Citation
The Role of Arginine 270 and 92 Residues in the Catalytic Mechanism of the Recombinant Bacillus subtilis Oxalate Decarboxylase

Material Information

Title:
The Role of Arginine 270 and 92 Residues in the Catalytic Mechanism of the Recombinant Bacillus subtilis Oxalate Decarboxylase
Creator:
WROCLAWSKA, EWA ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Active sites ( jstor )
Binding sites ( jstor )
Catalysis ( jstor )
Enzymes ( jstor )
Kinetic isotope effect ( jstor )
Kinetics ( jstor )
Malonates ( jstor )
Oxalates ( jstor )
Proteins ( jstor )
Purification ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Ewa Wroclawska. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
12/18/2004
Resource Identifier:
57731783 ( OCLC )

Downloads

This item has the following downloads:

wroclawska_e ( .pdf )

wroclawska_e_Page_42.txt

wroclawska_e_Page_57.txt

wroclawska_e_Page_08.txt

wroclawska_e_Page_23.txt

wroclawska_e_Page_76.txt

wroclawska_e_Page_02.txt

wroclawska_e_Page_05.txt

wroclawska_e_Page_55.txt

wroclawska_e_Page_10.txt

wroclawska_e_Page_21.txt

wroclawska_e_Page_60.txt

wroclawska_e_Page_58.txt

wroclawska_e_Page_63.txt

wroclawska_e_Page_46.txt

wroclawska_e_Page_07.txt

wroclawska_e_Page_75.txt

wroclawska_e_Page_20.txt

wroclawska_e_Page_72.txt

wroclawska_e_Page_36.txt

wroclawska_e_Page_03.txt

wroclawska_e_Page_27.txt

wroclawska_e_Page_16.txt

wroclawska_e_Page_39.txt

wroclawska_e_Page_56.txt

wroclawska_e_Page_19.txt

wroclawska_e_Page_50.txt

wroclawska_e_Page_44.txt

wroclawska_e_Page_43.txt

wroclawska_e_Page_28.txt

wroclawska_e_Page_15.txt

wroclawska_e_Page_49.txt

wroclawska_e_Page_35.txt

wroclawska_e_Page_17.txt

wroclawska_e_Page_45.txt

wroclawska_e_Page_64.txt

wroclawska_e_Page_53.txt

wroclawska_e_Page_33.txt

wroclawska_e_Page_31.txt

wroclawska_e_Page_71.txt

wroclawska_e_Page_12.txt

wroclawska_e_Page_67.txt

wroclawska_e_Page_65.txt

wroclawska_e_Page_69.txt

wroclawska_e_Page_18.txt

wroclawska_e_Page_30.txt

wroclawska_e_Page_37.txt

wroclawska_e_Page_41.txt

wroclawska_e_Page_74.txt

wroclawska_e_Page_06.txt

wroclawska_e_Page_59.txt

wroclawska_e_Page_01.txt

wroclawska_e_Page_04.txt

wroclawska_e_Page_66.txt

wroclawska_e_Page_22.txt

wroclawska_e_Page_62.txt

wroclawska_e_Page_09.txt

wroclawska_e_Page_29.txt

wroclawska_e_Page_34.txt

wroclawska_e_Page_48.txt

wroclawska_e_Page_51.txt

wroclawska_e_Page_32.txt

wroclawska_e_Page_26.txt

wroclawska_e_Page_68.txt

wroclawska_e_Page_52.txt

wroclawska_e_Page_73.txt

wroclawska_e_Page_25.txt

wroclawska_e_Page_61.txt

wroclawska_e_Page_24.txt

wroclawska_e_Page_47.txt

wroclawska_e_Page_54.txt

wroclawska_e_Page_40.txt

wroclawska_e_Page_13.txt

wroclawska_e_Page_38.txt

wroclawska_e_pdf.txt

wroclawska_e_Page_70.txt

wroclawska_e_Page_14.txt

wroclawska_e_Page_11.txt


Full Text












THE ROLE OF ARGININE 270 AND 92 RESIDUES IN THE CATALYTIC
MECHANISM OF THE RECOMBINANT BACILLUS SUBTILIS OXALATE
DECARBOXYLASE













By

EWA WROCLAWSKA


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

Ewa Wroclawska





































This thesis is dedicated to my parents and sister with many thanks for their love and
support.















ACKNOWLEDGMENTS

I would like to thank the following people for their help and support:

My teachers and mentors: Henryk Koroniak, KrzysztofKieliszewski,

Hanna Gasowska, Jim Deyrup

Thesis advisor Nigel Richards

My committee members Mike Scott and Tom Lyons

Co-workers from the Richards' research group, especially Drazenka

Svedruzic and Patricia Moussatche

Laurie Reinhardt for kinetic isotope effects

My family

My friends in Poland and Gainesville

Charlie Hughes
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TA BLE S ......... ............................ ........... .......... ....... viii

LIST OF FIGURES ......... ........................................... ............ ix

ABSTRACT ........ .............. ............. ...... ...................... xi

CHAPTER

1 INTRODUCTION, BACKGROUND AND AIMS ...............................................1

O x a lic A c id ........................................................ ................. .
O xalate D egrading Enzym es ............................................................................ 2
Oxalate D ecarboxylase (OxD c) ........................................................ ............. ..4
Properties ........................................................................................ .....4
The Two Crystal Structures of Oxalate Decarboxylase .......................................6
Catalytic Mechanism and Identity of the Active Site of Oxalate Decarboxylase ........9
Comparison of the Published Crystal Structures................................................9
Closed vs O pen Site.......................... .... .. .... .. ......... ................ 11
M echanism of Catalysis Early Proposals ...................................................... 12
Mechanism of Catalysis Based on the Heavy-Atom Kinetic Isotope Effects.....14
A active Site Identity ..... ............... .......... ................ ............................. 16
Characterization of Oxalate Decarboxylation an Overview of This Work ............17

2 M ATERIALS AND M ETHODS ........................................ ......................... 22

Expression of Recombinant Bacillus Subtilis Oxalate Decarboxylase....................22
W ild Type, R270A and R270K ........................................ ..... ............... 22
R 9 2 K .................................... ...... ............ ............ ... .................. ............... 2 3
Purification of Recombinant Bacillus Subtilis Oxalate Decarboxylase ...................23
B uffer and Solvent Filtration..................................... ............................ ........ 23
Cleaning-in-Place of the FPLC Columns.........................................................23
A nionic exchange colum ns ........................................ ....... ............... 23
Hydrophobic column.............................. ............ ............ ... 24
S am ple Ionic Strength .............................................................. .....................24
Fraction Concentration .................................... ...... ............... 24
Purification of the W ild Type OxDc ................................................ ............. 25









Purification of R270A, R270K and R92K Mutants of Oxalate Decarboxylase..25
Optimization of the Purification .................... ........... .. ...............26
Site-Directed Mutagenesis and Cloning of OxDc ................................................ 27
PCR Reactions ......... .. .. ........................................ 27
P lasm id P reparation ............ ... .................................................. ...... .... ..... 28
T ran sform action ..............................................................29
E nzy m e A ssay s .......................................... ................
Q uantitative A ssay ............. ...................................... .............. .. .... ..... .29
Qualitative Activity Assay .............................. .................... .. .. ...29
M ichaelis-M enten K inetics ............................................ ........................... 30
The pH D dependence .................................. .........................................30
Protein Concentration ........ ... .......... .......... .. ................ ............. 30
Inhibition Studies........................................................... .. .....31
Heavy-Atom Kinetic Isotope Effects ................. ....... ..... ......................... ...............31

3 RESULTS AND DISCU SSION ........................................... .......................... 32

Site-Directed M utagenesis........................................ ...................................... 32
Wild Type (WT) Oxalate Decarboxylase .............. ...........................................32
Expression and Purification ........................................................................33
R 2 7 0 A .......................................................................... 3 3
R 2 7 0 K .......................................................................... 3 4
R 9 2 K ............... ......................................................................3 5
Purification optimization....................... ..... .......................... 35
Expression optim ization ..................................................... ....... ......... 38
R e su lts ................................................................3 8
Steady-State K inetics................................................... 40
R 2 7 0 A .........................................................4 0
R 2 7 0 K .........................................................4 2
A activity ........................................... ..................... ........ 42
K inetic param eters.......... ............................................ ........ ........ 43
In h ib itio n stu d ie s .................................................................................... 4 4
T h e pH dep en den ce ................................................................................ 4 7
R 9 2 K ............................................................................... 4 8
A activity ........................................... ..................... ........ 48
K inetic param eters.......... ...................................... ............ ... ......... 48
Inhibition studies ............................................ .. .. .... .. ............ 50
The pH dependence ............................................. ................................... 51
H eavy-Atom Kinetic Isotope Effects................................................. .. ...... .... 51
R 270K ..................................................................... . .. .. 5 1
R 9 2 K .................................................................. ..................................... 5 2

4 CON CLU SION S .................................. .. .......... .. .............54

Expression and Purification of the Wild Type and Mutated OxDc............................54
Steady-State Kinetics of the Wild Type and Mutated OxDc ...................................55
H eavy-A tom K inetic Isotope Effects............................................... .....................57









Active Site Identity and Mechanism of Catalysis of OxDc.................... ........ 58
T he Future of the Project ........... ................. ................. .................... ............... 59

L IST O F R E F E R E N C E S .............................. ........................................ ........................... 6 1

B IO G R A PH IC A L SK E TCH ..................................................................... ..................64
















LIST OF TABLES


Table page

1 Activity decrease of the mutated OxDc in comparison to the native enzyme. ........13

2 13C and 180 kinetic isotope effects in the wild type oxalate decarboxylase
catalyzed reaction. ................................................................. .. ......... 16

3 Kinetic constants for the reactions catalyzed by the His-tagged wild type and
m utated O xD c. ..................................................... ................. 17

4 Primers for mutagenesis experiments. .......................................... ............... 27

5 W ild type OxDc characterization................................ ........................ ......... 33

6 Purification table for R92K mutant of OxDC. ................................. ...............40

7 Characterization of the R270A mutant of oxalate decarboxylase............................40

8 Kinetic characterization of the R270Kmutant of OxDc................ ..................42

9 Inhibition studies of R270K mutant of oxalate decarboxylase ..............................45

10 Kinetic parameters for R270K mutant of OxDc at pH 4.2 and 5.7........................47

11 Kinetic characterization of the R92K mutant of oxalate decarboxylase ................48

12 Inhibition studies of R92K mutant of OxDc. .....................................................50

13 Kinetic parameters for R92K mutant of OxDc at pH 4.2 and 5.7..........................51

14 13C and 180 kinetic isotope effects for R270K mutant of OxDc............................52

15 13C and 180 kinetic isotope effects for R92K mutant of OxDc..............................53

16 Comparison of kinetic characteristic of wild type OxDc and its active site
m utants. ....................................................................57

17 Summary of 13C and 180 kinetic isotope effects for the wild type and mutated
O x D c ............................................................................... 5 8
















LIST OF FIGURES


Figure page

1 Classes of enzymes that catalyze the degradation of oxalate in (a) plants,
(b) fungi, (c) bacteria .................. ............................. ........ .. ........ .. ..

2 Part of sequence alignment of oxalate decarboxylases from different organisms .....5

3 Structural similarity between the two domains of OxDc........................................6

4 Comparison of metal binding sites of OxDc.......................................................7

5 Structure of the OxDc monomer.................. ........... ......... .................

6 Comparison of the manganese ion binding sites of oxalate decarboxylase in the
tw o structures and a m odel ......................................................................... ... ... 10

7 Closure of the lid in oxalate decarboxylase ............. ............................................11

8 Catalytic mechanism proposed for oxalate decarboxylase by Anand et al..............13

9 Catalytic mechanism proposed for oxalate decarboxylase by Reinhardt et al.........14

10 Putative active sites of oxalate decarboxylase from the X-ray crystal structure......19

11 Overview of the QuikChange Site-Directed Mutagenesis method ..........................28

12 Purification results of the R270A mutant of OxDc...............................................34

13 R92K purification: fractions from the DEAE-Sepharose Fast Flow column..........36

14 R92K purification: fractions from Q-Sepharose Hi-Perfomance column ..............37

15 Expression results for R92K .............................................................................. 39

16 R 92K purification ....... ...................................................................... ... .... ..... .. 39

17 Michaelis-Menten kinetics of mutated oxalate decarboxylase: R270A .................41

18 Catalysis of format production by R270A mutant of OxDc as a function of
tim e ...................................... ..................................................... 4 2









19 Kinetic characterization of R270K mutant of OxDc..............................................44

20 R270K inhibition by m alonate.. ........................... ..............................................46

21 R270K inhibition by malonate Lineweaver Burk plot.......................................46

22 Catalysis of format production by R92K mutant of oxalate decarboxylase as
a function of tim e ................................................... .. ............ .............. .. 49

23 Michaelis-Menten kinetics of mutated oxalate decarboxylase: R92K...................49















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

THE ROLE OF ARGININE 270 AND 92 RESIDUES IN THE CATALYTIC
MECHANISM OF THE RECOMBINANT BACILLUS SUBTILIS OXALATE
DECARBOXYLASE

By

Ewa Wroclawska

December 2004

Chair: Nigel G. J. Richards
Major Department: Chemistry

Oxalate and its salts are widespread in nature and have many pathogenic effects on

humans and plants. Enzymes involved in the synthesis and degradation of oxalate are not

well understood. Oxalate decarboxylase (OxDc) is an enzyme that catalyzes the unique

conversion of oxalate to format and carbon dioxide without participation of any organic

cofactors. Two recently published crystal structures revealed that OxDc is a hexamer,

with two manganese ions per monomer and that it belongs to the bicupin superfamily of

proteins. The N-terminal metal binding site differs between the two crystal structures.

This putative active site was considered inactive in one proposal but capable of binding

the substrate and performing catalysis in the second proposal. The crystal structures

showed arginine residues (92 and 270) in the proximity of the metal centers in both N-

and C-terminal domains. Reinhardt et al proposed the role of Arg residue to be

facilitation of the decarboxylation process by polarizing the C-O bond of the oxalate

radical anion. Just et al showed that substitution of Arg92 with alanine and lysine









residues resulted, respectively, in deactivation or 100-fold decrease of activity compared

to that of the wild type. The same substitutions of Arg270 resulted in 100-fold activity

decrease for the alanine mutant and 50-fold for the lysine mutant. Purpose of this

research is a more detailed characterization of the arginine mutants. The role of arginine

residues in the polarization of the C-O bond was tested in both putative active sites with a

series of experiments using the R270A, R270K and R92K mutants of oxalate

decarboxylase. The characterization included steady state kinetics experiments and 13C

and 180 kinetic isotope effect measurements. The important feature of the proteins

investigated in this work was the lack of any His-tag, which previously used to facilitate

purification, caused instability and activity decrease. As a result of this work, the

significance of Arg270 and Arg92 to OxDc's activity was confirmed. However, the

residues in two different sites seemed to have influenced the catalytic ability of the

enzyme to a different extent based on the steady-state kinetics characterization. Their

involvement in the actual catalysis of the decarboxylation of oxalate was proven in the

heavy-atom kinetic isotope effects experiments. The observed isotope effects supported

the previously proposed mechanism of oxalate degradation that involves proton-coupled

single electron transfer and a formation of a radical intermediate.














CHAPTER 1
INTRODUCTION, BACKGROUND AND AIMS

Oxalic Acid

Oxalic acid is a highly toxic compound involved in many environmental,

geochemical and biological processes. 1,2 It is produced by microbes, fungi and plants as a

byproduct of degradation of oxaloacetate, glyoxylate and L-ascorbic acid. These

organisms possess catabolic pathways that can to some extent control the levels of

oxalate.3 4

Oxalate accumulation in plant tissues leads to many pathological conditions caused

primarily by its metal chelating ability. A number of essential minerals in the soil

precipitate after binding oxalate. Availability of phosphorus to plant roots is increased

due to the oxalate chelation of aluminum and calcium.2 Fungal pathogens can utilize

oxalate to their own advantage but also secrete it into the plant tissues during the initial

stages of pathogenesis, which causes cell degradation.5 Moreover, oxalate plays a role in

the regulation of osmotic potential and pH, as well as in calcium ion storage in plants.6

Oxalate is a key factor in the carbon cycle and in CO2 release from rotting wood.7

Fungi use oxalate manganese complexes to promote degradation of lignin, which affects

enzymes responsible for cell wall synthesis.8 For example, the fungus Whetzelinia

sclerotinium, uses oxalate to induce damage to sunflower plants. During pathogenesis,

the concentration of oxalate increases in the host tissue leading to leaf death.4

Humans and other vertebrates consume oxalic acid, found mainly in green leafy

plants such as spinach and rhubarb but also in black tea and ginger juice. However,









humans lack oxalate degrading enzymes. It was proposed that intestinal bacteria, such as

Oxalobacterformigenes, could be introduced into human gastrointestinal tract to

catabolize oxalate.3 These bacteria can, however, be eliminated from the gut flora by

extensive antibiotic treatments, which results in the increase of oxalate levels in humans.9

A number of pathological conditions such as hyperoxyluria, formation of kidney stones,

renal failure, cardiomyopathy and vulvodynia are caused by oxalate.10

Due to the problems related to oxalate accumulation, numerous efforts have been made to

reduce the amount of oxalate in food, including engineering transgenic plants to enable

them to express oxalate degrading enzymes.5 Structural, biochemical and mechanistic

information needs to be obtained for oxalate degrading enzymes to utilize them in

therapy, industry and agriculture.11'12 The most interesting aspect of oxalate degrading

enzymes is the variety of mechanisms they employ.13

Oxalate Degrading Enzymes

Oxalate degrading enzymes have potential uses in new therapeutic strategies for

lowering oxalate levels in biological fluids. For many years now, these enzymes have

been used in urine and blood testing for the presence of oxalate.14,15 Recently, re-

colonizing humans with intestinal bacteria that produce these enzymes has been used as a

preventive therapy.9 This new approach has been introduced, but has not been widely

recognized.

Three major enzymes have evolved in plants, fungi and bacteria. All of which

catalyze the degradation of oxalic acid. Each of these enzymes employs a different

mechanism, such as oxidation, decarboxylation in the presence of coenzyme A, or direct

decarboxylation [Figure 1].13










O
O I," Oxalate oxidase C2 H202

S02, 2H+
B. O
OB | Oxalate decarboxylase H 0
O cat. 02, H+
O O

C. O 0 Oxalyl-CoA decarboxylase C H SCoA

--O H-+ 2
0

Figure 1. Classes of enzymes that catalyze the degradation of oxalate in (a) plants, (b)
fungi, (c) bacteria.13

Oxalate oxidases (OXO) catalyze oxidation of oxalate into carbon dioxide and

hydrogen peroxide mostly in plants [Figure 1]. OXO requires molecular oxygen for its

activity but no organic cofactors.16 The main source of this enzyme is barley root.17 One

of the reaction products, H202, has been suggested to be involved in cell wall crosslinking

and can act as a fungicide. OXO has extreme thermal stability, and therefore can be

involved in the defense against biotic and abiotic stress in plants. The addition of oxidase

activity is one of the targets for transgenic plant engineering.18 OXO has been crystallized

and its structure determined for Hordeum vulgare protein.19 It belongs to the cupin

protein superfamily and has a manganese ion in its active site.17 There are three histidines

coordinating the metal ion with the fourth coordination site occupied by carboxylate from

a glutamate residue. These four amino acid residues are conserved in the sequences of

many metalloenzymes in the cupin superfamily.20 Their relevance to enzyme activity has

been confirmed through site-directed mutagenesis for another member of the

superfamily, oxalate decarboxylase.21 It was suggested that gene duplication facilitates

the evolution of new enzymes due to sequence divergence.22 This is how the bicupins

(oxalate decarboxylase) are formed from cupins (OXO).3 It has been suggested that gene









duplications must have occurred for OxDc to become a fully functional and structurally

developed enzyme during evolution from a single cupin OXO. Presumably, first, the

number of cupin genes doubled, and then the gene fusion occurred to produce the two-

domain bicupin.5

Formyl-CoA transferase (FRC), along with oxalyl-CoA decarboxylase (OXC), are

involved in the oxalate degradation pathway of Oxalobacterformigenes, a bacterium

involved in mammalian oxalate catabolism. The reaction catalyzed by FRC involves the

transfer of coenzyme A from format to oxalate producing oxalyl-CoA and format

[Figure 1].23

Oxalate Decarboxylase (OxDc)

Properties

The enzyme of interest in this study is oxalate decarboxylase (OxDc), which is

mainly found in fungi (Aspergillus niger, Flammulina velutipes, Sclerotinia

sclerotiorum) and more recently in the bacterium Bacillus subtilis.20 24 26 The most

thoroughly characterized OxDc's come from B. subtilis. The bacterium B. subtilis

reportedly possesses more than one gene encoding oxalate decarboxylase activity, YvrK

and YoaN.27

OxDc's found in fungi and bacteria have many common features. These features

are presumably responsible for their catalytic activity: the conversion of oxalate to

format and carbon dioxide [Fig. 1].24 The decarboxylation process does not require any

organic cofactors, such as coenzyme-A or ATP.28 OxDc's consume sub-stoichiometric

amounts of oxygen, relative to products, during turnover and are inactive in anaerobic

conditions. It was found that even though OxDc is sensitive to the presence of oxygen, it

is most likely only upon substrate binding.13









The enzyme is most active in acidic pH. The isoelectric point (pI) has been reported

to be 3.3 or 2.5 for the fungal OxDc and 6.1 for the bacterial one.5' 20 The latter enzyme is

stable between the pH of 4.0 and 7.5, while the optimum activity is between the pH 4.0

and 5.0.3,4 OxDc's stability is increased in the presence of o-phenylenediamine, a

compound used in the qualitative assay for oxalate decarboxylase turnover and in the

presence of surface active non-ionic detergents (Tween 20, Triton-X).24'29

Alignment of three sequences of oxalate decarboxylases from different organisms,

both fungal and bacterial, shows the conserved residues from both metal binding sites

[Figure 2].21'27

151 170...215 234
AnOxDc MRLDEGVIRE LHWHREAEWA.... NGTEFLLIFD DGNFSEESTF
FvOxDc MRLEAGAIRE LHWHKNAEWA... EGSEFILVFD SGAFNDDGTF
BsOxDc MRLKPGAIRE LHWHKEAEWA.....EGAEFLLVFD DGSFSENSTF
84 103...147 166

331 350.....401 420
AnOxDc AAAHLTINPG AIREMHWHPN......EEVEVLEIFR ADRFRDFSLF
FvOxDc AVAEVTVEPG ALRELHWHPT......TTLTYLEVFN TDRFADVSLS
BsOxDc ASALVTVEPG AMRELHWHPN......EPLVFLEIFK DDHYADVSLN
258 277....327 346


Figure 2. Part of sequence alignment of oxalate decarboxylases from different organisms:
Aspergillus niger (AnOxDc), Flammulina velutipes (FvOxDc), Bacillus
subtilis (BsOxDc). Conserved residues of interest in the active sites are shown
in bold letters.

The residues of interest in this project are Arg92 and Arg270 conserved in the N-

and C-terminal domains respectively. The positively charged arginines were predicted to

polarize C-O bond in oxalate. Also presented is conserved Glu333 from the C-terminal

active site that according to the crystal structure on which this work was based, can serve

as a general base during catalysis of OxDc.3'13 Conserved Glu162 from the N-terminal









domain, which was presumed to be catalytically relevant in the latest publication is also

shown.6

The Two Crystal Structures of Oxalate Decarboxylase

The structure of the fungal oxalate decarboxylase has been proposed using

sequence homology between the bicupins coming from different species and from a

structural model.21 The crystal structure of bacterial enzyme was determined in 2002 by

Anand et al and by Just et al in 2004.3 6

The first crystal structure of bacterial OxDc was solved at 1.75 A resolution in the

presence of formate3 B. subtilis OxDc crystallizes as a hexamer, which contains two

trimeric layers in which each monomer belongs to the bicupin structural family. One

domain of the monomer is at the C-terminus and the other at the N-terminus [Figure 3].

A B











Figure 3. Structural similarity between the two domains of OxDc.3 Blue color: N-
terminus, red color: C-terminus. (A) Domain I includes residues 56 233. (B)
Domain II includes residues 8 55 and 234 379. Reprinted with permission
of Anand et al, Biochemistry (2002) 41, 7659. Copyright 2002 American
Chemical Society.

Both cupin domains contain a manganese ion coordinated to one glutamate and

three histidine residues [Figure 4]. The molecular mass of the enzyme is 264 kDa for the

entire hexamer and 43.6 Da per monomer. It has the motifs characteristic to the cupin










superfamily: a P-sandwich, which consists of a six-stranded P-sheet and a five-stranded

P-sheet. The contact between the subunits is established by several a-helices [Figure 5].3


E333


Format


Figure 4. Comparison of metal binding sites of OxDc. (A) manganese binding site of
domain I. (B) manganese binding site of domain II. (C) Unknown metal site at
the protein surface. The metal was assigned to be magnesium for the purpose
of the X-ray refinement. Reprinted with permission of Anand et al,
Biochemistry (2002) 41, 7659. Copyright 2002 American Chemical Society.


H#42
C V..


L1S3












OWW. I


Ci~.:" 0.i
A~iyJ->&


-up


Figure 5. Structure of the OxDc monomer.3 (A) Stereoview of the Ca trace of OxDc. The
color changes from red to blue from N-terminus to C-terminus. (B) Structure
of the OxDc monomer, highlighting the secondary structural elements with 3-
sheets and a-helices colored as in panel C, 310 helices in cyan and loops in
yellow. (C) Topology diagram of OxDc, showing the domains I and II. The
six-stranded P-sheets that make up the front of the cupin barrel are in blue,
and the five-stranded P-sheets that make up the back are in red. The a-helices
are in green. Reprinted with permission of Anand et al, Biochemistry (2002)
41, 7659. Copyright 2002 American Chemical Society.


i
+*,; r
--
-1 ;'Y
L~7~ i
~9~dn I









The two metal binding sites described above are presumed to be the enzyme's

active sites, and there is an ongoing discussion as to which domain is actually responsible

for catalysis. In this model, there is a third metal binding site at the surface of the protein

that is most likely not involved in catalysis [Figure 4].3

The crystal structure described most recently for bacterial oxalate decarboxylase

was obtained based on the refinement of the coordinates from the first study as well as

additional, new 2.0 A resolution data. It differs from the previous one in the conformation

of one of the 310 helices, which creates a loop near the N-terminal metal binding site and

changes the identity of the second shell residues available to catalysis. Motifs defining

OxDc as a member of a cupin family, as well as metal coordinating residues in the

binding sites are exactly as described in the previous study. However, there are

significant changes in water occupancy of the active sites and their accessibility to the

substrate [Figure 6].6

Catalytic Mechanism and Identity of the Active Site of Oxalate Decarboxylase

Comparison of the Published Crystal Structures

In both published crystal structures of oxalate decarboxylase, there are two cupin

domains in the enzyme that are similar. Both metal binding sites are presumed to be

capable of performing the catalytic reaction.3

The second coordination sphere in the N-terminal site depends on the conformation

of a 310 helix that is different in both proposals. In both structures arginine residues

(Arg270 and Arg92) have the same position near the manganese ions. The position of

glutamate residue (E333) in the C-terminal site is the same in both proposals, while the

position of the glutamate in the N-terminus (E162) is not [Figure 6].










A Site 1 Closed
, lu 162 Iw; "









C Site I Open
S ,
2S- '-a











^iM


B Site 2 Closed


D Site 2 Open
.b


2$ -f-N~ 4
IB2\;'I3 '*-* H'^'. l


Y r -<


E Site 1 Model


vArge2








Figure 6. Comparison of the manganese ion binding sites of oxalate decarboxylase in the
two structures and a model.6 Metal binding sites of the closed structure (site 1
= A and site 2 = B) are shown next to the open structure sites ( site 1 = C; site
2 = D). In the molecular model (E) oxalate and dioxygen are bound to site 1
manganese ion and the need for displacement of Glu-162 is shown in
comparison with the experimental closed structure (A). Dashed lines represent
interionic distances. Reprinted with permission of Journal of Biol. Chem.
(2004) 279, 19867. Copyright 2004 ASBMB.

The most significant change in the N-terminal site is related to the movement of the

surface loop that includes the aforementioned 310 helix and is built from residues 161 to

165 (Ser, Glu, Asn, Ser, Thr) [Figure 7].3,6 These residues create a lid in the putative


active site structure.






























Figure 7. Closure of the lid in oxalate decarboxylase.6 Open (A) and closed (B)
hexameric structures are shown partially. Solvent-accessible surface of the
protein is shown in green, interior of the protein in pale yellow, solvent in
blue. The trajectory for substrate entry is indicated by a broken arrow.
Reprinted with permission of Journal of Biol. Chem. (2004) 279, 19867.
Copyright 2004 ASBMB.

Closed vs Open Site

The presence of the lid, identified in the 2004 structure, constitutes a dramatic

change in understanding of the role of the N-terminal site. In the first structure, the loop

created the channel allowing an easy access of the solvent to the Mn ion in the N-

terminus. The site was therefore called open. In the latest structure the loop occludes the

channel and prevents solvent access to the Mn-binding site. Therefore the site is called

closed. The conformational flip of the loop positions the glutamate 162 side chain in very

close proximity (under 2 A) to the metal and makes it a great candidate to be a part of the

catalytic mechanism as a proton donor.6 The conformational change is relevant, because









before this proposal only the C-terminal site was considered to be a good candidate to

perform catalysis in oxalate decarboxylase.

Mechanism of Catalysis Early Proposals

There have been a few mechanisms of catalysis proposed for oxalate decarboxylase

throughout the years. Initial characterization of the fungal OxDc included mutagenesis

studies, which supplied evidence for the crucial catalytic role of the manganese ions by

substituting metal binding residues. Removal of any of them in either of the sites resulted

in the complete inactivation of the enzyme. Therefore, it has been assumed that the two

manganese binding sites are acting in cooperation.21 The metal contents ofF. velutipes

recombinant enzyme were established in the same study as 2.5 Mn per monomer.

Emiliani and Riera have also found traces of hydrogen peroxide as an additional product

of the oxalate decarboxylase catalyzed reaction, which suggested a single electron

transfer to be involved in the mechanism.24 It has been established that the fungal enzyme

requires oxygen for the reaction, even though there is no net redox change during the

process.20 Subsequently, it was proposed that the decarboxylation cycle of the bacterial

enzyme included a percarbonate intermediate, however, this has never been proven.6

Publication of the first crystal structure from B. subtilis, as well as kinetic isotope

effect measurements, for the recombinant enzyme, have provided an interesting insight

into the nature of the bacterial OxDc.3' 13 It has been suggested that all the residues

necessary for catalysis are found only in the C-terminal metal binding site, while the

general base (Glu) was missing in the N-terminal one.3 The proposed mechanism

required an enzymatic source of acyl proton for format [Figure 8].










HO'0 HO
SGlu Glu
/4+ 3+
His Mn O His- Mn
His o His
His O His 0
0


HO0
Glu
/3+
His-Mn O
His o0 C
His 0
0


HO
HO HO.
Glu O O O
4+ OH Glu
His-Mn H OH I 4+
Os nHis MnC
His OH 0 His-Mn C
iHis i H
His His
0


Figure 8. Catalytic mechanism proposed for oxalate decarboxylase by Anand et al.3

Mutagenesis studies confirmed that mutating the conserved Arg270 and Glu333

from this active site resulted in a significant activity decrease [Table 1]. The proteins

were, however, purified with an N-terminal His-tag that caused their precipitation and

instability.

Table 1. Activity decrease of the mutated OxDc in comparison to the native enzyme.3
OxDc mutant Activity decrease
compared to the
native enzyme [fold]

E333A 4
Y340F 13
R270E 20

The role of Arg270 was predicted to form an ion pair with the second carboxylate

of the substrate in order to stabilize the charge division in the molecule and the position

of oxalate in the active site. This stabilization of the intermediate was expected to

facilitate the decarboxylation process.3










Mechanism of Catalysis Based on the Heavy-Atom Kinetic Isotope Effects

Heavy-atom kinetic isotope effect (KIE) studies on the wild type oxalate

decarboxylase have supplied new information about the mechanism of catalysis of the

enzyme [Figure 9]. The C-terminal site was, at the time, considered to be the only one

comprising of all the pieces necessary for catalysis.13


0

OHO
S H "0 Glu333 O HO GIu333


S3+ 0 0O
His-Mn 2+ 0
His \Gu. H +-H His-Mn
His Glu G u +
NHisN His Glu H-N+-H
N N Arg270 His H.N N
H H N Arg270
H H


CO2

0
O ^ Glu 333

HO0 O H HO, H u333

3+ 0 2+ C:
His-Mn His-Mn O-
His Glu H-N+H His \Glu H.N+-H
His His
His HN N Arg270 HN N Arg270
H H H H
Figure 9. Catalytic mechanism proposed for oxalate decarboxylase by Reinhardt et al.13

Kinetic isotope effects can be used to deduce the structure of the transition

state in enzyme mechanisms in which bond-breaking and bond-making events are rate

limiting.30 A KIE is defined as the change in the rate of reaction with an isotope labeled

substrate. It provides information on structural changes in going from the reactants'

ground states to the transition state. Changes in bond orders result in isotope effects. The









magnitude of KIE's depends on the extent to which the chemical step is rate limiting in

the catalysis. The intrinsic isotope effect may be masked if the rate limiting step is either

binding or conformational changes, and not chemistry.31-33 The isotope effects on

Vmax/KM (V/K KIE's) are associated with the steps up to and including the first

irreversible step in the mechanism.34 35

The proposed mechanism suggests that an electron is transferred from bound

substrate, oxalate, to the manganese and molecular oxygen complex. It seems that the

enzyme must stabilize the radical species before the C C bond cleavage. Positively

charged arginine residue and carboxylate group of a glutamate carrying a negative charge

can serve this purpose. After the decarboxylation, transfer of an electron and a proton to a

metal-bound format radical anion yields the final product [Figure 9].13

The values of isotope effects for the wild type oxalate decarboxylase

confirmed that the C-C bond cleavage is not the rate-limiting step [Table 2]. The values

for carbon dioxide production are lower than the typical carbon bond cleavage KIE's

values of 3 to 5 %. Results for format production suggest that a different step, prior to

the bond cleavage, was slower and rate-limiting. This assumption is supported by (i) the

catalytic dependence on the dioxygen presence, (ii) the presence of manganese, (iii) the

absence of organic cofactors, and (iv) no net redox change between substrates and

products. At pH 5.7 chemical steps are even more rate limiting than at pH 4.2. This is

probably due to the decrease in external commitments to catalysis. The bond order

increased for oxygen as the reaction proceeded (from 1.5 to 2), therefore the 80 KIE

values are inverse for CO2 production. The substrate-based radical formation was

expected to facilitate the cleavage of the C-C bond. The rate-limiting step predicted in









this mechanism is the oxidative transfer of an electron from oxalate to a Mn (III) -

dioxygen complex, coupled with the hydrogen abstraction.13 Arginine residues in the

enzyme's metal binding sites are expected to facilitate the reaction by polarizing the C-O

bond and provide stabilization to the charged intermediate as described for the previous

mechanism.

Table 2. 13C and 80 kinetic isotope effects in the wild type oxalate decarboxylase
catalyzed reaction.13
pH 13(V/K), % 13(V/K), % 18(V/K), % 18(V/K), %
CO2 format CO2 format

4.2 0.5 + 0.1 1.5 + 0.1 -0.2 + 0.2 1.1 + 0.2
5.7 0.8 0.1 1.9 + 0.1 -0.7 + 0.1 1.0 + 0.1

The wild type kinetic isotope effects experiments will be used as a model for

the mutant characterization. The aim of the experiments on the mutants is to establish

whether the mutation affects the stability of putative intermediates.

Active Site Identity

The most recent findings were published in 2004 along with the new crystal

structure.6 As described above, the conformational change resulted in opening the N-

terminal site for catalysis and positioning Glu162 as an equivalent of Glu333, present in

the C-terminal site. Hence, there is a general base available for catalysis in both metal

binding sites. Further mutagenesis experiments have been reported to support the new

thesis that the N-terminal site is more likely to be involved in the catalysis. Just et al have

observed a decrease in activity of the enzyme of 100 fold for the N-terminal site mutants

and 10 to 50 fold for the C-terminus mutants [Table 3].6 It was suggested that the C-

terminal site, with restricted access for both solvent and substrate, has a structural role in

the catalysis. The steady-state characterization of Arg92 and Arg270 mutants was









published but the proteins used for the experiments contained C-terminal His-tags. The

N-terminally His-tagged OxDc was found to be unstable leading to precipitation.3

Comparison of the specific activity values obtained by Just et al and those described in

this work suggest the unfavorable effect that also the C-terminal His-tag has on oxalate

decarboxylase. Specific activity of the wild type protein obtained in this study is about 40

U/mg while the enzyme used by Just et al only 21 U/mg.6 The activities of all mutants

described along with the second crystal structure are highly decreased. In this work, the

non-tagged mutated proteins were used, and all the results supply information about a

stable, non-precipitated enzyme, that is most likely at its highest possible activity.

Table 3. Kinetic constants for the reactions catalyzed by the His-tagged wild type and
mutated OxDc.6 Limit of detection: 0.03 U/mg
OxDc Specific activity KM kcat/KM
[U/mg] [mM] [M-s-1]

Wild type 21.0 16.4 952
R92A 0 not determined not determined
R92K 0.20 2 68
R270A 0.26 8 24
R270K 0.54 1 410

Characterization of Oxalate Decarboxylation An Overview of This Work

This project describes the entire process from mutagenesis to obtaining and

characterizing a number of mutants of the B. subtilis oxalate decarboxylase. Problems

arising from working with mutants are described, as well as attempts at overcoming

them. The importance of this study comes from numerous existing and predicted medical,

industrial and environmental applications, as well as an insight into the evolutionary

processes within the cupin superfamily of enzymes.18' 36

Oxalate decarboxylase has been studied for over 50 years and only in the past few

years has progress been made. The publication of two crystal structures, as well as kinetic









isotope effect measurements, have enabled more informed proposals of the catalytic

mechanism.3, 6,13 However, there are still many unanswered question and uncertainties

about the catalysis of oxalate decarboxylase, from the identity of the enzyme's active site

to the actual mechanism of catalysis.

The two published crystal structures differ enough for their authors to propose

opposite metal binding sites as the active site.3 6 In both of these studies, however, there

are many assumptions made that are clearly not final and need unambiguous evidence.

The mutagenesis results presented by Anand et al cannot be considered a thorough

analysis due to the general instability of the protein, which included a destabilizing N-

terminal His-tag.3 Just et al have performed a much more detailed mutagenesis study on a

stable, C-terminally His-tagged enzyme. Activity of all His-tagged enzymes is decreased

compared to the non-tagged proteins described in this work. Wild type enzyme obtained

by Just et al has specific activity of 21 U/mg compared to the average of 40 U/mg of the

enzyme in this study.6

This study has been based on the crystal structure published by Anand et al in the

year 2002 [Figure 10]. 3

Investigation of the catalytic mechanism and of the role of the protein environment

in controlling properties of the metal centers in B. subtilis oxalate decarboxylase was

conducted using site-directed mutagenesis, steady-state kinetics, and heavy atom isotope

effects.37-40

The main questions that need to be answered, and which this work is addressing are

(i) what is the role of conserved arginines in the N- and C-terminal active sites, (ii) which

manganese binding site of OxDc is responsible for catalytic activity of the enzyme, and









(iii) is the decarboxylation process facilitated by the electrostatic stabilization of the

transition state by the arginine side chain?


Leu-15,
Leu- 1 J al-82

t I Tyr-340
Glu-333
Val-321
Met-94



Tyr-200
Arg-92 ,
Arg-270


Figure 10. Putative active sites of oxalate decarboxylase from the X-ray crystal structure.
Residues potentially important for the catalytic mechanism are indicated in a
three letter code.

The mutations were performed on arginines in both metal binding sites (R270 and

R92) to try and answer the question of (i) their role or significance in the catalytic

mechanism, and (iii) their role in the enzyme's substrate selectivity.

The innovative approach in this research excluded the use of any His-tags to

facilitate the purification process. The His-tags have been shown to reduce both the

activity and the stability of the enzyme.6

All the mechanisms proposed thus far predict a significant role for Arg270 and

Arg92 in sustaining enzyme's activity. The hypothesis of this work has been that these

conserved arginines play a role in either catalysis or substrate selectivity of the enzyme.

The native enzyme exhibits high substrate selectivity towards oxalate.13 Therefore,

certain carboxylic acids and diacids of size similar to oxalate might be able to bind in the

active site of the mutated enzyme and inhibit it. Replacement of positively charged Arg









should have the most dramatic effect when substituted with a neutral and small alanine

residue, and much smaller, but detectable effect with lysine. The removal or decrease of

the charge would presumably result in, respectively, lack or decrease of enzyme's ability

to polarize the C-O bond and to facilitate the decarboxylation of oxalate. The kinetic

parameters, as well as the activity and stability of the enzyme were tested after

introducing these substitutions. The change in the turnover number was not expected if

the mutated residue affected just the selectivity of the enzyme. The value of KM was

expected to change: either decrease if the mutation allowed unproductive binding or

increase if the possible inhibitors competed with oxalate for the active site. Even though

the steady-state characterization has been published along with the latest crystal structure,

the proteins used had, as mentioned above, decreased activity due to the use of a C-

terminal His-tag.

Steady state kinetics provided information on changes in enzyme activity caused

by mutations of arginine residues. Changes in the turnover number prompted the

measurements of the kinetic isotope effects. If, as predicted, the role of arginine in the

catalysis is to facilitate decarboxylation by stabilizing the negative charge in the

intermediate, the KIE's should be observed and be altered compared to wild type OxDc.

Complete removal of the positive charge by mutating Arg to Ala was expected to almost

completely deactivate the enzyme. Substituting arginine side chain with an also positively

charged, but shorter lysine would lower the rate of decarboxylation resulting in increased

values for kinetic isotope effects. Differences in values for format and CO2 arise from

their influence on the rate-limiting step: proton-coupled electron transfer. The primary

13C and secondary 1O isotope effects were measured by the analysis of C02(g) by the









internal competition method using isotope ratio mass spectrometry. Therefore the results

obtained represent the isotopic substitution effects on V/Koxaiate, and are associated with

the first irreversible step of the mechanism and all the steps leading to it.34, 35 Analyzed

CO2(g) was obtained from direct isolation from the oxalate decarboxylase, catalyzed

partial conversion of oxalate or from the oxidation of initial oxalate, format and oxalate

from residual substrate after the partial reaction in anhydrous DMSO with iodine.

Isotopic ratios obtained are the R values (defined under the equation below) for these

compounds. Analysis of KIE results for mutated OxDc was based on a literature equation

for the wild type enzyme described below.13,41

ln(1 f) n(1 f)
applE = =
ln (1-f) Rs] n l[1-fR
Ro Ro

13or1 (V / K)C02 ( + x)appIE
2x
13or18 (V / K) format = x) appIE

x = Rc0 / Rformate
Rp = (Rco2 + Rformate )/ 2

f= fraction of reaction
R isotopic ratio determined by MS for:
Ro = initial oxalate
Rs = residual substrate
Rco2 = produced CO2
Rformate = produced format














CHAPTER 2
MATERIALS AND METHODS

Expression of Recombinant Bacillus Subtilis Oxalate Decarboxylase

Plasmid, previously produced by the insertion of the yvrk gene from B. subtilis into

the pET-9a expression vector, was transformed into the competent JM109 cells, and

sequenced to confirm the presence of the 1158 base pair long insert. This plasmid was

used to transform the E. coli strain BL21(DE3). Luria-Bertani broth containing

kanamycin (LBK, 1.0 L) was inoculated with yvrk:pET9a/BL21(DE3) in order to

express the yvrk-encoded protein.13

Wild Type, R270A and R270K

Cells were shaken at 37 C until the optical density (A600) of the cultures has

reached 1.7 and the cells were ready to be induced. The bacteria were subjected to heat

shock for 18 min followed by the addition of the inducing agent isopropyl thiogalactoside

(IPTG, 1 mM) and manganese chloride (MnC12, 5 mM). After being agitated for 4 h in 37

OC the cells were harvested by centrifugation (5000 rpm, 30 min, 4 OC), resuspended in

the Tris-HCl lysis buffer containing MnC12 and sonicated for 30 s. Lysate and cell debris

were separated by centrifugation (8000 rpm, 20 min, 4 OC). The pellet suspended in the

extraction buffer consisting of 1 M sodium chloride, 10 mM 2-mercaptothanol and 0.1 %

Triton-X was stirred overnight at room temperature and the supernatant was combined

with the original lysate.13 For the R270A and R270K mutants the expression protocol was

followed as for the wild type OxDc, only the lysis pellets were discarded instead of being

used for protein extraction.









R92K

The expression protocol was significantly changed for this low activity mutant.

Competent BL21(DE3) cells transformed with R92K plasmid were used to inoculate 25

mL of LBK in a 250 mL flask, which was shaken overnight at 30 OC until the cultures

reached OD600 of 3.2. Fresh LB (4 x 200 mL) without antibiotic in baffled flasks (2 L)

was inoculated with 2 mL of the overnight culture. The cells were grown at 30 OC until

the OD600 of 0.3, at which point the cells were heat-shocked for 5 min at 42 C and

induced with IPTG and MnC12 as for the wild type. The induction time varied from 2.5 to

3.5 h depending on when the cultures reached the OD600 of 1.6. At this time the cells

were harvested. Lysis, as well as lysis pellet extraction was performed as in the wild type

OxDc procedure.

Purification of Recombinant Bacillus Subtilis Oxalate Decarboxylase

Buffer and Solvent Filtration

In order to remove any particulate matter all the buffers and water solutions were

filtered through Millipore 0.45 |tm membrane filters before use in chromatography, 20 %

ethanol was filtered through 0.20 |tm membrane filter.

Cleaning-in-Place of the FPLC Columns

All the resins used in Fast Performance Liquid Chromatography (FPLC) system

(Akta prime; Amersham Biosciences) were cleaned before loading each sample

according to manufacturer's instructions.

Anionic exchange columns

The ionically bound proteins were removed by washing DEAE-Sepharose Fast

Flow (Sigma) and Q-Sepharose Hi-Performance (Amersham Pharmacia Biotech)









columns with 50 mL of 2 M NaC1 in a reversed flow direction (4 mL/min). To remove

precipitated proteins, as well as lipopropteins and proteins bound hydrophobically,

columns were washed (2 mL/min) in the reverse flow direction with 100 mL of 1 M

NaOH. A following wash with 20 % ethanol in reverse direction flow (4 mL/min)

eliminated the strongly hydrophobically bound proteins and lipoproteins. Before use the

columns were equilibrated with approximately 150 mL of a low ionic strength buffer.

Hydrophobic column

Phenyl-Sepharose Hi-Perormance (Amersham Pharmacia Biotech) column was

cleaned in reversed flow direction by washing with 0.5 M NaOH (1 mL/min; 10 min) and

20 % ethanol (4 mL/min; 25 min). Then it was washed in the forward flow direction with

the low ionic strength buffer and before use equilibrated with the high ionic strength

buffer (both 4 mL/min; 100 mL).

Sample Ionic Strength

The ionic strength of the loaded sample was adjusted for specific resins and buffer

sets by dilutions or addition of ammonium sulfate. Anionic exchange columns require

low conductivity (10 [tQ) and hydrophobic column requires samples with high ionic

strength (> 20 [tQ).

Fraction Concentration

Active fractions from the last column before dialysis were concentrated using the

Amicon Ultrafiltration Pressure Chamber on a Millipore Ultrafiltration Membrane with

an exclusion size of 30 K. The dialysis pool was concentrated in the Ultrafree-15

Centrifugal Filter Devices (Millipore).









Purification of the Wild Type OxDc

The purification system consists of two anion exchange and one hydrophobic FPLC

columns. First resin used was the weak anionic exchange DEAE-Sepharose in a 2.5 x 30

cm column onto which 1Ox diluted lysate and pellet extract sample was loaded. The

washing buffer used for all columns was 50 mM imidazole hydrochloride with 10 uM

MnCl2 at pH 7.0. The column was eluted with 0 to 1 M NaCl gradient (500 mL total).

Collected fractions were checked for their protein contents by UV spectroscopy (A280)

and assayed for their ability to catalyze the oxidation of o-phenylenediamine. Fractions,

which displayed activity were combined and solid ammonium sulfate was added to the

final concentration of 1.7 M. Precipitated proteins were removed by centrifugation and

the supernatant loaded onto the Phenyl-Sepharose Hi-Performance column. This time the

eluting gradient was 1.7 to 0 M (NH4)2SO4 (500 mL total). Active fractions were

recognized in the same way as after the DEAE column. They were pooled and diluted

1Ox before being loaded onto the anion exchange Q-Sepharose Hi-Performance column,

from which the protein of interest was eluted using the same buffers as described for the

DEAE column. The fractions containing the enzyme were concentrated by ultrafiltration

to the volume of 10 mL and dialyzed against the storage buffer (20 mM

hexamethylenetetramine hydrochloride, pH 6.0). The enzyme was concentrated and

divided into 100 uL aliquots stored at 80 oC.13

Purification of R270A, R270K and R92K Mutants of Oxalate Decarboxylase

During the purification of the mutated OxDc only two anionic exchange columns

were used: DEAE-Sepharose and Q-Sepharose with the same buffers and elution

gradients as described for the wild type. The ammonium sulfate step was omitted along









with the Phenyl-Sepharose column. The sample loaded onto the Q-Sepharose required 5x

dilution.

Optimization of the Purification

Difficulties in the purification of mutated oxalate decarboxylase according to the

wild type enzyme purification protocol resulted in making a few attempts to find resins

more suitable for purification than Phenyl Sepharose or to find optimal ammonium

sulfate concentration for the purification step based on protein precipitation. To test the

latter different amounts of ammonium sulfate (from 0 to 1.7 M final concentrations)

were added to the 750 ptL aliquots of a pool of active fractions from the DEAE-Sepharose

column. SDS gel chromatography was used to visualize the relative amounts of

precipitated proteins. Resins tested: not used previously anion exchange Q-Sepharose,

two hydrophobic ones, Butyl- and Octyl-Sepahrose (both Amersham Pharmacia Biotech),

as well as cation exchange SP-Sepharose (Sigma). Batch-wise purification experiments

were performed. In each case 0.4 mL of resin, not slurry, was placed in an Eppendorf

tube (1.5 mL) and washed ( 0.5 h with rocking) with start buffer: buffer A (pH 7.0, 50

mM imidazole, 10 [LM MnC12) for the anion exchange resin; buffer A including 0.6 mM

(NH4)2SO4 and acetate (pH 5.2, 10 ptL MnC12) for the cation exchange resin. Then the ion

exchange resins were washed (rocking for 5 min) with buffer A solutions containing

increasing amounts of sodium chloride (from 0 to 1 M). Hydrophobic resins were washed

in the same way with solutions of start buffer containing decreasing amounts of

ammonium sulfate (from 0.6 to 0 M). All fractions were visualized on SDS-PAGE gels to

establish binding to certain resins and influence of the gradient.









Site-Directed Mutagenesis and Cloning of OxDc

Site-directed mutagenesis was performed using the QuikChange Site-Directed

Mutagenesis Kit (Stratagene) according to manufacturer's guidelines, using primers in

Table 4. DNA sequence of each mutant was obtained from the DNA Sequencing Core

Laboratory ICBR at the University of Florida.

Table 4. Primers for mutagenesis experiments.
Primer name Primer sequence 5' to 3'
R270A sens CCC GGC GCC ATG GCT GAA CTG CAC TGG
R270A asens CCA GTG CAG TTC AGC CAT GGC GCC GGG
R270K sens CCC GGC GCC ATG AAA GAA CTG CAC TGG
R270K asens CCA GTG CAG TTC TTT CAT GGC GCC GGG
R92K sens CCA GGC GCG ATT AAA GAG CTT CAC TGG
R92K asens CCA GTG AAG CTC TTT AAT CGC GCC TGG

PCR Reactions

One polymer chain reaction of 100 ptL contained 5 ptL of 10x reaction buffer

(Stratagene), 125 ng of each primer (sens and asens), 20 ng of dsDNA template, 125 ng

of dNTP mix (Stratagene) and sterile distilled water. The PCR reaction was started from

a 30 s heating step to 95 OC, after which PfuTurbo DNA Polymerase (1 ptL of 2.5 U/[tL;

Stratagene) was added to the reaction mixture. Then followed the 16 cycles of 30 s

denaturation at 95 C, 1 min annealing at 55 C and 12 min extension (2min/kb) at 68 C.

XL 1-Blue Supercompetent cells were transformed with the digested (DpnI, Stratagene)

PCR product and plated on LB/kanamycin plates. The general method used in the

QuikChange Site-Directed Mutagenesis kit is presented in Figure 11.












Step 1
Plosmid Preparation






Step 2
Temperature Cyrclng M ge


Mpr ger
primers


Slep 3
Dge',t'on




Step 4
Transformation


4




,--i Muloted plosmid
(conaloins nicked
--- circular strands)







-\ <


Gene in plosmid with
target site () for mutation





Denature the plasmid and anneal the
oligonucleotide primers (f) containing
the desired mutation (x}


Using the nonstrand-displacing
action of PfuTurbo DNA polymerase,
extend and incorporate the
mutagenic primers resulting
in nicked circular strands




Digest Ihe .eril.jicd, nonmutoled
parental D014. renrpllie with Dpn I


Transform the circular, nicked dsDNA
into XL1-Blue supercompetent cells


After transformation, the XL1 -Blue
supercompetent cells repair the
nicks in the mutated plasmid

LEGEND
Porentol DNA plasmid
S Mulogenic primer
-- Mulated DNA plosmid


Figure 11. Overview of the QuikChange Site-Directed Mutagenesis method. Reprinted
with the permission of Stratagene.42

Plasmid Preparation

Plasmid was extracted either by the standard alkaline lysis, chloroform extraction


and PEG precipitation protocol required by the DNA Sequencing Core Laboratory ICBR


at the University of Florida or by using Wizard Plus Miniprep DNA Purification System


(Promega). In both cases plasmid DNA was obtained from 10 mL overnight cell cultures


grown in LBK.









Transformation

BL21(DE3) cells were transformed with each plasmid DNA extracted from the

XL1-Blue Supercompetent cells. They were added to the SOC media and incubated for 1

h with shaking at 225 rpm at 37 C. Transformed colonies were picked from the LBK

plates and placed in the 100 ptL aliquots of 50 % glycerol. Ready for expression cells

were frozen at 80 C.

Enzyme Assays

Quantitative Assay

The most basic quantitative assay utilizes the ability of oxalate decarboxylase to

oxidize o-phenylenediamine (o-PDA) to 2,3-diaminophenazine, which is supposedly a

side reaction during enzyme's turnover.29 Assay mixtures consisted of 50 mM potassium

acetate buffer (pH 4.2), 0.2 mM Triton X, 0.5 mM o-PDA, water and 50 mM potassium

oxalate (pH 4.2). Addition of 10 pL fraction from purification initiated the reaction.

Presence of the byproduct of the OxDc turnover is manifested by the appearance of the

yellow color, which confirms the presence of the active enzyme in the mixture.

Qualitative Activity Assay

Activity was measured in the end-point assay initiated by the addition of the

substrate. After certain time, specified for each protein to be from 3 to 30 min, raising the

pH up to 12 with NaOH terminated OxDc turnover. The levels of produced format were

established in a coupled assay. After an overnight incubation of the resulting mixture at

37 C with format dehydrogenase and NAD+ measurement of absorbance at 340 nm

gives the amount of produced format. The assay mixture consisted of 50 mM acetate

buffer pH 4.2, 0.2 mM Triton X, 0.5 mM o-PDA, 0 50 mM oxalate and 2 to 10 [iM









enzyme. All measurements were performed in triplicate. Simultaneously, coupled part of

the assay was performed on the mixtures containing from 0.1 to 8 mM formate.3

Quenching reaction at different time points from 3 to 60 minutes supplied information

about enzyme's ability of linear format production in time.

Michaelis-Menten Kinetics

Steady-state kinetic studies were performed towards the synthesis of format. The

concentration of substrate, oxalate, was varied for each protein (wild type and mutants) to

cover the entire area of the Michaelis Menten curves obtained (up to enzyme's

saturation). The assay mixtures differ from those in the activity assay only by the varied

oxalate concentrations. The reactions were started and quenched as in the activity assay,

time of reaction depended on the activity of the enzyme. The standard curve was obtained

as described before. The results were plotted using KaleidaGraph. Values of KM, Vmax

and kcat were obtained for all analyzed proteins.

The pH Dependence

The pH dependence of the steady state kinetic parameters for mutated OxDc-

catalyzed oxalate degradation was investigated.43 The pH range tested was based on the

wild type experiments, which established lack of enzyme stability below pH 2.8 and a 90

% activity decrease between pH 4.2 and 5.7. Buffers were used: for pH 4.2 acetate and

MES for pH 5.7. Oxalate concentrations were varied based on previously established

values of KM.

Protein Concentration

Protein concentration was determined by Lowry assay which engages Folin phenol

reagent or using the Coomasie PlusTM Protein assay Kit (Pierce) according to the

manufacturer's guidelines.44 Standard curve was created by using the same reagents to









detect known concentrations ofBSA from 0 to 120 mg/mL. The absorbance of the

standards and the samples was read at 600 nm for Lowry and 595 nm for Bradford assay.

Inhibition Studies

Mutants of oxalate decarboxylase, R270K and R92K, have been tested for activity

with oxalate analogs.45'46 The substrates used: pyruvate, oxamic acid, malonate, maleic

acid, succinic acid and glutarate were added to the Michaelis-Menten kinetics reaction

mixture described above at three or more different concentrations. The data points were

obtained in triplicate, and analyzed with KaleidaGraph.

Heavy-Atom Kinetic Isotope Effects

These experiments were performed entirely by Laurie A. Reinhardt Ph.D. and

Drazenka Svedruzic in the Institute for Enzyme Research and Department of

Biochemistry at the University of Wisconsin in Madison, Wisconsin in collaboration with

W. Wallace Cleland Ph.D.

The exact procedure has been published for the wild type OxDc and was followed

again for the mutated enzyme.13














CHAPTER 3
RESULTS AND DISCUSSION

Site-Directed Mutagenesis

The first mutation of oxalate decarboxylase targeted the C-terminal site, which at

the time was considered to be the more likely candidate for the catalytically active one.

Substitution of arginine (R270) for alanine with a small uncharged side chain has, as

expected, greatly decreased the activity of the enzyme. After the introduction of the

lysine side chain in place of arginine, oxalate decarboxylase retained the same charge

with the residue just slightly moved away from the manganese ion binding site. This is a

conservative substitution that changes the properties of the enzyme to a much smaller

extent.

The arginine in the C-terminal site proved to be important for enzyme's activity.

Therefore, the second manganese binding site was tested. Arg92, which is one of the two

conserved arginine residues in oxalate decarboxylase, is situated in the N-terminal site,

and was mutated to lysine. The mutant was characterized to compare the influence of the

mutations in both metal binding domains on the activity of oxalate decarboxylase. The

reason for the decrease in activity of the mutants could be either structural or catalytic.

Wild Type (WT) Oxalate Decarboxylase

Native oxalate decarboxylase was expressed and purified according to the

literature.13 Wild type OxDc was used as a reference and a control in all of the assays

performed on the OxDc mutants.









The expression of wild type OxDc has been by far the most effective compared to

all the mutated proteins. The purification process has given reproducible results and

yielded highly active enzyme [Table 5]. All of the recently published mutagenesis studies

on B. subtilis oxalate decarboxylase were performed on less active protein obtained via

different expression and purification systems.3' 17,27

Table 5. Wild type OxDc characterization
Specific activity Time of linear format KM kcat Vmax
[U/mg] production [min] [mM] [sec 1] [mM/min]

34 60 10 +1 63 +7 1.07 0.04

Expression and Purification

R270A

There were a number of problems with obtaining a sufficient amount of R270A to

perform assays. The expression levels were decreased compared to the wild type by about

70 %. It seemed, however, that the introduction of a large and potentially relevant change

in the active site had affected the ability to effectively produce mutated OxDc.

The purification protocol for wild type was not followed successfully. The level of

purity was different from one prep to another [Figure 12]. The main problem seemed to

be caused by the ammonium sulfate protein precipitation step combined with the Phenyl-

Sepharose column chromatography. Most of the active protein was precipitating along

with the impurities, and the remaining sample did not yield any enzyme with detectable

activity. The enzyme was most likely binding irreversibly to the hydrophobic resin. This

suggested that the mutation influenced not only the activity or substrate selectivity of the

enzyme, but also its structure. The majority of the protein may have folded differently

after the removal of the large positively charged amino acid side chain and when

submitted to the ammonium sulfate.










.0 i i









I-*


Figure 12. Purification results of the R270A mutant of OxDc. Samples on the SDS-
PAGE gel: molecular weight marker, WT, lysate, final R270A from 3
different preps.

A sufficient amount of R270A was obtained after reducing the purification protocol

to just two anion exchange chromatography steps, DEAE- and Q-Sepharose, and then

repeating the expression and purification cycle multiple times.

R270K

The expression and purification protocols used for wild type were applied to

R270K mutant more successfully than to R270A. Expression and solubility levels varied

from prep to prep but were always about 30 % below the WT levels. The purification has

been performed using two or three step chromatography. The resulting proteins differed

in purity, specific activity, total yield and total activity.

The purification of R270K was much easier than for R270A, but the WT procedure

could not be followed identically. Repeatedly in the three step chromatography

purification, protein was not recovered from the last column (Q-Sepharose) due to the

low yield at this stage of the process. The protein was, however, purified and it was

possible to establish the extent of inactivating effect the mutation had on the enzyme. The

only change introduced in the three step purification was a decrease in the concentration









of ammonium sulfate in the protein precipitation step after DEAE-Sepharose column

down to 1.4 M. The yields were higher from the two column chromatography

purifications and the level of purification was high enough for all the assays, as well as

kinetic isotope measurements, to be performed.

R92K

As for the previous mutants, the protocol for the wild type expression and

purification proved to be ineffective in obtaining active and pure R92K with reproducible

yields. Very similar problems occurred as for R270K. These problems included not

recovering protein from the last column, whenever Phenyl-Sepharose resin was used and

ammonium sulfate precipitation of a fraction of the active protein along with the

impurities. Purification optimization did not provide a new reliable procedure. Therefore,

for the first time, an attempt has been made to improve yield by changing the expression

conditions, not just the purification.

Purification optimization

The wild type protocol that proved to be ineffective for the discussed above

mutants was investigated to optimize the purification protocol for R92K.

DEAE-Sepharose gave the same results as for the wild type in both elution time

from the column and purity level [Figure 13]. It provided initial purification and yielded

active protein, and therefore no changes needed to be introduced.














97,400
66,200
45,000

31,000 -

21,500

14,400





Figure 13. R92K purification: fractions from the DEAE-Sepharose Fast Flow column.
Samples on the SDS-PAGE gel: MWM, lysate, wash, fractions: 6, 10, 20, 22,
24, 25, 27, 29, 30, 36.

The next step investigated in the purification was the ammonium sulfate protein

precipitation. Based on the SDS-PAGE analysis and the quantitative assay for OxDc it

was revealed that part of a sample of an active mutant precipitated at as low of a

concentration of ammonium sulfate as 0.6 M. Up to the concentration of 1.4 M,

approximately 10 to 20 % of the total sample co-precipitated with the impurities. The

addition of ammonium sulfate to the final concentration of 0.6 M guaranteed an ionic

strength high enough for the sample to bind to the Phenyl-Sepharose column and for the

protein of interest to stay in solution.

Since the Q-Sepharose resin seemed to have very low resolution on all samples,

after being previously used and cleaned numerous times, the new resin was used to test

its effectiveness. Q-Sepharose is an anionic exchange resin, so binding of the proteins

depends on their affinity to the charged groups on the resin, in this case a quaternary

amine. Elution is based solely on changing anionic interactions between bound proteins









and the resin by changing the salt (NaC1) concentration in the elution buffer. Different

proteins are released at different ionic strengths of the buffer. Surprisingly, it was

established that Q-Sepharose provided purification to the sample that was not treated with

ammonium sulfate at all [Figure 14]. However, even at a low level of (NH4)2804, the

sample was not separated from the impurities on this resin. In the presence of ammonium

sulfate ionic strength may be too high for the mutated OxDc to bind to the Phenyl-

Sepharose.





97,400 =.77_
66,200 Z. m,
45,000 -
31,000 -
21,500 ,
14,400 m


Figure 14. R92K purification: fractions from Q-Sepharose Hi-Perfomance column.
Samples on the SDS-Page gel: MWM, load, wash, fractions: 14, 16, 18, 19,
20, 21, 23, 24, 26, 28, 30, 44

Since Phenyl-Sepharose proved rather ineffective in purifying R92K, different

resins with other hydrophobic ligands were tested for the mutant binding, and its release

at a more narrow ionic strength range. Butyl- and Octyl-Sepharose did not provide

separation. In fact, they bound R92K irreversibly, and neither varied gradient elution, nor

extended washing with low ionic strength buffer, resulted in recovering any active

protein. Therefore, it has been established that Phenyl-Sepharose is the most effective

hydrophobic resin, but the purification yields more protein of sufficient purity in the

purification that utilizes just two anion exchange resins DEAE and Q-Sepharose.









Weak cation exchange CM (carboxymethyl) Sepharose resin has been previously

used for the purification of oxalate oxidase17. Strong cationic SP (sulphopropyl)

Sepharose resin was tested in similar conditions (acetate buffer, pH 5.2) for the ability of

purifying mutated OxDc R92K. The enzyme of interest did not bind to the resin and was

eluted in low salt concentrations. The impurities of lower molecular weight did bind, and

therefore, the cation exchange resin increased purity of the enzyme. However, the column

purification would require further optimization.

Expression optimization

The best conditions for mutated oxalate decarboxylase to be produced by E. coli

have been established through changing many variables in the expression process

compared to the wild type procedure. The first innovation was decreasing the volumes of

bacterial cultures per flask volume to improve aeration. Also, instead of Erlenmeyer,

baffled flasks were used.47 The time of bacterial growth as well as the temperature were

altered. Likewise, the heat shock and sonication times were adjusted. After introduction

of all these changes extraction of pellets yielded soluble protein for the first time for an

arginine mutant. The younger bacteria were more potent and produced enzyme at a

slower rate with higher influence of chaperonines on the correct protein folding.

Results

Both lysate and extract contained more enzyme of interest [Figure 15] than before,

and further studies proved that changes in bacterial growth conditions resulted in

production of more active protein. The total yield from 1 L of culture was approximately

2 mg with specific activity of 0.3 U/mg, whenever the wild type expression protocol was

followed. After introducing the new procedure the yield was 14 mg and the specific

activity was 3.5 U/mg.













97,400 4
66,200
45,000

31,000 '

21,500
14,400 .

Figure 15. Expression results for R92K. SDS-PAGE gel fractions: marker, lysate, extract,
lysate, extract.

The two column purification of the sample from the changed expression system

yielded highly purified protein. DEAE -Sepharose resin provided initial purification, and

high amounts of protein allowed the following experiments to proceed with only the most

active and pure fractions [Figure 16A].






97,400 97,400
66,200- 6' 2nn
45,000 Wo --
31,000 ,i .
21,500 ,,,,
14,400 F A 14B B


Figure 16. R92K purification: (A) fractions from DEAE-Sepharose: marker, WT, 2, 21,
23, 25, 26, 27, 28, 29, 30, 32, 36; (B) fractions form Q-Sepharose: marker,
load, wash, fractions: 4, 15, 20, 21, 23, 24, 25, 26, 27, 28, 30.

The Q-Sepharose used directly after DEAE-Sepharose provided further purification

[Figure 16B] but did not grant homogeneity. The pool of active protein was concentrated

to half the volume in the pressure chamber. Approximately one third of the protein

precipitated. Some of the precipitate was active but, as expected, most of it turned out to









be inactive, and the specific activity of the soluble sample increased [Table 6]. Values of

the total activity oflysate and extract in the purification table are not exact, as the activity

assay based on NAD+ consumption may be compromised by many impurities still present

at this point in the sample. Hence, values of the overall activity of the sample are more

reliable starting after the first step of chromatographic purification.

Table 6. Purification table for R92K mutant of OxDC.
Step Concentration Volume Total protein Total activity Specific activity
[mg/mL] [mL] [mg] [U] [U/mg]

Lysate 0.6 150 90 43 0.5
Extract 0.6 100 60 19 0.3
Qload 1.4 50 70 120 1.7
Q ool 0.6 40 26 57 2.2
Final 0.8 18 14.5 50 3.5

Steady-State Kinetics

R270A

The average purification yielded approximately 20 to 30 mg of R270A, with total

activities of 1 to 3 U from 2 L of bacterial culture. The obtained enzyme varied in

specific activity from 0.1 to 1.4 U/mg. Kinetic parameters were obtained for each

preparation cycle. The results shown in [Table 7] represent one of the preps.

Table 7. Characterization of the R270A mutant of oxalate decarboxylase.
Specific Time of linear KM kcat Vmax
activity format production [mM] [sec 1] [mM/min]
[U/mg] [min]

0.4 60 1.8 + 0.1 0.08 + 0.01 0.0124 0.0005

The mutation of arginine residues to alanine was predicted to inactivate the

enzyme,6 while the insufficient purification has lowered the specific activity of the

enzyme even further. The structure seemed to have been influenced by the mutation as

well. Therefore, kinetic isotope effects could not be measured, and obtaining reproducible










and reliable assay results proved impossible. It has been established that the mutation of

Arg270 from the C-terminal end of oxalate decarboxylase to alanine affects both the

structure and activity of the enzyme in a dramatic manner, causing about a 100-fold

decrease in activity.





0015

V[mM/min]

0 01 -



0 005



0 I 1 L L L I L
0 10 20 30 40 50 60
[oxalate]mM


Figure 17. Michaelis-Menten kinetics of mutated oxalate decarboxylase: R270A. Enzyme
activity was determined in triplicate, the curve was fit using the Michaelis-
Menten equation.

The time of linear format production by R270A was established to be 60 min,

which allowed the wide choice of time and enzyme concentrations to perform steady-

state kinetics experiments [Figure 17,18]. The rate of format production is 100 fold

slower and the value of kcat decreased significantly in comparison to the wild type

enzyme catalyzed reaction. This dramatic drop in the enzyme's ability to catalyze the

decarboxylation of oxalate may be caused by the arginine's role in catalysis. It is also

consistent with the assumption made by Just et al that the C-terminal metal binding site

influences the activity of the enzyme by structurally supporting the N-terminal site.6










Removing a charged residue would have changed the interaction between the sites and

disrupted the enzyme's catalytic ability.






06 -
[formate] iM
05 -

04 -

03 -

02

01

0
0 10 20 30 40 50 60 70
t [min]
Figure 18. Catalysis of format production by R270A mutant of OxDc as a function of
time32 Error bars come from the fit in KaleidaGraph.

R270K

Activity

The conservative mutation of Arg270 to lysine resulted in a 10-fold activity

decrease compared to the wild type OxDc [Table 8]. The format production is linear for

60 min [Figure 19B]. The mutation in the presumed active site has affected the enzyme's

activity by influencing either the catalytic mechanism or protein folding and metal

binding.

Table 8. Kinetic characterization of the R270Kmutant of OxDc.
Specific activity Time of linear format KM kcat Vmax
[U/mg] production [min] [mM] [sec 1] [mM/min]

4.9 60 1.7 + 0.2 7.8 + 0.8 0.045 + 0.001









Kinetic parameters

The decreased KM value [Figure 19A, Table 8] suggests that incorrect protein

folding has not disturbed the binding site. The decrease of KM might mean that the

mutant binds oxalate even more tightly than the wild type. The kcat value has decreased

for R270K compared to WT but was 100 times higher than that for R270A. The mutation

did not kill the enzyme activity. The catalysis was possible, but slower. The decrease in

KM and kcat values and no change in kcat/KM suggests unproductive binding. The

specificity constant (kcat/KM) is unaffected by mutation, as kcat and KM altered in a

compensating manner. The turnover number decreases when only a fraction of substrate

is bound productively and also tighter binding leads to a decreased value of KM. If the C-

terminal manganese binding site had catalytic ability, then moving the positive charge

further away from the bound substrate would have diminished this ability, as seen in

Table 16. If the role of the C-terminal domain were merely structural, the results [Table

16] confirmed this suggestion as well, as the disruption in the structure and charge of the

mutant is incomparably smaller than for the R270A. The catalysis performed dominantly

by the N-terminal site would not be affected to a very high degree by a conservative

mutation in a C-terminal domain, as observed for R270K. The values of kinetic

parameters kcat/KM and KM for R270K mutant of OxDc are comparable with those

published by Just et al [Table3]6, specific activity is, however, 10-fold higher.












A. B.
005 -
[formate] mM
V [mM/min] 6
0 04

003 4

002 3

2
0 01


0 10 20 30 40 50 60 0
0 10 20 30 40 50 60 70
[oxalate] mM time [min]
Figure 19. Kinetic characterization of R270K mutant of OxDc. (A.) Michaelis-Menten
kinetics of mutated oxalate decarboxylase: R270K. All data points were
determined in triplicate, the curve was fitted using the Michaelis-Menten
equation. (B.) Catalysis of format production by R270Kmutant of OxDc as a
function of time. Error bars come from the fit in KaleidaGraph.

Inhibition studies

The change in the turnover number and our prediction that both conserved

arginines may play a role in substrate selectivity lead to testing several compounds, all of

which contained one or two carboxylic groups, as alternative substrates for R270K. These

compounds might bind in the putative active site and inhibit the enzyme.

The enzyme's binding selectivity is not highly affected by the mutation [Table 9].

The wild type has been shown to be specific towards oxalate.13 The only potential

substrate that seemed to slightly inhibit the R270K mutant was malonate, which has been

previously observed to inhibit the wild type enzyme as well [Figure 20]. The Lineweaver-

Burk plot for increasing concentrations of malonate shows [Figure 21] that the lines

intercept on the y-axis. The KM increased for increasing concentrations of malonate. Vmax

stayed the same while Vmax/KM values decreased, all of which suggest competitive

inhibition [Table 9].









Table 9. Inhibition studies of R270K mutant of oxalate decarboxylase
Inhibitor Inhibitor KM [mM] Vmax [mM/min] (Vmax/KM)app KI [mM]
concentration
[mM]
Pyruvate 0 1.7 + 0.3 0.0049 + 0.001 0.030 + 0.001 No
50 3.1 + 0.8 0.099 + 0.006 0.03 + 0.01 inhibition
100 1.8 + 0.6 0.101 + 0.006 0.06 + 0.01
Oxamic 0 1.9 0.5 0.021 0.001 0.011 0.003 No
acid 25 1.7 + 0.2 0.0197 + 0.0006 0.012 + 0.001 inhibition
50 1.5 + 0.3 0.026 + 0.001 0.017 + 0.003
Malonate 0 4.3 0.3 0.073 + 0.001 0.018 + 0.001 ---
5 6.7 + 0.5 0.069 + 0.001 0.010 + 0.001 8.9
10 8.8 + 0.9 0.068 + 0.002 0.007 + 0.001 9.5
15 10.7 + 0.8 0.065 + 0.001 0.006 + 0.001 10.1
25 18.3 + 4.2 0.068 + 0.005 0.004 + 0.001 7.7
35 16 2 0.058 + 0.003 0.004 + 0.001 12.8
45 8.3
28 + 12 0.07 + 0.01 0.0025 + 0.001 8
50 13.7
50 20 + 7 0.053 + 0.007 0.0026 + 0.001
Maleic 0 6.6 + 0.5 0.052 + 0.001 0.008 + 0.001 No
acid 25 6 + 1 0.039 0.002 0.0065 0.0009 inhibition
50 4.9 0.9 0.031 0.002 0.006 0.001
Glutarate 0 3.6 + 0.9 0.014 + 0.001 0.004 + 0.001 No
25 3.1 0.3 0.0137 + 0.0004 0.0040 0.0004 inhibition
50 3.0 + 0.3 0.0124 0.0004 0.0040 0.0004

The equation used to calculate Ki for malonate was chosen based on the graphical

representation of data as well as on changes in the following kinetic parameters:

Ki= (KM[I]) / (KMapp-KM); where KI inhibition constant, KM Michaelis-Menten

constant in the absence of the inhibitor, KMapp -Michaelis-Menten constant in the

presence of different concentrations of the inhibitor, and [I] concentrations of

malonate.30 The mean value of Ki is 10.1 mM. Replotting data (1/KMapp VS 1/[I]) provided

a Ki value of 12.5 mM, in agreement with the calculated value, and confirmed that


R270K is inhibited competitively by malonate.




















0.07 --- V (50 mM malonate)
V [mM/min]
0.06 -
. -
0.05 + -


0.04 -


0.03 /


0.02


0.01



0 10 20 30 40 50 60


[oxalate] mM

Figure 20. R270K inhibition by malonate. Michaelis-Menten kinetics in the presence of
varied concentrations of malonate. Legend is shown above the graph.



1/V min/mM (0 mM malonate)
-- 1V (5 mM malonate)
--/V (10 mM malonate)
----1/V(25mM malonate)
---- 1/V (15 mM malonate)
200 1/V (35 mM malonate)
1N [min/mM]
V [mi/ ---- 1/V (50 mM malonate)


150 -

'- -

100







0 II-



0 0.1 0.2 0.3 0.4 0.5 0.6

1/[oxalate] [1/mM]

Figure 21. R270K inhibition by malonate Lineweaver Burk plot. Intercept on the y-axis
suggests competitive inhibition. Legend is shown above the graph.


--V mM/min/mg (0 mM malonate)
- -V (5 mM malonate)
-0- V (10 mM malonate)
--x--V(15mM malonate)
- V (25 mM malonate)
-- V (35 mM malonate)









Yet again, the two possible theories were considered. The catalysis of oxalate

decarboxylation in the C-terminus could have been slowed down by a smaller fraction of

substrate being able to bind in the active site, which was partially occupied by the

inhibitor.

Nonetheless, the inhibitor's binding could have disrupted the structural interaction

between the two domains in the monomer and, therefore, catalysis in the N-terminus was

diminished. The KI value for competitive binding of malonate by the wild type was

observed in our laboratory to be 6 mM. Increase of the KI value for the mutant suggests

stronger inhibition of this protein compared to the wild type, and possibly even confirms

its role in the substrate selectivity.

The pH dependence

The activity of the R270K mutant of oxalate decarboxylase decreased 50 % at the

pH of 5.7. Both KM and kcat values decreased about 5 fold, therefore the V/K value

remained unchanged [Table 10]. A single protonated oxalate was shown to be the correct

substrate for the oxalate decarboxylase.13 The results for R270K may mean that even

though oxalate is binding, 50 % of it do so in an unproductive manner. This may be

caused by the wrong protonation state of the oxalate at higher pH.

Table 10. Kinetic parameters for R270K mutant of OxDc at pH 4.2 and 5.7
pH KM [mM] Vmax [mM/min] Vmax/KM [min-1]

4.2 1.7 + 0.2 0.108 0.009 0.06 0.01
5.7 0.30 + 0.07 0.018 + 0.001 0.06 + 0.01









R92K

Activity

As in the C-terminal site, the activity of the conservative mutant of OxDc (R92K)

was expected to change. Activity was about 10-fold decreased compared to the wild type

[Table 16], almost exactly as for the R270K, the previously characterized conservative

mutant [Table 11]. Data presented by Just et al show a 100 fold activity decrease.6 Steady

state kinetics measurements were possible during the time of linear product formation by

a stable enzyme, which for R92K was 70 min [Figure 22]. The equal decrease in activity

of lysine mutants for both N- and C-terminal active sites suggests that they are both

equally important to catalysis, but does not assign them specific roles in the catalytic

mechanism.

Table 11. Kinetic characterization of the R92K mutant of oxalate decarboxylase.
Specific activity Time of linear KM kcat Vmax
[U/mg] format production [mM] [sec 1] [mM/min]
[min]

3.5 70 11 + 1 0.40 0.05 0.030 0.001

Kinetic parameters

The KM value was not altered for R92K, which suggested that binding of the

substrate was unaffected [Figure 24]. However, a large (100 fold) change in the turnover

number implies catalytic importance of this residue. The specificity constant, K i k, was

decreased for R92K 100-fold, which is as much as for the R270A mutant. The mutation

of arginine to lysine introduced the smallest possible structural change. Its dramatic

impact on the enzyme's specificity constant (kcat/KM) as well as the turnover number

suggested that the prediction of Just et al that the N-terminal metal binding site is the sole

or dominant catalytically active domain may have been correct [Table 11].6 Therefore, to







49


further investigate this issue, inhibition studies as well as kinetic isotope effect


measurements were performed.



formate
produced] mM

1.2


1


0.8


0.6


0.4


0.2


0
0 10 20 30 40 50 60 70 80

time [min]

Figure 22. Catalysis of format production by R92K mutant of oxalate decarboxylase as a
function of time. Errors bars come from the fit in KaleidaGraph.




V [mM/min]


10 20 30

[oxalate] mM


40 50


Figure 23. Michaelis-Menten kinetics of mutated oxalate decarboxylase: R92K. All data
points were determined in triplicate, the curve was fitted using the Michaelis-
Menten equation in KaleidaGraph.









Inhibition studies

Following the studies for R270K, the N-terminal site mutant was examined for

inhibition with carboxylic compounds comparable to oxalate in size. Even though both

wild type and R270K were inhibited by malonate, R92K was not affected by any of the

tested compounds [Table 12]. Interestingly malonate cannot compete or even bind to the

R92K mutant. No disruption in format production by the presence of malonate could

confirm the assumption of Just et al, that the N-terminal site is more stable and that it is

involved in the catalysis and not substrate recognition. They predicted that the newly

discovered lid, which prevents solvent access to the Mn ion in the N-terminus makes the

domain generally more stable and resistant.6 It is possible that substitution of lysine in

this manganese ion binding site did not create a big enough gap for the enzyme to bind

other substrates and be inhibited by them.

Table 12. Inhibition studies of R92K mutant of OxDc.
Inhibitor Inhibitor KM Vmax [mM/min] (Vmax/KM)app KI
concentration [mM] [mM]
[mM]
Pyruvate 0 4.4 0.5 0.019 0.0001 0.0043 0.0004 no
50 3 + 1 0.013 0.002 0.004 0.001 inhibition
100 1.5 + 0.5 0.009 + 0.005 0.006 0.003
Malonate 0 15 + 1 0.045 0.002 0.0030 0.0002 no
25 26 + 1 0.034 + 0.001 0.0013 0.0001 inhibition
50 25 +2 0.026 0.001 0.0010 0.0001
Maleic 0 10 + 1 0.031 + 0.001 0.0031 0.0003 no
acid 25 10 + 1 0.023 + 0.001 0.0023 0.0003 inhibition
50 13 + 1 0.019 + 0.001 0.0015 + 0.0001
Succinate 0 13 +3 0.032 + 0.003 0.0025 0.0006 no
25 15 + 1 0.033 0.001 0.0022 0.0002 inhibition
50 18 1 0.039 0.002 0.0022 0.0002
Glutarate 0 13 + 1 0.034 0.001 0.0030 0.0002 no
25 22 + 5 0.046 + 0.006 0.0021 0.0005 inhibition
50 25 +8 0.06 0.01 0.0024 0.0001









The pH dependence

The activity was measured at a higher pH of 5.7 to ensure that, as proved

experimentally for the wild type oxalate decarboxylase, the substrate for R92K is a half

protonated oxalate. The activity at pH 5.7 is 10 % of the R92K activity at the optimal pH

of 4.2 [Figure 26]. Also, the values of KM and Vmax decreased 100-fold and 10-fold

respectively, compared to the values at pH 4.2 [Table 13]. The activity was lower

because the substrate was not in a preferred protonation state anymore and an additional

step was necessary for the reaction to occur.

Table 13. Kinetic parameters for R92K mutant of OxDc at pH 4.2 and 5.7
pH KM [mM] Vmax [mM/min] Vmax/KM [min1]

4.2 11.5 + 0.7 0.030 + 0.001 0.0026 0.0002
5.7 0.9 + 0.1 0.0039 0.0001 0.0043 0.0004

Heavy-Atom Kinetic Isotope Effects

R270K

The measurements of the kinetic isotope effects for R270K have been based on the

experiments described previously for the wild type. The same model was used in the

analysis of data.13

The KIE measurements are invaluable in studying the role of amino acid residues

in the putative active sites.48 They provide information about the steps in the mechanism

of the catalyzed reaction, in addition to the transition state. The substitution of Arg to Lys

is conservative, the charge remains positive and the obtained values have not changed

drastically compared to WT, therefore the same kinetic model could be used for the data

analysis for the R270K mutant as for the wild type. If R270 indeed plays catalytic role by

polarizing the C-O bond and facilitating decarboxylation, then the observed isotope

effects should be higher than for the wild type. Such an increase in the KIE values was









observed. The change was statistically relevant but not large [Table 14]. The difference

between the effects on the format and the CO2 production can be estimated due to the

fact that isotope ratios for both products could be measured. As for the wild type, the

kinetic isotope effects were larger for the format end of oxalate. Thus, the arginine

residue possibly stabilized the transition state by polarization of the C O bond in the

oxalate radical anion. The isotope effects are higher for R270K than for the wild type.

The change was significant for CO2 and only slight for format. This indicates that the

bond breaking step became more rate-limiting in the reaction catalyzed by the R270K

mutant of oxalate decarboxylase.

The preliminary 180 isotope effects for R270K measured at pH 4.2 show that as for

the wild type there is no isotope effect for the CO2 production.

Table 14. 13C and 80 kinetic isotope effects for R270K mutant of OxDc.
pH 13(V/K), % 13(V/K), % 18(V/K), % 18(V/K), %
CO2 format CO2 format

4.2 0.8 0.1 1.6 0.1 -0.04 0.2 0.4 0.2
5.7 1.3 0.1 2.1 0.1 0.3 0.2 0.8 0.2

R92K

Preliminary results of kinetic isotope measurements were obtained for pH 4.2 and

can be considered in a general analysis [Table 15]. The conservative mutation of Arg to

also positively charged Lys implied the use of the same kinetic model as for the wild type

and the C-terminal R270K mutant, to analyze the data.

The 13C isotope effects for R92K mutant are practically the same as for the R270K

mutant higher than for the wild type. The change in the KIE values compared to those

for the wt is not large but statistically significant. Higher KIE values show that the

chemistry is slower and more rate-limiting after introducing the mutation, but still









possible. The results for carbon dioxide are smaller than for the standard carbon bond

cleavage, which confirms that breaking of the C-C bond in oxalate is not rate limiting.

The changes are bigger for the format end of oxalate rather than the CO2 one. Yet again

180 IE for CO2 was not present, however the effect for format increased significantly

compared to the wild type. This confirms the prediction that the Arg92 residue in the N-

terminal site is responsible for the polarization of the C-O bond of the oxalate radical

anion and thus facilitates decarboxylation. The fact that for both Arg mutants the same

13C isotope effects were observed proves that both N- and C-terminal sites are active

during catalysis.

Table 15. 13C and 80 kinetic isotope effects for R92K mutant of OxDc.
pH 13(V/K), % 13(V/K), % 18(V/K), % 18(V/K), %
CO2 format CO2 format

4.2 0.8 + 0.1 1.3 + 0.3 0.3 + 0.3 1.9 + 0.3














CHAPTER 4
CONCLUSIONS

Expression and Purification of the Wild Type and Mutated OxDc

Recombinant B. subtilis oxalate decarboxylase was expressed in E. coli with

success. The purifications yielded the most active mutated proteins as described in the

literature so far.3'6 This is most likely due to the lack of any His-tags in the protein. Both

N- and C-terminal His-tags were used in previous mutagenesis studies of OxDc and the

proteins were either unstable and precipitated or had lower activities than the ones

obtained in this work.

The mutated oxalate decarboxylase proved to be hard to both express and purify

according to the wild type enzyme protocol. The increase in yield of active enzyme was

achieved by changing several expression conditions. Decreasing the temperature during

cell growth might have affected higher expression of the manganese transporting

proteins4951 and therefore higher activity of OxDc may have been due to the increase of

manganese occupation per domain of OxDc. Decreasing the temperature also resulted in

slowing down the bacterial growth as well as the protein production. Induction of E. coli

cells in less dense cultures provided more potential to produce the enzyme. Increasing

aeration of bacterial cultures helped decrease fermentation and contributed to improving

manganese transport, but the major effect was on the ATP production. Oxygen is needed

for the biosynthesis of ATP used by the chaperonines, accessory proteins in the correct

protein folding process. Activation of chaperonines by heatshock was much more

effective when sufficient levels of ATP were present.30 Therefore, higher amounts of









active, correctly folded mutated OxDc may have been obtained. It seemed that the

bacteria, even though capable of producing wild type OxDc, had to be put in the optimal

conditions to yield any active enzyme with important mutations introduced.

Similar difficulties, in comparison to the wild type procedure, were encountered

during the purification of mutated oxalate decarboxylase. The major problem was caused

by the ammonium sulfate precipitation step combined with the Phenyl-Sepharose

column. The latter resin separates proteins based on their hydrophobicity. Its failure to

provide purification to the R270A mutant of oxalate decarboxylase, as well as binding it

more strongly than the wild type, lead to the conclusion that the mutation of Arg to Ala

may have decreased the levels of correctly folded enzyme. The changes in the expression

conditions helping the correct protein folding have been described above. The

purification of the samples from the improved expression system, using just anion

exchange chromatography, proved sufficient for the planned characterization

experiments.

Steady-State Kinetics of the Wild Type and Mutated OxDc

Kinetic characterization of mutated oxalate decarboxylase was performed in this

study and confirmed the hypothesis that both manganese ion binding sites are important

for enzyme's activity and catalysis [Table 13]. Just et al suggested that the N-terminal

site was dominant in the catalysis while the C-terminal one supported it in a structural

way. This assumption was based solely on the steady-state kinetics experiments on the

mutated OxDc.6 However, the results of this work, which include the kinetic isotope

effects measurements, show the significance of both metal binding sites to catalysis.

Binding of the substrate was slightly influenced by mutations in the C-terminal site.

Both alanine and lysine mutants of Arg270 had 10-fold decreased KM's. The binding









constant (KM ) for the N-terminal mutant, R92K was unaltered compared to the wild type.

This confirmed that the role of the C-terminus may be more structural, however did not

provide a proof excluding it from catalysis [Table 16].

The significance of steady-state kinetics results is not as high as that of the kinetic

isotope effects. The basic kinetic parameters were used for a general analysis and

comparison of the properties of the wild type and mutated enzyme. A decrease in

enzyme's turnover number proved the important catalytic role of the conserved arginine

residue in the N-terminal active site. The value of the kcat/KM of R92K decreased as much

as for the R270A. That means that the conservative mutation in this catalytically active

site had a dramatic impact on enzyme properties. Unproductive binding of oxalate was

observed for R270K, as the catalytic turnover and binding constant decreased but the

specificity constant remained unaffected by the mutation.

The decrease in activity at pH 5.7 compared to pH 4.2 that was similar to WT,

showed that the enzyme most likely used substrate in the same protonated form with and

without the mutations introduced.

Several acids and diacids, similar in size to oxalate, were tested as inhibitors of

oxalate degradation by R270K and R92K. The selectivity of the N-terminal mutant

remained high for oxalate. R270K, however, was shown to be able to bind malonate,

which, as in the wild type, competed for the enzyme's active site with oxalate. Binding of

malonate in the C-terminal manganese binding site slowed down catalysis by disrupting

the interactions between the two domains of the enzyme if the hypothesis of Just et al

were correct.6 The conservative change may not be sufficient, however, to see the real









effects of removing arginine residues. Further experiments should be conducted on the

OxDc with more significant substitutions in place of the Arg270 and Arg92.

As shown by Just et al, substitution of arginine with alanine had a very dramatic

effect on the enzyme's activity, which deteriorated to a very low level in R270A and was

extremely hard to detect.6 The activity of the R92A mutant was expected to be even

lower if N-terminal were the dominant catalytic site6, which would make it completely

impossible to characterize the protein. Therefore R92A was not prepared.

Table 16. Comparison of kinetic characteristic of wild type OxDc and its active site
mutants.
Specific KM [mM] Vmax [mM/min] kcat [sec-1] kcat/KM
activity
[U/mg]
Wild Type 34 10 + 1 1.07 0.04 63 + 7 6.3 0.8
R270A 0.4 1.8 + 0.1 0.0121 0.0009 0.08 + 0.01 0.040 0.005
R270K 4.9 1.7 + 0.2 0.045 + 0.001 7.8 + 0.8 4.5 + 0.7
R92K 3.5 11 + 1 0.030 0.001 0.40 0.05 0.040 0.001

Heavy-Atom Kinetic Isotope Effects

Kinetic isotope effect measurements were used to confirm the importance of

Arg270 and Arg92 in the catalytic mechanism of oxalate decarboxylase. Since steady-

state kinetics cannot prove the mechanism, just rule out some alternatives, heavy-atom

isotope effects were measured to investigate the influence the arginine residues have on

the catalytic ability of oxalate decarboxylase [Table 17].

In both investigated mutants of oxalate decarboxylase substitution of Arg to Lys

was introduced. The charge remained conserved and the same kinetic model was used for

the data analysis as for the wild type OxDc. The kinetic isotope effects were observed for

both mutants, which means that both Arg 92 and 270 are important to catalysis. The

changes were not large, which is consistent with the conservative mutation, but they were









significant. What is more, the 13V/K numbers were the same for both mutants, which

suggests that their participation in the catalytic mechanism is equal. The observed 13C

KIE's for CO2 were lower than for the typical carbon bond cleavage for both mutants,

which confirmed that the C-C bond cleavage in oxalate is not rate-limiting. The numbers

were higher for the format end of oxalate, which confirmed arginine residues' role in

facilitating decarboxylation by polarizing the C-O bond in the oxalate radical anion.

Table 17. Summary of 13C and 180 kinetic isotope effects for the wild type and mutated
OxDc
pH 13(V/K), % 13(V/K), % 1(V/K), % 1(V/K), %
CO2 format CO2 format

Wild type 4.2 0.5 0.1 1.5 0.1 -0.2 0.2 1.1 0.2
5.7 0.8 0.1 1.9 0.1 -0.7 0.1 1.0 0.1
R270K 4.2 0.8 0.1 1.6 0.1 -0.04 0.2 0.4 0.2
5.7 1.3 + 0.1 2.1 + 0.1 0.3 + 0.2 0.8 + 0.2
R92K 4.2 0.8 + 0.1 1.3 + 0.3 0.3 0.3 1.9 + 0.3

Active Site Identity and Mechanism of Catalysis of OxDc

The novel, nonoxidative decarboxylation without any organic cofactors in the

presence of oxygen is very unusual. Determination of the oxalate decarboxylase's crystal

structures has provided details necessary to elucidate the enzyme's mechanism of

catalysis. The first crystal structure predicted the C-terminal domain to be the

catalytically active one based on the presence of the general base only in this site.3 The

second crystal structure showed a slight change in the conformation of one of the helices,

that put a glutamate residue in the vicinity of the bound substrate. These studies have also

presented that the N-terminal active site is more available to the substrate and more stable

making it the dominant domain in the catalysis.6

The significance of the conserved arginine residues Arg92 and Arg270 was

confirmed in this work. Based on steady state kinetic experiments it was confirmed that









the N-terminal site is important during catalysis. Even the most conservative mutation of

arginine in the N-terminal domain lead to a dramatic decrease in enzyme turnover,

prompting the conclusion that it is the dominant, if not the only catalytically active site.

The C-terminal domain influences catalysis as well, possibly as proposed by Just et al, in

the structural way.6 The Arg residue found in the C-terminus was eventually confirmed to

play a crucial role in the catalysis when very similar values of heavy atom isotope effects

were observed for both R92K and R270K mutant [Table 17]. Removing the positive

charge of the arginine inactivated the enzyme, which is consistent with its assigned role

of polarizing the intermediate.

The Future of the Project

There are significant scientific and medical reasons for structural and mechanistic

characterization of oxalate decarboxylase.

Chemical insights concerning radical-mediated enzyme catalysis should be pursued

further through investigation of the identity of the radical species formed during steady-

state turnover of OxDc.3

The metal dependence of the enzyme has been previously shown for the wild type

oxalate decarboxylase.27 Metal occupancy of the active sites has not been established for

the mutants. It could provide additional information on the cause of the arginine mutants'

lack of or decrease in activity.

Evaluation of the effect of the protein environment in controlling the chemical

properties of the metal centers in bacterial OxDc should be performed by further site-

directed mutagenesis studies. The mutations of conserved glutamate residues from both

C- and N- terminal active sites should provide further proof of the relevance of each






60


manganese binding site to the catalysis. The characterization should include steady-state

kinetics and kinetic isotope effect measurements as for the arginine mutants.

This work has laid foundation for further characterization of recombinant B.subtilis

OxDc. The projects outlined above will be continued in the Richards' laboratories.
















LIST OF REFERENCES


1. Hodgkinson, A. Oxalic acid in Biology and Medicine; Academic Press: London,
1977.

2. Graustein, W. C., Cromack, K., and Sollins, P. (1997) Science 198, 1252-1254.

3. Anand, R., Dorrestein, P. C., Kinsland, C., Begley, T. P., and Ealick, S. E. (2002)
Biochemistry 41, 7659-7669.

4. Mehta, A., and Datta, A. (1991) J. Biol. Chem. 266, 23548-23553.

5. Dunwell, J. M., Khuri, S., and Gane, P. J. (2000) Microbiol. Mol. Biol. Rev. (200)
64, 153-179.

6. Just, V. J., Stevenson, C. E. M., Bowater, L., Tanner, A. Lawson, D. M., and
Bornemann, S. (2004) J. Biol. Chem. 279, 19867-19874.

7. Goodell, B., Jellison, J., Liu, J., Paszczynski, A., Fekete, F., Krishnamurthy, S.,
Jun, L., and Xu, G. (1997) J. Biotechnol. 53, 1333-162.

8. Shimada, M., Akamtsu, Y., Tokimatsu, T., Mii, K., and Hattori, T. (1997) J.
Biotechnol. 53, 103-113.

9. Sidhu H., Hoppe, B., Hesse, A., Tenbrock, K., Bromme, S., Rietschel E., and Peck,
A.B. (1998) Lancet 352, 1026-1029.

10. Williams, H. E., and Wandzilak, T. R. (1989) J. Urol. 141, 742-747.

11. Strasser, H., Burgstaller, W., and Schinner, F. (1994) FEMSMicrobiol. Lett. 119,
365-370.

12. Wei, Y., Zhang, Z., Andersen, C. H., Schmelzer, E., Gregersen, P. L., Collinge, D.
B., Smedegaard-Petersen, V., and Thordahl-Christensen, H. (1998) Plant Mol. Biol.
36, 101-112.

13. Reinhardt, L., A., Svedruzic, D., Chang, C. H., Cleland, W. W., and Richards, N.
G. J. (2003)J. Am. Chem. Soc. 125, 1244-1252.

14. Hesse, A., Bongartz, D., Heynck, H., and Berg, W. (1996) Clin. Biochem. 29, 467-
472.









15. Honow, R., Bongartz D., and Hesse, A. (1997) Clin. Biochem. Acta 261, 131-139.

16. Chiriboga, J. (1966) Arch. Biochem. Biophys. 116, 516-523.

17. Requena, L., and Bornemann, S. (1999) J. Biochem. 343, 185-190.

18. Dunwell, J. M. (1998) Biotechnol. Genetic Eng. Rev. 15, 1-32.

19. Woo, E.-J., Dunwell, J. M., Goodenough, P. W., Marvier, A. C., and Pickersgill,
R.W. (2000) Nat. Str. Biol. 7, 1036-1040.

20. Tanner, A., and Bornemann, S. (2000) J. Bacteriol. 182, 5271-5273.

21. Chakraborty, S., Chakraborty, N., Jain, D., Salunke, D. M., and Datta, A. (2002)
Prot. Sci. 11, 2138-2147.

22. Babbitt, P. C., and Gertl, J. A. (1997) J. Biol. Chem. 272, 30591-30594.

23. Ricagno, S., Jonsson, S., Richards, N., and Lindqvist, Y. (2003) EMBO Journal 22,
3210-3219.

24. Emiliani, E., and Riera, B. (1968) Biochim. Biophys. Acta 167, 414-421.

25. Magro, P., Marciano, P., and Di Lenna, P. ((1988) FEMSMacrobiol. Lett. 49, 49-
52.

26. Mehta, A., and Datta, A. (1991) J. Biol. Chem. 266, 23548-23553.

27. Tanner, A., Bowater, L., Rairhurst, S., and Bornemann, S. (2001) J. Biol. Chem.
276, 43627-43634.

28. Shimazono, H., and Hayaishi, 0. (1957) J. Biol. Chem. 227, 151-159.

29. Emiliani, E., and Bekes, P. (1964) Arch. Biochem. Biophys. 105, 488-493.

30. Fersht, A. Structure and Mechanism in Protein Science; W. H. Freeman and
Company: New York, 1999.

31. Cleland, W. W. (1987) Bioorg. Chem. 15, 283-302.

32. Cleland, W. W. (1982) Methods Enzymol. 87, 625-641.

33. Cleland, W. W. (1995) Methods Enzymol. 249, 341-367.

34. O'Leary, M. H. (1989) Annu. Rev. Biochem. 58, 377-401.

35. O'Leary, M. H. (1980) Methods Enzymol. 64, 83-104.

36. Dunwell, J. M., and Gane, P. J. (1998) J. Mol. Evol. 46, 147-154.









37. Plapp, B. V. (1995) Methods Enzymol. 249, 91-119.

38. Anderson, K. S., and Johnson, K. A. (1990) Chem. Rev. 90, 1131-1149.

39. Cannon, W. R., Singleton, S. F., and Benkovic, S. J. (1996) Nat. Str. Biol. 3, 821-
833.

40. Cleland, W. W. (1977) Adv. Enzymol. Relat. Areas Mol. Biol 45, 273-387.

41. Cleland W. W. (1979) Methods Enzymol. 63, 103-138.

42. QuikChange Site-Directed Mutagenesis kit manual, June 2004,
www. stratagene.com.

43. Cleland, W. W. (1982) Methods Enzymol. 87, 390-405.

44. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol.
Chem. 193, 265-275.

45. Schramm, V. L. (1999) Methods Enzymol. 308, 301-355.

46. Huang, C. Y. (1979) Methods Enzymol. 63, 486-500.

47. Laidler, K. J., and Peterman, B. F. (1979) Methods Enzymol. 63, 234-256.

48. Klinman, J. P. (1978) Adv. Enzymol. Relat. Areas Mol. Biol. 46, 415-494.

49. Kehres, D. G., and Maguire, M. E. (2003) FEMSMicrobiol. Rev. 27, 263-290.

50. Holm, R. H., Kennepohl, P., and Solomon, E. I. (1996) Chem. Rev. 96, 2239-2314.

51. Armstrong, R. N. (2000) Biochemistry 39, 13625-13632.















BIOGRAPHICAL SKETCH

Ewa Wroclawska was born in Poland in 1977. She studied chemistry at the

University of Adam Mickiewicz in Poznan, Poland, where she received her

undergraduate degree in bioorganic chemistry in May of 2001. Ewa Wroclawska was a

recipient of the European Union Erasmus Scholarship, which allowed her to conduct

undergraduate research at the Aristotle University of Thessaloniki in Greece for one

semester. In 2001 Ewa joined the biochemistry laboratory of Dr. Nigel Richards in the

Chemistry Department of University of Florida. After graduation, she is going to pursue

her career in research.




Full Text

PAGE 1

THE ROLE OF ARGININE 270 AND 92 RESIDUES IN THE CATALYTIC MECHANISM OF THE RECOMBINANT BACILLUS SUBTILIS OXALATE DECARBOXYLASE By EWA WROCLAWSKA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Ewa Wroclawska

PAGE 3

This thesis is dedicated to my parents and sister with many thanks for their love and support.

PAGE 4

ACKNOWLEDGMENTS I would like to thank the following people for their help and support: My teachers and mentors: Henryk Koroniak, Krzysztof Kieliszewski, Hanna Gasowska, Jim Deyrup Thesis advisor Nigel Richards My committee members Mike Scott and Tom Lyons Co-workers from the Richards research group, especially Drazenka Svedruzic and Patricia Moussatche Laurie Reinhardt for kinetic isotope effects My family My friends in Poland and Gainesville Charlie Hughes iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT.......................................................................................................................xi CHAPTER 1 INTRODUCTION, BACKGROUND AND AIMS.....................................................1 Oxalic Acid...................................................................................................................1 Oxalate Degrading Enzymes........................................................................................2 Oxalate Decarboxylase (OxDc)....................................................................................4 Properties...............................................................................................................4 The Two Crystal Structures of Oxalate Decarboxylase........................................6 Catalytic Mechanism and Identity of the Active Site of Oxalate Decarboxylase........9 Comparison of the Published Crystal Structures...................................................9 Closed vs Open Site.............................................................................................11 Mechanism of Catalysis Early Proposals.........................................................12 Mechanism of Catalysis Based on the Heavy-Atom Kinetic Isotope Effects.....14 Active Site Identity..............................................................................................16 Characterization of Oxalate Decarboxylation an Overview of This Work.............17 2 MATERIALS AND METHODS...............................................................................22 Expression of Recombinant Bacillus Subtilis Oxalate Decarboxylase.......................22 Wild Type, R270A and R270K...........................................................................22 R92K....................................................................................................................23 Purification of Recombinant Bacillus Subtilis Oxalate Decarboxylase.....................23 Buffer and Solvent Filtration...............................................................................23 Cleaning-in-Place of the FPLC Columns............................................................23 Anionic exchange columns..........................................................................23 Hydrophobic column....................................................................................24 Sample Ionic Strength.........................................................................................24 Fraction Concentration........................................................................................24 Purification of the Wild Type OxDc...................................................................25 v

PAGE 6

Purification of R270A, R270K and R92K Mutants of Oxalate Decarboxylase..25 Optimization of the Purification..........................................................................26 Site-Directed Mutagenesis and Cloning of OxDc......................................................27 PCR Reactions.....................................................................................................27 Plasmid Preparation.............................................................................................28 Transformation....................................................................................................29 Enzyme Assays...........................................................................................................29 Quantitative Assay...............................................................................................29 Qualitative Activity Assay..................................................................................29 Michaelis-Menten Kinetics.................................................................................30 The pH Dependence............................................................................................30 Protein Concentration..........................................................................................30 Inhibition Studies.................................................................................................31 Heavy-Atom Kinetic Isotope Effects..........................................................................31 3 RESULTS AND DISCUSSION.................................................................................32 Site-Directed Mutagenesis..........................................................................................32 Wild Type (WT) Oxalate Decarboxylase...................................................................32 Expression and Purification........................................................................................33 R270A..................................................................................................................33 R270K..................................................................................................................34 R92K....................................................................................................................35 Purification optimization..............................................................................35 Expression optimization...............................................................................38 Results..........................................................................................................38 Steady-State Kinetics..................................................................................................40 R270A..................................................................................................................40 R270K..................................................................................................................42 Activity.........................................................................................................42 Kinetic parameters........................................................................................43 Inhibition studies..........................................................................................44 The pH dependence......................................................................................47 R92K....................................................................................................................48 Activity.........................................................................................................48 Kinetic parameters........................................................................................48 Inhibition studies..........................................................................................50 The pH dependence......................................................................................51 Heavy-Atom Kinetic Isotope Effects..........................................................................51 R270K..................................................................................................................51 R92K....................................................................................................................52 4 CONCLUSIONS........................................................................................................54 Expression and Purification of the Wild Type and Mutated OxDc............................54 SteadyState Kinetics of the Wild Type and Mutated OxDc.....................................55 Heavy-Atom Kinetic Isotope Effects..........................................................................57 vi

PAGE 7

Active Site Identity and Mechanism of Catalysis of OxDc........................................58 The Future of the Project............................................................................................59 LIST OF REFERENCES...................................................................................................61 BIOGRAPHICAL SKETCH.............................................................................................64 vii

PAGE 8

LIST OF TABLES Table page 1 Activity decrease of the mutated OxDc in comparison to the native enzyme.........13 2 13 C and 18 O kinetic isotope effects in the wild type oxalate decarboxylase catalyzed reaction.....................................................................................................16 3 Kinetic constants for the reactions catalyzed by the His-tagged wild type and mutated OxDc..........................................................................................................17 4 Primers for mutagenesis experiments......................................................................27 5 Wild type OxDc characterization.............................................................................33 6 Purification table for R92K mutant of OxDC..........................................................40 7 Characterization of the R270A mutant of oxalate decarboxylase............................40 8 Kinetic characterization of the R270Kmutant of OxDc...........................................42 9 Inhibition studies of R270K mutant of oxalate decarboxylase................................45 10 Kinetic parameters for R270K mutant of OxDc at pH 4.2 and 5.7..........................47 11 Kinetic characterization of the R92K mutant of oxalate decarboxylase..................48 12 Inhibition studies of R92K mutant of OxDc............................................................50 13 Kinetic parameters for R92K mutant of OxDc at pH 4.2 and 5.7............................51 14 13 C and 18 O kinetic isotope effects for R270K mutant of OxDc..............................52 15 13 C and 18 O kinetic isotope effects for R92K mutant of OxDc................................53 16 Comparison of kinetic characteristic of wild type OxDc and its active site mutants.....................................................................................................................57 17 Summary of 13 C and 18 O kinetic isotope effects for the wild type and mutated OxDc........................................................................................................................58 viii

PAGE 9

LIST OF FIGURES Figure page 1 Classes of enzymes that catalyze the degradation of oxalate in (a) plants, (b) fungi, (c) bacteria.................................................................................................3 2 Part of sequence alignment of oxalate decarboxylases from different organisms.....5 3 Structural similarity between the two domains of OxDc...........................................6 4 Comparison of metal binding sites of OxDc..............................................................7 5 Structure of the OxDc monomer................................................................................8 6 Comparison of the manganese ion binding sites of oxalate decarboxylase in the two structures and a model......................................................................................10 7 Closure of the lid in oxalate decarboxylase.............................................................11 8 Catalytic mechanism proposed for oxalate decarboxylase by Anand et al..............13 9 Catalytic mechanism proposed for oxalate decarboxylase by Reinhardt et al.........14 10 Putative active sites of oxalate decarboxylase from the X-ray crystal structure......19 11 Overview of the QuikChange Site-Directed Mutagenesis method..........................28 12 Purification results of the R270A mutant of OxDc..................................................34 13 R92K purification: fractions from the DEAE-Sepharose Fast Flow column..........36 14 R92K purification: fractions from Q-Sepharose Hi-Perfomance column................37 15 Expression results for R92K....................................................................................39 16 R92K purification.....................................................................................................39 17 Michaelis-Menten kinetics of mutated oxalate decarboxylase: R270A...................41 18 Catalysis of formate production by R270A mutant of OxDc as a function of time...........................................................................................................................42 ix

PAGE 10

19 Kinetic characterization of R270K mutant of OxDc................................................44 20 R270K inhibition by malonate.................................................................................46 21 R270K inhibition by malonate Lineweaver Burk plot..........................................46 22 Catalysis of formate production by R92K mutant of oxalate decarboxylase as a function of time.....................................................................................................49 23 Michaelis-Menten kinetics of mutated oxalate decarboxylase: R92K.....................49 x

PAGE 11

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science THE ROLE OF ARGININE 270 AND 92 RESIDUES IN THE CATALYTIC MECHANISM OF THE RECOMBINANT BACILLUS SUBTILIS OXALATE DECARBOXYLASE By Ewa Wroclawska December 2004 Chair: Nigel G. J. Richards Major Department: Chemistry Oxalate and its salts are widespread in nature and have many pathogenic effects on humans and plants. Enzymes involved in the synthesis and degradation of oxalate are not well understood. Oxalate decarboxylase (OxDc) is an enzyme that catalyzes the unique conversion of oxalate to formate and carbon dioxide without participation of any organic cofactors. Two recently published crystal structures revealed that OxDc is a hexamer, with two manganese ions per monomer and that it belongs to the bicupin superfamily of proteins. The N-terminal metal binding site differs between the two crystal structures. This putative active site was considered inactive in one proposal but capable of binding the substrate and performing catalysis in the second proposal. The crystal structures showed arginine residues (92 and 270) in the proximity of the metal centers in both Nand C-terminal domains. Reinhardt et al proposed the role of Arg residue to be facilitation of the decarboxylation process by polarizing the C-O bond of the oxalate radical anion. Just et al showed that substitution of Arg92 with alanine and lysine xi

PAGE 12

residues resulted, respectively, in deactivation or 100-fold decrease of activity compared to that of the wild type. The same substitutions of Arg270 resulted in 100-fold activity decrease for the alanine mutant and 50-fold for the lysine mutant. Purpose of this research is a more detailed characterization of the arginine mutants. The role of arginine residues in the polarization of the C-O bond was tested in both putative active sites with a series of experiments using the R270A, R270K and R92K mutants of oxalate decarboxylase. The characterization included steady state kinetics experiments and 13 C and 18 O kinetic isotope effect measurements. The important feature of the proteins investigated in this work was the lack of any His-tag, which previously used to facilitate purification, caused instability and activity decrease. As a result of this work, the significance of Arg270 and Arg92 to OxDcs activity was confirmed. However, the residues in two different sites seemed to have influenced the catalytic ability of the enzyme to a different extent based on the steady-state kinetics characterization. Their involvement in the actual catalysis of the decarboxylation of oxalate was proven in the heavy-atom kinetic isotope effects experiments. The observed isotope effects supported the previously proposed mechanism of oxalate degradation that involves proton-coupled single electron transfer and a formation of a radical intermediate. xii

PAGE 13

CHAPTER 1 INTRODUCTION, BACKGROUND AND AIMS Oxalic Acid Oxalic acid is a highly toxic compound involved in many environmental, geochemical and biological processes. 1, 2 It is produced by microbes, fungi and plants as a byproduct of degradation of oxaloacetate, glyoxylate and L-ascorbic acid. These organisms possess catabolic pathways that can to some extent control the levels of oxalate. 3, 4 Oxalate accumulation in plant tissues leads to many pathological conditions caused primarily by its metal chelating ability. A number of essential minerals in the soil precipitate after binding oxalate. Availability of phosphorus to plant roots is increased due to the oxalate chelation of aluminum and calcium. 2 Fungal pathogens can utilize oxalate to their own advantage but also secrete it into the plant tissues during the initial stages of pathogenesis, which causes cell degradation. 5 Moreover, oxalate plays a role in the regulation of osmotic potential and pH, as well as in calcium ion storage in plants. 6 Oxalate is a key factor in the carbon cycle and in CO 2 release from rotting wood. 7 Fungi use oxalate manganese complexes to promote degradation of lignin, which affects enzymes responsible for cell wall synthesis. 8 For example, the fungus Whetzelinia sclerotinium, uses oxalate to induce damage to sunflower plants. During pathogenesis, the concentration of oxalate increases in the host tissue leading to leaf death. 4 Humans and other vertebrates consume oxalic acid, found mainly in green leafy plants such as spinach and rhubarb but also in black tea and ginger juice. However, 1

PAGE 14

2 humans lack oxalate degrading enzymes. It was proposed that intestinal bacteria, such as Oxalobacter formigenes, could be introduced into human gastrointestinal tract to catabolize oxalate. 3 These bacteria can, however, be eliminated from the gut flora by extensive antibiotic treatments, which results in the increase of oxalate levels in humans. 9 A number of pathological conditions such as hyperoxyluria, formation of kidney stones, renal failure, cardiomyopathy and vulvodynia are caused by oxalate. 10 Due to the problems related to oxalate accumulation, numerous efforts have been made to reduce the amount of oxalate in food, including engineering transgenic plants to enable them to express oxalate degrading enzymes. 5 Structural, biochemical and mechanistic information needs to be obtained for oxalate degrading enzymes to utilize them in therapy, industry and agriculture. 11,12 The most interesting aspect of oxalate degrading enzymes is the variety of mechanisms they employ. 13 Oxalate Degrading Enzymes Oxalate degrading enzymes have potential uses in new therapeutic strategies for lowering oxalate levels in biological fluids. For many years now, these enzymes have been used in urine and blood testing for the presence of oxalate. 14,15 Recently, re-colonizing humans with intestinal bacteria that produce these enzymes has been used as a preventive therapy. 9 This new approach has been introduced, but has not been widely recognized. Three major enzymes have evolved in plants, fungi and bacteria. All of which catalyze the degradation of oxalic acid. Each of these enzymes employs a different mechanism, such as oxidation, decarboxylation in the presence of coenzyme A, or direct decarboxylation [Figure 1]. 13

PAGE 15

3 H O O H SCoA O O O O O O O O O O O O O CO2 + H2O2CO2 + CO2 + Oxalate oxidaseO2, 2H+Oxalate decarboxylasecat. O2, H+Oxalyl-CoA decarboxylaseH+A.B.C. Figure 1. Classes of enzymes that catalyze the degradation of oxalate in (a) plants, (b) fungi, (c) bacteria. 13 Oxalate oxidases (OXO) catalyze oxidation of oxalate into carbon dioxide and hydrogen peroxide mostly in plants [Figure 1]. OXO requires molecular oxygen for its activity but no organic cofactors. 16 The main source of this enzyme is barley root. 17 One of the reaction products, H 2 O 2, has been suggested to be involved in cell wall crosslinking and can act as a fungicide. OXO has extreme thermal stability, and therefore can be involved in the defense against biotic and abiotic stress in plants. The addition of oxidase activity is one of the targets for transgenic plant engineering. 18 OXO has been crystallized and its structure determined for Hordeum vulgare protein. 19 It belongs to the cupin protein superfamily and has a manganese ion in its active site. 17 There are three histidines coordinating the metal ion with the fourth coordination site occupied by carboxylate from a glutamate residue. These four amino acid residues are conserved in the sequences of many metalloenzymes in the cupin superfamily. 20 Their relevance to enzyme activity has been confirmed through sitedirected mutagenesis for another member of the superfamily, oxalate decarboxylase. 21 It was suggested that gene duplication facilitates the evolution of new enzymes due to sequence divergence. 22 This is how the bicupins (oxalate decarboxylase) are formed from cupins (OXO). 3 It has been suggested that gene

PAGE 16

4 duplications must have occurred for OxDc to become a fully functional and structurally developed enzyme during evolution from a single cupin OXO. Presumably, first, the number of cupin genes doubled, and then the gene fusion occurred to produce the two-domain bicupin. 5 Formyl-CoA transferase (FRC), along with oxalyl-CoA decarboxylase (OXC), are involved in the oxalate degradation pathway of Oxalobacter formigenes, a bacterium involved in mammalian oxalate catabolism. The reaction catalyzed by FRC involves the transfer of coenzyme A from formate to oxalate producing oxalyl-CoA and formate [Figure 1]. 23 Oxalate Decarboxylase (OxDc) Properties The enzyme of interest in this study is oxalate decarboxylase (OxDc), which is mainly found in fungi ( Aspergillus niger, Flammulina velutipes, Sclerotinia sclerotiorum) and more recently in the bacterium Bacillus subtilis. 20, 24 26 The most thoroughly characterized OxDcs come from B. subtilis. The bacterium B. subtilis reportedly possesses more than one gene encoding oxalate decarboxylase activity, YvrK and YoaN. 27 OxDcs found in fungi and bacteria have many common features. These features are presumably responsible for their catalytic activity: the conversion of oxalate to formate and carbon dioxide [Fig. 1]. 24 The decarboxylation process does not require any organic cofactors, such as coenzyme-A or ATP. 28 OxDcs consume sub-stoichiometric amounts of oxygen, relative to products, during turnover and are inactive in anaerobic conditions. It was found that even though OxDc is sensitive to the presence of oxygen, it is most likely only upon substrate binding. 13

PAGE 17

5 The enzyme is most active in acidic pH. The isoelectric point (pI) has been reported to be 3.3 or 2.5 for the fungal OxDc and 6.1 for the bacterial one. 5, 20 The latter enzyme is stable between the pH of 4.0 and 7.5, while the optimum activity is between the pH 4.0 and 5.0. 3,4 OxDcs stability is increased in the presence of o-phenylenediamine, a compound used in the qualitative assay for oxalate decarboxylase turnover and in the presence of surface active non-ionic detergents (Tween 20, Triton-X). 24, 29 Alignment of three sequences of oxalate decarboxylases from different organisms, both fungal and bacterial, shows the conserved residues from both metal binding sites [Figure 2]. 21, 27 151 170 234 AnOxDc MRLDEGVIRE LHWHREAEWA. NGTEFLLIFD DGNFSEESTF FvOxDc MRLEAGAIRE LHWHKNAEWA EGSEFILVFD SGAFNDDGTF BsOxDc MRLKPGAIRE LHWHKEAEWA..EGAEFLLVFD DGSFSENSTF 84 103 166 331 350..401 420 AnOxDc AAAHLTINPG AIREMHWHPN...EEVEVLEIFR ADRFRDFSLF FvOxDc AVAEVTVEPG ALRELHWHPT...TTLTYLEVFN TDRFADVSLS BsOxDc ASALVTVEPG AMRELHWHPN...EPLVFLEIFK DDHYADVSLN 258 277.327 346 Figure 2. Part of sequence alignment of oxalate decarboxylases from different organisms: Aspergillus niger (AnOxDc), Flammulina velutipes (FvOxDc), Bacillus subtilis (BsOxDc). Conserved residues of interest in the active sites are shown in bold letters. The residues of interest in this project are Arg92 and Arg270 conserved in the Nand C-terminal domains respectively. The positively charged arginines were predicted to polarize C-O bond in oxalate. Also presented is conserved Glu333 from the C-terminal active site that according to the crystal structure on which this work was based, can serve as a general base during catalysis of OxDc. 3,13 Conserved Glu162 from the N-terminal

PAGE 18

6 domain, which was presumed to be catalytically relevant in the latest publication is also shown. 6 The Two Crystal Structures of Oxalate Decarboxylase The structure of the fungal oxalate decarboxylase has been proposed using sequence homology between the bicupins coming from different species and from a structural model. 21 The crystal structure of bacterial enzyme was determined in 2002 by Anand et al and by Just et al in 2004. 3, 6 The first crystal structure of bacterial OxDc was solved at 1.75 resolution in the presence of formate. 3 B. subtilis OxDc crystallizes as a hexamer, which contains two trimeric layers in which each monomer belongs to the bicupin structural family. One domain of the monomer is at the C-terminus and the other at the N-terminus [Figure 3]. Figure 3. Structural similarity between the two domains of OxDc. 3 Blue color: N-terminus, red color: C-terminus. (A) Domain I includes residues 56 233. (B) Domain II includes residues 8 55 and 234 379. Reprinted with permission of Anand et al, Biochemistry (2002) 41, 7659. Copyright 2002 American Chemical Society. Both cupin domains contain a manganese ion coordinated to one glutamate and three histidine residues [Figure 4]. The molecular mass of the enzyme is 264 kDa for the entire hexamer and 43.6 Da per monomer. It has the motifs characteristic to the cupin

PAGE 19

7 superfamily: a -sandwich, which consists of a six-stranded -sheet and a five-stranded -sheet. The contact between the subunits is established by several -helices [Figure 5]. 3 Figure 4. Comparison of metal binding sites of OxDc. 3 (A) manganese binding site of domain I. (B) manganese binding site of domain II. (C) Unknown metal site at the protein surface. The metal was assigned to be magnesium for the purpose of the X-ray refinement. Reprinted with permission of Anand et al, Biochemistry (2002) 41, 7659. Copyright 2002 American Chemical Society.

PAGE 20

8 Figure 5. Structure of the OxDc monomer. 3 (A) Stereoview of the C trace of OxDc. The color changes from red to blue from N-terminus to C-terminus. (B) Structure of the OxDc monomer, highlighting the secondary structural elements with -sheets and -helices colored as in panel C, 3 10 helices in cyan and loops in yellow. (C) Topology diagram of OxDc, showing the domains I and II. The six-stranded -sheets that make up the front of the cupin barrel are in blue, and the five-stranded -sheets that make up the back are in red. The -helices are in green. Reprinted with permission of Anand et al, Biochemistry (2002) 41, 7659. Copyright 2002 American Chemical Society.

PAGE 21

9 The two metal binding sites described above are presumed to be the enzymes active sites, and there is an ongoing discussion as to which domain is actually responsible for catalysis. In this model, there is a third metal binding site at the surface of the protein that is most likely not involved in catalysis [Figure 4]. 3 The crystal structure described most recently for bacterial oxalate decarboxylase was obtained based on the refinement of the coordinates from the first study as well as additional, new 2.0 resolution data. It differs from the previous one in the conformation of one of the 3 10 helices, which creates a loop near the N-terminal metal binding site and changes the identity of the second shell residues available to catalysis. Motifs defining OxDc as a member of a cupin family, as well as metal coordinating residues in the binding sites are exactly as described in the previous study. However, there are significant changes in water occupancy of the active sites and their accessibility to the substrate [Figure 6]. 6 Catalytic Mechanism and Identity of the Active Site of Oxalate Decarboxylase Comparison of the Published Crystal Structures In both published crystal structures of oxalate decarboxylase, there are two cupin domains in the enzyme that are similar. Both metal binding sites are presumed to be capable of performing the catalytic reaction. 3 The second coordination sphere in the N-terminal site depends on the conformation of a 3 10 helix that is different in both proposals. In both structures arginine residues (Arg270 and Arg92) have the same position near the manganese ions. The position of glutamate residue (E333) in the C-terminal site is the same in both proposals, while the position of the glutamate in the N-terminus (E162) is not [Figure 6].

PAGE 22

10 Figure 6. Comparison of the manganese ion binding sites of oxalate decarboxylase in the two structures and a model. 6 Metal binding sites of the closed structure (site 1 = A and site 2 = B) are shown next to the open structure sites ( site 1 = C; site 2 = D). In the molecular model (E) oxalate and dioxygen are bound to site 1 manganese ion and the need for displacement of Glu-162 is shown in comparison with the experimental closed structure (A). Dashed lines represent interionic distances. Reprinted with permission of Journal of Biol. Chem. (2004) 279, 19867. Copyright 2004 ASBMB. The most significant change in the N-terminal site is related to the movement of the surface loop that includes the aforementioned 3 10 helix and is built from residues 161 to 165 (Ser, Glu, Asn, Ser, Thr) [Figure 7]. 3,6 These residues create a lid in the putative active site structure.

PAGE 23

11 Figure 7. Closure of the lid in oxalate decarboxylase. 6 Open (A) and closed (B) hexameric structures are shown partially. Solvent-accessible surface of the protein is shown in green, interior of the protein in pale yellow, solvent in blue. The trajectory for substrate entry is indicated by a broken arrow. Reprinted with permission of Journal of Biol. Chem. (2004) 279, 19867. Copyright 2004 ASBMB. Closed vs Open Site The presence of the lid, identified in the 2004 structure, constitutes a dramatic change in understanding of the role of the N-terminal site. In the first structure, the loop created the channel allowing an easy access of the solvent to the Mn ion in the N-terminus. The site was therefore called open. In the latest structure the loop occludes the channel and prevents solvent access to the Mn-binding site. Therefore the site is called closed. The conformational flip of the loop positions the glutamate 162 side chain in very close proximity (under 2 ) to the metal and makes it a great candidate to be a part of the catalytic mechanism as a proton donor. 6 The conformational change is relevant, because

PAGE 24

12 before this proposal only the C-terminal site was considered to be a good candidate to perform catalysis in oxalate decarboxylase. Mechanism of Catalysis Early Proposals There have been a few mechanisms of catalysis proposed for oxalate decarboxylase throughout the years. Initial characterization of the fungal OxDc included mutagenesis studies, which supplied evidence for the crucial catalytic role of the manganese ions by substituting metal binding residues. Removal of any of them in either of the sites resulted in the complete inactivation of the enzyme. Therefore, it has been assumed that the two manganese binding sites are acting in cooperation. 21 The metal contents of F. velutipes recombinant enzyme were established in the same study as 2.5 Mn per monomer. Emiliani and Riera have also found traces of hydrogen peroxide as an additional product of the oxalate decarboxylase catalyzed reaction, which suggested a single electron transfer to be involved in the mechanism. 24 It has been established that the fungal enzyme requires oxygen for the reaction, even though there is no net redox change during the process. 20 Subsequently, it was proposed that the decarboxylation cycle of the bacterial enzyme included a percarbonate intermediate, however, this has never been proven. 6 Publication of the first crystal structure from B. subtilis, as well as kinetic isotope effect measurements, for the recombinant enzyme, have provided an interesting insight into the nature of the bacterial OxDc. 3, 13 It has been suggested that all the residues necessary for catalysis are found only in the C-terminal metal binding site, while the general base (Glu) was missing in the N-terminal one. 3 The proposed mechanism required an enzymatic source of acyl proton for formate [Figure 8].

PAGE 25

13 O O O O Mn4+O OH His His His Glu O O O O Mn3+O OH His His His Glu O O Mn3+O OH His His His Glu O O H Mn3+O OH His His His Glu O O H Mn4+O OH His His His Glu OH2Mn4+O OH His His His Glu CO O CO O CO O H O OH CO O Figure 8. Catalytic mechanism proposed for oxalate decarboxylase by Anand et al. 3 Mutagenesis studies confirmed that mutating the conserved Arg270 and Glu333 from this active site resulted in a significant activity decrease [Table 1]. The proteins were, however, purified with an N-terminal His-tag that caused their precipitation and instability. Table 1. Activity decrease of the mutated OxDc in comparison to the native enzyme. 3 OxDc mutant Activity decrease compared to the native enzyme [fold] E333A 4 Y340F 13 R270E 20 The role of Arg270 was predicted to form an ion pair with the second carboxylate of the substrate in order to stabilize the charge division in the molecule and the position of oxalate in the active site. This stabilization of the intermediate was expected to facilitate the decarboxylation process. 3

PAGE 26

14 Mechanism of Catalysis Based on the Heavy-Atom Kinetic Isotope Effects Heavy-atom kinetic isotope effect (KIE) studies on the wild type oxalate decarboxylase have supplied new information about the mechanism of catalysis of the enzyme [Figure 9]. The C-terminal site was, at the time, considered to be the only one comprising of all the pieces necessary for catalysis. 13 O O O O Mn3+O OH His His His Glu H O O Glu333 N+ N Arg270 N H H H H H O O O O Mn2+O OH His His His Glu OH O Glu333 N+ N Arg270 N H H H H H O C O Mn2+O OH His His His Glu O O Glu333 H N+ N Arg270 N H H H H H O O H Mn3+O OH His His His Glu O O Glu333 N+ N Arg270 N H H H H H -"H" abstraction+-CO2Figure 9. Catalytic mechanism proposed for oxalate decarboxylase by Reinhardt et al. 13 Kinetic isotope effects can be used to deduce the structure of the transition state in enzyme mechanisms in which bond-breaking and bond-making events are rate limiting. 30 A KIE is defined as the change in the rate of reaction with an isotope labeled substrate. It provides information on structural changes in going from the reactants ground states to the transition state. Changes in bond orders result in isotope effects. The

PAGE 27

15 magnitude of KIEs depends on the extent to which the chemical step is rate limiting in the catalysis. The intrinsic isotope effect may be masked if the rate limiting step is either binding or conformational changes, and not chemistry. 31-33 The isotope effects on V max /K M (V/K KIEs) are associated with the steps up to and including the first irreversible step in the mechanism. 34, 35 The proposed mechanism suggests that an electron is transferred from bound substrate, oxalate, to the manganese and molecular oxygen complex. It seems that the enzyme must stabilize the radical species before the C C bond cleavage. Positively charged arginine residue and carboxylate group of a glutamate carrying a negative charge can serve this purpose. After the decarboxylation, transfer of an electron and a proton to a metal-bound formate radical anion yields the final product [Figure 9]. 13 The values of isotope effects for the wild type oxalate decarboxylase confirmed that the C-C bond cleavage is not the rate-limiting step [Table 2]. The values for carbon dioxide production are lower than the typical carbon bond cleavage KIEs values of 3 to 5 %. Results for formate production suggest that a different step, prior to the bond cleavage, was slower and rate-limiting. This assumption is supported by (i) the catalytic dependence on the dioxygen presence, (ii) the presence of manganese, (iii) the absence of organic cofactors, and (iv) no net redox change between substrates and products. At pH 5.7 chemical steps are even more rate limiting than at pH 4.2. This is probably due to the decrease in external commitments to catalysis. The bond order increased for oxygen as the reaction proceeded (from 1.5 to 2), therefore the 18 O KIE values are inverse for CO 2 production. The substrate-based radical formation was expected to facilitate the cleavage of the C-C bond. The rate-limiting step predicted in

PAGE 28

16 this mechanism is the oxidative transfer of an electron from oxalate to a Mn (III) dioxygen complex, coupled with the hydrogen abstraction. 13 Arginine residues in the enzymes metal binding sites are expected to facilitate the reaction by polarizing the C-O bond and provide stabilization to the charged intermediate as described for the previous mechanism. Table 2. 13 C and 18 O kinetic isotope effects in the wild type oxalate decarboxylase catalyzed reaction. 13 pH 13 (V/K), % CO 2 13 (V/K), % formate 18 (V/K), % CO 2 18 (V/K), % formate 4.2 0.5 0.1 1.5 0.1 -0.2 0.2 1.1 0.2 5.7 0.8 0.1 1.9 0.1 -0.7 0.1 1.0 0.1 The wild type kinetic isotope effects experiments will be used as a model for the mutant characterization. The aim of the experiments on the mutants is to establish whether the mutation affects the stability of putative intermediates. Active Site Identity The most recent findings were published in 2004 along with the new crystal structure. 6 As described above, the conformational change resulted in opening the N-terminal site for catalysis and positioning Glu162 as an equivalent of Glu333, present in the C-terminal site. Hence, there is a general base available for catalysis in both metal binding sites. Further mutagenesis experiments have been reported to support the new thesis that the N-terminal site is more likely to be involved in the catalysis. Just et al have observed a decrease in activity of the enzyme of 100 fold for the N-terminal site mutants and 10 to 50 fold for the C-terminus mutants [Table 3]. 6 It was suggested that the C-terminal site, with restricted access for both solvent and substrate, has a structural role in the catalysis. The steady-state characterization of Arg92 and Arg270 mutants was

PAGE 29

17 published but the proteins used for the experiments contained C-terminal His-tags. The N-terminally His-tagged OxDc was found to be unstable leading to precipitation. 3 Comparison of the specific activity values obtained by Just et al and those described in this work suggest the unfavorable effect that also the C-terminal His-tag has on oxalate decarboxylase. Specific activity of the wild type protein obtained in this study is about 40 U/mg while the enzyme used by Just et al only 21 U/mg. 6 The activities of all mutants described along with the second crystal structure are highly decreased. In this work, the non-tagged mutated proteins were used, and all the results supply information about a stable, non-precipitated enzyme, that is most likely at its highest possible activity. Table 3. Kinetic constants for the reactions catalyzed by the His-tagged wild type and mutated OxDc. 6 Limit of detection: 0.03 U/mg OxDc Specific activity [U/mg] K M [mM] k cat /K M [M -1 s -1 ] Wild type 21.0 16.4 952 R92A 0 not determined not determined R92K 0.20 2 68 R270A 0.26 8 24 R270K 0.54 1 410 Characterization of Oxalate Decarboxylation An Overview of This Work This project describes the entire process from mutagenesis to obtaining and characterizing a number of mutants of the B. subtilis oxalate decarboxylase. Problems arising from working with mutants are described, as well as attempts at overcoming them. The importance of this study comes from numerous existing and predicted medical, industrial and environmental applications, as well as an insight into the evolutionary processes within the cupin superfamily of enzymes. 18, 36 Oxalate decarboxylase has been studied for over 50 years and only in the past few years has progress been made. The publication of two crystal structures, as well as kinetic

PAGE 30

18 isotope effect measurements, have enabled more informed proposals of the catalytic mechanism. 3, 6, 13 However, there are still many unanswered question and uncertainties about the catalysis of oxalate decarboxylase, from the identity of the enzymes active site to the actual mechanism of catalysis. The two published crystal structures differ enough for their authors to propose opposite metal binding sites as the active site. 3, 6 In both of these studies, however, there are many assumptions made that are clearly not final and need unambiguous evidence. The mutagenesis results presented by Anand et al cannot be considered a thorough analysis due to the general instability of the protein, which included a destabilizing N-terminal His-tag. 3 Just et al have performed a much more detailed mutagenesis study on a stable, C-terminally His-tagged enzyme. Activity of all His-tagged enzymes is decreased compared to the non-tagged proteins described in this work. Wild type enzyme obtained by Just et al has specific activity of 21 U/mg compared to the average of 40 U/mg of the enzyme in this study. 6 This study has been based on the crystal structure published by Anand et al in the year 2002 [Figure 10]. 3 Investigation of the catalytic mechanism and of the role of the protein environment in controlling properties of the metal centers in B. subtilis oxalate decarboxylase was conducted using site-directed mutagenesis, steady-state kinetics, and heavy atom isotope effects. 37-40 The main questions that need to be answered, and which this work is addressing are (i) what is the role of conserved arginines in the Nand C-terminal active sites, (ii) which manganese binding site of OxDc is responsible for catalytic activity of the enzyme, and

PAGE 31

19 (iii) is the decarboxylation process facilitated by the electrostatic stabilization of the transition state by the arginine side chain? Leu-153 Val-82 Met-94 Tyr-200 Arg-92 Glu-333 Val-321 Arg-270 Tyr-340 Figure 10. Putative active sites of oxalate decarboxylase from the X-ray crystal structure. Residues potentially important for the catalytic mechanism are indicated in a three letter code. The mutations were performed on arginines in both metal binding sites (R270 and R92) to try and answer the question of (i) their role or significance in the catalytic mechanism, and (iii) their role in the enzymes substrate selectivity. The innovative approach in this research excluded the use of any His-tags to facilitate the purification process. The His-tags have been shown to reduce both the activity and the stability of the enzyme. 6 All the mechanisms proposed thus far predict a significant role for Arg270 and Arg92 in sustaining enzymes activity. The hypothesis of this work has been that these conserved arginines play a role in either catalysis or substrate selectivity of the enzyme. The native enzyme exhibits high substrate selectivity towards oxalate. 13 Therefore, certain carboxylic acids and diacids of size similar to oxalate might be able to bind in the active site of the mutated enzyme and inhibit it. Replacement of positively charged Arg

PAGE 32

20 should have the most dramatic effect when substituted with a neutral and small alanine residue, and much smaller, but detectable effect with lysine. The removal or decrease of the charge would presumably result in, respectively, lack or decrease of enzymes ability to polarize the C-O bond and to facilitate the decarboxylation of oxalate. The kinetic parameters, as well as the activity and stability of the enzyme were tested after introducing these substitutions. The change in the turnover number was not expected if the mutated residue affected just the selectivity of the enzyme. The value of K M was expected to change: either decrease if the mutation allowed unproductive binding or increase if the possible inhibitors competed with oxalate for the active site. Even though the steady-state characterization has been published along with the latest crystal structure, the proteins used had, as mentioned above, decreased activity due to the use of a C-terminal His-tag. Steady state kinetics provided information on changes in enzyme activity caused by mutations of arginine residues. Changes in the turnover number prompted the measurements of the kinetic isotope effects. If, as predicted, the role of arginine in the catalysis is to facilitate decarboxylation by stabilizing the negative charge in the intermediate, the KIEs should be observed and be altered compared to wild type OxDc. Complete removal of the positive charge by mutating Arg to Ala was expected to almost completely deactivate the enzyme. Substituting arginine side chain with an also positively charged, but shorter lysine would lower the rate of decarboxylation resulting in increased values for kinetic isotope effects. Differences in values for formate and CO 2 arise from their influence on the rate-limiting step: proton-coupled electron transfer. The primary 13 C and secondary 18 O isotope effects were measured by the analysis of CO 2 (g) by the

PAGE 33

21 internal competition method using isotope ratio mass spectrometry. Therefore the results obtained represent the isotopic substitution effects on V/K oxalate and are associated with the first irreversible step of the mechanism and all the steps leading to it. 34, 35 Analyzed CO 2 (g) was obtained from direct isolation from the oxalate decarboxylase, catalyzed partial conversion of oxalate or from the oxidation of initial oxalate, formate and oxalate from residual substrate after the partial reaction in anhydrous DMSO with iodine. Isotopic ratios obtained are the R values (defined under the equation below) for these compounds. Analysis of KIE results for mutated OxDc was based on a literature equation for the wild type enzyme described below. 13, 41 2//21/21/1ln1ln)1(ln)1ln(22181321813formateCOformateCORRRpRRxappIExformateKVorappIExxCOKVorRoRpffRoRsffappIE f = fraction of reaction R isotopic ratio determined by MS for: Ro = initial oxalate Rs = residual substrate R CO2 = produced CO 2 R formate = produced formate

PAGE 34

CHAPTER 2 MATERIALS AND METHODS Expression of Recombinant Bacillus Subtilis Oxalate Decarboxylase Plasmid, previously produced by the insertion of the yvrk gene from B. subtilis into the pET-9a expression vector, was transformed into the competent JM109 cells, and sequenced to confirm the presence of the 1158 base pair long insert. This plasmid was used to transform the E. coli strain BL21(DE3). Luria-Bertani broth containing kanamycin (LBK, 1.0 L) was inoculated with yvrk:pET9a/BL21(DE3) in order to express the yvrk-encoded protein. 13 Wild Type, R270A and R270K Cells were shaken at 37 C until the optical density (A 600 ) of the cultures has reached 1.7 and the cells were ready to be induced. The bacteria were subjected to heat shock for 18 min followed by the addition of the inducing agent isopropyl thiogalactoside (IPTG, 1 mM) and manganese chloride (MnCl 2 5 mM). After being agitated for 4 h in 37 C the cells were harvested by centrifugation (5000 rpm, 30 min, 4 C), resuspended in the TrisHCl lysis buffer containing MnCl 2 and sonicated for 30 s. Lysate and cell debris were separated by centrifugation (8000 rpm, 20 min, 4 C). The pellet suspended in the extraction buffer consisting of 1 M sodium chloride, 10 mM 2-mercaptothanol and 0.1 % Triton-X was stirred overnight at room temperature and the supernatant was combined with the original lysate. 13 For the R270A and R270K mutants the expression protocol was followed as for the wild type OxDc, only the lysis pellets were discarded instead of being used for protein extraction. 22

PAGE 35

23 R92K The expression protocol was significantly changed for this low activity mutant. Competent BL21(DE3) cells transformed with R92K plasmid were used to inoculate 25 mL of LBK in a 250 mL flask, which was shaken overnight at 30 C until the cultures reached OD 600 of 3.2. Fresh LB (4 x 200 mL) without antibiotic in baffled flasks (2 L) was inoculated with 2 mL of the overnight culture. The cells were grown at 30 C until the OD 600 of 0.3, at which point the cells were heat-shocked for 5 min at 42 C and induced with IPTG and MnCl 2 as for the wild type. The induction time varied from 2.5 to 3.5 h depending on when the cultures reached the OD 600 of 1.6. At this time the cells were harvested. Lysis, as well as lysis pellet extraction was performed as in the wild type OxDc procedure. Purification of Recombinant Bacillus Subtilis Oxalate Decarboxylase Buffer and Solvent Filtration In order to remove any particulate matter all the buffers and water solutions were filtered through Millipore 0.45 m membrane filters before use in chromatography, 20 % ethanol was filtered through 0.20 m membrane filter. Cleaning-in-Place of the FPLC Columns All the resins used in Fast Performance Liquid Chromatography (FPLC) system (kta prime; Amersham Biosciences) were cleaned before loading each sample according to manufacturers instructions. Anionic exchange columns The ionically bound proteins were removed by washing DEAE-Sepharose Fast Flow (Sigma) and Q-Sepharose Hi-Performance (Amersham Pharmacia Biotech)

PAGE 36

24 columns with 50 mL of 2 M NaCl in a reversed flow direction (4 mL/min). To remove precipitated proteins, as well as lipopropteins and proteins bound hydrophobically, columns were washed (2 mL/min) in the reverse flow direction with 100 mL of 1 M NaOH. A following wash with 20 % ethanol in reverse direction flow (4 mL/min) eliminated the strongly hydrophobically bound proteins and lipoproteins. Before use the columns were equilibrated with approximately 150 mL of a low ionic strength buffer. Hydrophobic column Phenyl-Sepharose Hi-Perormance (Amersham Pharmacia Biotech) column was cleaned in reversed flow direction by washing with 0.5 M NaOH (1 mL/min; 10 min) and 20 % ethanol (4 mL/min; 25 min). Then it was washed in the forward flow direction with the low ionic strength buffer and before use equilibrated with the high ionic strength buffer (both 4 mL/min; 100 mL). Sample Ionic Strength The ionic strength of the loaded sample was adjusted for specific resins and buffer sets by dilutions or addition of ammonium sulfate. Anionic exchange columns require low conductivity (10 ) and hydrophobic column requires samples with high ionic strength (> 20 ). Fraction Concentration Active fractions from the last column before dialysis were concentrated using the Amicon Ultrafiltration Pressure Chamber on a Millipore Ultrafiltration Membrane with an exclusion size of 30 K. The dialysis pool was concentrated in the Ultrafree-15 Centrifugal Filter Devices (Millipore).

PAGE 37

25 Purification of the Wild Type OxDc The purification system consists of two anion exchange and one hydrophobic FPLC columns. First resin used was the weak anionic exchange DEAE-Sepharose in a 2.5 x 30 cm column onto which 10x diluted lysate and pellet extract sample was loaded. The washing buffer used for all columns was 50 mM imidazole hydrochloride with 10 uM MnCl 2 at pH 7.0. The column was eluted with 0 to 1 M NaCl gradient (500 mL total). Collected fractions were checked for their protein contents by UV spectroscopy (A 280 ) and assayed for their ability to catalyze the oxidation of o-phenylenediamine. Fractions, which displayed activity were combined and solid ammonium sulfate was added to the final concentration of 1.7 M. Precipitated proteins were removed by centrifugation and the supernatant loaded onto the Phenyl-Sepharose Hi-Performance column. This time the eluting gradient was 1.7 to 0 M (NH 4 ) 2 SO 4 (500 mL total). Active fractions were recognized in the same way as after the DEAE column. They were pooled and diluted 10x before being loaded onto the anion exchange Q-Sepharose Hi-Performance column, from which the protein of interest was eluted using the same buffers as described for the DEAE column. The fractions containing the enzyme were concentrated by ultrafiltration to the volume of 10 mL and dialyzed against the storage buffer (20 mM hexamethylenetetramine hydrochloride, pH 6.0). The enzyme was concentrated and divided into 100 uL aliquots stored at 80 C. 13 Purification of R270A, R270K and R92K Mutants of Oxalate Decarboxylase During the purification of the mutated OxDc only two anionic exchange columns were used: DEAE-Sepharose and Q-Sepharose with the same buffers and elution gradients as described for the wild type. The ammonium sulfate step was omitted along

PAGE 38

26 with the Phenyl-Sepharose column. The sample loaded onto the Q-Sepharose required 5x dilution. Optimization of the Purification Difficulties in the purification of mutated oxalate decarboxylase according to the wild type enzyme purification protocol resulted in making a few attempts to find resins more suitable for purification than Phenyl Sepharose or to find optimal ammonium sulfate concentration for the purification step based on protein precipitation. To test the latter different amounts of ammonium sulfate (from 0 to 1.7 M final concentrations) were added to the 750 L aliquots of a pool of active fractions from the DEAE-Sepharose column. SDS gel chromatography was used to visualize the relative amounts of precipitated proteins. Resins tested: not used previously anion exchange Q-Sepharose, two hydrophobic ones, Butyland Octyl-Sepahrose (both Amersham Pharmacia Biotech), as well as cation exchange SP-Sepharose (Sigma). Batch-wise purification experiments were performed. In each case 0.4 mL of resin, not slurry, was placed in an Eppendorf tube (1.5 mL) and washed ( 0.5 h with rocking) with start buffer: buffer A (pH 7.0, 50 mM imidazole, 10 M MnCl 2 ) for the anion exchange resin; buffer A including 0.6 mM (NH 4 ) 2 SO 4 and acetate (pH 5.2, 10 L MnCl 2 ) for the cation exchange resin. Then the ion exchange resins were washed (rocking for 5 min) with buffer A solutions containing increasing amounts of sodium chloride (from 0 to 1 M). Hydrophobic resins were washed in the same way with solutions of start buffer containing decreasing amounts of ammonium sulfate (from 0.6 to 0 M). All fractions were visualized on SDS-PAGE gels to establish binding to certain resins and influence of the gradient.

PAGE 39

27 Site-Directed Mutagenesis and Cloning of OxDc Site-directed mutagenesis was performed using the QuikChange Site-Directed Mutagenesis Kit (Stratagene) according to manufacturers guidelines, using primers in Table 4. DNA sequence of each mutant was obtained from the DNA Sequencing Core Laboratory ICBR at the University of Florida. Table 4. Primers for mutagenesis experiments. Primer name Primer sequence 5 to 3 R270A sens CCC GGC GCC ATG GCT GAA CTG CAC TGG R270A asens CCA GTG CAG TTC AGC CAT GGC GCC GGG R270K sens CCC GGC GCC ATG AAA GAA CTG CAC TGG R270K asens CCA GTG CAG TTC TTT CAT GGC GCC GGG R92K sens CCA GGC GCG ATT AAA GAG CTT CAC TGG R92K asens CCA GTG AAG CTC TTT AAT CGC GCC TGG PCR Reactions One polymer chain reaction of 100 L contained 5 L of 10x reaction buffer (Stratagene), 125 ng of each primer (sens and asens), 20 ng of dsDNA template, 125 ng of dNTP mix (Stratagene) and sterile distilled water. The PCR reaction was started from a 30 s heating step to 95 C, after which PfuTurbo DNA Polymerase (1 L of 2.5 U/L; Stratagene) was added to the reaction mixture. Then followed the 16 cycles of 30 s denaturation at 95 C, 1 min annealing at 55 C and 12 min extension (2min/kb) at 68 C. XL1-Blue Supercompetent cells were transformed with the digested (DpnI, Stratagene) PCR product and plated on LB/kanamycin plates. The general method used in the QuikChange Site-Directed Mutagenesis kit is presented in Figure 11.

PAGE 40

28 Figure 11. Overview of the QuikChange Site-Directed Mutagenesis method. Reprinted with the permission of Stratagene. 42 Plasmid Preparation Plasmid was extracted either by the standard alkaline lysis, chloroform extraction and PEG precipitation protocol required by the DNA Sequencing Core Laboratory ICBR at the University of Florida or by using Wizard Plus Miniprep DNA Purification System (Promega). In both cases plasmid DNA was obtained from 10 mL overnight cell cultures grown in LBK.

PAGE 41

29 Transformation BL21(DE3) cells were transformed with each plasmid DNA extracted from the XL1-Blue Supercompetent cells. They were added to the SOC media and incubated for 1 h with shaking at 225 rpm at 37 C. Transformed colonies were picked from the LBK plates and placed in the 100 L aliquots of 50 % glycerol. Ready for expression cells were frozen at 80 C. Enzyme Assays Quantitative Assay The most basic quantitative assay utilizes the ability of oxalate decarboxylase to oxidize o-phenylenediamine (o-PDA) to 2,3-diaminophenazine, which is supposedly a side reaction during enzymes turnover. 29 Assay mixtures consisted of 50 mM potassium acetate buffer (pH 4.2), 0.2 mM Triton X, 0.5 mM o-PDA, water and 50 mM potassium oxalate (pH 4.2). Addition of 10 L fraction from purification initiated the reaction. Presence of the byproduct of the OxDc turnover is manifested by the appearance of the yellow color, which confirms the presence of the active enzyme in the mixture. Qualitative Activity Assay Activity was measured in the end-point assay initiated by the addition of the substrate. After certain time, specified for each protein to be from 3 to 30 min, raising the pH up to 12 with NaOH terminated OxDc turnover. The levels of produced formate were established in a coupled assay. After an overnight incubation of the resulting mixture at 37 C with formate dehydrogenase and NAD + measurement of absorbance at 340 nm gives the amount of produced formate. The assay mixture consisted of 50 mM acetate buffer pH 4.2, 0.2 mM Triton X, 0.5 mM o-PDA, 0 50 mM oxalate and 2 to 10 M

PAGE 42

30 enzyme. All measurements were performed in triplicate. Simultaneously, coupled part of the assay was performed on the mixtures containing from 0.1 to 8 mM formate. 13 Quenching reaction at different time points from 3 to 60 minutes supplied information about enzymes ability of linear formate production in time. Michaelis-Menten Kinetics Steady-state kinetic studies were performed towards the synthesis of formate. The concentration of substrate, oxalate, was varied for each protein (wild type and mutants) to cover the entire area of the Michaelis Menten curves obtained (up to enzymes saturation). The assay mixtures differ from those in the activity assay only by the varied oxalate concentrations. The reactions were started and quenched as in the activity assay, time of reaction depended on the activity of the enzyme. The standard curve was obtained as described before. The results were plotted using KaleidaGraph. Values of K M V max and k cat were obtained for all analyzed proteins. The pH Dependence The pH dependence of the steady state kinetic parameters for mutated OxDc-catalyzed oxalate degradation was investigated. 43 The pH range tested was based on the wild type experiments, which established lack of enzyme stability below pH 2.8 and a 90 % activity decrease between pH 4.2 and 5.7. Buffers were used: for pH 4.2 acetate and MES for pH 5.7. Oxalate concentrations were varied based on previously established values of K M Protein Concentration Protein concentration was determined by Lowry assay which engages Folin phenol reagent or using the Coomasie Plus TM Protein assay Kit (Pierce) according to the manufacturers guidelines. 44 Standard curve was created by using the same reagents to

PAGE 43

31 detect known concentrations of BSA from 0 to 120 mg/mL. The absorbance of the standards and the samples was read at 600 nm for Lowry and 595 nm for Bradford assay. Inhibition Studies Mutants of oxalate decarboxylase, R270K and R92K, have been tested for activity with oxalate analogs. 45, 46 The substrates used: pyruvate, oxamic acid, malonate, maleic acid, succinic acid and glutarate were added to the Michaelis-Menten kinetics reaction mixture described above at three or more different concentrations. The data points were obtained in triplicate, and analyzed with KaleidaGraph. Heavy-Atom Kinetic Isotope Effects These experiments were performed entirely by Laurie A. Reinhardt Ph.D. and Drazenka Svedruzic in the Institute for Enzyme Research and Department of Biochemistry at the University of Wisconsin in Madison, Wisconsin in collaboration with W. Wallace Cleland Ph.D. The exact procedure has been published for the wild type OxDc and was followed again for the mutated enzyme. 13

PAGE 44

CHAPTER 3 RESULTS AND DISCUSSION Site-Directed Mutagenesis The first mutation of oxalate decarboxylase targeted the C-terminal site, which at the time was considered to be the more likely candidate for the catalytically active one. Substitution of arginine (R270) for alanine with a small uncharged side chain has, as expected, greatly decreased the activity of the enzyme. After the introduction of the lysine side chain in place of arginine, oxalate decarboxylase retained the same charge with the residue just slightly moved away from the manganese ion binding site. This is a conservative substitution that changes the properties of the enzyme to a much smaller extent. The arginine in the C-terminal site proved to be important for enzymes activity. Therefore, the second manganese binding site was tested. Arg92, which is one of the two conserved arginine residues in oxalate decarboxylase, is situated in the N-terminal site, and was mutated to lysine. The mutant was characterized to compare the influence of the mutations in both metal binding domains on the activity of oxalate decarboxylase. The reason for the decrease in activity of the mutants could be either structural or catalytic. Wild Type (WT) Oxalate Decarboxylase Native oxalate decarboxylase was expressed and purified according to the literature. 13 Wild type OxDc was used as a reference and a control in all of the assays performed on the OxDc mutants. 32

PAGE 45

33 The expression of wild type OxDc has been by far the most effective compared to all the mutated proteins. The purification process has given reproducible results and yielded highly active enzyme [Table 5]. All of the recently published mutagenesis studies on B. subtilis oxalate decarboxylase were performed on less active protein obtained via different expression and purification systems. 3, 17, 27 Table 5. Wild type OxDc characterization Specific activity [U/mg] Time of linear formate production [min] K M [mM] k cat [sec ] V max [mM/min] 34 60 10 1 63 7 1.07 0.04 Expression and Purification R270A There were a number of problems with obtaining a sufficient amount of R270A to perform assays. The expression levels were decreased compared to the wild type by about 70 %. It seemed, however, that the introduction of a large and potentially relevant change in the active site had affected the ability to effectively produce mutated OxDc. The purification protocol for wild type was not followed successfully. The level of purity was different from one prep to another [Figure 12]. The main problem seemed to be caused by the ammonium sulfate protein precipitation step combined with the Phenyl-Sepharose column chromatography. Most of the active protein was precipitating along with the impurities, and the remaining sample did not yield any enzyme with detectable activity. The enzyme was most likely binding irreversibly to the hydrophobic resin. This suggested that the mutation influenced not only the activity or substrate selectivity of the enzyme, but also its structure. The majority of the protein may have folded differently after the removal of the large positively charged amino acid side chain and when submitted to the ammonium sulfate.

PAGE 46

34 97,40066,20045,00021,00031,000MWMWild typelysatePure R270APure R270APure R270A Figure 12. Purification results of the R270A mutant of OxDc. Samples on the SDS-PAGE gel: molecular weight marker, WT, lysate, final R270A from 3 different preps. A sufficient amount of R270A was obtained after reducing the purification protocol to just two anion exchange chromatography steps, DEAEand Q-Sepharose, and then repeating the expression and purification cycle multiple times. R270K The expression and purification protocols used for wild type were applied to R270K mutant more successfully than to R270A. Expression and solubility levels varied from prep to prep but were always about 30 % below the WT levels. The purification has been performed using two or three step chromatography. The resulting proteins differed in purity, specific activity, total yield and total activity. The purification of R270K was much easier than for R270A, but the WT procedure could not be followed identically. Repeatedly in the three step chromatography purification, protein was not recovered from the last column (Q-Sepharose) due to the low yield at this stage of the process. The protein was, however, purified and it was possible to establish the extent of inactivating effect the mutation had on the enzyme. The only change introduced in the three step purification was a decrease in the concentration

PAGE 47

35 of ammonium sulfate in the protein precipitation step after DEAE-Sepharose column down to 1.4 M. The yields were higher from the two column chromatography purifications and the level of purification was high enough for all the assays, as well as kinetic isotope measurements, to be performed. R92K As for the previous mutants, the protocol for the wild type expression and purification proved to be ineffective in obtaining active and pure R92K with reproducible yields. Very similar problems occurred as for R270K. These problems included not recovering protein from the last column, whenever Phenyl-Sepharose resin was used and ammonium sulfate precipitation of a fraction of the active protein along with the impurities. Purification optimization did not provide a new reliable procedure. Therefore, for the first time, an attempt has been made to improve yield by changing the expression conditions, not just the purification. Purification optimization The wild type protocol that proved to be ineffective for the discussed above mutants was investigated to optimize the purification protocol for R92K. DEAE-Sepharose gave the same results as for the wild type in both elution time from the column and purity level [Figure 13]. It provided initial purification and yielded active protein, and therefore no changes needed to be introduced.

PAGE 48

36 97,40066,20045,00031,00021,50014,400MWMlysatewash1062030362224252729 Figure 13. R92K purification: fractions from the DEAE-Sepharose Fast Flow column. Samples on the SDS-PAGE gel: MWM, lysate, wash, fractions: 6, 10, 20, 22, 24, 25, 27, 29, 30, 36. The next step investigated in the purification was the ammonium sulfate protein precipitation. Based on the SDS-PAGE analysis and the quantitative assay for OxDc it was revealed that part of a sample of an active mutant precipitated at as low of a concentration of ammonium sulfate as 0.6 M. Up to the concentration of 1.4 M, approximately 10 to 20 % of the total sample co-precipitated with the impurities. The addition of ammonium sulfate to the final concentration of 0.6 M guaranteed an ionic strength high enough for the sample to bind to the Phenyl-Sepharose column and for the protein of interest to stay in solution. Since the Q-Sepharose resin seemed to have very low resolution on all samples, after being previously used and cleaned numerous times, the new resin was used to test its effectiveness. Q-Sepharose is an anionic exchange resin, so binding of the proteins depends on their affinity to the charged groups on the resin, in this case a quaternary amine. Elution is based solely on changing anionic interactions between bound proteins

PAGE 49

37 and the resin by changing the salt (NaCl) concentration in the elution buffer. Different proteins are released at different ionic strengths of the buffer. Surprisingly, it was established that Q-Sepharose provided purification to the sample that was not treated with ammonium sulfate at all [Figure 14]. However, even at a low level of (NH 4 ) 2 SO 4 the sample was not separated from the impurities on this resin. In the presence of ammonium sulfate ionic strength may be too high for the mutated OxDc to bind to the Phenyl-Sepharose. 97,40066,20045,00031,00021,50014,400MWMloadwash182023212624284414161930 Figure 14. R92K purification: fractions from Q-Sepharose Hi-Perfomance column. Samples on the SDS-Page gel: MWM, load, wash, fractions: 14, 16, 18, 19, 20, 21, 23, 24, 26, 28, 30, 44 Since Phenyl-Sepharose proved rather ineffective in purifying R92K, different resins with other hydrophobic ligands were tested for the mutant binding, and its release at a more narrow ionic strength range. Butyland Octyl-Sepharose did not provide separation. In fact, they bound R92K irreversibly, and neither varied gradient elution, nor extended washing with low ionic strength buffer, resulted in recovering any active protein. Therefore, it has been established that Phenyl-Sepharose is the most effective hydrophobic resin, but the purification yields more protein of sufficient purity in the purification that utilizes just two anion exchange resins DEAE and Q-Sepharose.

PAGE 50

38 Weak cation exchange CM (carboxymethyl) Sepharose resin has been previously used for the purification of oxalate oxidase 17 Strong cationic SP (sulphopropyl) Sepharose resin was tested in similar conditions (acetate buffer, pH 5.2) for the ability of purifying mutated OxDc R92K. The enzyme of interest did not bind to the resin and was eluted in low salt concentrations. The impurities of lower molecular weight did bind, and therefore, the cation exchange resin increased purity of the enzyme. However, the column purification would require further optimization. Expression optimization The best conditions for mutated oxalate decarboxylase to be produced by E. coli have been established through changing many variables in the expression process compared to the wild type procedure. The first innovation was decreasing the volumes of bacterial cultures per flask volume to improve aeration. Also, instead of Erlenmeyer, baffled flasks were used. 47 The time of bacterial growth as well as the temperature were altered. Likewise, the heat shock and sonication times were adjusted. After introduction of all these changes extraction of pellets yielded soluble protein for the first time for an arginine mutant. The younger bacteria were more potent and produced enzyme at a slower rate with higher influence of chaperonines on the correct protein folding. Results Both lysate and extract contained more enzyme of interest [Figure 15] than before, and further studies proved that changes in bacterial growth conditions resulted in production of more active protein. The total yield from 1 L of culture was approximately 2 mg with specific activity of 0.3 U/mg, whenever the wild type expression protocol was followed. After introducing the new procedure the yield was 14 mg and the specific activity was 3.5 U/mg.

PAGE 51

39 97,40066,20045,00031,00021,50014,400MWMlysateextractlysateextract Figure 15. Expression results for R92K. SDS-PAGE gel fractions: marker, lysate, extract, lysate extract. The two column purification of the sample from the changed expression system yielded highly purified protein. DEAE Sepharose resin provided initial purification, and high amounts of protein allowed the following experiments to proceed with only the most active and pure fractions [Figure 16A]. A97,40066,20045,00031,00021,50014,400MWMWild type221232526272829303236 B97,40066,20045,00031,00021,50014,400MWMloadwash415202123242526272830 Figure 16. R92K purification: (A) fractions from DEAE-Sepharose: marker, WT, 2, 21, 23, 25, 26, 27, 28, 29, 30, 32, 36; (B) fractions form Q-Sepharose: marker, load, wash, fractions: 4, 15, 20, 21, 23, 24, 25, 26, 27, 28, 30. The Q-Sepharose used directly after DEAE-Sepharose provided further purification [Figure 16B] but did not grant homogeneity. The pool of active protein was concentrated to half the volume in the pressure chamber. Approximately one third of the protein precipitated. Some of the precipitate was active but, as expected, most of it turned out to

PAGE 52

40 be inactive, and the specific activity of the soluble sample increased [Table 6]. Values of the total activity of lysate and extract in the purification table are not exact, as the activity assay based on NAD + consumption may be compromised by many impurities still present at this point in the sample. Hence, values of the overall activity of the sample are more reliable starting after the first step of chromatographic purification. Table 6. Purification table for R92K mutant of OxDC. Step Concentration [mg/mL] Volume [mL] Total protein [mg] Total activity [U] Specific activity [U/mg] Lysate 0.6 150 90 43 0.5 Extract 0.6 100 60 19 0.3 Q load 1.4 50 70 120 1.7 Q pool 0.6 40 26 57 2.2 Final 0.8 18 14.5 50 3.5 Steady-State Kinetics R270A The average purification yielded approximately 20 to 30 mg of R270A, with total activities of 1 to 3 U from 2 L of bacterial culture. The obtained enzyme varied in specific activity from 0.1 to 1.4 U/mg. Kinetic parameters were obtained for each preparation cycle. The results shown in [Table 7] represent one of the preps. Table 7. Characterization of the R270A mutant of oxalate decarboxylase. Specific activity [U/mg] Time of linear formate production [min] K M [mM] k cat [sec ] V max [mM/min] 0.4 60 1.8 0.1 0.08 0.01 0.0124 0.0005 The mutation of arginine residues to alanine was predicted to inactivate the enzyme, 6 while the insufficient purification has lowered the specific activity of the enzyme even further. The structure seemed to have been influenced by the mutation as well. Therefore, kinetic isotope effects could not be measured, and obtaining reproducible

PAGE 53

41 and reliable assay results proved impossible. It has been established that the mutation of Arg270 from the C-terminal end of oxalate decarboxylase to alanine affects both the structure and activity of the enzyme in a dramatic manner, causing about a 100-fold decrease in activity. 00.0050.010.0150102030405060V[mM/min] [oxalate]mM Figure 17. Michaelis-Menten kinetics of mutated oxalate decarboxylase: R270A. Enzyme activity was determined in triplicate, the curve was fit using the Michaelis-Menten equation. The time of linear formate production by R270A was established to be 60 min, which allowed the wide choice of time and enzyme concentrations to perform steady-state kinetics experiments [Figure 17,18]. The rate of formate production is 100 fold slower and the value of k cat decreased significantly in comparison to the wild type enzyme catalyzed reaction. This dramatic drop in the enzymes ability to catalyze the decarboxylation of oxalate may be caused by the arginines role in catalysis. It is also consistent with the assumption made by Just et al that the C-terminal metal binding site influences the activity of the enzyme by structurally supporting the N-terminal site. 6

PAGE 54

42 Removing a charged residue would have changed the interaction between the sites and disrupted the enzymes catalytic ability. 0 0.1 0.2 0.3 0.4 0.5 0.6 0 10203040506070[formate] mMt [min] Figure 18. Catalysis of formate production by R270A mutant of OxDc as a function of time 32 Error bars come from the fit in KaleidaGraph. R270K Activity The conservative mutation of Arg270 to lysine resulted in a 10-fold activity decrease compared to the wild type OxDc [Table 8]. The formate production is linear for 60 min [Figure 19B]. The mutation in the presumed active site has affected the enzymes activity by influencing either the catalytic mechanism or protein folding and metal binding. Table 8. Kinetic characterization of the R270Kmutant of OxDc. Specific activity [U/mg] Time of linear formate production [min] K M [mM] k cat [sec ] V max [mM/min] 4.9 60 1.7 0.2 7.8 0.8 0.045 0.001

PAGE 55

43 Kinetic parameters The decreased K M value [Figure 19A, Table 8] suggests that incorrect protein folding has not disturbed the binding site. The decrease of K M might mean that the mutant binds oxalate even more tightly than the wild type. The k cat value has decreased for R270K compared to WT but was 100 times higher than that for R270A. The mutation did not kill the enzyme activity. The catalysis was possible, but slower. The decrease in K M and k cat values and no change in k cat /K M suggests unproductive binding. The specificity constant (k cat /K M ) is unaffected by mutation, as k cat and K M altered in a compensating manner. The turnover number decreases when only a fraction of substrate is bound productively and also tighter binding leads to a decreased value of K M If the C-terminal manganese binding site had catalytic ability, then moving the positive charge further away from the bound substrate would have diminished this ability, as seen in Table 16. If the role of the C-terminal domain were merely structural, the results [Table 16] confirmed this suggestion as well, as the disruption in the structure and charge of the mutant is incomparably smaller than for the R270A. The catalysis performed dominantly by the N-terminal site would not be affected to a very high degree by a conservative mutation in a C-terminal domain, as observed for R270K. The values of kinetic parameters k cat /K M and K M for R270K mutant of OxDc are comparable with those published by Just et al [Table3] 6 specific activity is, however, 10-fold higher.

PAGE 56

44 0 0.01 0.02 0.03 0.04 0.05 0 10 20 30 405060V [mM/min] [oxalate] mM A. 01234560102030 40 50 6070[formate] mM time [min] B. Figure 19. Kinetic characterization of R270K mutant of OxDc. (A.) Michaelis-Menten kinetics of mutated oxalate decarboxylase: R270K. All data points were determined in triplicate, the curve was fitted using the Michaelis-Menten equation. (B.) Catalysis of formate production by R270Kmutant of OxDc as a function of time. Error bars come from the fit in KaleidaGraph. Inhibition studies The change in the turnover number and our prediction that both conserved arginines may play a role in substrate selectivity lead to testing several compounds, all of which contained one or two carboxylic groups, as alternative substrates for R270K. These compounds might bind in the putative active site and inhibit the enzyme. The enzymes binding selectivity is not highly affected by the mutation [Table 9]. The wild type has been shown to be specific towards oxalate. 13 The only potential substrate that seemed to slightly inhibit the R270K mutant was malonate, which has been previously observed to inhibit the wild type enzyme as well [Figure 20]. The Lineweaver-Burk plot for increasing concentrations of malonate shows [Figure 21] that the lines intercept on the y-axis. The K M increased for increasing concentrations of malonate. V max stayed the same while V max /K M values decreased, all of which suggest competitive inhibition [Table 9].

PAGE 57

45 Table 9. Inhibition studies of R270K mutant of oxalate decarboxylase Inhibitor Inhibitor concentration [mM] K M [mM] V max [mM/min] (V max /K M ) app K I [mM] Pyruvate 0 50 100 1.7 0.3 3.1 0.8 1.8 0.6 0.0049 0.001 0.099 0.006 0.101 0.006 0.030 0.001 0.03 0.01 0.06 0.01 No inhibition Oxamic acid 0 25 50 1.9 0.5 1.7 0.2 1.5 0.3 0.021 0.001 0.0197 0.0006 0.026 0.001 0.011 0.003 0.012 0.001 0.017 0.003 No inhibition Malonate 0 5 10 15 25 35 45 50 4.3 0.3 6.7 0.5 8.8 0.9 10.7 0.8 18.3 4.2 16 2 28 12 20 7 0.073 0.001 0.069 0.001 0.068 0.002 0.065 0.001 0.068 0.005 0.058 0.003 0.07 0.01 0.053 0.007 0.018 0.001 0.010 0.001 0.007 0.001 0.006 0.001 0.004 0.001 0.004 0.001 0.0025 0.001 0.0026 0.001 --8.9 9.5 10.1 7.7 12.8 8.3 13.7 Maleic acid 0 25 50 6.6 0.5 6 1 4.9 0.9 0.052 0.001 0.039 0.002 0.031 0.002 0.008 0.001 0.0065 0.0009 0.006 0.001 No inhibition Glutarate 0 25 50 3.6 0.9 3.1 0.3 3.0 0.3 0.014 0.001 0.0137 0.0004 0.0124 0.0004 0.004 0.001 0.0040 0.0004 0.0040 0.0004 No inhibition The equation used to calculate K I for malonate was chosen based on the graphical representation of data as well as on changes in the following kinetic parameters: K I = (K M [I]) / (K Mapp -K M ); where K I inhibition constant, K M Michaelis-Menten constant in the absence of the inhibitor, K Mapp Michaelis-Menten constant in the presence of different concentrations of the inhibitor, and [I] concentrations of malonate. 30 The mean value of K I is 10.1 mM. Replotting data (1/K Mapp vs 1/[I]) provided a K I value of 12.5 mM, in agreement with the calculated value, and confirmed that R270K is inhibited competitively by malonate.

PAGE 58

46 00.010.020.030.040.050.060.070102030405060 V mM/min/mg (0 mM malonate) V (5 mM malonate) V (10 mM malonate) V (15 mM malonate) V (25 mM malonate) V (35 mM malonate) V (45 mM malonate) V (50 mM malonate)V [mM/min][oxalate] mM Figure 20. R270K inhibition by malonate. Michaelis-Menten kinetics in the presence of varied concentrations of malonate. Legend is shown above the graph. 05010015020000.10.20.30.40.50.6 1/V min/mM (0 mM malonate) 1/V (5 mM malonate) 1/V (10 mM malonate) 1/V (25 mM malonate) 1/V (15 mM malonate) 1/V (35 mM malonate) 1/V (45 mM malonate) 1/V (50 mM malonate)1/V [min/mM] 1/[oxalate] [1/mM] Figure 21. R270K inhibition by malonate Lineweaver Burk plot. Intercept on the y-axis suggests competitive inhibition. Legend is shown above the graph.

PAGE 59

47 Yet again, the two possible theories were considered. The catalysis of oxalate decarboxylation in the C-terminus could have been slowed down by a smaller fraction of substrate being able to bind in the active site, which was partially occupied by the inhibitor. Nonetheless, the inhibitors binding could have disrupted the structural interaction between the two domains in the monomer and, therefore, catalysis in the N-terminus was diminished. The K I value for competitive binding of malonate by the wild type was observed in our laboratory to be 6 mM. Increase of the K I value for the mutant suggests stronger inhibition of this protein compared to the wild type, and possibly even confirms its role in the substrate selectivity. The pH dependence The activity of the R270K mutant of oxalate decarboxylase decreased 50 % at the pH of 5.7. Both K M and k cat values decreased about 5 fold, therefore the V/K value remained unchanged [Table 10]. A single protonated oxalate was shown to be the correct substrate for the oxalate decarboxylase. 13 The results for R270K may mean that even though oxalate is binding 50 % of it do so in an unproductive manner. This may be caused by the wrong protonation state of the oxalate at higher pH. Table 10. Kinetic parameters for R270K mutant of OxDc at pH 4.2 and 5.7 pH K M [mM] V max [mM/min] V max /K M [min -1 ] 4.2 1.7 0.2 0.108 0.009 0.06 0.01 5.7 0.30 0.07 0.018 0.001 0.06 0.01

PAGE 60

48 R92K Activity As in the C-terminal site, the activity of the conservative mutant of OxDc (R92K) was expected to change. Activity was about 10-fold decreased compared to the wild type [Table 16], almost exactly as for the R270K, the previously characterized conservative mutant [Table 11]. Data presented by Just et al show a 100 fold activity decrease. 6 Steady state kinetics measurements were possible during the time of linear product formation by a stable enzyme, which for R92K was 70 min [Figure 22]. The equal decrease in activity of lysine mutants for both Nand C-terminal active sites suggests that they are both equally important to catalysis, but does not assign them specific roles in the catalytic mechanism. Table 11. Kinetic characterization of the R92K mutant of oxalate decarboxylase. Specific activity [U/mg] Time of linear formate production [min] K M [mM] k cat [sec ] V max [mM/min] 3.5 70 11 1 0.40 0.05 0.030 0.001 Kinetic parameters The K M value was not altered for R92K, which suggested that binding of the substrate was unaffected [Figure 24]. However, a large (100 fold) change in the turnover number implies catalytic importance of this residue. The specificity constant, K M /k cat was decreased for R92K 100-fold, which is as much as for the R270A mutant. The mutation of arginine to lysine introduced the smallest possible structural change. Its dramatic impact on the enzymes specificity constant (k cat /K M ) as well as the turnover number suggested that the prediction of Just et al that the N-terminal metal binding site is the sole or dominant catalytically active domain may have been correct [Table 11]. 6 Therefore, to

PAGE 61

49 further investigate this issue, inhibition studies as well as kinetic isotope effect measurements were performed. 0 0.2 0.4 0.6 0.8 1 1.2 010203040506070 80 time [min] [formate produced] mM Figure 22. Catalysis of formate production by R92K mutant of oxalate decarboxylase as a function of time. Errors bars come from the fit in KaleidaGraph. 0 0.005 0.01 0.015 0.02 0.025 0 1020304050 [oxalate] mM V [mM/min] Figure 23. Michaelis-Menten kinetics of mutated oxalate decarboxylase: R92K. All data points were determined in triplicate, the curve was fitted using the Michaelis-Menten equation in KaleidaGraph.

PAGE 62

50 Inhibition studies Following the studies for R270K, the N-terminal site mutant was examined for inhibition with carboxylic compounds comparable to oxalate in size. Even though both wild type and R270K were inhibited by malonate, R92K was not affected by any of the tested compounds [Table 12]. Interestingly malonate cannot compete or even bind to the R92K mutant. No disruption in formate production by the presence of malonate could confirm the assumption of Just et al, that the N-terminal site is more stable and that it is involved in the catalysis and not substrate recognition. They predicted that the newly discovered lid, which prevents solvent access to the Mn ion in the N-terminus makes the domain generally more stable and resistant. 6 It is possible that substitution of lysine in this manganese ion binding site did not create a big enough gap for the enzyme to bind other substrates and be inhibited by them. Table 12. Inhibition studies of R92K mutant of OxDc. Inhibitor Inhibitor concentration [mM] K M [mM] V max [mM/min] (V max /K M ) app K I [mM] Pyruvate 0 50 100 4.4 0.5 3 1 1.5 0.5 0.019 0.0001 0.013 0.002 0.009 0.005 0.0043 0.0004 0.004 0.001 0.006 0.003 no inhibition Malonate 0 25 50 15 1 26 1 25 2 0.045 0.002 0.034 0.001 0.026 0.001 0.0030 0.0002 0.0013 0.0001 0.0010 0.0001 no inhibition Maleic acid 0 25 50 10 1 10 1 13 1 0.031 0.001 0.023 0.001 0.019 0.001 0.0031 0.0003 0.0023 0.0003 0.0015 0.0001 no inhibition Succinate 0 25 50 13 3 15 1 18 1 0.032 0.003 0.033 0.001 0.039 0.002 0.0025 0.0006 0.0022 0.0002 0.0022 0.0002 no inhibition Glutarate 0 25 50 13 1 22 5 25 8 0.034 0.001 0.046 0.006 0.06 0.01 0.0030 0.0002 0.0021 0.0005 0.0024 0.0001 no inhibition

PAGE 63

51 The pH dependence The activity was measured at a higher pH of 5.7 to ensure that, as proved experimentally for the wild type oxalate decarboxylase, the substrate for R92K is a half protonated oxalate. The activity at pH 5.7 is 10 % of the R92K activity at the optimal pH of 4.2 [Figure 26]. Also, the values of K M and V max decreased 100-fold and 10-fold respectively, compared to the values at pH 4.2 [Table 13]. The activity was lower because the substrate was not in a preferred protonation state anymore and an additional step was necessary for the reaction to occur. Table 13. Kinetic parameters for R92K mutant of OxDc at pH 4.2 and 5.7 pH K M [mM] V max [mM/min] V max /K M [min -1 ] 4.2 11.5 0.7 0.030 0.001 0.0026 0.0002 5.7 0.9 0.1 0.0039 0.0001 0.0043 0.0004 Heavy-Atom Kinetic Isotope Effects R270K The measurements of the kinetic isotope effects for R270K have been based on the experiments described previously for the wild type. The same model was used in the analysis of data. 13 The KIE measurements are invaluable in studying the role of amino acid residues in the putative active sites. 48 They provide information about the steps in the mechanism of the catalyzed reaction, in addition to the transition state. The substitution of Arg to Lys is conservative, the charge remains positive and the obtained values have not changed drastically compared to WT, therefore the same kinetic model could be used for the data analysis for the R270K mutant as for the wild type. If R270 indeed plays catalytic role by polarizing the C-O bond and facilitating decarboxylation, then the observed isotope effects should be higher than for the wild type. Such an increase in the KIE values was

PAGE 64

52 observed. The change was statistically relevant but not large [Table 14]. The difference between the effects on the formate and the CO 2 production can be estimated due to the fact that isotope ratios for both products could be measured. As for the wild type, the kinetic isotope effects were larger for the formate end of oxalate. Thus, the arginine residue possibly stabilized the transition state by polarization of the C O bond in the oxalate radical anion. The isotope effects are higher for R270K than for the wild type. The change was significant for CO 2 and only slight for formate. This indicates that the bond breaking step became more rate-limiting in the reaction catalyzed by the R270K mutant of oxalate decarboxylase. The preliminary 18 O isotope effects for R270K measured at pH 4.2 show that as for the wild type there is no isotope effect for the CO 2 production. Table 14. 13 C and 18 O kinetic isotope effects for R270K mutant of OxDc. pH 13 (V/K), % CO 2 13 (V/K), % formate 18 (V/K), % CO 2 18 (V/K), % formate 4.2 0.8 0.1 1.6 0.1 -0.04 0.2 0.4 0.2 5.7 1.3 0.1 2.1 0.1 0.3 0.2 0.8 0.2 R92K Preliminary results of kinetic isotope measurements were obtained for pH 4.2 and can be considered in a general analysis [Table 15]. The conservative mutation of Arg to also positively charged Lys implied the use of the same kinetic model as for the wild type and the C-terminal R270K mutant, to analyze the data. The 13 C isotope effects for R92K mutant are practically the same as for the R270K mutant higher than for the wild type. The change in the KIE values compared to those for the wt is not large but statistically significant. Higher KIE values show that the chemistry is slower and more rate-limiting after introducing the mutation, but still

PAGE 65

53 possible. The results for carbon dioxide are smaller than for the standard carbon bond cleavage, which confirms that breaking of the C-C bond in oxalate is not rate limiting. The changes are bigger for the formate end of oxalate rather than the CO 2 one. Yet again 18 O IE for CO 2 was not present, however the effect for formate increased significantly compared to the wild type. This confirms the prediction that the Arg92 residue in the N-terminal site is responsible for the polarization of the C-O bond of the oxalate radical anion and thus facilitates decarboxylation. The fact that for both Arg mutants the same 13 C isotope effects were observed proves that both Nand C-terminal sites are active during catalysis. Table 15. 13 C and 18 O kinetic isotope effects for R92K mutant of OxDc. pH 13 (V/K), % CO 2 13 (V/K), % formate 18 (V/K), % CO 2 18 (V/K), % formate 4.2 0.8 0.1 1.3 0.3 0.3 0.3 1.9 0.3

PAGE 66

CHAPTER 4 CONCLUSIONS Expression and Purification of the Wild Type and Mutated OxDc Recombinant B. subtilis oxalate decarboxylase was expressed in E. coli with success. The purifications yielded the most active mutated proteins as described in the literature so far. 3, 6 This is most likely due to the lack of any His-tags in the protein. Both Nand C-terminal His-tags were used in previous mutagenesis studies of OxDc and the proteins were either unstable and precipitated or had lower activities than the ones obtained in this work. The mutated oxalate decarboxylase proved to be hard to both express and purify according to the wild type enzyme protocol. The increase in yield of active enzyme was achieved by changing several expression conditions. Decreasing the temperature during cell growth might have affected higher expression of the manganese transporting proteins 49-51 and therefore higher activity of OxDc may have been due to the increase of manganese occupation per domain of OxDc. Decreasing the temperature also resulted in slowing down the bacterial growth as well as the protein production. Induction of E. coli cells in less dense cultures provided more potential to produce the enzyme. Increasing aeration of bacterial cultures helped decrease fermentation and contributed to improving manganese transport, but the major effect was on the ATP production. Oxygen is needed for the biosynthesis of ATP used by the chaperonines, accessory proteins in the correct protein folding process. Activation of chaperonines by heatshock was much more effective when sufficient levels of ATP were present. 30 Therefore, higher amounts of 54

PAGE 67

55 active, correctly folded mutated OxDc may have been obtained. It seemed that the bacteria, even though capable of producing wild type OxDc, had to be put in the optimal conditions to yield any active enzyme with important mutations introduced. Similar difficulties, in comparison to the wild type procedure, were encountered during the purification of mutated oxalate decarboxylase. The major problem was caused by the ammonium sulfate precipitation step combined with the Phenyl-Sepharose column. The latter resin separates proteins based on their hydrophobicity. Its failure to provide purification to the R270A mutant of oxalate decarboxylase, as well as binding it more strongly than the wild type, lead to the conclusion that the mutation of Arg to Ala may have decreased the levels of correctly folded enzyme. The changes in the expression conditions helping the correct protein folding have been described above. The purification of the samples from the improved expression system, using just anion exchange chromatography, proved sufficient for the planned characterization experiments. SteadyState Kinetics of the Wild Type and Mutated OxDc Kinetic characterization of mutated oxalate decarboxylase was performed in this study and confirmed the hypothesis that both manganese ion binding sites are important for enzymes activity and catalysis [Table 13]. Just et al suggested that the N-terminal site was dominant in the catalysis while the C-terminal one supported it in a structural way. This assumption was based solely on the steady-state kinetics experiments on the mutated OxDc. 6 However, the results of this work, which include the kinetic isotope effects measurements, show the significance of both metal binding sites to catalysis. Binding of the substrate was slightly influenced by mutations in the C-terminal site. Both alanine and lysine mutants of Arg270 had 10-fold decreased K M s. The binding

PAGE 68

56 constant (K M ) for the N-terminal mutant, R92K was unaltered compared to the wild type. This confirmed that the role of the C-terminus may be more structural, however did not provide a proof excluding it from catalysis [Table 16]. The significance of steady-state kinetics results is not as high as that of the kinetic isotope effects. The basic kinetic parameters were used for a general analysis and comparison of the properties of the wild type and mutated enzyme. A decrease in enzymes turnover number proved the important catalytic role of the conserved arginine residue in the N-terminal active site. The value of the k cat /K M of R92K decreased as much as for the R270A. That means that the conservative mutation in this catalytically active site had a dramatic impact on enzyme properties. Unproductive binding of oxalate was observed for R270K, as the catalytic turnover and binding constant decreased but the specificity constant remained unaffected by the mutation. The decrease in activity at pH 5.7 compared to pH 4.2 that was similar to WT, showed that the enzyme most likely used substrate in the same protonated form with and without the mutations introduced. Several acids and diacids, similar in size to oxalate, were tested as inhibitors of oxalate degradation by R270K and R92K. The selectivity of the N-terminal mutant remained high for oxalate. R270K, however, was shown to be able to bind malonate, which, as in the wild type, competed for the enzymes active site with oxalate. Binding of malonate in the C-terminal manganese binding site slowed down catalysis by disrupting the interactions between the two domains of the enzyme if the hypothesis of Just et al were correct. 6 The conservative change may not be sufficient, however, to see the real

PAGE 69

57 effects of removing arginine residues. Further experiments should be conducted on the OxDc with more significant substitutions in place of the Arg270 and Arg92. As shown by Just et al, substitution of arginine with alanine had a very dramatic effect on the enzymes activity, which deteriorated to a very low level in R270A and was extremely hard to detect. 6 The activity of the R92A mutant was expected to be even lower if N-terminal were the dominant catalytic site 6 which would make it completely impossible to characterize the protein. Therefore R92A was not prepared. Table 16. Comparison of kinetic characteristic of wild type OxDc and its active site mutants. Specific activity [U/mg] K M [mM] V max [mM/min] k cat [sec -1 ] k cat /K M Wild Type 34 10 1 1.07 0.04 63 7 6.3 0.8 R270A 0.4 1.8 0.1 0.0121 0.0009 0.08 0.01 0.040 0.005 R270K 4.9 1.7 0.2 0.045 0.001 7.8 0.8 4.5 0.7 R92K 3.5 11 1 0.030 0.001 0.40 0.05 0.040 0.001 Heavy-Atom Kinetic Isotope Effects Kinetic isotope effect measurements were used to confirm the importance of Arg270 and Arg92 in the catalytic mechanism of oxalate decarboxylase. Since steady-state kinetics cannot prove the mechanism, just rule out some alternatives, heavy-atom isotope effects were measured to investigate the influence the arginine residues have on the catalytic ability of oxalate decarboxylase [Table 17]. In both investigated mutants of oxalate decarboxylase substitution of Arg to Lys was introduced. The charge remained conserved and the same kinetic model was used for the data analysis as for the wild type OxDc. The kinetic isotope effects were observed for both mutants, which means that both Arg 92 and 270 are important to catalysis. The changes were not large, which is consistent with the conservative mutation, but they were

PAGE 70

58 significant. What is more, the 13 V/K numbers were the same for both mutants, which suggests that their participation in the catalytic mechanism is equal. The observed 13 C KIEs for CO 2 were lower than for the typical carbon bond cleavage for both mutants, which confirmed that the C-C bond cleavage in oxalate is not rate-limiting. The numbers were higher for the formate end of oxalate, which confirmed arginine residues role in facilitating decarboxylation by polarizing the C-O bond in the oxalate radical anion. Table 17. Summary of 13 C and 18 O kinetic isotope effects for the wild type and mutated OxDc pH 13 (V/K), % CO 2 13 (V/K), % formate 18 (V/K), % CO 2 18 (V/K), % formate 4.2 0.5 0.1 1.5 0.1 -0.2 0.2 1.1 0.2 Wild type 5.7 0.8 0.1 1.9 0.1 -0.7 0.1 1.0 0.1 4.2 0.8 0.1 1.6 0.1 -0.04 0.2 0.4 0.2 R270K 5.7 1.3 0.1 2.1 0.1 0.3 0.2 0.8 0.2 R92K 4.2 0.8 0.1 1.3 0.3 0.3 0.3 1.9 0.3 Active Site Identity and Mechanism of Catalysis of OxDc The novel, nonoxidative decarboxylation without any organic cofactors in the presence of oxygen is very unusual. Determination of the oxalate decarboxylases crystal structures has provided details necessary to elucidate the enzymes mechanism of catalysis. The first crystal structure predicted the C-terminal domain to be the catalytically active one based on the presence of the general base only in this site. 3 The second crystal structure showed a slight change in the conformation of one of the helices, that put a glutamate residue in the vicinity of the bound substrate. These studies have also presented that the N-terminal active site is more available to the substrate and more stable making it the dominant domain in the catalysis. 6 The significance of the conserved arginine residues Arg92 and Arg270 was confirmed in this work. Based on steady state kinetic experiments it was confirmed that

PAGE 71

59 the N-terminal site is important during catalysis. Even the most conservative mutation of arginine in the N-terminal domain lead to a dramatic decrease in enzyme turnover, prompting the conclusion that it is the dominant, if not the only catalytically active site. The C-terminal domain influences catalysis as well, possibly as proposed by Just et al, in the structural way.6 The Arg residue found in the C-terminus was eventually confirmed to play a crucial role in the catalysis when very similar values of heavy atom isotope effects were observed for both R92K and R270K mutant [Table 17]. Removing the positive charge of the arginine inactivated the enzyme, which is consistent with its assigned role of polarizing the intermediate. The Future of the Project There are significant scientific and medical reasons for structural and mechanistic characterization of oxalate decarboxylase. Chemical insights concerning radical-mediated enzyme catalysis should be pursued further through investigation of the identity of the radical species formed during steady-state turnover of OxDc. 3 The metal dependence of the enzyme has been previously shown for the wild type oxalate decarboxylase. 27 Metal occupancy of the active sites has not been established for the mutants. It could provide additional information on the cause of the arginine mutants lack of or decrease in activity. Evaluation of the effect of the protein environment in controlling the chemical properties of the metal centers in bacterial OxDc should be performed by further site-directed mutagenesis studies. The mutations of conserved glutamate residues from both Cand Nterminal active sites should provide further proof of the relevance of each

PAGE 72

60 manganese binding site to the catalysis. The characterization should include steady-state kinetics and kinetic isotope effect measurements as for the arginine mutants. This work has laid foundation for further characterization of recombinant B.subtilis OxDc. The projects outlined above will be continued in the Richards laboratories.

PAGE 73

LIST OF REFERENCES 1. Hodgkinson, A. Oxalic acid in Biology and Medicine; Academic Press: London, 1977. 2. Graustein, W. C., Cromack, K., and Sollins, P. (1997) Science 198, 1252-1254. 3. Anand, R., Dorrestein, P. C., Kinsland, C., Begley, T. P., and Ealick, S. E. (2002) Biochemistry 41, 7659-7669. 4. Mehta, A., and Datta, A. (1991) J. Biol. Chem. 266, 23548-23553. 5. Dunwell, J. M., Khuri, S., and Gane, P. J. (2000) Microbiol. Mol. Biol. Rev. (200) 64, 153-179. 6. Just, V. J., Stevenson, C. E. M., Bowater, L., Tanner, A. Lawson, D. M., and Bornemann, S. (2004) J. Biol. Chem. 279, 19867-19874. 7. Goodell, B., Jellison, J., Liu, J., Paszczynski, A., Fekete, F., Krishnamurthy, S., Jun, L., and Xu, G. (1997) J. Biotechnol. 53, 1333-162. 8. Shimada, M., Akamtsu, Y., Tokimatsu, T., Mii, K., and Hattori, T. (1997) J. Biotechnol. 53, 103-113. 9. Sidhu H., Hoppe, B., Hesse, A., Tenbrock, K., Bromme, S., Rietschel E., and Peck, A.B. (1998) Lancet 352, 1026-1029. 10. Williams, H. E., and Wandzilak, T. R. (1989) J. Urol. 141, 742-747. 11. Strasser, H., Burgstaller, W., and Schinner, F. (1994) FEMS Microbiol. Lett. 119, 365-370. 12. Wei, Y., Zhang, Z., Andersen, C. H., Schmelzer, E., Gregersen, P. L., Collinge, D. B., Smedegaard-Petersen, V., and Thordahl-Christensen, H. (1998) Plant Mol. Biol. 36, 101-112. 13. Reinhardt, L., A., Svedruzic, D., Chang, C. H., Cleland, W. W., and Richards, N. G. J. (2003) J. Am. Chem. Soc. 125, 1244-1252. 14. Hesse, A., Bongartz, D., Heynck, H., and Berg, W. (1996) Clin. Biochem. 29, 467-472. 61

PAGE 74

62 15. Honow, R., Bongartz D., and Hesse, A. (1997) Clin. Biochem. Acta 261, 131-139. 16. Chiriboga, J. (1966) Arch. Biochem. Biophys. 116, 516-523. 17. Requena, L., and Bornemann, S. (1999) J. Biochem. 343, 185-190. 18. Dunwell, J. M. (1998) Biotechnol. Genetic Eng. Rev. 15, 1-32. 19. Woo, E.-J., Dunwell, J. M., Goodenough, P. W., Marvier, A. C., and Pickersgill, R.W. (2000) Nat. Str. Biol. 7, 1036-1040. 20. Tanner, A., and Bornemann, S. (2000) J. Bacteriol. 182, 5271-5273. 21. Chakraborty, S., Chakraborty, N., Jain, D., Salunke, D. M., and Datta, A. (2002) Prot. Sci. 11, 2138-2147. 22. Babbitt, P. C., and Gertl, J. A. (1997) J. Biol. Chem. 272, 30591-30594. 23. Ricagno, S., Jonsson, S., Richards, N., and Lindqvist, Y. (2003) EMBO Journal 22, 3210-3219. 24. Emiliani, E., and Riera, B. (1968) Biochim. Biophys. Acta 167, 414-421. 25. Magro, P., Marciano, P., and Di Lenna, P. ((1988) FEMS Macrobiol. Lett. 49, 49-52. 26. Mehta, A., and Datta, A. (1991) J. Biol. Chem. 266, 23548-23553. 27. Tanner, A., Bowater, L., Rairhurst, S., and Bornemann, S. (2001) J. Biol. Chem. 276, 43627-43634. 28. Shimazono, H., and Hayaishi, O. (1957) J. Biol. Chem. 227, 151-159. 29. Emiliani, E., and Bekes, P. (1964) Arch. Biochem. Biophys. 105, 488-493. 30. Fersht, A. Structure and Mechanism in Protein Science; W. H. Freeman and Company: New York, 1999. 31. Cleland, W. W. (1987) Bioorg. Chem. 15, 283-302. 32. Cleland, W. W. (1982) Methods Enzymol. 87, 625-641. 33. Cleland, W. W. (1995) Methods Enzymol. 249, 341-367. 34. OLeary, M. H. (1989) Annu. Rev. Biochem. 58, 377-401. 35. OLeary, M. H. (1980) Methods Enzymol. 64, 83-104. 36. Dunwell, J. M., and Gane, P. J. (1998) J. Mol. Evol. 46, 147-154.

PAGE 75

63 37. Plapp, B. V. (1995) Methods Enzymol. 249, 91-119. 38. Anderson, K. S., and Johnson, K. A. (1990) Chem. Rev. 90, 1131-1149. 39. Cannon, W. R., Singleton, S. F., and Benkovic, S. J. (1996) Nat. Str. Biol. 3, 821-833. 40. Cleland, W. W. (1977) Adv. Enzymol. Relat. Areas Mol. Biol 45, 273-387. 41. Cleland W. W. (1979) Methods Enzymol. 63, 103-138. 42. QuikChange Site-Directed Mutagenesis kit manual, June 2004, www.stratagene.com. 43. Cleland, W. W. (1982) Methods Enzymol. 87, 390-405. 44. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275. 45. Schramm, V. L. (1999) Methods Enzymol. 308, 301-355. 46. Huang, C. Y. (1979) Methods Enzymol. 63, 486-500. 47. Laidler, K. J., and Peterman, B. F. (1979) Methods Enzymol. 63, 234-256. 48. Klinman, J. P. (1978) Adv. Enzymol. Relat. Areas Mol. Biol. 46, 415-494. 49. Kehres, D. G., and Maguire, M. E. (2003) FEMS Microbiol. Rev. 27, 263-290. 50. Holm, R. H., Kennepohl, P., and Solomon, E. I. (1996) Chem. Rev. 96, 2239-2314. 51. Armstrong, R. N. (2000) Biochemistry 39, 13625-13632.

PAGE 76

BIOGRAPHICAL SKETCH Ewa Wroclawska was born in Poland in 1977. She studied chemistry at the University of Adam Mickiewicz in Poznan, Poland, where she received her undergraduate degree in bioorganic chemistry in May of 2001. Ewa Wroclawska was a recipient of the European Union Erasmus Scholarship, which allowed her to conduct undergraduate research at the Aristotle University of Thessaloniki in Greece for one semester. In 2001 Ewa joined the biochemistry laboratory of Dr. Nigel Richards in the Chemistry Department of University of Florida. After graduation, she is going to pursue her career in research. 64