Engineering a Highly Enantioselective Horseradish Peroxidase by
Directed Evolution
by
MASSACHUSETTS INSTITUTE
OF TECHNOLOGY
Eugene Antipov
AUG 16 2010
B.E. Chemical Engineering (2002)
B.S. Biochemistry (2002)
University of Delaware
LIBRARIES
ARCHIVES
Submitted to the Department of Biological Engineering
in Partial Fulfillment of the Requirements for the Degree of
DOCTOR OF PHILOSOPHY in Biological Engineering
at the
Massachusetts Institute of Technology
June 2009
@2009 Massachusetts Institute of Technology
All rights reserved
Signature of Author
Department of Biological Eigineering
May 18, 2009
-
.
A
-
Certified by
Alexander M. Klibanov
Novartis Endowed Chair Professor of Chemistry and Bioengineering
Thesis Supervisor
Accepted by
Peter Dedon
Associate Head, Department of Biological Engineering
This doctoral thesis has been examined by a Committee of the Department of Biological
Engineering as follows:
Professor Dane K. Wittrup
/.
Committee Chairman
Professor Alexander M. Klibanov
Research Supervisor
Professor Michael B. Yaffe
i/Committee Member
Engineering a Highly Enantioselective Horseradish Peroxidase by
Directed Evolution
by
Eugene Antipov
Submitted to the Department of Biological Engineering
on May 22, 2009 in Partial Fulfillment of the
Requirements for the Degree of
Doctor of Philosophy in Biological Engineering
ABSTRACT
There is an ever-growing demand for enantiopure chemical compounds, particularly new
pharmaceuticals. Enzymes, as natural biocatalysts, possess many appealing properties as
robust asymmetric catalysts for synthetic chemistry. However, their enantioselectivity
toward most synthetically useful, non-natural substrates is typically low. Therefore,
improving enzymatic enantioselectivity toward a given substrate is a practically
important but arduous task. Here we report a highly efficient selection method for
enhanced enzymatic enantioselectivity based on yeast surface display and fluorescenceactivated cell sorting (FACS). By exploiting the aforementioned method, in just three
rounds of directed evolution we both greatly increased (up to 30-fold) and also reversed
(up to 70-fold) the enantioselectivity of the commercially useful enzyme, horseradish
peroxidase (HRP), toward a chiral phenol. In doing so, we discovered that mutations
close to the active site not only preserve HRP catalytic activity but impact its
enantioselectivity far greater than distal mutations. We thus examined how a single
mutation near the active site (Argl78Glu) greatly enhances (by 25-fold) the
enantioselectivity of yeast surface-bound HRP. Using kinetic analysis of enzymatic
oxidation of various substrate analogs and molecular modeling of enzyme-substrate
complexes, this enantioselectivity enhancement was attributed to changes in the transition
state energy due to electrostatic repulsion between the carboxylates of the enzyme's Glu178 and the substrate's D enantiomer. In addition, the effect of yeast surface
immobilization and influence of a fluorescent dye on controlling the enantioselectivity of
the discovered HRP variants was investigated. Soluble variants were also shown to have
marked improvements in enantioselectivity, which were rationalized by computational
docking studies.
Thesis Supervisor: Alexander M. Klibanov
Title: Novartis Endowed Chair Professor of Chemistry and Bioengineering
ACKNOWLEDGMENTS
I would like to extend my gratitude to my thesis advisor, Prof. Alexander M.
Klibanov. Thanks to his high standards I am now a more thoughtful and careful
researcher.
I would also like to thank the members of my thesis committee, Prof. Dane K.
Wittrup and Prof. Michael B. Yaffe, for the support and guidance that they offered
throughout my research. In particular, I am very grateful to Dane for allowing me to
carry out my experiments in his lab even when there was limited space.
I am thankful to the past and present members of the Klibanov and Wittrup
groups, especially Raj Chakrabarti, Nebojsa Milovic, Vikas Sharma, Mini Thomas,
Alisha Weight, Ben Hackel, Charles Wescott, and Bryan Hsu for helpful discussions.
I am also grateful to my collaborators:
Bruce Tidor and especially Art Cho.
Dasa Lipovsek, Kathryn Armstrong,
Finally, I thank my friends and my family for their love and support throughout
the years. I owe much gratitude to Steve Sazinsky for his excellent listening skills, his
scientific advice, and especially his friendship. Without Laura's optimism, love, and
support, I don't know if I would have been able to accomplish this work: thank you for
being part of my life. Most importantly, I am forever grateful to my beloved sister:
thank you for making it all possible.
To my Mom and Dad
TABLE OF CONTENTS
Chapter I: Introduction
Page
7
Chapter II: Selection of horseradish peroxidase variants with enhanced
enantioselectivity by yeast surface display
A.
B.
C.
D.
Introduction
Results and Discussion
Materials and Methods
References
15
19
39
47
Chapter III: Highly L and D enantioselective variants of horseradish peroxidase
discovered by an ultra high-throughput selection method
A.
B.
C.
D.
Introduction
Results and Discussion
Materials and Methods
References
50
52
70
78
Chapter IV: How a single-point mutation in horseradish peroxidase markedly
enhances enzymatic enantioselectivity
A.
B.
C.
D.
Introduction
Results and Discussion
Materials and Methods
References
80
82
97
101
Appendix A: Effect of induction medium supplementation on horseradish
peroxidase activity and display levels
103
Appendix B: Enantioselectivities of L and D selective variants discovered in
each round of directed evolution toward substrates 1, 2, 3, and 4
107
Chapter I: Introduction
Numerous therapeutic drugs, plant-protecting agents, fragrances and most natural
products are chiral molecules, many of which exert a specific biological effect in only
one enantiomeric form (1). As a result, there is a rapidly rising demand for enantiopure
bioactives (2). In 2000, for example, the total chiral drug sales exceeded the $100 billion
mark for the first time, representing one-third of all drug sales worldwide (3). Despite
the increasing demand, the chemical and pharmaceutical industries still use classical
antipode separation, such as enantioselective liquid chromatography, to separate racemic
product mixtures (3).
This process requires stoichiometric amounts of an appropriate
optically active reagent, as well as large amounts of organic solvents (4). Hence, to meet
the rising demand, a more cost-effective and environmentally-friendly technology is
needed.
Asymmetric catalysis is an attractive alternative to classical separation methods
(5).
Some asymmetric catatysts that are currently being developed include transition
metal catalysts (6), organocatalysts (7) and biocatalysts (8). Biocatalysts, in particular,
are increasingly emerging as a key component in the toolbox of process chemists (9, 10).
Enzymes, as natural biocatalysts, exhibit many properties which make them
attractive candidates to resolve or create chiral centers. They already perform difficult
enantioselective and regioselective transformations without tedious protection steps, and
can accelerate reaction rates by as much as 1012 fold. Moreover, because they operate
under mild conditions and because their selectivity results in few by-products, enzymes
are environmentally friendly. Although enzymes often exhibit high enantioselectivities
toward their natural substrates, most industrially relevant and commercially useful
7
substrates are non-natural (11).
biocatalysts
for
synthetic
Therefore, in order to create superior practical
chemistry
it
is
necessary
to
enhance
enzymatic
enantioselectivity toward these artificial substrates.
The main goal of this thesis project was to engineer horseradish peroxidase (HRP)
using directed evolution (12) to enhance its enantioselectivity. HRP is already a highly
active and versatile enzyme (13), thus raising its enatiopreference toward synthetically
useful substrates would increase its utility as an asymmetric biocatalyst.
HRP is comprised of a single polypeptide of 308 amino acid residues. There are
four disulfides, a buried salt bridge, and eight N-linked glycosylations.
HRP also
contains a heme prosthetic group and two calcium atoms, which are essential for the
functional and structural integrity of the enzyme (13). Although there is no X-ray crystal
structure of the plant HRP, there is a recently solved crystal structure (2.0 A resolution)
of the recombinant enzyme produced in Escherichia coli in a non-glycosylated form
(Figure 1.1) (14).
As a heme-dependent oxygenase, HRP uses hydrogen peroxide to oxidize many
diverse compounds.
investigated (15).
Its catalytic mechanism (Figure 1.2) has been extensively
HRP is particularly appealing as a practical biocatalyst due to
numerous asymmetric processes it can catalyze (16). Among other synthetically useful
reactions, HRP catalyzes the oxidation of a variety of chiral phenols, albeit typically with
low enantioselectivity (17).
Although the enantioselectivity of HRP was found to be markedly enhanced by
solvent composition and by the history or formulation of the enzyme sample (18-19),
there are few published reports of applying protein engineering to augment HRP's
enantioselectivity (20).
This lack of success stems from the difficulty of obtaining
recombinant HRP using standard expression systems (21-23).
This presents a
particularly significant challenge to improving HRP's enantioselectivity using directed
evolution, as the success of this protein engineering approach is highly dependent upon
the expression of many enzyme variants (12).
In Chapter II of my thesis, I describe an expression system whereby active HRP
variants are displayed on the surface of Saccharomyces cerevisiae yeast as fusion
proteins to yeast native surface protein Aga2p.
In addition, I demonstrate that the
enantioselectivity of HRP as well as its display levels can be readily measured and
characterized by analytical flow cytometry.
A major bottleneck to discovering enantioselective enzyme variants using
directed evolution has been the availability of a genuinely high-throughput screen or
selection method for enzymatic enantioselectivity (24). Typically, improved enzyme
variants are isolated using low- to medium-throughput screening techniques based on
agar plate or microplate assays, where enantioselectivity is evaluated by HPLC, gaschromatography, NMR, or mass spectroscopy (25). Despite the fact that these screening
technologies are amenable to automation, only a limited number of mutants, normally not
exceeding several thousand to several hundred thousand enzyme variants, can be
screened for enhanced enantioselectivity (26). In Chapter II, a high-throughput selection
method for HRP enantioselectivity, based on yeast surface display and fluorescenceactivated cell sorting (FACS), is presented. As shown in Chapter II, its application to a
library of some two million HRP clones yielded enzyme variants with markedly
improved enantioselectivities. Chapter III expands on the previous study and describes a
more efficient selection method with its throughput increased by at least 2-fold. As
reported in Chapter III, this ultra high-throughput selection method for enhanced HRP
enantioselectivity is validated by the discovery of enzyme variants with up to two orders
of magnitude higher selectivity toward either substrate enantiomer in just three rounds of
directed evolution.
Also, in Chapter III an improved expression system is described to obtain soluble
HRP from S. cerevisiae. These improvements dramatically elevated the secretion of HRP
from micrograms to milligrams of functional enzyme per liter of culture, which allowed
purification and further characterization of enantioselective HRP variants. The results of
these characterization studies are presented in Chapter III.
While the ability to rationally predict mutations that improve selectivity would be
of great value in the rational design of highly enantioselective enzymes, insufficient
mechanistic details governing enzymatic enantioselectivity limit such approaches.
To this end, I aimed to elucidate how a single-point mutation in HRP markedly enhances
its enantioselectivity by some 20-fold. The results of this investigation, presented in
Chapter IV, suggest that molecular modeling in combination with in vitro kinetic assays
and substrate analog studies can provide useful mechanistic insights into explaining
enzymatic enantioselectivity.
For the convenience of the reader, each research chapter contains its own
introduction, results and discussion, methods, and reference sections. Note that each of
the following chapters has resulted in a publication: Chapter II in Chem. Biol. 2007, 14,
1 (2007); Chapter III in Proc. NatL. Acad. Sci. U.S.A. 2008, 105, 17694; and Chapter IV
has been submitted to J. Am. Chem. Soc.
Figure 1.1.
Three-dimensional representation of the X-ray crystal structure of HRP
(PDB: 1ATJ). HRP's backbone is shown in ribbon with the heme moiety in red and
calcium ions in blue.
11
HO
I
H
NH 3
0
O
H
I
k2
.
N
,OH
0 IV
IV+
Fe
compound I
-
Fe
-
compound II
HO
H
H20
H20 2
Fe
resting state
'+
H20
H2
+
NH3
OH
Figure 1.2. The catalytic cycle of HRP with tyrosinol as a reducing substrate. Hydrogen
peroxide (H20 2 ) initiates the peroxidase catalytic cycle by a rapid two-electron oxidation
of the ferric resting-state HRP to compound I, which is a porphyrin-7t-cation radical.
Two successive single-electron transfers from tyrosinol reduce compound I first to
compound II and then back to the resting state. Both of these electron transfer steps yield
highly reactive phenoxy radicals. The rate constants ki and k2 represent the rate of
compound I formation and reduction, respectively; k3 represent the rate of compound II
reduction.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Agranat I, Caner H, Caldwell J (2002) Putting chirality to work: The strategy of
chiral switches. Nat Rev DrugDiscov 1:753-768.
Rouhi M (2004) Chiral chemistry: Traditional methods thrive despite numerous
hurdles, including tough luck, slow commercialization of catalytic processes.
Chem Eng News 82:47-62.
Stinson SC (2000) Chiral drugs. Chem Eng News 78:55-78.
Stinson SC (1999) Chiral drugs: 101. Chem Eng News 77:101-120.
Jacobsen EN, et al. (1999) Comprehensive Asymmetric Catalysis (Springer,
Berlin).
Noyori R (2002) Asymmetric catalysis: Science and opportunities (Nobel lecture).
Angew Chem Int Ed 41:2008-2022.
List B (2002) Proline-catalyzed asymmetric reactions. Tetrahedron 58:55735590.
Drauz K, Waldmann H (2002) Enzyme Catalysis in Organic Synthesis: A
Comprehensive Handbook (Weinheim: VCH).
Rouhi M (2002) Chiral roundup. Chem Eng News 80:43-50.
Pollard DJ, Woodley J (2006) Biocatalysis for pharmaceuticals intermediates: The
future is now. Trends Biotechnol 25:66-73.
Panke S, Wubbolts M (2005) Advances in biocatalytic synthesis of
pharmaceutical intermediates. Curr Opin Chem Biol 9:188-194.
Arnold FH (2001) Combinatorial and computational challenges for biocatalyst
design. Nature 409:253-257.
Veitch NC (2004) Horseradish peroxidase: A modern view of a classical enzyme.
Phytochemistry 65:240-259.
Henriksen A, Smith AT, Gajhede M (1999) The structures of the horseradish
peroxidase C-ferulic acid complex and the ternary complex with cyanide suggest
how peroxidases oxidize small phenolic substrates. J Biol Chem 274:3500535011.
Veitch NC, Smith AT (2000) Horseradish peroxidase. Adv Inorg Chem 51:107162.
van Deurzen MPJ, van Rantwijk F, Sheldon RA (1997) Selective oxidations
catalyzed by peroxidases. Tetrahedron53:13183-13220.
Gilabert MA, et al. (2004) Stereospecificity of horseradish peroxidise. J Biol
Chem 385:1177-1184.
Xie Y, Das PK, Caaviero JMM, Klibanov AM (2002) Unexpectedly enhanced
stereoselectivity of peroxidase-catalyzed sulfoxidation in branched alcohols.
Biotechnol Bioeng 79:105-111.
Yu JH, Klibanov AM (2006) Co-lyophilization with D-proline greatly enhances
peroxidase's stereoselectivity in non-aqueous medium. Biotechnol Lett 28:555558.
Ozaki S, Ortiz de Montellano PR (1994) Molecular engineering of horseradish
peroxidase. Highly enantioselective sulfoxidation of aryl alkyl sulfides by the
Phe-41--Leu mutant. JAm Chem Soc 116:4487-4488.
21.
22.
23.
24.
25.
26.
Smith AT, et al. (1990) Expression of a synthetic gene for horseradish peroxidase
C in Escherichiacoli and folding and activation of the recombinant enzyme with
Ca2 + and heme. JBiol Chem 265:13335-133343.
Morawski B, Lin Z, Cirino P, Joo H, Bandara G, Arnold FH (2000) Functional
expression of horseradish peroxidase in Saccharomyces cerevisiae and Pichia
pastoris.ProteinEng 13:377-384.
Levin G, Mendive G, Targovnik HM, Cascone 0, Miranda MV (2005)
Genetically engineered horseradish peroxidise for facilitated purification from
baculovirus cultures by cation-exchange chromatography. J Biotechnol 118:363369.
Reetz MT (2006) Directed evolution of enantioselective enzymes as catalysts for
organic synthesis. Adv Catal 49:1-69.
Reetz MT (2003) Select protocols of high-throughput ee-screening systems for
assaying enantioselective enzymes. Meth Mol Biol 230:283-290.
Lye GJ, Ayazi-Shamlou P, Baganz F, Dalby PA, Woodley JM (2003) Accelerated
design of bioconversion processes using automated microscale processing
techniques. Trends Biotechnol 21:29-37.
Chapter II: Selection of horseradish peroxidase variants with enhanced
enantioselectivity by yeast surface display
A. Introduction
Enzymes are attractive catalysts for applications in organic chemistry, primarily due to
their exquisite stereospecificity, and especially the ability to recognize and produce
particular enantiomers of chiral molecules (1). Because only a small fraction of reactions
of interest to synthetic chemists are catalyzed by naturally evolved enzymes, recent years
have seen a major effort to create enzymes with altered activity, usually by reengineering existing enzymes.
Directed evolution has proven to be a particularly
powerful approach to engineering enzymes (2, 3), as well as other proteins (4), with
improved properties, because it does not require a detailed knowledge of protein structure
and function. Instead, large libraries of proteins with varying sequence, structure, and
function are created, followed by screening and selection for variants with desired
properties.
The major challenge in directed evolution of enzymes is the creation of a stable
linkage between genotype (the DNA encoding a particular enzyme variant) and
phenotype (enzymatic activity).
The most direct approach, which separates different
enzyme variants in wells of microtiter plates, can be applied to almost any enzyme but
limits the throughput of screening to 103-104 variants per library (5). The alternative, invitro selection using display technologies can process libraries of 106-1010 variants but
limits the types of enzymes and chemical reactions that can be explored (6). An indirect
approach to in-vitro selection for enzymatic activity uses binding to transition-state
analogs as indication of potential activity (7).
Selection methods that test the
performance of variants in actual enzymatic reactions generally require that the reaction
product be trapped on the surface of phage or bacterial cell or inside bacterial cells.
Alternatively, water-in-oil emulsion droplets have been used to co-localize reaction
products with the bacteria that produce each enzyme variant (8). To date, all selection
methods applied to enzymes that act on small molecules rely on bacterial or in vitro
expression of enzyme variants, thus precluding directed evolution of numerous
eukaryotic enzymes with extensive disulfide bonding or post-translational modification.
Whereas the advantages of a selection system based on a eukaryotic organism for the
evolution of such enzymes are clear, the only example of a selection for enzymatic
activity in a eukaryote described so far involves a negative selection of homing
endonucleases unable to cleave their DNA targets in yeast (9). We set out to extend this
work by designing a yeast-based selection system that can be applied to enzymes not
involved in the biology of the yeast cell.
We report a new method for in vitro selection of enzymatic activity from large
libraries of variants displayed on the surface of the yeast Saccharomyces cerevisiae and
separated by fluorescence-activated cell sorting (FACS). As previously demonstrated for
antibody fragments (10, 11), extracellular receptor domains (12, 13), and a eukaryotic
lipase in a high-throughput screen (14), yeast surface display is well suited to eukaryotic
proteins. We used the model enzyme horseradish peroxidase (HRP), which contains four
disulfide bonds, as well as a heme prosthetic group, and cannot be expressed in a soluble
form in bacteria (15).
Due to their potential applications in synthetic chemistry, HRP and other
peroxidases have been subjected to directed evolution using random or directed
mutagenesis, DNA shuffling, and high-throughput screening to identify mutants with
higher thermostability (16, 17) or altered specificity (18). In addition, a bacterial catalase
has been mutated and screened (19), and a catalytic antibody raised against a transitionstate analog has been discovered by phage display (20) for increased peroxidase activity.
We focused on the enantioselectivity of HRP during its catalysis of radical
dimerization of two chiral phenolic substrates-tyrosinol supplied in solution and
tyrosine found naturally on the yeast surface.
Wild-type HRP shows a marginal
preference for L-tyrosinol over D-tyrosinol conjugated to the Alexa Fluor*488 fluorescent
dye.
Using two separate selection strategies, we both enhanced and inverted HRP
enantioselectivity.
Enzymes with altered enantioselectivity have been engineered previously (21, 22)
by screening libraries made by either error-prone PCR (23), mutagenesis of specific
active-site residues (24-27), or a combination of randomization methods (28-30) and
DNA shuffling (31-33). Enantioselective variants selected contained between one and 11
mutations per gene; there were critical mutations found in, or near, the active site, as well
as at a distance from it (34).
To
identify
the
most
efficient
randomization
strategy to
manipulate
enantioselectivity of HRP, we started our selections with two different HRP-based
libraries. The first one was constructed by using error-prone PCR and introduced a range
of mutations throughout the HRP gene. The second library focused on five residues at, or
close to, the HRP active site, and was designed to sample all possible sequence
permutations at those five positions. We found that in selections for selectivity for L- and
for D-tyrosinol linked to the fluorescent dye, only the active-site-directed library yielded
variants with a significantly enhanced enantioselectivity.
While it is tempting to
conclude that the failure of the error-prone PCR library to yield variants with improved
enantioselectivity is due to the requirement for simultaneous mutation of at least three
residues close to the active site, which would be a rare event in the error-prone PCR
library, other possibilities exist; for instance, uneven or insufficient coverage of the errorprone PCR library, or selectivity and detection issues in the screen could also account for
the current results.
.....
.....
...............
B. Results and Discussion
Yeast surface display of wild-type horseradish peroxidase
To demonstrate that HRP displayed on the yeast surface retains enzymatic activity, the
synthetic gene encoding the wild-type enzyme (Figure 2. 1A) was cloned into a pCTderived yeast surface display vector, pCT2, downstream from the gene for Aga2 protein
and upstream from a c-Myc tag (Figure 2.2), and transformed into S. cerevisiae. The
presence of HRP on the yeast surface was established by fluorescently labeling the cells
with antibody against c-Myc conjugated to Alexa Fluor*633 dye (A633) (Figure 2.2).
The enzymatic activity of surface-displayed HRP was confirmed by incubating the yeast
cells with two substrates, hydrogen peroxide and tyrosinol (shown in green below) linked
to a second fluorescent dye, Alexa Fluor*488 (A488).
So
s3
H2N
so;
H2
Ss
H2N
H2
00
00
H
NH
HO
HO
H
H
L-tyrosinol-A488
D-tyrosinol-A488
Due to HRP-catalyzed oxidation of tyrosinol-A488, the yeast cells can presumably be
labeled with the A488 fluorescent dye, whereby HRP-produced free radicals may react
with tyrosines of trans-membrane
surface proteins (Figure 2.3A).
After the
aforementioned incubation, the yeast cells were characterized using analytical flow
cytometry
(Figure
2.3B).
:.:::::..:..:::..:.::.:..:.:
............................................................................
A
0
1
L
T
P
T
T
Y
D
N
6
10
C
P
N V
3
V
I
V
R
D
T
I
V
N
9
L
1
3
D
P
TTCCACGACTGCTTTfTTAACGOTTGCCTCATTTAACAACAACATCGAAGAGAGGTTGACCAAAC-1
C X A N
F N D C F V M G C D A 8 t L L D 9 T T S F I T t X D A
41
A
A
3
1
R
L
1, 2
40
GGcG:.qqQGGATTTCCTGTGAT
S
A
R
4
P
F
V
1
70
60
50
I
M
30
20
0
Z&91
G
D
81
L
R
K
9
A
A
V
E
S5A A
G R
ft V,
S L
0
A
P
P
A
T
TCCGCAGACGCTCCTGCGCAAAg05TCTT?5
AAV
1 A A 0 0 3 V T L A G 0 7
3
V
C
A
T
TCTOCGA
S D L
V
A
L
G
S
0
L
D
L
A
M
A
9
L
8
L
0
T
L
T
R
G
L
C
P
T
A
P
P
T
r
T
L
P
0
L
K
V
G
N
S
L
A
L
V
D
V
U
R
P
V
120
D
V
3
R
N V G
L
N
R
S
160
ISO
Y
N
F S
9
T
G
L
P
D
V
T
L
N
r
D
L
R
T
P
T
I
F
D
9
M
Y
Y
V
V
L
T
E
E
Q
240
230
220
T
200
190
180
L
5
110
TTGAAACAECATATTACATAACCACTCCAACACTACT
F G 9 N 0 C R F I N D R L
210
201
L
140
170
161
L
D
100
130
121
V
C
90
AAAGTCTCA cAA~7AAACAGCCCAAGCCACTGACAAACCCACTATGAACATTTCTAATACACACAAACAm
X G L I Q S D 0 9 L F S S P N A T D T I P L V R 8 F A V S T 0 T F F N A F
270
260
250
241
tC
V E A
280
BanHZ
&AAGATGGGAAACATACACCCTTACGGAACAGACAAGTOAACTTA0G0TACCA??
V N S N 8S
I R L N C R
I T P L T G T 0 GQ
D R M GN
308
I00
290
281
B
Figure 2.1. Horseradish peroxidase. (A) Wild-type protein sequence and DNA sequence
used to construct HRP libraries.
Unique restriction-enzyme recognition sites used in
HRP cloning and library construction are underlined. The wild-type protein sequence is
shown in capital letters. Residues 68, 69, 72, 73, and 74, which were randomized to
make library HRP-C, are shown in red. (B) Crystal structure (PDB ID: 1HCH). The
HRP main chain and the side chains of disulfide-bonded cysteines are shown in grey.
Residues 68, 69, 72, 73, and 74, which were randomized to make library HRP-C, are
shown
in
green;
the
heme
prosthetic
20
group
is
shown
in
red.
.
c-Myc
Anching
protein
AgapAgp
HRPplamid
Yeast cell
Figure 2.2. Yeast surface display of HRP. HRP gene fused to Aga2 and c-Myc tag is
secreted from yeast and captured on the outside surface of the yeast cell through AgalpAga2p disulfide bonding. Antibodies against the c-Myc tag are used to label those yeast
cells
that display HRP
on their
surfaces
with
the A633
fluorescent
dye.
..
. .
...
.........
.
A
....
.
A48
HO
M
NH
I
HOA
N
OH
H
IOH
HO
L- or D-
H
-
I-
--
7v.. .so
YuSJSuu
104.
Negative control
D-tyrosinol
Wild-type HRP
D-tyrosinol
Wild-type HRP
L-tyrosinol
103.
02
0J 102.
U.
100
101
102
103
104 100
101
102
103 104 100
101
102
103
104
FL2 (expression)
Figure 2.3.
Analysis of enzymatic activity and expression of HRP displayed on the
surface of yeast. (A) Active HRP attaches its native substrate tagged with the A488
fluorescent dye (L or D-tyrosinol-A488) to acceptor substrate (tyrosine) on the yeast
surface. The cells are then labeled with anti-c-Myc antibodies conjugated to the A633
dye to determine HRP expression levels. (B) Analytical flow cytometry of negativecontrol yeast with no HRP gene and yeast displaying wild-type HRP labeled with L- and
D-tyrosinol-A488 (y-axis) and fluorescently labeled antibodies against the c-Myc epitope
tag (x-axis).
..
......
- ....
........
. ....
. .........
. .....
The presence of double-labeled cells in the yeast transformed with wild-type
HRP, but not in the negative-control yeast transformed with the same plasmid missing the
HRP gene, demonstrates that yeast-surface-displayed HRP can oxidize tyrosinol-A488.
This approach of detecting HRP activity, first proposed conceptually as "enzyme
screening by covalent attachment of products via enzyme display" by Becker and Kolmar
(6), is similar to the method used by Yin et al (20), who detected HRP-catalyzed
modification of a phage-displayed antibody fragment with biotinylated tyramine.
Whereas both selections rely on the incorporation of a phenolic substrate into protein
associated with a display particle, using a similar chemical reaction, the two approaches
differ in two significant ways. First, our use of a eukaryotic display organism allowed us
to study HRP, which cannot be expressed in an active form in bacteria used to express
protein in phage display. Second, our use of tyrosinol, a chiral substrate, also allowed us
to study the enantioselectivity of HRP.
As shown in Figure 2.3B and Table 2.1, L-tyrosinol-A488 appears to be
incorporated into yeast only slightly more efficiently than D-tyrosinol-A488, with the
enantioselectivity, E(L/D), of 1.2.
The fact that the subpopulation of yeast transformed with wild-type HRP with a
low A633 signal, and thus presumably a low level of HRP expression (35, 36), is still
labeled with tyrosinol-A488 (Figure 2.3B) suggests trans-labeling, i.e., that HRP
displayed on a yeast cell attaches the fluorescent substrate onto a different yeast cell. A
significant amount of trans-labeling would disturb the linkage between genotype and
phenotype and thus preclude selection for HRP enantioselectivity. We quantified the
amount of trans-labelingby labeling a mixture of a yeast strain displaying HRP and a C-
terminal c-Myc tag and a yeast strain displaying bovine trypsin inhibitor I (BPTI) and a
C-terminal Flag tag. Most of the cells (62%) expressing the c-Myc tag (and HRP) but
only 7% of the cells expressing the Flag tag (and BPTI) were labeled with the A488 dye.
In addition, as shown in Figure 2.3B, yeast transformed with wild-type HRP with a high
A633 signal, and thus a high level of HRP expression, incorporates more tyrosinol-A488
than the yeast with low-level expression of HRP; the amount of tyrosinol-A488
incorporated is roughly proportional to the level of expression. The combination of a low
number of trans-labeledcells and the high efficiency of cis-labeling by HRP-expressing
cells provides a high enough cis-to-trans (i.e., signal-to-noise) ratio to select new HRP
variants based on enzymatic activity by using yeast surface display.
Table 2.1. HRP variants selected from the active-site-directed library, HRP-C, with
enhanced L and D enantioselectivities
L enantioselective variants
E(L/D)
Frequency
HRP variant
Sequence
Wild-type
FGNANSA
CL8.01
LA..ELY
52%
9± 2
CL8.09
WA..AM.
17%
1.9 ± 0.1
CL8.02
.A..VVT
13%
3.3 ± 0.7
CL8.03
HA..ARD
13%
1.4 ±0.1
CL8.16
RH..WTT
4%
NA
1.2 ± 0.1
D enantioselective variants
E(D/L)
Frequency
HRP variant
Sequence
Wild-type
FGNANSA
CD8.02
EP..KA.
76%
3.4± 0.2
CD8.14
RP..HWT
10%
0.6 ±0.1
CD8.01
WV..FWS
5%
NA
CD8.07
MV..PMG
5%
NA
CD8.11
HS..GM.
5%
NA
0.9± 0.1
E(L/D) and E(D/L) are L and D enantioselectivities, respectively.
The sequence in the
randomized region (68-74) is shown, and the residues randomized are underlined. NA,
not active (less than 10% of wild-type HRP activity). All experiments were conducted in
duplicate with the mean and standard
deviation
values given in the table.
Construction of HRP-based libraries
We used two different HRP-based libraries to compare the effectiveness of two common
approaches to generating sequence variation in libraries for in vitro evolution: random vs.
active-site-directed mutagenesis.
The randomly mutagenized library, HRP-E, was generated by error-prone PCR
amplification of the entire HRP gene. Mutations could occur anywhere in the gene;
between zero and 17 DNA mutations per clone were observed in the sequences of 24
randomly chosen clones from the unscreened library, with a median of three.
The
perceived advantage of this approach is that it samples all possible types of mutations,
namely, (i) single, double, and multiple mutations; (ii) those in the active site and distal
from it; and (iii) those both affecting substrate binding and catalysis directly and through
subtle changes in enzyme structure.
The disadvantage of this approach is that any
physical library generated by random mutagenesis is only a small subset of all possible
libraries generated by this method, because it is impossible to sample all possible
permutations of multiple mutations for all but the shortest sequences.
Library HRP-E
contains approximately 1.6 x 106 unique sequences.
The
active-site-directed
library,
HRP-C,
was
generated
by
exhaustive
randomization of five positions at or near the active site: Phe-68, Gly-69, Asn-72, Ser73, and Ala-74 (Figure 2.1B), allowing any of the 19 non-Cys amino acid residues to
occur at each of the five positions. (Cysteine was excluded to avoid possible disruption
of folding, dimerization, or aggregation of HRP through unpaired cysteines under the
oxidizing conditions found in the yeast secretory apparatus.)
The limited number of
residues randomized allows an exhaustive sampling of the 2.5 x 106 possible sequence
permutations. The proximity of the randomized sites to the active site (Figure 2.1 B)
ensures that many of the mutations will have a significant effect on enzyme activity;
however, it leads to the risk that the mutations may be too drastic to preserve activity.
To compare the effectiveness of the two strategies in creating enantioselective
HRP variants, we performed in vitro selections for substrate enantioselectivity of HRP
using libraries HRP-E and HRP-C in parallel, under the same conditions and selection
pressure, and then analyzed the most improved clones selected from each library.
Selection of enantioselective HRP variants
Each library underwent two parallel selections by FACS, one for enhanced L
enantioselectivity (E(L/D)) and one for enhanced D enantioselectivity (E(D/L)) with
alternating rounds of positive and negative selection. For example, the selection for
E(D/L) alternated between FACS of populations with the highest incorporation of Dtyrosinol-A488 (selection rounds 1, 3, 5, and 7; Figure 2.4A) and FACS of populations
with low incorporation of L-tyrosinol-A488 (selection rounds 2, 4, 6, and 8; Figure 2.4B).
Between 21 and 24 clones from each selected population were sequenced,
enantioselectivity of all of the clones that appeared in the selected population more than
once was determined (Figure 2.5, Tables 2.1 and 2.2).
.. ..
. ........
........
..
- -
-
.... ..........
- ---- ....
Round 6
L-tyrosinol
Round 5
D-tyrosino
A
LL
100
101
102
103
104100
101
102
103 104
FL2 (expression)
Figure 2.4. Selection for HRP enantioselectivity. (A and B) Populations shown were
selected from the active-site-directed library, HPR-C, for enantioselectivity for D- over Ltyrosinol-A488 (E(D/L)).
(A) Positive selection in round 5. The 0.5 % of HRP-
expressing cells with the highest D-tyrosinol-A488 signal was collected. (B) Negative
selection in round 6. The 90% of HRP-expressing cells with the lowest L-tyrosinol-A488
signal was collected.
. .
Wild-type HRP
A
CL8.01
B
150
25.
125
20
100
15
75
10
50
25
.a
0
0
1
2
3
1
4
L
14-
70
12-
60
10-
50
8-
40
6-
30
20
4-
10
2
Figure 2.5.
4
Rev73
CD8.02
04
0
3
t (min)
t (min)
c
2
0
1
3
2
t (min)
4
0
1
2
3
4
t (min)
Determination of enantioselectivity of wild-type HPR (A), CL8.01 (B),
CD8.02 (C), and Rev73 (D).
Enantioselectivity is defined as the ratio of the initial
oxidation rates of L-tyrosinol-A488 and D-tyrosinol-A488. The initial oxidation rates are
measured as temporal changes in the fluorescence of the yeast cells labeled with the
A488 dye as a result of the HRP enzymatic activity. Bullets and a solid line represent a
fluorescence signal of the HRP-displaying yeast cells labeled with D-tyrosinol-A488;
triangles and a dashed line correspond to the L-tyrosinol-A488 fluorescence signal. The
fluorescence signal is recorded by a flow cytometer as mean fluorescence units (MFU).
Table 2.2. HRP variants selected from a randomly mutagenized library, HRP-E,
with enhanced L and D enantioselectivities
L enantioselective variants
Mutations
HRP variant
Wild-type
Frequency
E(L/D)
74%
1.2 ±0.1
EL8.02
132V, G213D
17%
1.2 ±0.1
EL8.08
S126G
4%
ND
EL8.22
V235M
4%
ND
Frequency
E(D/L)
D enantioselective variants
Mutations
HRP variant
0.9± 0.1
Wild-type
ED8.05
Wild-type
ED8.02
ED8.01
ED8.04
ED8.06
ED8.07
ED8.08
ED8.10
ED8.11
ED8.16
ED8.17
ED8.18
S73L
G69A, S126G, A134T, P146S
L215R, T257S, N268S
N9S, 122L
S216N
P261A
E249G
R93G
F221L
N214S, N307D
S73T
S126G
19%
14%
14%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
0.8 ±0.1
0.9± 0.1
0.9 ± 0.1
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
E(L/D) and E(D/L) are L and D enantioselectivities, respectively. ND, not determined. All
experiments were conducted in duplicate with the mean and standard deviation values
given in the table.
Both selections for enhanced E(L/D) and E(D/L) from the active-site-directed
library, HRP-C (Table 2.1), yielded a single HRP variant that was enriched more than
any other selected clone, and whose enantioselectivity exceeded that of all other clones in
the selected population and that of wild-type HRP. Variant CD8.02, whose sequence was
found in 76% of the clones in the final population selected for enhanced E(D/L), has a
3.4-fold preference for D- over L-tyrosinol-A488, i.e., a 3.8-fold improvement over wildtype HRP. Similarly, clone CL8.01, whose sequence was found in 52% of the clones in
the final population selected for higher E(L/D), has a 9-fold preference for L- over Dtyrosinol-A488, which is a 7.5-fold improvement over wild-type HRP.
In contrast, the most highly represented clones selected from the randomly
mutagenized library, HRP-E (Table 2.2), for higher E(D/L), were ED8.05 and ED8.02,
which represented 19% and 14%, respectively, of selected clones; their enantioselectivity
was indistinguishable from that of the wild-type enzyme. Similarly, the selection from
the aforementioned library for higher E(L/D) produced no HRP variants with an improved
enantioselectivity than that of wild-type HRP; the sequence of the latter was found in
74% of the sequenced clones from that selected population.
In summary (Figure 2.6), the active-site-directed library yielded variants with
higher E(L/D) and E(D/L) values; however the error-prone PCR library failed to identify
any HRP variants with a significant change in enantioselectivity.
12
10
U1
A
B
10
8
8
S.
-j
64
6
2
2
0]
0
WtHRP EL8.02 CL8.03 CLS.09 CLS.02 CL8.01
Figure 2.6. L and
D
wtHRP ED8.05 CD8.14 CD8.02 Rev68
Rev73
enantioselectivities (A and B, respectively) of HRP variants selected
from the HRP-C and HRP-E libraries. All experiments were conducted in duplicate with
the mean and standard deviation values given in the table.
Mutational analysis of enantioselective variant CD8.02
The enantioselectivities of CD8.02-based mutants, in which one of the four mutations at a
time was reverted back to the wild-type sequence, are shown in Table 2.3. Two of the
mutants, Rev69 (P69G) and Rev72 (K72N), lost most of enzymatic activity which
Mutant Rev68 (E68F)
precluded accurate determination of their enantioselectivities.
remained sufficiently active, but its E(D/L) was half that for CD8.02 (1.7 compared to
3.4). In contrast, mutant Rev73 (A73S) had an even higher enantioselectivity than its
parent clone, E(D/L) = 5.5, which corresponds to a 6-fold improvement over wild-type
HRP.
That a single mutation from the selected CD8.02 sequence back to the wild-type
at positions 68, 69, or 72 abolishes activity or reduces enantioselectivity suggests that, in
the context of the selected CD8.02 sequence (including Ala-73), Glu-68, Pro-69, and
Lys-72 are all required for catalytic activity and high E(D/L).
Such a requirement for
three non-wild-type residues is one possible explanation for the failure of the error-prone
PCR library in this selection, but others exist as well.
Whereas error-prone PCR is
relatively efficient at sampling single and double mutations throughout the HRP gene, the
odds of generating a particular combination of three mutations at specific sites are low.
In contrast, the HRP-C library focused attention on and essentially enumerated all
combinations of mutations at positions 68, 69, 72, 73, and 74.
This complementary
strategy enabled the discovery of the highly enantioselective mutant CD8.02, which
requires three simultaneous changes from the wild-type sequence for its favorable
phenotype.
Table 2.3. Properties of single-site revertants of variant CD8.02
HRP variant
Sequence
E(D/L)
Wild-type
FGNANSA
0.9 ± 0.1
CD8.02
EP..KA.
3.4 ± 0.2
Rev68
.P..KA.
1.7
Rev69
E.. .KA.
NA
Rev72
EP...A.
NA
Rev73
EP..K..
5.5 ± 0.4
NA, not active (less than 10% of wild-type HRP activity).
0.1
The residues
randomized to generate the HRP-C library are underlined. Errors were derived
from two independent experiments.
It is not possible to generalize to other cases from this one example, but the relative
effectiveness of whole-gene error-prone methods that provide excellent coverage of
single and probably double mutants versus focused enumeration methods of all multiple
mutants at a small number of sites remains an open research question. Other strategies
not employed here are also possible. The efficiency of constructing multiple mutations
nearby in three-dimensional space using a focused procedure might be particularly useful
near the enzyme active and binding sites, where there may be a high level of
cooperativity.
For identifying more distributed and less cooperative mutations,
approaches similar to the error-prone PCR library utilized here might have an advantage,
particularly if single and double mutants can be combined to produce further
enhancements.
In principle, mutant Rev73, which had a higher enantioselectivity than CD8.02,
should have been encoded in the HRP-C library and isolated in the selection for enhanced
E(D/L). Two possible explanations for not selecting Rev73 from the library are that a 5fold over-sampling of the theoretical sequence space in the physical library was not
sufficient to include a copy of each possible sequence, and that Rev73 had other
properties that were selected against, such as a lower expression level in yeast.
Nevertheless, we expect that thoroughly sampled, active-site-directed libraries should
provide an advantage in in vitro evolution of activity for enzymes whose structure and
location of the active site are known. This hypothesis is supported by other enzyme
directed evolution studies (37).
Further directions for yeast-based in vitro evolution of enzyme activity
The use of FACS to capture variants of interest requires physical association of product
with the yeast cell that harbors the gene that codes for the enzyme variant. In this study,
we ensured the stability of this genotype-phenotype linkage by utilizing tyrosine, as a
substrate, which is ubiquitous on the surface of yeast. However, the use of yeast surface
display is not limited to the study of enzymes whose substrates are natural components of
the yeast cell wall. We propose that this method can be applied to other bimolecular
reactions by tethering one of the synthetic substrates to the surface of yeast (analogously
to the tyrosine naturally present on yeast cell wall in the HRP example), and by adding
the second substrate in solution (like tyrosinol-A488 in the HRP example).
A
generalizable method for covalently attaching a small molecule to yeast surface was
recently demonstrated for biotin, which was attached to the yeast surface through a PEG
linker with an NHS functional group (38). Furthermore, the HRP-catalyzed production
of free radicals subsequently captured by a cell may be a generic means for detecting the
reaction products of other enzymes that unmask a pro-substrate, which then serves as a
substrate for HRP (6).
Conclusion
To our knowledge, we present the first application of yeast surface display to in vitro
selection of altered enzymatic activity. The method immobilizes one of the reaction
substrates, as well as a library of enzyme variants, on the surface of live yeast cells. A
second, fluorescent substrate is then supplied in solution and is utilized by the active
enzyme variants to label those cells that express such active enzyme variants. Labeled
yeast cells are subsequently captured by FACS. The use of a eukaryotic organism to
display the enzyme under selection makes possible in vitro evolution of a number of
enzymes that cannot be expressed in a soluble and active form in bacteria, such as highly
disulfide-bonded enzymes.
We used a combination of positive and negative selections to identify variants of
HRP that are enantioselective for L- or D-tyrosinol-A488, and we succeeded at enhancing
and even reversing the enantioselectivity from that of the slight preference for the L
enantiomer shown by wild-type HRP to a substantial preference for the D enantiomer, a
4-fold change in enantioselectivity.
In a separate selection, we improved the
enantioselectivity by 8-fold for L-tyrosinol-A488 compared to wild-type HRP.
A
comparison of selections from two different HRP-based libraries revealed that an activesite-directed library yielded variants with a large change in enantioselectivity, whereas a
randomly mutagenized library failed to yield improved clones; this difference could be
due to the superior sampling of multiple mutations in the vicinity of the active site by the
active-site-directed library.
The immobilized substrate used in our selection was tyrosine, present naturally in
proteins associated with the yeast cell wall.
Owing to a recent development in
derivatization of the yeast surface with a variety of small molecules, the scope of enzyme
yeast surface display can be extended to using any nontoxic substrate that can be
conjugated to a standard linker.
C. Materials and Methods
Synthesis and cloning of wild-type HRP gene
The gene for wild-type HRP, redesigned to introduce a number of unique restriction sites
without altering the protein sequence (Figure 2.1A), was synthesized by GenScript
(Piscataway, NJ).
A new yeast surface display vector, pCTcon2, was derived from
plasmid pCTcon (10) by replacing the DNA encoding the (Gly-Gly-Gly-Ser) 3 linker with
a less repetitive DNA sequence (5'-GGTGGAGGAGGCTCTGGTGGAGGCGGTAGCG
GAGGCGGAGGGTCG-3'),
again
without mutating
the encoded
peptide-linker
sequence. The synthetic HRP gene and the pCTcon2 plasmid were digested with NheI
and BamHI, and the HRP gene was ligated into the BamHI-NheI backbone of pCTcon2.
The resulting plasmid, pCT2-HRP, was transformed into the yeast surface display strain
of S. cerevisiae,EBY100 (10).
Construction of the HRP-based library, HRP-E, by using error-prone PCR
Library HRP-E was made by amplifying the HRP insert in pCT2-HRP in the presence of
nucleotide analogues as described previously (10).
Co-transformation of EBY100 with
the BamHI-EcoRI backbone of pCT2con and the amplified, mutated HRP gene following
the published method (10) yielded a library of 1.6 x 106 clones in EBY1O.
DNA
sequencing of 24 library clones revealed 0-17 mutations per clone (at the nucleotide
level), with a median of three mutations per clone. Two of the 24 sequenced clones had
the wild-type HRP gene sequence.
Selection of the five active-site positions for randomization
Groups of five residue positions were chosen based on structural proximity to the active
site, and computational protein design techniques (the dead-end elimination (39) and A*
algorithms (40)) were used to determine allowed sequences for the wild-type backbone
structure within 15 kcal/mol of the wild-type energy, which corresponds to the free
energy of unfolding of an extremely stable protein (41). The pairwise energy function for
these calculations was the sum of van der Waals, solvent-accessible surface area (42),
and a Coulombic electrostatic term with a dielectric constant of four times the distance
between each pair of atoms (43).
These calculations highlighted multiple sets of
candidate positions that were calculated to allow many sequences. Further analysis of the
built structures shows that they did not fill the active site with heavy atoms or consist of
many charged residues. We then used three other metrics to choose the positions for
randomization: mutual information between positions in our sequence alignments, amino
acid frequency in known genes, and ease of synthesis. Mutual information was used to
seek interactive positions, which are of special interest. The mutual information (44)
between positions highlights pairs that might be structurally dependent on each other, and
therefore might be forced to mutate in unison. A sequence alignment of HRP genes was
taken from Pfam (45), and our wild-type HRP gene sequence was aligned by eye with the
most highly homologous of the 309 seed alignment sequences to create a sequence
alignment of 310 sequences.
Highly conserved positions in this alignment were
considered to be risky for randomization. We chose to mutate Phe-68, Gly-69, Asn-72,
Ser-73, and Ala-74, which the computational protein design techniques indicated would
allow mutation. These five residues had low conservation in our sequence alignment and
moderate mutual information; in addition, their close proximity in sequence made them
easy to modify with a single randomized oligonucleotide.
Construction of the HRP-based library, HRP-C, by using active-site-directed
saturation mutagenesis
Library HRP-C was constructed by replacing the BstBI-EagI fragment of the wild-type
HRP gene (Figure 2.1A) with a synthetic DNA fragment randomized at amino acid
positions 68, 69, 72, 73, and 74. The synthetic DNA fragment was assembled from two
defined oligonucleotides, cl (5'-GTTGTGACGCATCGATCTTGTTAGACAACA
CAACATCATTTCGAACAGAGAAAGATGCG-3') and c2 (5'-CTGCGCAGGATA
CAGTTCTTGGGCATGCACTCTCCACGGCCGCCTTCATTCTGTCAATCACAGGA
AATCCGC-3'), and one randomized oligonucleotide, rC (5'-CATCATTTCGAACAG
AGAAAGATGCG1 1AACGCA1 11 CGCGGATTTCCTGTGATTGACAGAATG-3',
where "1" stands for equimolar mixture of codons encoding for the 19 non-cysteine
amino acid residues. The triphosphoramidite codon mixture was purchased from Glen
Research (Sterling, VA), and the randomized oligonucleotide was synthesized manually
by Trilink (San Diego, CA).
The oligonucleotides were assembled using KOD Hot Start Polymerase
(Novagen, San Diego, CA).
First, 20 pmol of oligonuclotide c2 and 10 pmol of
oligonuclotide rC were combined with 1 U of KOD Hot Start Polymerase in 50 IL of
KOD mix (1 x KOD buffer, 0.2 mM dNTP mix, 1 mM MgSO 4 , 1 M betaine, and 3%
DMSO). The oligonucleotides were denatured for 2 min at 95"C; subjected to ten cycles
of 30 sec at 94"C, 30 sec at 58"C, and 1 min at 68"C; and, finally, incubated for 10 min at
68"C. Twenty pmol of cl was added to the mixture in 2 pL, and the thermocycling
program was repeated as described above. The resulting double-stranded DNA fragment
was ethanol precipitated, and 2 pg of the product were amplified 10-fold by limiting the
amounts of the PCR primers pci (5'-GTTGTGACGCATCGATCTTGTTAGAC-3')
pc2 (5'-CTGCGCAGGATACAGTTCTTGGGC-3'),
and
and by using 20 cycles of the
program described above with 30 U KOD Hot Start Polymerase in 1.5 mL of the KOD
mix. The amplified DNA fragment ("HRP-C insert") was again ethanol precipitated and
resuspended in ddH2 0 at 0.6 pg/pL.
The pCT2-HRP plasmid missing the BstBI-EagI fragment (Figure 2.1A) was
prepared by a sequence of restriction digests (EagI, BssHII, and BstBI), followed by
purification using Qiagen PCR purification columns and ethanol precipitation.
The
gapped pCT2-HRP plasmid and HRP-C insert, which overlapped in sequence with the
ends of the gapped pCT2-HRP plasmid by 41 nucleotides both upstream of the BstBI
restriction site and downstream of the EagI restriction site, were co-transformed into the
EBY100 cells following the established protocol (10). A total of 20 pg of the gapped
pCT2-HRP and 30 pg of the HRP-C insert were transformed into 1 mL electrocompetent
EBY100, yielding a yeast surface display library of about 9.0 x 107 independent
transformants, which is larger than the 2.5 x 106 possible sequence permutations
permitted by library design.
Of the 24 clones from the HRP-C library that were
sequenced, 22 conformed to the library design, and two showed protein truncations due
to frameshift mutations.
Synthesis of L- and D-tyrosinol-A488 substrates
Each enantiomer of tyrosinol (Sigma-Aldrich, St. Louis, MO) dissolved in 50 mM
sodium borate buffer, pH 8.6, was added in 10-fold molar excess to Alexa Fluor*488
succinimidyl ester (Molecular Probes/Invitrogen, Carlsbad, CA). The L and D mixtures
were stirred at room temperature for 3 h.
Each fluorescently tagged enantiomer of
tyrosinol was then purified by reverse-phase HPLC using a 9.4 x 250 mm, 5 pM Zorbax
Rx-C8 column (Agilent Technologies, Santa Clara, CA) with 0.1% TFA water as loading
buffer and 0.1% TFA acetonitrile as mobile phase. The product was eluted with a 30 min,
4 mL/min gradient of 10-30% mobile phase.
Labeling of HRP libraries displayed on the yeast surface
A yeast culture containing either 10 copies of each clone or 2 x 106 cells, whichever was
the larger, was induced at the cell density of 4 x 105 per mL by growing the culture in
90% SG-CAA, 10% SD-CAA, 3.6 mM 6-aminolevulinic acid, and 0.2 mM ferric citrate,
for 18 h at 30"C. (The positive effect of induction medium supplementation on HRP
activity and display levels is shown in Appendix A). Two million induced yeast cells
were washed with 1 mL of PBS containing 0.5% BSA, followed by 0.5 mL of PBS with
0.1% BSA. The cells were resuspended in 200 ptL of PBS containing 0.003% H2 0 2 and
15 ptM L- or D-tyrosinol-A488. Yeast populations for FACS were incubated for 30 min
at 30*C, whereas samples used to determine enantioselectivities of selected clones were
incubated for 2-8 min at room temperature. Labeling reactions were stopped by adding a
10-fold excess of PBS containing 0.5% BSA and 10 mM ascorbic acid, and washed by
0.5 mL of PBS with 0.1% BSA.
Samples for FACS analysis were then labeled as
previously described (10) with anti-c-Myc monoclonal antibody, 9E10 (Covance,
Princeton, NJ), and with goat anti-mouse Alexa-PE polyclonal antibodies, and then were
resuspended in 0.5 mL of PBS with 0.1% BSA.
Sorting of HRP libraries displayed on the yeast surface using FACS
Double-labeled yeast cells were sorted on a Becton Dickinson Aria high-speed cell sorter
(Franklin Lakes, NJ) with 488 nm and 635 nm lasers at the rates of 6,000-10,000 cells per
s. Gates were adjusted to collect the yeast cells positive for A633 signal that also had the
highest A488 signal (for the positive selection rounds 1, 3, 5, and 7) or the lowest A488
signal (for the negative selection rounds 2, 4, 6, and 8). Of the A633-positive cells, the
1% of the cells with the highest A488 signal was collected in round 1, whereas 0.5% of
the cells with the highest A488 signal was collected in rounds 3, 5, and 7. Conversely, of
the A633-positive cells, 3% of the cells with the highest A488 signal was excluded in
round 2, and 10% of the cells with the highest A488 signal was excluded in rounds 4, 6,
and 8. Selected cells were collected in 0.5 mL of SD-CAA, pH 4.5, containing 50 pg/mL
kanamycin, 100 U/mL penicillin G, and 200 U/mL streptomycin, then grown to
saturation in 5 mL of the same media with shaking at 30"C for 2 days before the cells
were induced and labeled for the next round of sorting.
Isolation of selected HRP variants
After eight rounds of selection, plasmid DNA was extracted from 1 mL of each saturated
culture using the Zymoprep Yeast Plasmid Miniprep Kit (Zymo Research, Orange, CA)
and transformed into E. coli XL1-Blue competent cells (Stratagene, La Jolla, CA).
Plasmids from 21-23 colonies from each selection were sequenced; those encoding
unique HRP variants were re-transformed into EBY100 for characterization of
enantioselectivity.
Characterization of HRP variants
To determine the enantioselectivities of selected HRP variants, yeast cells transformed
with each variant of interest were labeled in parallel with L- and D-tyrosinol-A488 as
described above for 0-4 min. Each time point sample was analyzed by using a Coulter
Epics XL flow cytometer (Fullerton, CA). The mean fluorescence of the A488 dye was
plotted against time to determine the initial reaction rates with each substrate (Figure
2.5), and the enantioselectivity was calculated as E(L/D) = (initial oxidation rate of Ltyrosinol-A488)/(initial oxidation rate of D-tyrosinol-A488) and E(D/L) = (initial
oxidation rate of D-tyrosinol-A488)/(initial oxidation rate of L-tyrosinol-A488). Clearly,
E(L/D) x E(D/L) = 1.
Mutagenesis of enantioselective variant CD8.02
The four non-wild-type residues in variant CD8.02-Glu-68, Pro-69, Lys-72, and Ala73-were
mutated back to wild-type HRP by using the same strategy as in the
construction of library HRP-C, except that oligonucleotides with defined sequences were
used instead of the randomized oligonucleotide, rC.
Mutant Rev68 (E68F) was
constructed by using oligonucleotide 5-'CATCATTTCGAACAGAGAAAGATGCGTTT
CCTAACGCAAAGGCAGCGCGCGGATTTCCTGTGATTGACAGAATG-3';
mutant
Rev69 (P69G) was constructed by using oligonucleotide 5'-CATCATTTCGAACAG
AGAAAGATGCGGAGGGGAACGCAAAGGCAGCGCGCGGATTTCCTGTGATTG
ACAGAATG-3'; mutant Rev72 (K72N) was constructed by using oligonucleotide 5'CATCATTTCGAACAGAGAAAGATGCGGAGCCTAACGCAAACGCAGC
GCGCGGATTTCCTGTGATTGACAGAATG-3';
and mutant Rev73
(A73S)
was
constructed by using oligonucleotide 5'-CATCATTTCGAACAGAGAAAGATG
CGGAGCCTAACGCAAAGTCGGCGCGCGGATTTCCTGTGATTGACAGAATG-3'.
D. References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Drauz K, Waldmann H (2002) Enzyme Catalysis in Organic Synthesis: A
Comprehensive Handbook, 2nd edition (Weinheim: VCH).
Cherry JR, Fidantsef AL (2003) Directed evolution of industrial enzymes: an
update. Curr Opin Biotechnol 14:438-443.
Bloom JD et al. (2005) Evolving strategies for enzyme engineering. Curr Opin
Struct Biol 15:447-452.
Schaffitzel C, Pluckthun A (2001) Protein-fold evolution in the test tube. Trends
Biochem Sci 26:577-579.
Aharoni A, Griffiths AD, Tawfik DS (2005) High-throughput screens and
selections of enzyme-encoding genes. Curr Opin Chem Biol 9:210-216.
Becker S, Schmoldt HU, Adams TM, Wilhelm S, Kolmar H (2004) Ultra-highthroughput screening based on cell-surface display and fluorescence-activated cell
sorting for the identification of novel biocatalysts. Curr Opin Biotechnol 15:323329.
Hilvert D (2000) Critical analysis of antibody catalysis. Annu Rev Biochem
69:751-793.
Aharoni A, Amitai G, Bernath K, Magdassi S, Tawfik DS (2005) Highthroughput screening of enzyme libraries: thiolactonases evolved by fluorescenceactivated sorting of single cells in emulsion compartments. Chem Biol 12:12811289.
Chames P, et al. (2005) In vivo selection of engineered homing endonucleases
using double-strand break induced homologous recombination. Nucleic Acids Res
33: e178.
Colby DW, et al. (2004) Engineering antibody affinity by yeast surface display.
Methods Enzymol 388:348-358.
Feldhaus MJ, Siegel RW (2004) Yeast display of antibody fragments: a discovery
and characterization platform. JImmunol Methods 290:69-80.
Kim YS, Bhandari R, Cochran JR, Kuriyan J, Wittrup KD (2006) Directed
evolution of the epidermal growth factor receptor extracellular domain for
expression in yeast. Proteins 62:1026-1035.
Shusta EV, Holler PD, Kieke MC, Kranz DM, Wittrup KD (2000) Directed
evolution of a stable scaffold for T-cell receptor engineering. Nat Biotechnol
18:754-759.
Shiraga S, Ishiguro M, Fukami H, Nakao M, Ueda M (2005) Creation of Rhizopus
oryzae lipase having a unique oxyanion hole by combinatorial mutagenesis in the
lid domain. Appl MicrobiolBiotechnol 68:779-785.
Veitch NC (2004) Horseradish peroxidase: a modem view of a classic enzyme.
Phytochemistry 65:249-259.
Morawski B, Quan S, Arnold FH (2001) Functional expression and stabilization
of horseradish peroxidase by directed evolution in Saccharomyces cerevisiae.
Biotechnol Bioeng 76:99-107.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
Cherry JR, et al. (1999) Directed evolution of a fungal peroxidase. Nat Biotechnol
17:379-384.
Iffland A, Tafelmeyer P, Saudan C, Johnsson K (2000) Directed molecular
evolution of cytochrome c peroxidase. Biochemistry 39:10790-10798.
Ni J, Sasaki Y, Tokuyama S, Sogabe A, Tahara Y (2002) Conversion of a typical
catalase from Bacillus sp. TE124 to a catalase-peroxidase by directed evolution. J
Biosci Bioeng 93:31-36.
Yin J, Mills JH, Schultz PG (2004) A catalysis-based selection for peroxidase
antibodies with increased activity. JAm Chem Soc 126:3006-3007.
Reetz MT (2004) Changing the enantioselectivity of enzymes by directed
evolution. Methods Enzymol 388:238-256.
Reetz MT (2004) Controlling the enantioselectivity of enzymes by directed
evolution: practical and theoretical ramifications. Proc Natl Acad Sci U S A
101:5716-5722.
Henke E, Bornscheuer UT (1999) Directed evolution of an esterase from
Pseudomonasfluorescens.Random mutagenesis by error-prone PCR or a mutator
strain and identification of mutants showing enhanced enantioselectivity by a
resorufin-based fluorescence assay. Biol Chem 380:1029-1033.
Koga Y, Kato K, Nakano H, Yamane T (2003) Inverting enantioselectivity of
Burkholderia cepacia KWI-56 lipase by combinatorial mutation and highthroughput screening using single-molecule PCR and in vitro expression. J Mol
Biol 331:585-592.
Park S, et al. (2005) Focusing mutations into the P. fluorescens esterase binding
site increases enantioselectivity more effectively than distant mutations. Chem
Biol 12:45-54.
Funke SA, et al. (2005) Combination of computational prescreening and
experimental library construction can accelerate enzyme optimization by directed
evolution. ProteinEng Des Sel 18:509-514.
Reetz MT, Wang LW, Bocola M (2006) Directed evolution of enantioselective
enzymes: iterative cycles of CASTing for probing protein-sequence space. Angew
Chem Int Ed Engl 45:1236-1241.
May 0, Nguyen PT, Arnold FH (2000) Inverting enantioselectivity by directed
evolution of hydantoinase for improved production of L-methionine. Nat
Biotechnol 18:317-320.
Liebeton K, et al. (2000). Directed evolution of an enantioselective lipase. Chem
Biol 7:709-718.
Horsman GP, Liu AM, Henke E, Bornscheuer UT, Kazlauskas RJ (2003)
Mutations in distant residues moderately increase the enantioselectivity of
Pseudomonas fluorescens esterase towards methyl 3bromo-2-methylpropanoate
and ethyl 3phenylbutyrate. Chemistry (Easton)9:1933-1939.
Peters MW, Meinhold P, Glieder A, Arnold FH (2003) Regio- and
enantioselective alkane hydroxylation with engineered cytochromes P450 BM-3.
JAm Chem Soc 125:13442-13450.
van Loo B, et al. (2004) Directed evolution of epoxide hydrolase from A.
radiobacter toward higher enantioselectivity by error-prone PCR and DNA
shuffling. Chem Biol 11:981-990.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
Kubo T, Peters MW, Meinhold P, Arnold FH (2006) Enantioselective epoxidation
of terminal alkenes to (R)- and (S)-epoxides by engineered cytochromes P450
BM-3. Chemistry 12:1216-1220.
Morley KL, Kazlauskas RJ (2005) Improving enzyme properties: when are closer
mutations better? Trends Biotechnol 23:231-237.
Shusta EV, Kieke MC, Parke E, Kranz DM, Wittrup KD (1999) Yeast
polypeptide fusion surface display levels predict thermal stability and soluble
secretion efficiency. JMol Biol 292:949-956.
Graff CP, Chester K, Begent R, Wittrup KD (2004) Directed evolution of an anticarcinoembryonic antigen scFv with a 4-day monovalent dissociation half-time at
37 'C. ProteinEng Des Sel 17:293-304.
Chica RA, Doucet N, Pelletier JN (2005) Semi-rational approaches to engineering
enzyme activity: combining the benefits of directed evolution and rational design.
Curr Opin Biotechnol 16:378-384.
Rakestraw JA, Baskaran AR, Wittrup KD (2006) A flow cytometric assay for
screening improved heterologous protein secretion in yeast. Biotechnol Prog
22:1200-1208.
Desmet J, De Maeyer M, Hazes B, Lasters I (1992) The dead-end elimination
theorem and its use in protein side-chain positioning. Nature 356:539-542.
Leach AR, Lemon AP (1998) Exploring the conformational space of protein side
chains using dead-end elimination and the A* algorithm. Proteins33:227-239.
Creighton TE (1984) Proteins: Structures and molecular properties (New York:
W. H. Freeman and Company).
Sitkoff D, Sharp KA, Honig B (1994) Accurate calculation of hydration freeenergies using macroscopic solvent models. JPhys Chem 98:1978-1988.
Warshel A, Levitt M (1976) Theoretical studies of enzymic reactions--dielectric,
electrostatic and steric stabilization of carbonium-ion in the reaction of lysozyme.
JMol Biol 103:227-249.
MacKay DJC (2003) Information theory, inference, and learning algorithms
(Cambridge University Press).
Bateman A, et al. (2000) The Pfam protein families database. Nucleic Acids Res
28:263-266.
Chapter III: Highly L and D enantioselective variants of horseradish
peroxidase discovered by an ultra high-throughput selection method
A. Introduction
There is an ever-growing demand for enantiopure chemical compounds, in particular for
new pharmaceuticals (1).
While enzymes, being chiral molecules, offer undeniable
benefits as asymmetric catalysts in organic synthesis, their enantioselectivity for desired
unnatural substrates is often insufficient for practical applications (2).
Improving
enzymatic enantioselectivity toward a given substrate is thus a practically important but
arduous task. Established strategies for achieving this goal without genetically modifying
the enzyme include solvent engineering (3), bioimprinting (4), optimization of reaction
conditions (5), and coupling biocatalysis with chemical catalysis (6).
Recently, much emphasis has been placed on protein engineering, particularly by
directed evolution, as an effective approach to create enzymes with improved properties
(7, 8). Although effective in improving such enzyme properties as catalytic activity and
thermal stability, this approach has been far less successful in evolving enzymes with
higher enantioselectivity (9). In particular, an efficient search of protein sequence space
with respect to enantioselectivity and development of high-throughput selection methods
for assaying enantioselectivity remain major challenges (10, 11). Consequently, enzyme
enantioselectivities achieved thus far using directed evolution typically have been quite
modest (11).
These problems are particularly severe for such a complex (albeit catalytically
powerful and versatile) enzyme as horseradish peroxidase (HRP).
50
Among other
reactions, HRP catalyzes oxidation of numerous phenols with hydrogen peroxide but
typically does that with low enantioselectivity (12). It contains multiple disulfides, Nlinked glycosylations, and a catalytically essential heme moiety, making the enzyme
refractory to expression in prokaryotes (13). Therefore, to screen large libraries of HRP
variants, a eukaryotic system, such as yeast, must be employed (14).
In the present study, we have developed and validated a highly efficient selection
method based on yeast surface display and fluorescence-activated cell sorting (FACS)
that has led to raising HRP's enantioselectivity up to two orders of magnitude toward
either substrate enantiomer at will. These marked improvements in enantioselectivity
have been demonstrated and rationalized for both the surface-bound and soluble
enzymes.
B. Results and Discussion
We recently demonstrated that the enantioselectivity of HRP displayed on the cell surface
of yeast can be readily determined using fluorescent phenolic substrates (15). Employing
this methodology in the present study, we determined the enantioselectivity of wild-type
HRP toward a representative chiral phenol, tyrosinol, linked to two different positional
isomers of the Alexa Fluor*488 fluorescent dye (1 and 2 in Figure 3.1).
so;:
H2N
so
O
sH2
H2N
sO0
NH2
cOO
Scoo
HN
o
HN
O
HO*
2
OH
coo
HO
H
OH
N
HO
HO
3
*
H
N
coo
HO
0
HO
4
Figure 3.1. Chemical structures of the reducing substrates used in the present study to
assess HRP's enantioselectivity (asterisks designate stereogenic centers); the tyrosinol
(chiral) portion is shown in green.
The enantioselectivity, E(L/D) (defined herein as the initial rate of the enzymatic
oxidation of the L enantiomer divided by that of the D enantiomer), of wild-type yeastsurface-bound HRP was negligible for both substrates: 1.6 ± 0.5 and 0.8 ± 0.1 for 1 and
2, respectively (the first two entries in Table 3.1), in agreement with that for other chiral
phenolic substrates (12). We then endeavored to enhance it toward both enantiomers of
the substrates 1 and 2 by means of directed evolution.
Despite some progress in the first main step of the directed evolution
methodology, creation of genetic diversity in the target gene in the form of gene libraries
(16), the development of an effective high-throughput selection method for enzymatic
enantioselectivity remains daunting (11, 17). While FACS has shown much promise as a
high-throughput selection method (18), it requires a stable link between genotype (the
DNA
encoding
a
particular
enzyme
variant)
and phenotype
(the
enzyme's
enantioselectivity) if the selection is to be carried out on the entire gene pool at once (19).
We reasoned that such a link could be created when two fluorescent enantiomeric
substrates are simultaneously oxidized by HRP that is displayed on the surface of yeast.
In this scheme, an enantiomeric pair of chiral phenolic HRP substrates is conjugated to
two different fluorescent dyes (Dye 1 and Dye 2 in Figure 3.2). The enzymatic oxidation
of these conjugates yields phenolic free radicals that are captured by the cell surface,
thereby creating yeast cells stained with two different colored dyes. The ratio of the
fluorescence intensities (Dye 1/Dye 2) of these cells should correlate to the enzyme's
enantioselectivity (to be exact, the enzyme's selectivity toward the same chiral fragment
of the substrate since the dyes are different), thus establishing the genotype-phenotype
link required for FACS analysis.
Table 3.1. Enantioselectivities of L and D selective yeast-bound HRP variants
toward 1 and 2 discovered in three rounds of directed evolution
HRP variant
E(L/D)b
Substrate
0.6
1.6
0.7
1.6
0.5
4
0.8
0.1
Wild-type
2.6
Wild-type
21
3
25
LIIIc
85
6
1.8
0.5
49
1
LIIIc
4.2
0.6
0.4
0.1
10
1
DIIId
0.8
0.3
2.4
0.8
0.3
DIIId
2.0
0.4
154± 4
0.013
aVL
0.4
0.003
and VD are the initial rates of oxidation of the L enantiomer and D enantiomer,
respectively, reported in Mean Fluorescence Units (MFU) per min. All experiments were
conducted at least in triplicate with the mean and standard deviation values given in the
table. See Methods for details.
bEnantioselectivity, E(L/D), is defined as VL/ VD. Note that E(L/D) x E(D/L) = 1.
cThe L selective variant discovered in three rounds of directed evolution.
dThe D selective variant discovered in three rounds of directed evolution.
I.........
..........
..............
...
.........
..
........
........
............
._. - - -
NH
NH
HO
NH
m_
NH
HO
HO
H
C.
COH
OH
OH
OH
HRP
HRP
Diffusion
H202
HO
HO
OH
OH
c-Myc
Tyr
SS
ss
Agalp
Yeast cell wall
OH
H
OH
c-Myc
Tyr
sII s
ss
OH
Agalp
Yeast ce wall
Figure 3.2. Schematic representation of the ultra high-throughput selection method for
yeast-surface-bound HRP variants with enhanced L or D enantioselectivity.
HRP,
expressed as a fusion protein to the c-Myc tag and Aga2p mating agglutinin protein, is
displayed on the yeast surface via disulfide bridges between the Aga2p and Agalp
proteins. Enzymatic oxidation of the L and D enantiomers of tyrosinol (shown in green)
conjugated to fluorescent dyes (Dye 1 and Dye 2) yields phenoxy radicals that then nonenzymatically react with Tyr residues of membrane-bound proteins; this reaction leads to
labeled cells with fluorescence intensity dependent upon the enantioselectivity of HRP.
The enzymatic activity is normalized via fluorescently labeled antibodies against the cMyc tag (magenta star). Multiparameter FACS is used to isolate cells with the highest
ratio of fluorescence intensities (Dye 1/Dye 2 or Dye 2/Dye 1) encoding L or D selective
HRP variants.
Furthermore, this two-color selection method affords a "dual selection" -
selecting
enzyme variants with reactivity for the desired enantiomer, while simultaneously
excluding those with reactivity toward the undesired enantiomer, a key attribute of an
ultra high-throughput selection method.
To test this idea, we covalently attached one enantiomer of tyrosinol to the Alexa
Fluor*488 dye (A488, obtained commercially as a mixture of two positional isomers) and
the other to the Alexa Fluor*647 dye (A647, used by us herein for screening purposes
only).
We then determined HRP's activity and selectivity toward these fluorescent
conjugates by incubating yeast cells displaying wild-type HRP on their surface with the
oxidizing substrate H2 0 2 and an equimolar mixture of L-tyrosinol conjugated to A488
and D-tyrosinol to A647, followed by FACS analysis. The resultant dual-parameter dot
plot of A488 (y-axis) vs. A647 (x-axis) fluorescence intensity exhibits a clustered signal
in the middle (Figure 3.3) indicating that (i) the enzyme was active toward both the
enantiomeric substrates and (ii) the cells were labeled with both dyes to an equal
intensity. The same pattern of fluorescence intensities (i.e., a clustered signal) was
observed with D-tyrosinol attached to A488 and L-tyrosinol to A647 (data not shown)
under otherwise identical conditions, indicating that the identity of the dye had little
effect on the enzyme's enantiopreference.
Consequently, using the aforementioned
substrate pairs along with FACS, one could screen HRP-based libraries and select HRP
variants with higher L or D enantioselectivity by isolating cells above or below the
diagonal cluster, respectively, as depicted schematically in Figure 3.3.
............
::::..::
......................
..............
104 ~
0
-
C
100
10
102
10
4
A647 Fluorescence
(D-tyrosinol-A647)
Figure 3.3. Multiparameter FACS analysis of surface-bound, wild-type HRP incubated
with L-1 + L-2, D-tyrosinol-A647, and H2 0 2 .
The regions outlined by trapezoids
schematically represent library sort gates, used to isolate L selective (cells with high
A488 and low A647 fluorescence) and D selective (cells with low A488 and high A647
fluorescence) HRP variants.
To validate our high-throughput selection method, in the first two rounds of
evolution we examined the effectiveness of two types of libraries, those produced by
active-site-targeted saturated mutagenesis and random mutagenesis, in creating both L
and D selective HRP variants. Each round of evolution included a round of mutagenesis
of the variant with highest enantioselectivity, followed by screening and selection. In the
third round of evolution, we used a library that was created by randomizing an area of the
gene at, or near, the enzyme active site. This modification of the experimental approach
was prompted by the findings of the first two rounds of evolution, namely that (i) the
mutations that impacted enantioselectivity discovered through random mutagenesis were
located close to the active site and (ii) the locations of these mutations did not seem to be
obvious targets for saturated mutagenesis, underscoring the difficulty in identifying
residues whose alteration would affect the enantioselectivity.
The foregoing libraries assayed by our selection method systematically yielded
HRP variants with enhanced L and D enantioselectivity for both substrates 1 and 2 (Figure
3.4). As seen in Figure 3.4A, the L enantioselectivity of surface-bound HRP variants
toward 1 improved steadily and markedly with each round of evolution giving rise to
E(L/D) values of 4.5
0.2, 29
1, and 49
1, respectively. Interestingly, the most
enantioselective variants produced by random mutagenesis in the 1st and 2nd rounds (LIr
and LIIr, respectively) both have single mutations near the active site with a high impact
on enantioselectivity. By modifying our strategy to create genetic diversity as explained
in the preceding paragraph, in the 3rd round of evolution we discovered a highly L
selective variant, LIII, with a total of eight mutations (listed in the legend to Figure 3.4).
-,
I'll
..
I.
-
I'll 11
.
I
I -
---
-
-
.-
-- I'll,
25-
'-
-
-
-
-
-
-
-
-
8
0
0
wt Lir
Lis rs LIOr Lill
wt Lir Lis LIrs LIlr Lill
80-
4-
D
C
40-
2B
0 , .I
wt
Dis
Dils
, 0- -
Dill
wt
Dis
Dlls
Dill
Figure 3.4. Enantioselectivities of L selective variants toward 1 and 3 (A) and 2 and 4
(B), as well as D selective variants toward 1 and 3 (C) and 2 and 4 (D), discovered in each
round of evolution (for numerical values, please see Appendix B). Red, yellow, blue, and
green bar colors represent substrates 1, 2, 3, and 4, respectively; hatched and solid bars
designate surface-bound and soluble HRP, respectively. The L and D letters designate
the direction of enantiopreference; the Roman numerals after the letters define the round
of directed evolution; the r and/or s letters after the Roman numerals indicate whether
these variants were isolated from the random or saturated mutagenesis libraries,
respectively.
Mutations: LIr (Argl78Gln), LIs (Phe68Leu, Gly69Ala, Asn72Glu,
Ser73Leu, Ala74Tyr) (14), LIrs (LIr + LIs), LIIr (LIrs + Glnl47Arg), LIII (LIIr +
Asnl58Asp), DIs (Phe68Glu, Gly69Pro, Asn72Lys) (14), DIs (DIs + Asnl37Arg,
Alal40His,
Phel42Lys,
Phel43Met),
59
and
DIII
(Dlls
+
Serl67Ile).
-
-
- I -- Aff-
Inspection of the initial rates of oxidation of 1 catalyzed by this variant reveals
that a high E(L/D) value, 49 + 1, was achieved by increasing the reactivity of the L
enantiomer (85 ± 6 MFU/min for LIII vs. 2.6 ± 0.6 MFU/min for the wild-type enzyme;
Table 3.1), whereas the oxidation rate of its D counterpart remained similarly low (1.8 ±
0.5 and 1.6 ± 0.7 MFU/min, respectively; Table 3.1). Likewise, the enantioselectivities
of the L selective HRP variants toward 2 rise steadily with each consecutive round of
evolution (Figure 3.4B). The LIII variant exhibits a 13-fold increase in E(L/D) toward 2
compared with the wild-type enzyme (Table 3.1).
However, because our selection
strategy centered on simultaneously isolating variants with high enantiopreference toward
both L-1 and L-2, the LIII variant has a lower E(L/D) value toward 2 compared to that of
LIr (Figure 3.4B).
We also discovered HRP variants whose D enantioselectivity rose consistently
with each round of evolution for both 1 (Figure 3.4C) and 2 (Figure 3.4D). The most
enantioselective variant identified, DIII, has an E(D/L) value of 77 ± 1 toward 2, i.e., a 64fold improvement compared with the wild-type enzyme (Table 3.1).
Like LIII, this
variant also has eight mutations, all near the active site (listed in the legend to Figure
3.4). As seen in Table 3.1, the enhancement in enantioselectivity of DIII toward 2 stems
from both a faster oxidation of the D enantiomer and a slower oxidation of the L
enantiomer. The enantioselectivity of DIII toward 1 is enhanced in the same way but to a
lesser degree (Table 3.1).
These findings validate the notion that our experimental
methodology affords a "dual selection", i.e., accelerating the evolution of a new function
while eliminating the native one. Moreover, our results differ favorably from those with
other enzymes evolved for higher enantioselectivity that typically exhibit lower specific
activities relative to their parents (20, 21). In contrast, all but one of our most L and D
selective variants exhibit higher specific activities than the wild-type enzyme (Table 3.1).
To determine (i) how the enantioselectivity of the discovered HRP variants
depends on cell-surface immobilization and (ii) the effect of the fluorescent dye on the
enantioselectivity of these variants, we expressed and purified the corresponding soluble
enzyme species. To maintain the glycosylation pattern of the surface-bound HRP, we
employed the same yeast host for heterologous expression of the soluble enzyme.
Because HRP is expressed poorly in this yeast (14, 22), we used a strain of S. cerevisiae
wherein the folding chaperones protein disulfide isomerase (PDI) and immunoglobulin
heavy-chain binding protein (BiP) are overexpressed. This strain, in combination with an
optimized induction medium (see Methods) and secretory leader sequence (23),
dramatically elevated the secretion of HRP from micrograms to several milligrams of
functional enzyme per liter of culture (Figure 3.5).
kDa
1
2
3
250 150 -
<--
glycosylated
HRP
50-*
4-. aglycosylated
37 -e
HRP
4-Endo H
25 -
w
Figure 3.5. Purified wild-type HRP from S. cerevisiae analyzed on a 12% SDS-PAGE.
Lane 1 contains protein standards (with their molecular masses shown in kDa). Lane 2
contains Endo H, an enzyme that removes N-linked glycosylation. Lanes 3 and 4 contain
purified yeast wild-type HRP and its Endo H-deglycosylated derivative, respectively.
The gel was stained with Coomassie Blue.
Following purification of the soluble enantioselective HRP variants, their E
values were measured. As seen in Figures 3.4A and 3.4B, the soluble wild-type enzyme
has the same low enantioselectivity as its predecessor displayed on the yeast surface:
E(L/D) values of 1.4 ± 0.1 and 0.7 ± 0.1 for 1 and 2, respectively.
Moreover, the
enantioselectivities of the soluble variants, as of the surface-bound ones, increase with
each round of directed evolution, although the E values of the soluble enzyme are several
times lower (Figure 3.4). This phenomenon is not uncommon (24) and consistent with
the basic rule of directed evolution "one gets what one selects for". Specifically, our
selection method was applied to the HRP enzyme fused to a large, highly glycosylated,
Aga2p-Agalp protein complex (a total of 1,150 amino acid residues) integrated into the
cell wall (as schematically depicted in Figure 3.2). The attachment of HRP to the Aga2pAgalp protein complex likely restricts the number of conformations that the enzyme can
adopt which may affect enantioselectivity. In contrast, in analyzing the enantioselectivity
of the soluble enzyme we used a much smaller protein consisting of HRP linked merely
to two affinity purification tags (a total of 330 amino acid residues).
To elucidate the role of the fluorescent dye portions of 1 and 2 in the
enantioselectivities of the discovered HRP variants, we measured the E values of the
soluble wild-type HRP, as well as LIII and DIII, with tyrosinol. As seen in Table 3.2, the
LIII variant, highly enantioselective toward 1 and 2, is one-half as enantioselective with
tyrosinol as the wild-type enzyme. On the other hand, DIII, the variant with the highest
preference for D-1 and D-2, is also D enantioselective with tyrosinol (Table 3.2).
Table 3.2. Enantioselectivities of soluble HRP variants toward tyrosinol and Nacetyl-tyrosinol
E(L/D)a
HRP variant
tyrosinol
N-acetyl-tyrosinol
Wild-type
5.3
0.4
2.7
0.1
LIIIb
2.7
0.1
0.8
0.1
DIIIc
0.4
0.1
1.7
0.1
aSee
footnote b to Table 3.1.
bSee
footnote c to Table 3.1.
cSee footnote d to Table 3.1.
As seen in Table 3.2, however, LIII and DIII exhibit inverted enantioselectivities
toward N-acetyl-tyrosinol compared to tyrosinol suggesting the influence of the positive
charge of the substrate. Therefore, attaching the fluorescent dye to tyrosinol plays an
important role in determining the enantioselectivity of the discovered HRP variants, once
again confirming that "one gets what one selects for".
Separately, we uncovered significant differences in enantioselectivities of the
surface-bound enzyme depending on the structural isomer of the dye attached to tyrosinol
(1 and 2). For example, DIII is far more enantioselective toward 2 than toward 1 (E(D/L)
of 77 ± 1 vs. 3.1 ± 0.5), whereas LIII, on the contrary, strongly prefers 1 to 2 (E(L/D) of
49 ± 1 vs. 10 ± 1) (Table 3.1).
The same trends hold for their soluble counterparts,
although the differences in enantioselectivities are more modest (Figure 3.4).
As is
evident from Figure 3.1, the structural difference between 1 and 2 arises from the
attachment point of tyrosinol: particularly, it is five carbon atoms away from the carboxyl
group of the benzoate moiety in 1, whereas in 2 it is four carbons away. Hence one
explanation of the difference in E values for the regioisomers might be the location of the
negatively charged carboxyl group with respect to the stereogenic center.
It is also
possible, however, that the fused phenyl rings of the A488 dye play a role in
enantioselectivity.
To
distinguish
between
these
two
alternatives,
we
measured
the
enantioselectivities of the soluble HRP variants toward analogs of 1 and 2 that lack the
fused phenyl rings of the dye but still retain the benzoate group (3 and 4, respectively, in
Figure 3.1).
(Our experimental methodology does not allow characterization of the
surface-bound enzyme using these non-fluorescent substrates.) The wild-type enzyme
exhibits no enantiodiscrimination with either 3 or 4, as evidenced by the E values of unity
for both regioisomers (Figures 3.4A and 3.4B). However, as with the substrates 1 and 2,
the enantioselectivity toward 3 and 4 is enhanced with each round of evolution for both
the L and D selective variants (Figure 3.4). Furthermore, the L selective variants are more
enantioselective with 3 than with 4, while the opposite is true for the D selective ones,
consistent with the data obtained for the corresponding variants with 1 and 2 (Figure 3.4).
These results point to the disposition of the carboxyl group as the main determinant in the
enantioselectivity of the discovered variants.
To examine how the location of the negatively charged carboxylate vis-i-vis the
stereogenic center can give rise to vastly different enantioselectivities of the isolated
variants with 3 and 4 (used instead of 1 and 2, respectively, due to their much simpler
structures), we employed molecular modeling to simulate complexes of these
regioisomers with the wild-type enzyme.
As depicted in Figure 3.6, the L and D
enantiomers of both 3 and 4 exhibit distinct binding modes: in particular, the locations of
their carboxyl groups in the active site differ for each enantiomer. The most striking
difference in enantioselectivity between 3 and 4 is seen with the Us variant, which is
marginally L selective with 3 (E(L/D) = 1.4 ± 0.1; Figure 3.4A) but highly D selective
with 4 (E(L/D) = 0.1 ± 0.3; Figure 3.4B). Of the five mutations of the Us variant, one
seems particularly influential, namely Asn72Glu. Inspection of Figures 3.6A and 3.6C
reveals that the Asn72 residue is located close to the carboxyl group of D-4, and even
closer to that of D-3 (but not for their L counterparts): the distance between that amino
acid residue's amide nitrogen and the carboxylate's oxygens is 5.07 and 4.25
A,
respectively. Therefore, replacing Asn72 with the negatively charged Glu should weaken
the binding of D-3 to the enzyme in the transition state due to electrostatic repulsion,
leading to a slower oxidation of D-3 and, in turn, imparting L enantioselectivity.
Furthermore, D-4's carboxylate is closer to Arg-178 than D-3's (Figures 3.6A and 3.6C).
Therefore, the Asn72Glu mutation is more likely to electrostatically repel the D-4
substrate into a new orientation, resulting in a salt bridge with Arg-178 in the transition
state, thus enhancing the binding affinity of D-4 and making the Us variant highly D
selective with 4. If this hypothesis is correct, the Us variant should become L selective
with 4 instead when the putative salt bridge is eliminated. Indeed, when the Argl78Gln
mutation is introduced, the resultant LIrs variant becomes L selective with 4 (E(L/D) = 5.0
0.2), as well as with 3 (E(L/D) = 9.2 ± 0.2) (Figures 3.4A and 3.4B).
In the case of the D selective variants, the differences in enantioselectivities with 3
and 4 are more modest compared to the L selective ones: e.g., for DIII the E(D/L) values
are 3.7 ± 0.1 with 3 and 5.2 ± 0.1 with 4 (Figures 3.4C and 3.4D). Nevertheless, the
enantioselectivity enhancement in these variants still could be attributed to the position of
the substrate's carboxylate with respect to the enzyme mutations. For example, the DIs
variant has three mutations, including Phe68Glu and Asn72Lys. The latter one is likely
to accelerate the oxidation of D-3 and D-4 by increasing their binding affinities for the
enzyme in the transition state through either the establishment of a salt bridge or
hydrogen bonding between the positively charged Lys and the carboxylate of D-3 (Figure
3.6A) or D-4 (Figure 3.6C). On the other hand, the Phe68Glu mutation may lower the
oxidation rate of the L enantiomers of 3 and 4 due to electrostatic repulsion of the
negatively charged carboxyl group (Figures 3.6B and 3.6D, respectively).
A
1
Asn72
n72
Arg18
Arg178
C
Asn72
Asn72
~*Aj9
178
Figure 3.6. Modeled complexes of wild-type HRP with D-3 (A), L-3 (B), D-4 (C), and L4 (D). For clarity, only the active site of the enzyme is shown with the heme moiety in
orange, substrate in blue, and some mutated residues in green. Distances indicated are in
A. The Arg-178 residue is shown in a double rotamer configuration as it appears in the
crystal structure (29); only one rotamer configuration was used in docking experiments.
See Methods for details of how these models were built.
In closing, we have developed an ultra high-throughput selection method for
enzyme enantioselectivity, based on yeast cell surface display paired with FACS,
validated by discovering highly enantioselective variants of HRP toward tyrosinol
conjugated to the A488 fluorescent dye. We have found that the enantioselectivity of the
isolated HRP variants depends upon the attachment of tyrosinol to the benzoate moiety of
this dye and specifically on the position of the carboxylate. The discovered HRP variants
are several times more enantioselective when bound to the cell surface than when
solubilized.
Such surface-bound enzymes with enhanced enantioselectivity toward
commercially useful substrates may be used as naturally immobilized asymmetric
biocatalysts in chemical reactors.
C. Materials and Methods
Syntheses
Tyrosinol-A488 (1, 2). The mixture of 1 and 2 for each enantiomer was synthesized as
previously described (15). 1 and 2 were separated by reverse-phase HPLC by using a 9.4
x 250 mm 5 ptM Zorbax Rx-C8 column (Agilent Technologies, Santa Clara, CA) with
100 mM triethylammonium acetate buffer (pH 7.0) (Calbiochem, San Diego, CA) as a
loading buffer and acetonitrile as a mobile phase. The products were eluted with a 30min, 4 mL/min gradient of 10-20% (v/v) acetonitrile with retention times of 9 and 13 min
for 1 and 2, respectively. Each product then underwent a second purification using the
same conditions. The identity 1 and 2 was confirmed by electrospray ionization (ESI)MS. All chemicals from here onward were from Sigma-Aldrich Chemical Co. (St. Louis,
MO), unless stated otherwise, and were of the highest purity available from that vendor.
Tyrosinol-A647. Each enantiomer of tyrosinol dissolved in 50 mM Na borate buffer (pH
8.6) was added in 10-fold molar excess to Alexa Fluor*647 carboxylic acid succinimidyl
ester (Invitrogen, Carlsbad, CA). The mixture was stirred at overnight room temperature,
and the product was purified as described in the preceding paragraph, except that it was
eluted with a 40-min, 4 mL/min gradient of 15-19% (v/v) acetonitrile.
3 and 4 were prepared by reacting L- or D-tyrosinol with mono-methyl isophthalate or
terephthalate, followed by ester hydrolysis.
In all cases, a solution of mono-methyl
phthalate (90 mg, 0.5 mmol), 4-DMAP (43 mg, 0.35 mmol), and EDC HCl (115 mg, 0.6
mmol) in DMF (20 mL) was incubated at room temperature for 30 min and then added
dropwise to a solution of L- or D-tyrosinol HCl (204 mg, 1.0 mmol) and triethylamine (2.5
mmol) in DMF (10 mL). The resulting reaction mixture was stirred overnight at room
temperature, evaporated, and re-dissolved in 0.1 M HCl (20 mL).
The mixture was
extracted with ethyl acetate (3 x 60 mL), washed with saturated aqueous NaHCO 3 (3 x 60
mL), brine (60 mL), dried over anhydrous Na 2SO 4 , and evaporated. The resulting residue
was treated with 5 mL of 0.4 M NaOH in tetrahydrofuran/water (3:1, v/v) for 2 h. After
the removal of tetrahydrofuran by evaporation, the solution was acidified to pH -2.0 and
extracted with ethyl acetate (3 x 5 mL). The combined organic fractions were dried over
anhydrous Na 2SO 4 and evaporated to give crude product. The pure product was then
obtained by recrystallization from water. For 3: 'H NMR (300 MHz, CD 3 0D) 3 2.72
(dd, J= 7.6, 13.6, 1H), 2.92 (dd, J= 6.5, 13.6, 1H), 3.62 (m, 2H), 4.24 (m, lH), 6.70 (d, J
= 8.4, 2H), 7.10 (d, J= 8.4, 2H), 7.53 (dd, J= 7.8, 7.8, 1H), 7.96 (ddd, J= 7.8, 1.8, 1.2,
1H), 8.14 (ddd, J= 7.8, 1.5, 1.4, 1H), 8.43 (dd, J= 1.7, 1.7, 1H). For 4: 1H NMR (300
MHz, CD 30D) 3 2.77 (dd, J= 8.1, 13.8, 1H), 2.92 (dd, J= 6.3, 13.8, 1H), 3.64 (m, 2H),
4.29 (m, 1H), 6.70 (d, J= 8.0, 2H), 7.10 (d, J= 8.7, 2H), 7.81 (d, J= 8,7, 2H), 8.08 (d, J
=
8.7, 2H).
N-Acetyl-tyrosinol was synthesized as described in the literature (25). To a mixture of
tyrosinol HCl (250 mg, 1.23 mmol) and triethylamine (3.0 mmol) in 10 mL of dry ethyl
acetate on ice, acetyl chloride (1.5 mmol) in 10 mL of dry ethyl acetate was added
dropwise with stirring. After the addition was completed (~30 min), the reaction mixture
was incubated for 2 h on ice. The white precipitate of triethylamine HCl was filtered and
washed with ethyl acetate (2 x 20 mL). To remove unreacted tyrosinol, the combined
filtrates were then washed with 0.1 M HCl (3 x 60 mL), brine (60 mL), dried over
anhydrous Na 2 SO 4 , and evaporated. The purity of the product was confirmed by reversephase HPLC and determined to be greater than 95%.
Enantioselectivity of yeast surface-bound HRP
The initial rates of substrate oxidation with hydrogen peroxide catalyzed by surfacebound HRP were measured by suspending 1x106 HPR-displaying yeast cells in 100 tL of
PBS buffer (pH 7.4) containing 15 pM 1 or 2 and 150 pM H2 0 2 , in parallel for both
enantiomers. Periodically, 20 ptL of the L and D substrate mixtures were withdrawn into
1 mL of PBS containing 0.5% BSA and 10 mM ascorbic acid to quench the reactions.
The fluorescently labeled cells were then washed with 0.5 mL of PBS containing 0.1%
BSA and labeled with mouse anti-c-Myc monoclonal 9E10 (Covance, Princeton, NJ) and
phycoerythrin-goat anti-mouse antibodies (Sigma), as described previously (26).
The
cells with the same HRP display levels were then analyzed using a Coulter (Fullerton,
CA) Epics XL flow cytometer. The mean fluorescence of Alexa Fluor*488 (MFU) for
each enantiomer was plotted as a function of time to determine initial reaction rates.
Construction of HRP libraries
Random mutagenesis libraries were created using a protocol adapted from ref. 26.
Briefly, for both D and L selective enzyme libraries, the HRP gene coding for the best
variant was amplified in the presence of nucleotide analogs using forward (5'GGTGGAGGAGGCTCTGGTGGAGGCGGTAGCGGAGGCGGAG GGTCGGCTAGC
-3')
and
reverse
(5'-CAGATCTCGAGCTATTACAAGTCCTCTTC
AGAAATAAGCTTTTGTTCGGATCC-3')
primers (IDT, Coralville, IA) under the
following conditions. A mixture of 2 ng of pCT2-HRP plasmid (15), 1 pM each reverse
and forward primers, 0.2 mM dNTPs, 2 mM MgCl 2 , 2 pM 8-oxo-dGTP, 2 pM dPTP, and
0.05 U p.L Taq polymerase in 1x Taq polymerase reaction buffer was subjected to a
thermocycling program which comprised of 1 min at 95 'C, followed by 15 cycles of 1
min at 94 *C, 30 s at 60 *C, and 2 min at 72 *C. The mutagenic HRP gene was further
amplified 10- to 100- fold in a total volume of 1.5-2 mL using the aforementioned
thermocycling program with 20 additional cycles, and then gel-purified. The libraries
were obtained by transforming the mutagenic HRP gene, along with the BamH1-NheI
backbone of pCT2con-HRP plasmid, into EBY100 following a published method (26).
In the last round of evolution, the libraries were created as described above except that
different primers (5'-GTCCTAACGTCTCAAACATAGTACGGGACACTATTGTCAA
TGAGTTACGATCGGACCC-3', 5'GTACGCAGATCGAAGTCGACCAAGGCGCTTA
GGTTGCCATTAAGGGGACATAGTCC-3')
and
the
AvrII-PflFI
pCT2con-HRP
backbone were used. Each library contained ~107 unique sequences with a mutation
frequency of 1-3 mutations per gene.
Saturated mutagenesis libraries were constructed by replacing the BsmI-AflI fragment
of LIrs and DIs genes with a DNA fragment where Asn137, Leul38, Ala140, Phe142 and
Phel43 were exhaustively randomized.
This DNA fragment was assembled using
oligonucleotides (5'-CAGGAGGTCCCTCTTGGAGGGTTCCTTTGGGACGTCGAGA
CAGCCTACAAGCATTTTTAGATCTCGCGAATGCG-3', 5'-CCACCGCTGAGGGC
AACGAGATCAGAAGAACGGTTTAAACCAACATTTCTAAAAGAATCCTTAAGT
TGTGGAAGTG-3', 5'-ATTTTTAGATCTCGCGAATGCGNNBNNBCCANNBCCAN
NBNNBACACTTCCACAACTTAAGGAT-3')
(TriLink, San Diego, CA) and Phusion
High-fidelity DNA polymerase (NEB, Ipswich, MA) under the PCR conditions suggested
by the manufacturer of the polymerase following a literature procedure (15).
73
To make
LIs or DIIs libraries, this insert, along with either the pCT2-LIrs or pCT2-DIs plasmid
missing the BsmI-AflII fragment, was then used in the EBY100 transformation step as
described in ref. 15.
Selection by FACS
HRP libraries were grown and induced as previously described (15).
Freshly induced
cells (the number of cells used was at least 10-fold greater than the population diversity)
were washed with PBS containing 0.5% BSA, followed by another wash with PBS alone.
The cells were resuspended in a PBS solution containing 15 p.M L-tyrosinol-A488, 15
p.M D-tyrosinol-A647, and 150 pM H2 0 2 , followed by a 30-min shaking at 30 'C. The
labeling reaction was then quenched by the addition of ascorbic acid to a final
concentration of 10 mM; the cells were washed with PBS containing 0.5% BSA, labeled
with mouse anti-c-Myc monoclonal antibody and phycoerythrin-goat anti-mouse
antibodies, and sorted as previously described (15). To minimize the influence of the dye
on HRP's enantioselectivity, the aforementioned fluorescent substrates were replaced
with D-tyrosinol-A488 and L-tyrosinol-A647 in every other round of sorting. After six
rounds of sorting, plasmid DNA encoding the HRP variants was extracted from the sorted
libraries using the Zymoprep Yeast Plasmid Miniprep Kit (Zymo Research, Orange, CA)
and transformed into XL10-Gold ultracompetent E. coli cells (Stratagene, La Jolla, CA).
The plasmids, encoding individual HRP variants, were then isolated from E. coli cells
and transformed into yeast strain EBY100 for further characterization.
Soluble HRP purification
The HPR genes were subcloned into the 4m5.3 plasmid (27) backbone with the Gall-10
promoter and appS4 leader sequence (23). These HRP-containing plasmids, along with
the BiP-overexpressing plasmid pMR-1341 (CEN-URA3) (27), were transformed into a
PDI-overexpressing strain of S. cerevisiae, YVH10 (26).
The resultant colonies were
grown to an OD 600 of 5-7 in 1 L of synthetic defined (SD) medium (2% dextrose, 0.34%
yeast nitrogen base without (NH4 )2 SO 4 , 0.8% casamino acids (VWR, West Chester, PA),
50 mM Na phosphate buffer adjusted to pH 6.6) in a 2.5-L fully baffled Tunair flask
(Shelton Scientific, Shelton, CT) by shaking at 250 rpm at 30 'C. To induce expression
of the HRP variants, the cells were centrifuged to remove the supernatant and
resuspended in 1 L of yeast peptone galactose (YPG) medium (1% Bactoyeast extract,
2% Bactopeptone, 1.8% galactose, 0.2% dextrose, 50 mM Na phosphate buffer adjusted
to pH 6.6) in a 2-L glass Erlenmeyer flask. This medium was supplemented with 50
ptg/mL kanamycin, 100 U/mL penicillin G, 200 U/mL streptomycin, 0.034%, thiamine
HCl, 0.084% 6-aminolevulinic acid, 0.1 mM ferric citrate, and 0.5 pM hemin (100x stock
solution was freshly prepared by dissolving hemin in equal parts of ethanol and 0.04 M
aqueous NaOH) (Frontier Scientific, Logan, UT). The flasks were shaken at 250 rpm at
20 'C for 72 h, with 0.02% 6-aminolevulinic acid added every 24 h. The supernatant was
then separated by centrifugation, filter-sterilized, and concentrated to ~100 mL with a
30K MWCO Amicon Stirred Cell (Millipore, Bedford, MA). PBS (400 mL) was then
added, and the sample was concentrated to a final volume of 50 mL. The protein was
purified by anti-FLAG (Sigma) affinity chromatography using the protocol provided by
the manufacturer. Final protein, digested with EndoH (NEB) to remove glycosylations,
was seen as a single band by Coomassie Blue staining of an SDS-PAGE gel under
reducing conditions. Purified yields had RZ (Reinheitszahl) values of 1.5-2.5 and were
1-5 mg/L of initial culture as determined spectrophotometrically using the extinction
coefficient of 100 mM-'cm-' at 403 nm.
Enantioselectivity of soluble HRP
The initial rates of substrate oxidation by soluble HRP were measured by monitoring the
rising absorbance of the products as previously described (29). In a typical experiment,
the enzyme (0.01-1 pM) was added directly to a spectrophotometric cuvette containing
300 pL of a reaction mixture consisting of H2 0 2 (0.15 mM for 1 or 2, or 10 mM
otherwise) and reducing substrate (15 pM for 1 or 2, or 1 mM otherwise) in a PBS buffer
at room temperature. To measure the initial reaction rates, the increases in absorbance as
a function of time were recorded at 513 nm for 1 and 2; at 290 nm for 3 and 4; and at 315
nm for tyrosinol and N-acetyl-tyrosinol.
Computational docking of HRP variants
The complexes of wild-type HRP (PDB: 7ATJ) (30) with the enantiomers of 3 and 4
were obtained using a docking method that combines quantum mechanical calculations
with Schrddinger's Glide version 4.5. The substrates were geometry-optimized first in
molecular mechanics with Macromodel using the OPLS2001 force field and then in
quantum mechanics with Jaguar using the Poisson-Boltzmann implicit solvent model of
aqueous environment simulation.
Quantum mechanics was represented by density
functional theory with the B3LYP functional (31) and 6-31G* basis set (32).
The
presence of iron in the enzyme's heme moiety requires the use of quantum chemical
calculations for the complex region that involves electron transfer (33).
variation of the previously described
QM/MM
Therefore, a
(quantum mechanical/molecular
mechanical) docking algorithm (34, 35) was used. To properly sample binding modes,
"restricted docking", wherein the phenol ring of ferulic acid in the 7ATJ complex served
as the restriction point, was performed in addition to standard Glide sampling to generate
a total of 10 diverse poses of each enantiomer of 3 and 4. To accurately score these
poses, QM/MM single-point energy calculations without geometry optimization were
carried out, treating the heme moiety and the substrate as a quantum region.
Upon
convergence of QM/MM, the atomic charges were fitted for atoms involved in
calculations using the ESP (electrostatic potential) method.
The poses, with fitted
charges, were then ranked using Glide's score-in-place function and the lowest binding
energy pose was selected.
D. References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Rouhi M (2004) Chiral chemistry: Traditional methods thrive despite numerous
hurdles, including tough luck, slow commercialization of catalytic processes.
Chem Eng News 82:47-62.
Shin H-D, Guo X, Chen R (2006) in Bioprocessingfor Value Added Products
from Renewable Resources, ed Yang ST (Elsevier BV, Amsterdam), pp 351-371.
Klibanov AM (2001) Improving enzymes by using them in organic solvents.
Nature 409:241-246.
Klibanov AM (1995) Enzyme memory - What is remembered and why? Nature
374: 596
Phillips RS (1996) Temperature modulation of the stereochemistry of enzymatic
catalysis: Prospects for exploitation. Trends Biotechnol 14:13-16.
Turner NJ (2003) Controlling chirality. Curr Opin Biotechnol 14:401-406.
Hult K, Berglund P (2003) Engineered enzymes for improved organic synthesis.
Curr Opin Biotechnol 14:395-400.
Arnold FH (2001) Combinatorial and computational challenges for biocatalyst
design. Nature 409:253-257.
Tao HY, Cornish VW (2002) Milestones in directed enzyme evolution. Curr Opin
Chem Biol 6:858-864.
Reetz MT (2004) Controlling the enantioselectivity of enzymes by directed
evolution: Practical and theoretical ramifications. Proc Natl Acad Sci USA
101:5716-5722.
Reetz MT (2006) Directed evolution of enantioselective enzymes as catalysts for
organic synthesis. Adv Catal 49:1-69.
Gilabert MA, et al. (2004) Stereospecificity of horseradish peroxidase. Biol Chem
385:1177-1184.
Veitch NC (2004) Horseradish peroxidase: a modern view of a classic enzyme.
Phytochemistry 65:249-259.
Morawski B, et al. (2000) Functional expression of horseradish peroxidase in
Saccharomyces cerevisiae and Pichiapastoris.ProteinEng 13:377-384.
Lipovsek D, et al. (2007) Selection of horseradish peroxidase variants with
enhanced enantioselectivity by yeast surface display. Chem Biol 14:1176-1185.
Sen S, Venkata Dasu V, Mandal B (2007) Developments in directed evolution for
improving enzyme functions. Appl Biochem Biotechnol 143:212-223.
Boersma YL, Droge MJ, Quax WJ (2007) Selection strategies for improved
biocatalysts. FEBSJ274:2181-2195.
Farinas ET (2006) Fluorescence activated cell sorting for enzymatic activity.
Comb Chem High Throughput Screen 9:321-328.
Bershtein S, Tawfik DS (2008) Advances in laboratory evolution of enzymes.
Curr Opin Chem Biol 12:151-158.
Reetz MT, et al. (2007) Learning from directed evolution: Further lessons from
theoretical investigations into cooperative mutations in lipase enantioselectivity.
ChemBioChem 8:106-112.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
van Loo B, et al. (2004) Directed evolution of epoxide hydrolase from A.
radiobacter toward higher enantioselectivity by error-prone PCR and DNA
shuffling. Chem Biol 11:981-990.
Morawski B, Quan S, Arnold FH (2001) Functional expression and stabilization
of horseradish peroxidase by directed evolution in Saccharomyces cerevisiae.
BiotechnolBioeng 76:99-107.
Rakestraw AJ, Sazinsky SL, Piatesi A, Antipov E, Wittrup KD. (2008) Directed
evolution of a secretory leader for the improved expression of heterologous
proteins and full-length antibodies in Saccharomyces cerevisiae. Biotechnol
Bioeng, in press (doi: 10.1002/bit.22338).
Nakamura Y, et al. (2006) Enhancement of activity of lipase-displaying yeast
cells and their application to optical resolution of (R,S)-1-benzyloxy-3-chloro-2propyl monosuccinate. Biotechnol Prog22:998-1002.
Dymicky M (1976) N-Acetyl-L-tyrosine ethyl ester. Organic Preparationsand
ProceduresInternational8:219-222.
Chao G, et al. (2006) Isolating and engineering human antibodies using yeast
surface display. Nature Protocols 1:755-768.
Midelfort KS, et al. (2004) Substantial energetic improvement with minimal
structural perturbation in a high affinity mutant antibody. JMol Biol 343:685-701.
Shusta EV, Raines RT, Pluckthun A, Wittrup KD (1998) Increasing the secretory
capacity of Saccharomyces cerevisiae for production of single-chain antibody
fragments. Nat Biotechnol 16:773-777.
Rojas AM, Gonzalez PA, Antipov E, Klibanov AM (2007) Specificity of a DNAbased (DNAzyme) peroxidative biocatalyst. Biotechnol Lett 29:227-232.
Henriksen A, Smith AT, Gajhede M (1999) The structures of the horseradish
peroxidase C-ferulic acid complex and the ternary complex with cyanide suggest
how peroxidases oxidize small phenolic substrates. J Biol Chem 274:3500535011.
Becke AD (1993) A new mixing of Hartree-Fock and local density-functional
theories. J Chem Phys 98:1372-1377.
Rassolov VA, Pople JA, Ratner MA, Windus TL (1998) 6-31G* basis set for
atoms K through Zn. J Chem Phys 109:1123-1229.
Cho AE (2007) Effect of quantum mechanical charges in binding sites of
metalloproteins. BioChipJ 1:70-75.
Cho AE (2008) Quantum mechanical calculations for binding sites of
metalloproteins. BioChipJ 2:148-153.
Cho AE, Rinaldo D (2009) Extension of QM/MM docking and its applications to
metalloproteins. J Comput Chem, in press (doi: 10.1002/jcc.21270).
Chapter IV: How a single-point mutation in horseradish peroxidase
markedly enhances enzymatic enantioselectivity
A. Introduction
While enzymes typically exhibit exquisite enantioselectivities toward their natural
substrates, most synthetically useful substrates are non-natural (1, 2). Therefore, there
has been much effort to enhance enzymatic enantioselectivity toward these artificial
substrates to create superior practical biocatalysts for organic and industrial chemistry (35). While the ability to rationally predict mutations that improve selectivity would be of
great value, insufficient mechanistic details governing enzymatic enantioselectivity limit
such approaches. Directed evolution, which requires no knowledge of enzyme structure
and/or mechanism, in principle provides a promising alternative protein engineering
strategy to enhance enantioselectivity (6). Its success, however, depends on the efficient
search of protein sequence space using high-throughput screening or selection methods,
whose development remains daunting (7, 8).
Therefore, semi-rational enzyme
engineering strategies, where the search space is reduced by targeting for mutations only
those residues likely to improve enzyme function, thereby resulting in smaller libraries,
are emerging as powerful tools for augmenting enzymatic enantioselectivity (9-11).
Knowing these residues and how they exert their influence could also help the rational
design of highly enantioselective enzymes (12-16).
Oxidoreductases, such as horseradish peroxidase (HRP), are particularly attractive
biocatalysts due to numerous asymmetric processes they catalyze (17, 18). Among other
synthetically useful reactions, HRP catalyzes the oxidation of a variety of chiral phenols
with hydrogen peroxide, albeit typically with low enantioselectivity (19).
We recently developed an efficient directed-evolution method based on yeast
surface display and fluorescence-activated cell sorting (FACS) that enabled us to
dramatically improve HRP's enantioselectivity toward certain phenols (20). Using this
experimental platform, we discovered an HRP variant with a single Arg-to-Gln mutation
at position 178 (Arg178Gln) exhibiting an 18-fold greater enantioselectivity toward
fluorescent substrate 2 (tyrosinol covalently linked to one of the structural isomers of the
Alexa Fluor*488 dye) than the native enzyme (20).
+
sO
sO
H3N
0
.
NH2
coo
HN
0
OH
2
OH
In the present study, we elucidate mechanistically how a single mutation at
position 178 of yeast surface-bound HRP leads to a marked improvement in
enantioselectivity toward 2. Moreover, using in vitro kinetic assays, substrate analogs,
and molecular modeling, we show that a 25-fold enhancement in enantioselectivity
exhibited by the Argl78Glu variant of HRP is mostly due to a change in the transition
state energy stemming from the electrostatic repulsion between the carboxylates of the D
enantiomer of the substrate and the Glu-178 residue of the enzyme.
B. Results and Discussion
HRP is a highly glycosylated enzyme that contains four disulfide bonds and a
catalytically essential heme prosthetic group (21, 22). Due to this structural complexity,
heterologous expression of the enzyme in prokaryotes is impaired, thus necessitating a
eukaryotic system to produce active HRP (21, 22). Previously, we showed that yeast
surface display (23) was well suited for the expression, engineering, and characterization
of HRP, in particular using fluorescent phenolic substrates (20, 24). Yeast display allows
quantitative measurements of HRP expression and activity by flow cytometry and also
enables convenient characterization of enzyme variants without soluble expression and
purification of each individual clone (20, 24). Using this system, we discovered a number
of HRP variants with enhanced enantioselectivities toward both the L and D enantiomers
of 2 (20).
In particular, a single-point mutant isolated from a random mutagenesis
library, Arg178Gln, greatly preferred the L enantiomer while the wild-type enzyme was
virtually non-enantioselective (20).
To understand the mechanism by which the single mutation at position 178
endows HRP with the keen enantioselectivity, we first investigated the relationship
between the latter and the nature of the amino acid residue at this position. To this end,
we mutated this residue to each of the other 19 standard amino acids and measured the
enantioselectivity of every respective surface-bound HRP variant toward the optical
isomers of substrate 2. As seen in Figure 4.1, the enantioselectivity, E(L/D), of the wildtype enzyme displayed on the cell surface of yeast was negligible: 0.8 ± 0.2; however,
when the wild-type's Arg-178 was replaced with Gln, the enantioselectivity jumped to 14
± 1 in agreement with our previous results (20).
0
0
4-
Figure 4.1. Enantioselectivities of yeast surface-bound wild-type HRP and its 19 amino
acid variants at position 178 toward a fluorescent phenolic substrate (2). Oxidation of 2
by surface-bound HRP yields fluorescently labeled cells whose fluorescence intensity is a
direct estimate of the product amount of the enzymatic reaction. The reaction rates are
defined as temporal changes in the fluorescent intensity of HRP-displaying yeast cells
measured by flow cytometry. Asterisks designate catalytically inactive variants.
The E(L/D) values depicted in Figure 4.1 also show that replacing the positively
charged Arg-178 with the aromatic residues Tyr, Phe, or Trp produced no active enzyme
variants, presumably due to steric constraints imposed by the bulky side chains. [This
supposition is supported by computational docking of these HRP variants with 2 which
indicates that the substitution of Arg-178 with bulky aromatic residues makes the active
site inaccessible to 2 (data not shown).] Similarly, the Argl78Cys variant exhibited no
catalytic activity, probably due to mispairing of disulfide bonds within HRP. However,
all other Argl78X variants were enzymatically active, and their E(L/D) analysis afforded
some interesting conclusions.
First, preserving the positive charge at position 178
retained the enzyme's low enantioselectivity-Arg178Lys's E(L/D) = 1.8 ± 0.3-whereas
all mutations abolishing the positive charge at that position increased the E(L/D) values
(Figure 4.1).
Second, reversal of the charge via the introduction of the negatively
charged Asp or Glu residues greatly raised the enantioselectivity to 13 + 1 and 20 ± 3,
respectively. These results suggest that electrostatic interactions mediated by residue 178
are the main determinant of HRP's enantiopreference toward 2.
To further explore the role of electrostatics, we examined the kinetics governing
the highest, 25-fold improvement in enantioselectivity observed with the charge-reversed
Arg178Glu variant. Since the E(L/D) of HRP directly depends on the oxidation rates of
both enantiomers of 2, it can be enhanced by either increasing the oxidation rate of the L
enantiomer, or decreasing that of the D enantiomer, or both. To determine which of these
scenarios actually occurs when Arg-178 is replaced with Glu, we measured the initial
reaction rates of the native and mutant enzymes with both substrate enantiomers. Table
4.1 shows that the wild-type enzyme, consistent with its E(L/D) value being close to
unity, oxidizes L-2 and D-2 with similar rates: 5.0 ± 0.2 and 6.3 ± 0.3 MFU/min,
respectively. Interestingly, the Argl78Glu variant has almost the same oxidation rate of
L-2 as the native enzyme (6.2 ± 0.6 and 5.0 ± 0.2 MFU/min, respectively), whereas the
oxidation of the D enantiomer is some 20-fold slower than that by the wild-type (0.3 ± 0.1
MFU/min and 6.3 ± 0.3 MFU/min, respectively).
Thus, the 25-fold rise in
enantioselectivity attained by Argl78Glu HRP is predominantly due to the plunged
reactivity of D-2. Moreover, taken together with the mutagenesis analysis (Figure 4.1),
these data suggest that it is largely electrostatic interactions between the D enantiomer
and the residue at position 178 in the enzyme's transition state that control the
enantioselectivity.
To test this hypothesis, we explored which functional groups of D-2 are involved
in this putative electrostatic interaction. Under our experimental pH (7.4), two types of
anionic groups-a carboxylate and two sulfonates-may play such a role. One plausible
mechanism by which Argl78Glu HRP can acquire high enantioselectivity is that the
negatively charged carboxylate and/or sulfonates stabilize the transition state of the wildtype enzyme and D-2 by forming a salt bridge with the positively charged Arg-178.
Replacing the latter with any amino acid residue other than Lys would eliminate this
stabilizing interaction and hence lower the oxidation of the D enantiomer. Note, however,
that the oxidation rate of the L enantiomer is similar to that of its D counterpart for the
wild-type enzyme and is almost unaffected by the Arg178Glu mutation. Therefore, the L
enantiomer must form a very different transition state with wild-type HRP than the D
enantiomer for its oxidation rate by the Argl 78Glu variant to remain unaltered, while that
of the D enantiomer's is slashed some 20-fold.
Table 4.1. Initial rates and enantioselectivities of oxidation of 2, 5, and 6 by yeast
surface-bound wild-type and Arg178Glu HRP
HRP variant
VL, MFU/mina
Substrate
VD,
MFU/mina
E(L/D)
0.8 ± 0.2
Wild-type
5.0
0.2
6.3
0.3
Arg1 78Glu
6.2
0.6
0.3
0.1
Wild-type
7.3
0.2
10
1
0.7
0.1
Arg178Glu
6.4
0.2
0.5
0.1
13
1
Wild-type
9.2
0.5
8.4
0.2
1.1
Arg178Glu
21
1
8.9
0.3
2.4-± 0.1
aVL
20
3
0.1
and VD are the initial rates of oxidation of the L and D enantiomer, respectively,
reported in Mean Fluorescence Units (MFU) per min.
fluorescence intensity of 3
x 104
MFUs represent mean
HRP-displaying yeast cells that captured fluorescent
products during the time of the enzymatic reaction.
approximately 3 x 104 HRP molecules on its surface (25).
Each yeast cell displays
The initial rates are not
absolute as fluorescence intensity varies for each substrate; however, their ratios giving
the E(L/D)values are unaffected by these variations. All experiments were conducted at
least in triplicate with the mean and standard deviation values given in the table. See
Methods for experimental details.
Another mechanism also consistent with the hypothesis that the transition states
for both enantiomers with the wild-type enzyme have similar energies involves the
aforementioned anionic groups of D-2 preventing the formation of a stable activated
complex between this enantiomer and the Argl78Glu variant due to an electrostatic
repulsion with Glu-178. To distinguish between these alternatives, we have employed
molecular modeling to obtain structures of wild-type HRP complexed with each
enantiomer of 2.
As seen in Figure 4.2, L-2 and D-2 bind similarly to the wild-type enzyme with
calculated binding energies of -5.41 kcal/mol and -5.89 kcal/mol, respectively.
This
similarity is consistent with our observation that the oxidation of L-2 by the wild-type
enzyme is just slightly slower than that of D-2 (Table 4.1). Figure 4.2 also reveals that
the sulfonate groups of both substrate enantiomers are located in proximity to Arg-178:
the distances between the oxygen of each sulfonate group and the closest nitrogen of the
guanidinium group are 3.5 A and 2.8
A for the L enantiomer and
3.1
A and 2.8 A for the
D enantiomer. This structural information argues against the high enantioselectivity of
Argl78Glu HRP being due to the loss of a salt bridge with the sulfonates of D-2, since
substitution of Arg- 178 would have led to elimination of any potential salt bridges in both
the L and D transition states with no consequent selectivity.
Figure 4.2. Modeled complexes of wild-type HRP with L-2 (A, C) and D-2 (B, D). (A)
and (B) show the front view of the active site; (C) and (D), respectively, show the active
site rotated 900 clockwise along the z-axis. For clarity, HRP's backbone is shown in
ribbon with the heme moiety in orange, substrate in blue, and Arg-178 in green ball-andstick. Distances shown are in A. See Methods for details of how these molecular models
were built.
These docking studies also shed light on the role of the substrate's carboxylate in
mediating enantioselectivity of HRP. As seen in Figures 4.2A and 4.2C, the carboxyl
group of L-2 points away from the guanidinium group of Arg-178 such that they are
separated by the planar fused aromatic rings. In contrast, although the orientation of the
D-2's carboxyl group is conducive to making a salt bridge with this guanidinium group, a
relatively large distance between them, 5.3 A (Figure 4.2B), makes this scenario unlikely.
It appears, therefore, that it is the electrostatic repulsion between the carboxylate or
sulfonates of D-2 and Glu-178 that plays a dominant role in imparting HRP's
enantioselectivity toward 2.
To determine which of the anionic groups of D-2 plays the main role in this
repulsion, we measured the enantioselectivity of the native and Argl78Glu enzymes
toward 3, a substrate analog of 2 lacking the sulfonates. As seen in Table 4.1, wild-type
0
H3N
NH 2
0
H3 N
NH
COOCH 3
COO
0
HN
0
HN
2
OH
OH
1
5
s
6
OH
OH
HRP catalyzes the oxidation of 5 with a slight D enantiopreference: E(L/D)
0.7 ± 0.1.
In contrast, the Argl78Glu variant is keenly L selective toward 2 with an E(L/D) value of
13 ± 1 (Table 4.1), thus representing a 19-fold rise in enantioselectivity compared to the
native enzyme.
Therefore, the sulfonates of 2 seem insignificant in controlling the
enantiopreference of Argl78Glu HRP given the similar magnitudes of the improvement
in the E(L/D) toward 5 and 2 (19-fold and 25-fold, respectively).
89
This conclusion is
consistent with our molecular modeling predictions that the sulfonates are close to Arg178 for both enantiomers (Figure 4.2) and therefore their interactions with Glu-178 are
unlikely to induce enantioselectivity. (It is also possible, however, that the sulfonates are
not involved because they form intramolecular salt bridges with the neighboring
protonated amino groups.)
Inspection of Table 4.1 also reveals that while the oxidation of D-5 by the
Argl78Glu variant is some 20-fold slower than by the wild-type enzyme (0.5 ± 0.1 and
10 ± 1 MFU/min, respectively), the oxidation rates of the L enantiomer are similar for
both enzymes.
Therefore, the 19-fold increase in enantioselectivity exhibited by the
Argl78Glu variant toward 5 is entirely due to a drop in the reactivity of the D enantiomer,
as is also the case with 2.
These results suggest that it is an electrostatic repulsion
between the carboxylate of the substrate and the introduced Glu that is responsible for the
enhanced enantiopreference of Argl78Glu HRP as compared to its wild-type
predecessor.
To probe these interactions further, we proceeded to model complexes of wildtype HRP with L-5 and D-5. The binding modes thus obtained indicate that the carboxyl
groups of the two enantiomers are oriented differently vis-i-vis Arg-178. For example, a
planar aromatic ring system of L-5 prevents its carboxylate from interacting with Arg-178
(Figure 4.3A). In contrast, the carboxyl group of D-5 is positioned to directly interact
with Arg-178 (Figure 4.3B).
In this orientation, the carboxylate is also likely to
experience an electrostatic repulsion with Glu-178 which would, in turn, weaken the
binding of D-5 in the transition state, thereby making the Arg178Glu variant highly L
selective, as is actually observed.
Figure 4.3. Modeled complexes of wild-type HRP with L-5 (A), D-5 (B), L-6 (C), and D6 (D). For clarity, HRP's backbone is shown in ribbon with the heme moiety in orange,
substrate in blue, and Arg-178 in green ball-and-stick. Distances shown are in
Methods for details of how these molecular models were built.
A.
See
To ascertain whether this electrostatic repulsion is indeed important in
determining the enantioselectivity toward 5, we measured the initial oxidation rates of
both substrate enantiomers with the wild-type and Arg178Glu enzymes in the presence of
a high salt concentration. As seen in Table 4.2, both enzymes are more active at 1 M
NaCl than at 137 mM NaCl. The enantioselectivity of wild-type HRP in these high-salt
and low-salt buffered solutions were the same (0.9 ± 0.2 and 0.7 ± 0.1, respectively),
indicating that the putative electrostatic attraction between the carboxylate of D-5 and
Arg-178 is insensitive to the salt concentration. In contrast, the enantioselectivity of the
Argl78Glu variant was nearly 3-fold lower in the high-salt than in the low-salt solution:
the E(L/D) values are 4.9 ± 0.6 and 13 ± 1, respectively. Furthermore, Table 4.2 shows
that this drop in the E(L/D) stems from an increase in the oxidation rate of the D
enantiomer consistent with the proposed electrostatic repulsion between the carboxylates
of D-5 and Glu-178, which expectedly was partially alleviated by the presence of high
salt.
We thus reasoned that eliminating this repulsion by neutralizing the negative
charge of the substrate's carboxylate should increase the oxidation rate of the D
enantiomer and hence restore the wild-type-like level of HRP's enantioselectivity.
Computational docking of methyl esters of L-5 and D-5 (i.e., L-6 and D-6, respectively) to
the wild-type enzyme yielded binding modes similar to those observed with their
respective ionized carboxylate counterparts, suggesting that the esterification would
affect only the proposed electrostatic repulsion (Figure 4.3).
Table 4.2.
The effect of salt (NaCl) concentration on enantioselectivity of yeast
surface-bound wild-type and Arg178Glu HRP toward 5
YhighL
E(L/D)c
HRP variant
Substrate
Wild-type
5
2.1 ± 0.1
1.7
0.2
0.9
0.2
Argl78Glu
5
2.0 ± 0.2
5.5
0.2
4.9
0.6
ajigh L/
okwL
ow
La
kighD
powDb
is the ratio of the initial rates of oxidation of the L enantiomer measured in a
phosphate buffer with final NaCl concentrations of 1 M and 137 mM, respectively. All
experiments were conducted at least in triplicate with the mean and standard deviation
values given in the table.
b ,ighD
/
JowD
is the ratio of the initial rates of oxidation of the D enantiomer measured in
a phosphate buffer with final NaCl concentrations of 1 M and 137 mM, respectively. All
experiments were conducted at least in triplicate with the mean and standard deviation
values given in the table.
cEnantioselectivity, E(L/D), is measured in a phosphate buffer with a final 1 M NaCl
concentration.
To test these computer-modeling-based predictions, we synthesized L-6 and D-6
and measured their initial oxidation rates catalyzed by the wild-type and Arg178Glu
enzymes (rows 3 and 4 in Table 4.1). As predicted, wild-type HRP exhibits similar
enantioselectivities with 5 and 6: E(L/D) are 0.7 ± 0.1 and 1.1 ± 0.1, respectively.
Importantly, protecting the carboxyl group (in substrate 6 compared to 5) indeed restores
the oxidation rate of the D enantiomer by the Argl78Glu variant to that of the wild-type
enzyme (8.9
0.3 and 8.4 ± 0.2 MFU/min, respectively; Table 4.1). This significant
rise in the oxidation rate of D-6 by Argl78Glu HRP points to the electrostatic repulsion
between the carboxylates of the D enantiomer of 2 or 5 and Glu- 178 in the transition state
as the defining mechanism of the enhanced enantioselectivity of this enzyme variant. It
should be noted that the enantioselectivity of the Arg178Glu variant toward 6 differs
from that of the wild-type enzyme (2.4 ± 0.1 vs. 1.1 ± 0.1, respectively) caused by a
surprising doubling in the oxidation rate of the L enantiomer (Table 4.1).
In conclusion, we have found herein that eliminating the positive charge in the
side chain of residue 178 moderately increases the enantioselectivity of yeast surfacebound HRP toward 2. The computational modeling and kinetic analysis using high salt
have also indicated that the observed increase in the enantioselectivity of the chargeneutral Arg178X variants does not arise from a loss of a salt bridge between the D
enantiomer of substrate 2 or 5 and Arg-178. The observed improvement in E(L/D) of
these variants may be explained by the presence of a cation-7i interaction between Arg178 and Phe-179 (Figure 4.4). Elimination of this interaction via replacement of Arg-178
by any other residue with the exception of Lys could destabilize the transition-state
complex of the D enantiomer and the wild-type enzyme, which would lower the oxidation
rate of the D enantiomer and result in higher L enantioselectivity.
More research is
needed to validate this hypothesis.
We have also showed herein that replacing the positively charged amino acid at
position 178 with a negatively charged one provides the greatest improvement on the
enantioselectivity of yeast surface-bound HRP toward 2.
molecular
modeling,
we
have
rationalized
that
a
Aided by structure-based
25-fold
enhancement
in
enantioselectivity for the charge-reversed Arg178Glu HRP is primarily caused by a
slower oxidation rate of the D enantiomer which, in turn, is due to the electrostatic
repulsion between the carboxyl groups of this enantiomer and Glu- 178 of the enzyme in
the transition state.
Overall, our analysis suggests that molecular modeling in
combination with in vitro kinetic assays and substrate analog studies can provide useful
mechanistic insights into enzyme enantioselectivity and how to improve it.
.
.
...........
Figure 4.4. Location of Phe-179 with respect to Arg-178 and D-2. The carboxyl group
of D-2 is oriented to directly interact with Phe- 179. Elimination of the putative cation- R
interaction between Arg-178 and Phe-179 could free the latter residue's electron-rich side
chain to engage in the unfavorable electrostatic interaction with D-2 carboxylate. Phe179 was also previously found to be important for the binding of aromatic substrates (26).
For clarity, HRP's backbone is shown in ribbon with the heme moiety in orange, D-2 in
blue,
and
Arg-178
and
Phe- 179
in
green
ball-and-stick.
C. Materials and Methods
Materials
All chemicals were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO) unless
stated otherwise and were of the highest purity available from the vendor.
The
enantiomers of substrate 2 were synthesized as previously described (20).
The
enantiomers of substrate 5 were prepared by reacting L- or D-tyrosinol with 5carboxyrhodamine 110 succinimidyl ester (AnaSpec, San Jose, CA) according to the
following procedure (27). The fluorescent dye (2 mg, 4 pimol) was added to a solution of
L- or D-tyrosinol (2.4 mg, 12 pmol) and triethylamine (24 pmol) in DMF (1 mL). The
resulting mixture was stirred at room temperature for 3 h, evaporated, and re-dissolved in
10% (v/v) acetonitrile/water (1 mL). The product was purified by reverse-phase HPLC
using a 9.4 x 250 mm 5 pM SB-Phenyl column (Agilent Technologies, Santa Clara, CA)
with 100 mM triethylammonium acetate buffer (pH 7.0) (Calbiochem, San Diego, CA) as
a loading buffer and acetonitrile as a mobile phase. The product was eluted with a 30min, 4 mL/min gradient of 10%-100% acetonitrile. The enantiomers of substrate 6 were
prepared by dissolving those of dry crude substrate 2 product in 1% H2 S0
anhydrous methanol (3 mL).
4
(v/v) in
The mixture was refluxed for 2 days, evaporated, and
neutralized to pH 7 by saturated aqueous NaHCO 3 . The product was purified by reversephase HPLC under the same conditions as used to purify 5. The identity of all the
substrates was confirmed by electrospray ionization (ESI)-MS.
Mutations at position 178 of HRP were made using the QuikChange site-directed
mutagenesis kit (Stratagene, La Jolla, CA).
HRP variants were displayed on the cell
surface of the Saccharomyces cerevisiae yeast according to the published procedure (20,
24).
Briefly, the HRP-containing plasmids were transformed into the yeast surface
display strain of S. cerevisiae, EBY100, using the Frozen-EZ Yeast Transformation II kit
(Zymo Research, Orange, CA). The resultant colonies were grown to an OD 600 of 5-7 in
5 mL of synthetic defined (SD) medium (2% dextrose, 0.34% yeast nitrogen base without
(NH 4 )2 SO 4 , 0.8% casamino acids (VWR, West Chester, PA), and 50 mM Na phosphate
buffer adjusted to pH 6.6) by shaking at 250 rpm at 30 'C. To induce expression of HRP,
the cells were centrifuged to remove the supernatant and resuspended in 5 mL of the SD
medium where dextrose was replaced with galactose and supplemented with 50 pig/mL
kanamycin, 100 U/mL penicillin G, 200 U/mL streptomycin, 0.034%, thiamine HCl,
0.084% 6-aminolevulinic acid, and 0.1 mM ferric citrate. The cultures were shaken at
250 rpm at 30 'C for 19-21 h. The induced cells were then washed with phosphatebuffered saline (PBS) containing 0.5% BSA, followed by another wash with PBS alone,
and used directly in the enzymatic reactions.
Enzymatic reactions
The initial oxidation rates catalyzed by surface-bound HRP were determined by
suspending 1x106 HRP-displaying yeast cells in 100 tL of PBS solution (pH 7.4, 137
mM or 1 M NaCl) containing 15 pM fluorescent substrate and 150 pM H2 0 2 in parallel
for both enantiomers. Three data points for each sample were collected by periodically
withdrawing 30 ptL of the L and
D
substrate mixtures into 1 mL of PBS containing 0.5%
BSA and 10 mM ascorbic acid to quench the reactions. The fluorescently labeled yeast
cells from each data point were then analyzed using a Coulter Epics XL flow cytometer
(Fullerton, CA). The mean fluorescence of 30,000 cells was plotted as a function of time
to determine the initial reaction rates. Enantioselectivity, E(L/D), was calculated as the
ratio of the initial rate of the enzymatic oxidation of the L enantiomer divided by that of
the D enantiomer.
The initial reaction rates for the wild-type and Arg178Glu HRP were determined
by monitoring the enzymatic reaction above except that mean fluorescence of each data
point was acquired from analyzing HPR-displaying cells with the same enzyme surface
concentration, which were identified using fluorescently labeled antibodies against the cMyc epitope tag fused to HRP. To this end, the cells from each data point of enzymatic
reaction were washed with 0.5 mL of PBS with 0.1% BSA and labeled with mouse antic-Myc monoclonal 9E10 (Covance, Princeton, NJ) and phycoerythrin-goat anti-mouse
antibodies, as described previously (20, 24), and analyzed using a Coulter Epics XL flow
cytometer. The mean fluorescence of 30,000 cells with the same surface concentration of
HRP was then plotted as a function of time to determine the initial reaction rates.
Computational modeling
Molecular models of HRP-substrate complexes were built on the basis of the published
X-ray crystal structure of HRP and its complex with ferulic acid (28), which was obtained
by retrieving the heavy atom coordinates (entry 7ATJ) from the Brookhaven Protein Data
Bank. The complexes of HRP with the substrates described in this study were generated
by using a docking method that integrates quantum mechanical calculations with
Schrddinger Glide version 4.5. Protein preparation wizard of Schr6dinger software was
used to prepare the original PDB file for docking and further modeling. With heavy
atoms fixed, hydrogen atoms were added and their positions were optimized using the
IMPACT (29) molecular minimization tool. The substrates were geometry-optimized
first in molecular mechanics with Macromodel using the OPLS2005 force field and then
in quantum mechanics using the Poisson-Boltzmann implicit solvent model of aqueous
environment simulation. Quantum mechanics were represented by density functional
theory with B3LYP functional (30) and 6-31G* basis set (31).
Current docking methods generally employ force-field-based energy scoring with
various search algorithms (32, 33).
This approach, however, is inadequate to model
enzymes that contain metal ions in the active site (34, 35). The presence of iron in the
HRP's heme group requires the use of quantum chemical calculations of the complex
region that involves electron transfer in order to correctly predict binding modes.
Therefore, a modified version of the previously described QM/MM
(quantum
mechanics/molecular mechanics) docking algorithm (36) was used according to the
following procedure. A total of 10 diverse poses were generated for each substrate: 5
poses were generated with Schr6dinger Glide version 4.5 and 5 more poses were
generated using "restricted docking", wherein the phenol ring of ferulic acid in the
complex with wild-type HRP served as the restriction point within prescribed tolerance (2
A of RMSD for the carbon atoms of the phenol ring of the substrate). In order to
accurately score these poses, QM/MM single-point energy calculations without geometry
optimization were carried out, treating the heme group and the substrate as a quantum
region. Upon convergence of these calculations, the atomic charges were fitted for atoms
in the quantum region using the ESP (electrostatic potential) method. Binding energy
calculations using Glide's score-in-place function were then performed to identify the
lowest binding energy pose.
100
D. References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Schoemaker HE, Mink D, Wubbolts MG (2003) Dispelling the myths-biocatalysis in industrial synthesis. Science 299:1694-1697.
Panke S, Wubbolts M (2005) Advances in biocatalytic synthesis of
pharmaceutical intermediates. Curr Opin Chem Biol 9:188-194.
Klibanov AM (2001) Improving enzymes by using them in organic solvents.
Nature 409:241-246.
Turner NJ (2003) Controlling chirality. Curr Opin Biotechnol 14:401-406.
Reetz MT (2004) Controlling the enantioselectivity of enzymes by directed
evolution: Practical and theoretical ramifications. Proc Natl Acad Sci USA
101:5716-5722.
Arnold FH (2001) Combinatorial and computational challenges for biocatalyst
design. Nature 409:253-257.
Tao HY, Cornish VW (2002) Milestones in directed enzyme evolution. Curr Opin
Chem Biol 6:858-864.
Reetz MT (2006) Directed evolution of enantioselective enzymes as catalysts for
organic synthesis. Adv Catal 49:1-69.
Lutz S, Patrick WM (2004) Novel methods for directed evolution of enzymes:
quality, not quantity. Curr Opin Biotechnol 15:291-297.
Chaparro-Riggers JF, Polizzi KM, Bommarius AS (2007) Better library design:
data-driven protein engineering. Biotechnol J 2:180-191.
Fox RJ et al. (2007) Improving catalytic function by ProSAR-driven enzyme
evolution. Nat Biotechnol 25:33 8-344.
Rotticci D, Rotticci-Mulder JC, Denman, S, Norin T, Hult K (2001) Improved
enantioselectivity of a lipase by rational protein engineering. ChemBioChem
2:766-770.
Park S, et al. (2005) Focusing mutations into the P. fluorescens esterase binding
site increases enantioselectivity more effectively than distant mutations. Chem
Biol 12:45-54.
Ijima Y, Matoishi K, Terao Y, Doi N, Yanagawa H, Ohta H (2005) Inversion of
enantioselectivity of asymmetric biocatalytic decarboxylation by site-directed
mutagenesis based on the reaction mechanism. Chem Commun 877-879.
Ema T, Fujii T, Ozaki M, Korenaga T, Sakai T (2005) Rational control of
enantioselectivity of lipase by site-directed mutagenesis based on the mechanism.
Chem Commun 4650-4651.
Reetz MT, et al. (2007) Learning from directed evolution: Further lessons from
theoretical investigations into cooperative mutations in lipase enantioselectivity.
ChemBioChem 8:106-112.
Azevedo AM, Martins VC, Prazers DM, Vojinovic V, Cabral JM, Fonseca LP
(2003) Horseradish peroxidase: a valuable tool in biotechnology. Biotechnol Annu
Rev 9:199-247.
Van Deurzen MPJ, Van Rantwijk F, Sheldon RA (1997) Selective oxidations
catalyzed by peroxidases. Tetrahedron 53:13183-13220.
Gilabert MA, et al. (2004) Stereospecificity of horseradish peroxidase. J Biol
Chem 385:1177-1184.
101
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
Antipov E, Cho AE, Wittrup KD, Klibanov AM (2008) Highly L and D
enantioselective variants of horseradish peroxidase discovered by an ultrahighthroughput selection method. ProcNati Acad Sci USA 105:17694-17699.
Veitch NC (2004) Horseradish peroxidase: a modem view of a classic enzyme.
Phytochemistry 65:249-259.
Ryan BJ, Carolan N, O'Fagain C (2006) Horseradish and soybean peroxidases:
comparable tools for alternative niches? Trends Biotechnol 24:355-363.
Gai SA, Wittrup KD (2007) Yeast surface display for protein engineering and
characterization. Curr Opin Struct Biol 17:467-473.
Lipovsek D, et al. (2007) Selection of horseradish peroxidase variants with
enhanced enantioselectivity by yeast surface display. Chem Biol 14:1176-1185.
Boder ET, Wittrup KD (1997) Yeast surface display for screening combinatorial
polypeptide libraries. Nat Biotechnol 15:553-557.
Veitch NC, Gao Y, Smith AT, White CG (1997) Identification of a critical
phenylalanine residue in horseradish peroxidase, Phe179, by site-directed
mutagenesis and 'H-NMR: Implications for complex formation with aromatic
donor molecules. Biochemistry 36:14751-14761.
Takakusa H, Kikuchi K, Urano Y, Kojima H, Nagano T (2003) A novel design
method of ratiometric fluorescent probes based on fluorescence resonance energy
transfer switching by spectral overlap integral. Chem Eur J 9:1479-1485.
Henriksen A, Smith AT, Gajhede M (1999) The structures of the horseradish
peroxidase C-ferulic acid complex and the ternary complex with cyanide suggest
how peroxidases oxidize small phenolic substrates. J Biol Chem 274:3500535011.
Banks JL, et al. (2005) Integrated modeling program, applied chemical theory
(IMPACT). J Comput Chem 26:1752-1780.
Becke AD (1993) A new mixing of Hartree-Fock and local density-functional
theories. J Chem Phys 98:1372-1377.
Rassolov VA, Pople JA, Ratner MA, Windus TL (1998) 6-31G* basis set for
atoms K through Zn. J Chem Phys 109:1123-1229.
Abagyan R, Totrov M (2001) High-throughput docking for lead generation. Curr
Opin Chem Biol 5:375-382.
Halperin I,Ma BY, Wolfson H, Nussinov R (2002) Principles of docking: An
overview of search algorithms and a guide to scoring functions. Proteins 47:409443.
Cho AE (2007) Effect of quantum mechanical charges in binding sites of
metalloproteins. BioChip J 1:70-75.
Cho AE (2008) Quantum mechanical calculations for binding sites of
metalloproteins. BioChip J 2:148-153.
Cho AE, Rinaldo D (2008) Extension of QM/MM docking and its applications to
metalloproteins. J Comput Chem, in press (doi: 10.1002/jcc.21270).
102
Appendix A: Effect of induction medium supplementation on HRP
activity and display levels
To characterize the catalytic activity of wild-type HRP displayed on the surface of yeast,
we used a non-chiral reducing substrate, tyramine, linked to the A488 fluorescent dye
(tyramine-A488).
Incubation of the yeast cells displaying wild-type HRP, expressed
under regular induction conditions (galactose induction medium, 19 hours, 30'C), with
tyramine-A488 in the absence of hydrogen peroxide resulted in fluorescently labeled
cells (Figure A.1A). Since an oxidizing substrate (e.g., hydrogen peroxide) is necessary
to support HRP catalytic activity, this led to the hypothesis that such a substrate was
released by yeast to activate HRP-catalyzed oxidation of tyramine-A488.
Furthermore,
the addition of hydrogen peroxide actually reduced the observed cell labeling (Figure
A. 1A), indicating that HRP enzymatic activity was inhibited by the addition of hydrogen
peroxide. Intriguingly, no cell labeling was observed when the same experiment was
performed in the presence of catalase, an enzyme that breaks down hydrogen peroxide
(data not shown). This led to the conclusion that the oxidizing substrate released by yeast
was hydrogen peroxide.
Because overexpression of HRP is expected to lead to
significantly reduced levels of endogenous heme, the levels of yeast catalase, a hemecontaining enzyme, are also expected to be significantly reduced. This reduction in the
levels of yeast catalase would lead to excess hydrogen peroxide, explaining the release of
this molecule from the HRP-displaying cells. We reasoned that we could increase the
activity of yeast catalases, and therefore decrease the amount of hydrogen peroxide
released by the cells, by restoring the amount of endogenous heme. To this end, we
103
supplemented the induction medium with heme synthesis intermediates: protoporphyrin
IX, Fe , and 6-aminolevulinic acid.
As shown in Figure A.1B, the supplementation of the induction medium indeed
decreased the levels of endogenous hydrogen peroxide, which also led to a 10-fold
increase in HRP catalytic activity. The resultant increase in catalytic activity is likely due
to a rise in HRP display levels. As illustrated in Figure A.2, when the induction medium
is supplemented with the aforementioned heme synthesis precursors, the average number
of HRP molecules displayed on the cell surface increases, as evidenced by a right shift of
the leading peak in flow cytometry histograms. Furthermore, a decrease in the amplitude
of the trailing peak (non-displaying cells) indicates that supplementation with heme
synthesis precursors also increases the percentage of cells displaying HRP.
104
50
500 -
A
40
*withoutH202
LM 30
Ewith
H202
400 -
*withoutH2O2
300 -
with H202
20
200 -
10 -
100 0
0
0
Figure A.1.
5
10
B
t
0
15
Time (min)
5
10
15
Characterization of HRP catalytic activity with non-chiral fluorescent
substrate, tyramine-A488.
Wild-type HRP is expressed in (A) galactose induction
medium or in (B) galactose induction medium supplemented with protoporphyrin IX
(150 pg/mL), ferric citrate (100 pM), and 6-aminolevulinic acid (250 pg/mL). HRPdisplaying cells were incubated with tyramine-A488 for the indicated period of time
(with or without hydrogen peroxide). The HRP-catalyzed reaction was then quenched
and the cells were analyzed using analytical flow cytometry.
105
.......................
. .........
N
A
Ba
653%k
62.3%
S3.6%
..0E
E
z
Phycoerythrin Fluorescence
Figure A.2.
Effect of medium supplementation on display levels of HRP.
Flow
cytometry analysis of wild-type HRP after incubation with antibodies labeled with
phycoerythrin.
(A) Galactose induction media.
(B) Galactose induction media
(C) Galactose induction media
supplemented with porphyrin IX (150
pg/mL).
supplemented with porphyrin IX (150
ptg/mL), ferric citrate (100
aminolevulinic acid (250 pg/mL).
106
pM), and 6-
Appendix B: Enantioselectivities of L and D selective variants
discovered in each round of directed evolution toward substrates 1, 2, 3,
and 4
The values shown in the tables below are represented in a graphical format in Figure 3.4.
Table B.1. Enantioselectivities of L selective yeast-bound HRP variants toward 1
and 2
HRP variant
E(L/D)
E(L/D)
(1)
(2)
Wild-type
1.6 ±0.5
LIr
4.5 ± 0.2
LIs
4.2 ± 0.3
2.3
0.1
LIrs
24± 1
5.4
0.2
LIIr
29± 1
8.3
0.2
LIII
49± 1
107
0.8
14
0.1
1
10 ±1
Table B.2. Enantioselectivities of D selective yeast-bound HRP variants toward 1
and 2
E(D/L)
E(D/L)
(1)
(2)
Wild-type
0.6 ±0.3
1.2 ±0.1
DIs
1.6 ±0.1
5.1 ±0.2
DIIs
3.1 ±0.5
31± 4
DIII
3.1 ±0.5
77 ±1
HRP variant
Table B.3. Enantioselectivities of L selective soluble HRP variants toward 1, 2, 3,
and 4
E(L/D)
E(L/D)
E(L/D)
E(L/D)
(1)
(2)
(3)
(4)
Wild-type
1.4 ± 0.1
0.7 ± 0.1
1.0 ± 0.1
1.0 ± 0.2
LIr
2.4 ± 0.1
2.0 ± 0.2
6.1 ± 0.2
5.8 ± 0.3
LIs
3.7 ± 0.1
1.9 ± 0.1
1.4± 0.1
0.1
0.3
LIrs
6.3 ± 0.3
2.2 ± 0.2
9.2
0.2
5.0
0.2
LIfr
7.8 ± 0.3
3.0 ± 0.1
12
1
7.8
0.3
LIII
9.5 ± 0.2
3.5 ± 0.1
17
1
9.1
0.1
HRP variant
108
Table B.4. Enantioselectivities of D selective soluble HRP variants toward 1, 2, 3,
and 4
E(D/L)
E(D/L)
E(D/L)
E(D/L)
(1)
(2)
(3)
(4)
Wild-type
0.7 ±0.1
1.4 ±0.1
1.0 ±0.1
1.0 ±0.2
DIs
1.4 ±0.1
2.3 ±0.1
2.0 ±0.1
1.9 ±0.3
DIs
3.0 ±0.1
5.7 ±0.1
2.8 ±0.2
3.1 ±0.3
DIII
3.4 ±0.1
13 ±1
3.7 ±0.1
5.2 ±0.1
HRP variant
109
The values shown in the table below are represented in a graphical format in Figure 4.1.
Table B.5. Enantioselectivities of yeast-bound Argl78X variants toward 2
Amino acid residue at position 178
E(L/D)
(2)
Ala
4 ±1
Arg (wild-type)
0.8 ±0.2
Asn
6 ±1
Asp
13
1
Glu
20
3
Gln
14
1
Gly
4
1
His
6
1
Ile
10
1
Ile
10
1
Leu
8 ±1
Lys
1.8 ± 0.3
Met
9 ±1
Cys
Phe
Pro
4 ±1
Ser
6 ±1
Thr
6 ±1
Trp
Tyr
10
Val
110
2