassium phosphate at
pH 7.2. Use
of a buffer at too low a concentration (ie, less than 5 mM} may result in
increased sensitivity
of the system where minor p13 shifts due to the background (after adding the
enzyme or the
substrate) may lead to inconsistent readings. In that event, correction of the
pH in each
distorted well may be required.
In practicing the present invention, the enantioselectivity corresponds with
the faster
turnover (development of a color change in the reaction mixture) of the
preferred isomer. It
is well within the skills of the. ordinary practicioner to determine the
appropriate time frames
for use with certain reactions. If the enzymatic reaction proceeds slowly, a
greater amount of
time is required to monitor the; progress of the reaction. In certain cases,
the same biocatalyst
may be utilized to evaluate a library of substrates, and the nature of the
substrates will dictate
the time frame of the assay. Although the actual biocatalytie transformation
may occur under
different conditions, such as different stirring conditions, the method
provided herein
generates estimates of initial reaction rates. This may serve to indicate the
proper
enzymeaubstrate ratio required for the biotransf~rmation of one substrate
versus another.
For cases in which only one of the isomers is available, the method functions
most efficiently
if the available isomer is the slower reacting one. Otherwise, both wells may
change color
almost simultaneously and no conclusion can be obtained from the experiment.
As would be
understood by the skilled artisan, it is not necessary for both isomers to run
on a single
screen.
In a . preferred embodiment, a commercial kit including . multiple enzymes of
a
particular class, such as a lipase, protease or esterase, may be organized
into a microplate
array in a buffer containing BTB, for example. In a preferred embodiment, the
buffer has a
Kpi of approximately 20 mM at pH 7.2. A first round of hierarchical screening
is
accomplished using all the enzymes in the kit to identify those enzymes
reacting with the
racemic substrate. At this point, the enzymes having the fastest reaction
times are selected
for fizrther analysis. This first round prevents the unnecessary use of the
oftentimes scarce
pure isomers to screen out negatives.
The following Examples are for illustrative purposes only and are not
intended, nor
should they be construed as limiting the invention in any manner. Those
skilled in the art
9
CA 02332638 2000-12-28
WO 0010I842 PCT/US99/15400
will appreciate that variations and modifications can be made without
violating the spirit or
scope of the invention.
EXAMPLES
Example 1
Materials ahd Methods
The enzymes used were obtained from Amano Pharmaceutical (Japan) and
ThermoGen, Inc. (Chicago, USA). PSL was dissolved in buffer (IO mglml) and
centrifuged
before use to remove insolubl<; material. Both single isomers (R and S} of
methyl mandelate,
1-indanol, i-phenylethanol, and a-hydroxy-y-butyrolactone were purchased from
Aldrich
(Milwaukee, USA) together with R-mandelonitrile acetate (R-S) and its
corresponding
racemic mixture. These were <ierivatized by acetylation to obtain both isomers
of 1, 3, 4 and
6 together with the racemic 5. R- and S-ethyl 3-hydroxy-3-phenylpropionate (2)
were
purchased from Fluka (Switzerland) and used as is. All of the isomers were
stocked in
MeCN at 60 rng/ml. The phosphate salts and the indicators bromothymol blue and
phenol
red were obtained from Aldrich (Milwaukee, USA).
)Example 2
pH Dnpehdent Assay for Hydrolase Activity
Bromothymol Blue (B~CB) and Phenol Red were initially studied for use in this
assay,
as the pKa values of these indicators are the closest to that of the phosphate
buffer to be
utilized and color transitions resulting therefrom show high contrast.
Preliminary
experiments showed that the concentratian of indicator dye in the reaction had
no effect on
the reaction rate, suggesting that -the indicator was not acting as an
inhibitor of enzyme
activity (data not shown). Control experiments using BSA as the protein source
caused no
change in indicator color and. established that pH changes in solution were
the result of
enzyme catalyzed hydrolysis. Further tests of reaction solutions containing
enzymes and
indicators without substrates established that color changes in the solutions
were not the
result of buffer salts or the enzymes themselves. The basic principle
governing this reaction
is shown below:
CA 02332638 2000-12-28
WO 00/01842 PCT/US99l1S400
Hydrolase
RC02Et ~~. RC02 + H+
KPi buffer (pKa 7.20)
Iri + H+ pH = b-8 InH
.~ >
blue Bromothymol Blue (pKa ?.30) yellow
red Phenol Red (pKa=8.00) yellow
In order to prove the concept of monitoring enantioselectivity using this
method, we
chose an array of enantiomeric substrates described in the literature to be
efficiently resolved
by well-known enzymes. A widely utilized lipase biocatalyst, has been isolated
from
Pseudomonas cepacia ("PSL" available from Amano Pharmaceutical, Japan). This
enzyme
has been utilized for enantio~selective hydrolysis as well as
transesterifications in organic
solvents. (Xie, Z.-F. Tetrahedron: Asymmetry 1991, 2, 733.) The substrates
shown below
display enantioselectivities in the desired range (from high to very high).
OAc OH OAc OAc O
CO I=t
w C02Me ~. 2 ~ ~ w w CN Ac0
i ~~ Ac I i I i
I 2 3 4 5
E value 35 127 412 684 26 high
fast isomer S S R R S R
reaction acetylation hydrolysis acetylation acetylation acetylation acetylatio
The enantioselectivity {E) values are shown together with the fast-reacting
isomer and
the kind of reaction described in the literature. Compounds 1 through 6 shown
above are
ready to use in the hydrolytic assay. The substrates used were: methyl
mandelate acetate (1),
E = 35 for the acetylation of butyl ester (Ebert, et al. Tetrahedron:
Asymmetry 1992, 3, 903)
and E = 18 for the methyl ester using the related AK lipase from Amano
(Miyazawa, et ai., S.
.l. Chem. Soc. Perkin Trans. l', 1992, 2263); ethyl 3-hydroxy-3-
phenylpropionate (2), (Boaz,
N. W. J. Org. Chem., 1992, ~~7, 4289); 1-indanol acetate (3) (Margolin, et al.
J. Am. Chem.
Soc., 1991, 113, 4693); 1-phenylethanol acetate (4) (Laumen, et al. .7. Chem.
Soc. Chem.
Commun., 1988, 598; Laumerc, et al. J. Chem. Soc. Chem. Commun., 1988, 1459;
Seernayer,
et al. Tetrahedron Asymm., 1992, 3, 827; Bianchi, et al. J. Org. Chem., 1988,
53, 5531;
Terao, et al. Chem. Pharm. Bull., 1989, 37, 1653; Bianchi,et al. Tetrahedron
Asymm., 1993,
4, 777; Keumi, et al. Chem. Lett., 1991, 1989; Gutman, et aI. Tetrahedron
Asymm., 1993, 4,
839); mandelonitrile acetate (5) (van Almsick, et al. J. Chem. Soc. Chem.
Commun., 1989,
1391; Effenberger, et al. Lie~~igs Ann. Chem., 1991, 47; Inagaki, et al. J.
Am. Chem. Soc.,
1991, 113, 9360; Inagaki, et al. J. Org. Chem., 1992, 57, 5643); a-hydraxy-y-
butyrolactone
11
CA 02332638 2000-12-28
WO 00/01842 PCT/US99115400
acetate (6) (Naoyuki, et al. E~ur. Patent App. 1992, app # 91110749.8;
Miyazawa, et al. Eur.
Patent App. 1991, app # 90124577.9). In the case of mandelonitrile, only the R
isomer was ,
available and it was compared to the racemic mixture. The comparison is valid
if the single
isomer (R) is the slow reacting one so the racemic mixture will change color
faster.
Compounds 1, 3, 4, and 6 2vre only available as alcohols (both isomers) as
well as (~)5
racemic mixture), and therefore must be acetylated prior to the assay. Since
many of them
are available as alcohols, the E values have been described for the
transesterification reaction,
not for hydrolysis.
The reactions were set up in a 96-well microplate and the total volume was 200
~L
split as follows: 100~.L of PS:L enzyme (1 mg, from a 10 mg/mL solution in
buffer and spun
off to avoid turbidity), 10 ~.L of substrate solution (60 mg/mL for a final
concentration of 3
mg/mL) and 90 ~.L of 20 mM buffer solution pH = 7.20 containing 0.001 % of
indicator dye.
The amount of PSL enzyme was 1 mg per well, which allowed a convenient time-
window of
no more than 48 h for the turnover of one of the isomers.
The concentration of the buffer employed was 20mM, weak enough to be saturated
by
the carboxylic acid being produced and drive the pH to the turnover point of
the indicator.
Five mM and 10 mM KPi buffers were also tested, and change color at a faster
rate, but the
sensitivity of the system was too high and minor pH shifts due to the
background (after
adding the enzyme or the substrate) led to inconsistent readings.
Ten microliters of sinl;le isomers I-4 and 6 (or R-isomer 5 and racemic 5)
solution
were added, an amount representing (at 60 mg/mL) 0.6 mg per 200 ~.L well (3
mg/mL final
concentration). Since the molecular weight of the substrates is between 144
and 208, the
molar concentration of the substrates ranges from 14-20 rnM. This low
concentration allowed
for the use of very little pure e;nantiomer, so for example, 1 mL stock
solution would last for
100 experiments.
As for the indicators used, Bromothymol Blue was the most extensively
employed,
although phenol red displayed good contrast, and two of the time points
obtained are shown
for comparison. Figure 1 depicts the results obtained with both indicators and
PSL enzyme.
As can be seen in Figure 1, the enantioselectivity described above corresponds
with the
faster turnover (development of yellow color) of the preferred isomer in all
cases studied.
Since the hydrolysis proceeds slowly, the time for monitoring the progress of
the reaction is
large. While substrates 2, 3, 4 and 6 are hydrolyzed in I-3 h, substrates 1
and 5 required
longer reaction times. In this'. case, the same biocatalyst is used to
evaluate a library of
substrates, therefore the nature of these will dictate the time frame of the
assay.
12
CA 02332638 2000-12-28
WO 00/01842 PCT/US99/I5400
An advantage of the method is rioted in the case of mandelonitrile acetate
(S), with
only one of the isomers available. Comparison of these results to those
obtained using the
racemic mixture yields the conclusion 'that the racemate turns color much
faster that the R
isomer. This is most likely due to the faster hydrolysis of the S isomer
present in the
racernate; thus, the skilled ari:isan is led to the conclusion that the enzyme
is S-selective. It is
not necessary to count on bol:h isomers to run the screen, although a
successful result may be
more difficult to obtain in the: absence of one of the isomers.
In the case of using Phenol Red as indicator, Figure Z shows essentially the
same
results. A control without en~:yme or substrate is also shown. In this case
the pH of the buffer
is slightly higher (7.4) in order to obtain the red initial color. This could
be the cause of the
substrate S-1 being hydrolyzed apparently faster than in the bromothymol blue
case. Thus,
phenol red is a good choice if the reaction has to be carried out at slightly
basic pH.
Example 3
Identification of,Enantioselcctive Es:zymes from apt Ehzyme Library
An application of the method is illustrated by the hierarchical screening of a
ThermoCat~ library of thernaophilic esterases against a-hydroxy-y-
butyrolactone (6). This
compound is a useful in the preparation of 4-substituted-2-hydroxy-butanoates
and other
optically and physiologically active compounds. The resolution can be
visualized following
either transesterification or hydrolysis of the corresponding acetate. The
conclusions from
the hydrolysis may be translated into the transesterification, at least in
terms of
enantioselectivity. However, ahe stability of the enzyme in organic solvent
may jeopardize the
resolution.
The commercial hydrolase kit from ThermoGen consists of 20 different
hydrolases
(E001 to E020) which all hydrolyze ester bonds and are organized in a 5x4
microplate array
using BTB-containing 20mN( KPi buffer at pH 7.2 and substrate at 3 mg/mL, as
explained
above. Total volume is 210 pL, and every one of kit enzymes (i0 units each)
were dissolved
in 1 mL of buffer, from which 200 p.L was added to the microplate wells (2
units per
experiment). This first round of hierarchical screening involved all the
enzymes in the kit was
utilized for identification of enzymes that react with the racemic substrate
such that the
enzymes having the fastest reaction times would be selected. As such, the pure
isomers were
not required for screen against negatives.
13
CA 02332638 2000-12-28
WO 00/01842 PCT/US99/15400
Figure 2 depicts the process, and in this first round several enzymes were
identified
as potential candidates for re;>olution of the hydroxylactone: E004, E009,
E011, E013E015,_
E017b and E018b were selected for the enantioselectivity assay shown at the
bottom of
Figure 2. In this case, single; isomers of the acetate were tested side by
side using the same
protocol as above. The control (substrate but no enzyme) and the PSL
experiment are also
shown in the same experiment. It is apparent that enzymes E004, E009, E011 and
E013 are
good candidates for further study, because a color change corresponding to the
R isomer
occurred in 20 minutes, as does PSL. On the other hand, E017b appears to react
more
slowly. E018b does not appear to be selective according to the observations
during the f rst 3
hours of reaction. Thus, this method allows for the identification of
potentially selective
enzymes that should be studied further, and for identification of non-
selective enzymes that
should not be further considered.
Example 4
Identification of a DNA Molecule EncodiHg are Enzyrsze having Ester-
Hydrolyzing Activity
Provided herein is a method for isolating a DNA molecule encoding an enzyme
having ester-hydrolyzing activity. First, a host cell is transformed with a
DNA fragment
encoding an enzyme having ester-hydrolyzing activity. Those host cells
expressing the
enzyme are then identified by selecting for expression of the ester-
hydrolyzirig enzyme in the
host cell by detecting activity against an ester containing substrate in the
presence of a pH
indicator that is provides a color change following a drop in the pH of the
reaction mixture.
The ester-hydrolyzing activity is detected in a pH-dependent reaction by
preparing a crude
iysate of the transformed cell, incubating a sample of the crude lysate in a
buffer containing a
pH-dependent indicator dye and a substrate to form-a mixture; and monitoring
the mixture for
a color change.
The DNA molecule encoding the enzyme is then isolated from the host cell and
the
sequence encoding the hydrolase determined by DNA sequencing. It may be
advantageous to
subject the DNA fragment to multiple rounds of selection following enzymatic
digestion in
order to isolate smaller DNA fragments for analysis. Using this method,
multiple host cell
populations may be screened simultaneously, with increased speed and accuracy
as compared
to systems presently available.
14
CA 02332638 2000-12-28
WO 00/01842 PCT/US99/15400
lExample 5
Identification of a DNA Molecule Encoding Mutated Ester Hydrolyzing Enzyme
A method for isolating a DNA molecule encoding an enzyme having ester-
hydrolyzing activity where the activity is altered as compared to the
naturally-occurnng
enzyme is also provided. To identify such an enzyme, a host cell transformed
with a
fragment of a DNA encoding an enzyme having ester-hydrolyzing activity is
exposed to
conditions causing mutagenesis of the DNA molecule. For instance, the host
cell is exposed
to ethylmethane sulfanoate, N methyl-N'-nitro-11l'-nitrosoguanidine, or
ultraviolet light.
Those host cells expressing the enzyme are then identified by selecting far
expression of the
ester-hydrolyzing enzyme in t:he host cell by detecting activity against an
ester containing
substrate in the presence of a p~H indicator that is provides a color change
following a drop in
the pH of the reaction mixture.. The ester-hydrolyzing activity is detected in
a pH-dependent
reaction by preparing a crude lysate of the transformed cell, incubating a
sample of the crude
lysate in a buffer containing a pH-dependent indicator dye and a substrate to
form a mixture,
and monitoring the mixture for a color change. Host cells demonstrating
altered ester-
hydrolyzing activity as comp;~red to the wild-type enzyme are then selected.
The DNA
molecule encoding the enzyme is then isolated from the host cell. To determine
the nature of
the mutation, the portion of the DNA molecule encoding the enzyme is
sequenced.
hJxample 6
Identification of Mutations Resulting in Expression of Ester-Hydrolyzing
Enzymes Having
Altered Activity
A method fox isolating a DNA molecule encoding an enzyme having ester-
hydrolyzing activity where the activity is altered as compared to the
naturally-occurnng ester-
hydrolyzing enzyme is provided. A D:LVA molecule encoding an enzyme having
ester-
hydrolyzing activity is mutated by random or directed rnutagenesis or to form
a mutated
DNA molecule. A host cell is then transformed with the mutated DNA molecule.
Those host
cells expressing the enzyme are then identified by selecting for expression of
the ester-
hydrolyzing enzyme in the hosit cell by detecting activity against an ester
containing substrate
in the presence of a pH indicator that provides for a color change following a
drop in the pH
of the reaction mixture. The ester-hydrolyzing activity is detected in a pH-
dependent reaction
by preparing a crude Iysate of the transformed cell, incubating a sample of
the crude lysate in
a buffer containing a pH-dependent indicator dye and a substrate to form a
mixture, and
CA 02332638 2000-12-28
WO 00/01842 PCT/US99/15400
monitoring the mixture for a color change. The DNA encoding the mutated ester-
hydrolyzing enzyme is then isolated. Using this method, those mutations that
confer an
altered activity upon the enzyme are then identified:
While a preferred form of the invention has been shown in the drawings and
described, since variations in the preferred form will be apparent to those
skilled in the art,
the invention should not be construed as limited to the specific form shown
and described,
but instead is as set forth in the claims.
16
