Examination of Mutants That Alter Oxygen Sensitivity and Co2/O2 Substrate Specificity of the Ribulose 1,5-Bisphosphate Carboxyla

Home , Enzyme catalysis

EXAMINATION OF MUTANTS THAT ALTER OXYGEN SENSITIVITY AND CO2/O2 SUBSTRATE SPECIFICITY OF THE RIBULOSE 1,5-BISPHOSPHATE CARBOXYLASE/OXYGENASE (RUBISCO) FROM ARCHAEOGLOBUS FULGIDUS

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Nathaniel E. Kreel, B.S.

*****

The Ohio State University 2008

Dissertation Committee:

Professor Dr. F. Robert Tabita, Advisor Approved by

Professor Dr. Charles E. Bell

Professor Dr. Charles L. Brooks

Professor Dr. Michael Ibba ______Advisor Ohio State Biochemistry Graduate Program

ABSTRACT

The archaeon Archaeoglobus fulgidus contains a gene (rbcL2) that encodes the enzyme ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco), the enzyme necessary for biological reduction and assimilation of CO2 to organic carbon. Based on sequence homologies and phylogenetic differences, archaeal Rubiscos represent a special class of Rubisco, termed form III, that distinguishes it from the previously characterized form I and form II enzymes. Form III Rubisco retains many features characteristic of all forms of Rubisco, yet exhibits many interesting and unique differences that might be exploited to learn more about structure-function relationships for this protein. For

example, recombinant A. fulgidus RbcL2 was shown to possess an extremely high kcat value (23 s-1) and optimal activity was reached at temperatures up to 93°C. Furthermore, this protein was unusual in that exposure or assay in the presence of O2 (in the presence of high levels of CO2) resulted in substantial loss (90%) in activity compared to assays

performed under strictly anaerobic conditions. Kinetic studies indicated that A. fulgidus

RbcL2 possessed an unusually high affinity for O2. Comparative bioinformatic analyses

of available archaeal Rubisco sequences suggested the potential importance of several

unique residues, as did further analyses within the context of available forms I/II/III

Rubisco structures. Two residues unique to archaeal Rubisco, Met-295 and Ser-363,

were of particular interest due to their proximity to known active site residues.

ii Moreover, it was shown that recombinant mutant M295D, S363I and S363V A. fulgidus

enzymes were less sensitive to oxygen compared to the wild-type protein. The unique

oxygen sensitivity of this form III archaeal Rubisco is being used as a model to provide

clues as to how Rubisco has evolved to become more stable in the presence of oxygen in

more evolutionarily advanced form I and form II proteins. In addition, the same

mutational changes were made at identical sites in Thermococcus kodakaraensis RbcL which shares a 72% amino acid sequence identity with A. fulgidus RbcL2. Although this

Rubisco enzyme from T. kodakaraensis is not as oxygen sensitive as the Rubisco from A. fulgidus, there are many similarities in oxygen sensitivity not observed in other form I and form II Rubiscos. A. fulgidus RbcL2 is able to complement growth in the double

Rubisco knockout strain SBI/II- from Rhodobacter capsulatus. For the first time, using

the A. fulgidus RbcL2 enzyme to complement growth in this knockout strain has enabled the wild-type and mutant enzymes to be studied in vivo.

iii

Dedicated to all who have traveled along in this journey with me, through the ups and downs, the good times and the not so good times, there is no way I could have made it without your love and support. No names need to be mentioned, you know who you are.

ACKNOWLEDGMENTS

I would like to thank my advisor, F. Robert Tabita, for creating a working

environment where several different avenues of research could be investigated.

I thank Dr. Stephanie Scott and Dr. Cedric Bobst for their consistent technical

assistance throughout my thesis.

I am grateful to my coworkers, both past and present, for giving me scientific and computer advice.

This research was supported by a grant from the National Institute of Health.

VITA

May 6, 1978 ...... Born – Fayetteville, NC

2000...... B.S. Biology, East Carolina University

2000-2001 ...... Biochemistry Research Technician Novartis Biotechnology

2001 – present...... Graduate Research Associate, The Ohio State University

PUBLICATIONS

Research Publications

1. Kreel, N.E. and Tabita, F.R. 2007. Substitutions at Methionine 295 of the Archaeoglobus fulgidus ribulose-1,5-bisphosphate carboxylase/oxygenase affects oxygen binding and CO2/O2 specificity. J. Biol Chem. 282:1341- 51. 2. Tabita, F.R., Satagopan, S., Hanson, T.E., Kreel, N.E., and Scott, S.S. Distinct form I, II, III and IV Rubisco proteins from the three kingdoms of life provide clues about Rubisco evolution and structure/function relationships. J. Exp. Botany. In press 2008. 3. Tabita, F.R., Hanson, T.E., Satagopan, S., Witte, B.H., and Kreel, N.E. The evolution, structure and function of Rubisco and its homolog the Rubisco-like protein. Manuscript submitted. 2008. 4. Kreel, N.E. and Tabita, F.R. Activity and carboxylation specificity factor of mutants of ribulose 1,5-bisphosphate carboxylase/oxygenase from Archaeoglobus fulgidus. Manuscript submitted. 2008.

FIELDS OF STUDY

Major Field: Biochemistry

TABLE OF CONTENTS

Abstract...... ii

Dedication...... iv

Acknowledgments...... v

Vita...... vi

List of Tables ...... ix

List of Figures...... x

List of Abbreviations ...... xiv

Chapters:

1. Introduction...... 1

2. Kinetic properties of Archaeoglobus fulgidus RbcL2; site-directed mutangenesis studies to probe the basis for oxygen sensitivity...... 21

Introduction...... 21 Materials and Methods...... 26 Results...... 32 Discussion...... 59

3. Further studies of mutant ribulose 1,5-bisphosphate carboxylase/oxygenase proteins from Archaeoglobus fulgidus and a related archaeon...... 66

Introduction...... 66 Materials and Methods...... 70 Results...... 82 Discussion...... 118

vii 4. Summary...... 134

Future Experiments...... 145

References...... 148

viii

LIST OF TABLES

Table Page

2.1 Kinetic properties of purified, recombinant wild-type and mutant M295D Rubisco from A. fulgidus RbcL2...... 51

3.1 Plasmids and bacterial strains ...... 81

3.2 Amino acid comparison in hydrophobic pocket of form I, II and III Rubiscos...... 83

3.3 Kinetic properties of purified wild-type and mutant enzymes from A. fulgidus RbcL2 ...... 87

3.4 Carboxylase activity of heat stable extracts from A. fulgidus RbcL2 wild-type and mutant enzymes under anaerobic and oxygen exposed conditions...... 94

3.5 Carboxylase activity of heat stable extracts from A. fulgidus RbcL2 wild-type and mutant enzymes under anaerobic and oxygen exposed conditions...... 99

3.6 Carboxylase activity of crude soluble extracts of A. fulgidus RbcL2 wild-type and mutant enzymes from R. capsulatus SBI/II- grown under photoheterotrophic conditions ...... 107

3.7 Kinetic properties of purified wild-type Rubisco from T. kodakaraensis RbcL ...... 116

3.8 Carboxylase activity of heat stable extracts from T. kodakaraensis RbcL wild-type and mutant enzymes under anaerobic and oxygen exposed conditions ...... 117

4.1 Carboxylase activity at 30°C of crude soluble extracts from R. rubrum CbbM wild-type and mutant enzymes...... 142

LIST OF FIGURES

Figure Page

1.1 Calvin-Benson-Bassham (CBB) reductive pentose phosphate pathway ...... 4

1.2 Phylogenetic tree of Rubisco ...... 6

1.3 Crystal structures of forms I, II and III Rubiscos ...... 8

1.4 Carboxylase and oxygenase mechanism of Rubisco ...... 11

1.5 Active site of Rubisco...... 14

1.6 Metabolism of AMP to produce RuBP in archaeal organisms...... 18

2.1 Amino acid sequence alignment of Form I, II and III ribulose 1, 5-bisphosphate carboxylase/oxygenase ...... 23

2.2 SDS-PAGE gel of A. fulgidus RbcL2 Rubisco purification ...... 33

2.3 Carboxylase activity at varying temperatures for A. fulgidus RbcL2 ...... 34

2.4 Requirement and inhibitory effects of salt for carboxylase activity of A. fulgidus RbcL2...... 35

2.5 Recovery of carboxylase activity of A. fulgidus RbcL2 wild-type upon oxygen exposure ...... 37

2.6 Reversibility of oxygen inhibition of A. fulgidus RbcL2...... 39

2.7 Double reciprocal plot of carboxylase activity of wild-type A. fulgidus RbcL2 ...... 41

x 2.8 Secondary replot used to calculate the Ki for O2 for wild-type A. fulgidus RbcL2 ...... 42

2.9 Quarternary structure prediction of A fulgidus RbcL2 ...... 45

2.10 Comparison of carboxylase activity of A. fulgidus RbcL2 wild-type and mutant enzymes upon oxygen exposure ...... 47

2.11 Percent activity retained of A. fulgidus RbcL2 wild-type and mutant Rubiscos...... 48

2.12 Recovery of carboxylase activity of A. fulgidus RbcL2 M295D upon oxygen exposure ...... 49

2.13 Double reciprocal plot of carboxylase activity of A. fulgidus RbcL2 M295D...... 51

2.14 Secondary replot used to calculate the Ki for O2 for A. fulgidus RbcL2 M295D...... 52

2.15 Retention of carboxylase activity in the presence of oxygen for A. fulgidus RbcL2 wild-type and M295D...... 52

2.16 Anion exchange chromatographic separation of Rubisco reaction products for A. fulgidus RbcL2 wild-type and M295D ...... 55

2.17 Wild-type and mutant A. fulgidus RbcL2 carboxylase activity remaining after exposure to oxygen ...... 58

2.18 Predicted side-chain interactions with Met-295 in wild-type A. fulgidus RbcL2 and the mutant M295D enzyme ...... 64

3.1 Retention of carboxylase activity in the presence of oxygen of wild-type and mutant enzymes from A. fulgidus RbcL2 ...... 86

3.2 Far UV circular dichroism of A. fulgidus RbcL2 wild-type and mutant enzymes ...... 89

3.3 Far UV circular dichroism of A. fulgidus RbcL2 wild-type under anaerobic and oxygen exposed conditions...... 90

3.4 Non-denaturing PAGE gel of A. fulgidus RbcL2 wild-type and mutant enzymes ...... 92

xi 3.5 Model structure of A. fulgidus RbcL2 depicting position and side chain interactions of Ser-363 in a hydrophobic pocket...... 98

3.6 Growth curve of R. capsulatus SBI/II- complemented with A. fulgidus RbcL2 wild-type and mutant enzymes under photoheterotrophic conditions...... 103

3.7 SDS-PAGE and Western immunoblot of crude extracts from R. capsulatus SBI/II- complemented with A. fulgidus RbcL2 grown under photoheterotrophic conditions ...... 104

3.8 Growth curve of R. capsulatus SBI/II- complemented with A. fulgidus RbcL2 wild-type and mutant enzymes under photoautotrophic conditions...... 105

3.9 SDS-PAGE and Western immunoblot of crude extracts from R. capsulatus SBI/II- complemented with A. fulgidus RbcL2 grown under photoautotrophic conditions ...... 106

3.10 Partial amino acid sequence alignment of form III A. fulgidus RbcL2 and T. kodakaraensis RbcL ...... 108

3.11 SDS-PAGE gel of T. kodakaraensis RbcL Rubisco purification...... 110

3.12 Non-denaturing PAGE gel of A. fulgidus RbcL2 and T. kodakaraensis RbcL ...... 111

3.13 Anion exchange chromatographic separation of Rubisco reaction products for T. kodakaraensis RbcL wild-type and A. fulgidus RbcL2 wild-type under varying conditions...... 114

3.14 Model structure of monomer large subunit of A. fulgidus RbcL2 highlighting amino acids of a hydrophobic pocket surrounding Ser-363...... 121

3.15 Comparison of side-chain interactions with highly conserved arginine residue in Synechococcus PCC6301 and T. kodakaraensis Rubisco enzymes ...... 129

3.16 Predicted side-chain interactions with Met-298 in wild-type T. kodakaraensis RbcL and the mutant M298D enzyme...... 131

xii 3.17 Comparson of side chain interactions of serine with glycine and threonine residues in the solved crystal structure of T. kodakaraensis RbcL with corresponding residues in the model structure of A. fulgidus RbcL2 ...... 133

4.1 Non-denaturing PAGE and Western immunoblot of crude soluble extracts from E. coli of R. rubrum CbbM wild-type and mutant enzymes ...... 142

4.2 Comparison of side-chain interactions with Ala-305 in the solved crystal structure from wild-type R. rubrum CbbM and the predicted interaction of mutant A305D...... 144

xiii

LIST OF ABBREVIATIONS

Rubisco – Ribulose 1,5-bisphosphate carboxylase/oxgyenase

RuBP – Ribulose 1,5-bisphosphate

CBB – Calvin-Benson-Bassham reductive pentose phosphate pathway

CABP – 2-Carboxyarabinitol 1,5-bisphosphate

IPTG – Isopropyl-β-thioglucanoside

PMSF - Phenylmethylsulfonyl fluoride

EDTA – Ethelyndiamine tetraacetic acid

MWCO – Molecular Weight Cut Off

BSA – Bovine Serum Albumin

PDB – Protein Data Bank

SDS-PAGE – Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis

PGA – 2-Phosphoglyceric acid

PCA – Protocatechuic acid

PCD – Protocatechuic acid 3,4-dioxygense

PG – 2-Phosphoglyceric acid

PRPP – 5-Phospho-D-ribose-1-pyrophosphate

AMP – Adenosine monophosphate

PYE – Peptone Yeast Extract

xiv

CHAPTER 1

INTRODUCTION

Several eukaryotic and prokaryotic organisms are able to obtain all essential carbon by directly assimilating and reducing CO2. Although there are five major pathways to assimilating carbon dioxide into reduced organic matter, the major mechanism by which CO2 is assimilated in nature is via the Calvin-Benson-Basham

(CBB) reductive pentose phosphate pathway (Figure 1.1). There are two unique enzymatic reactions in this pathway that allow CO2 to serve as the sole carbon source for growth; the first of these is catalyzed by ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco), the enzyme that catalyzes the actual CO2 fixation reaction. The other unique enzyme is phosphoribulokinase (PRK), which catalyzes the synthesis of the substrate or CO2-acceptor compound for Rubisco, ribulose-1,5-

bisposphate (RuBP). Rubisco is the key enzyme in the CBB pathway and is responsible

for the assimilation of exogenous carbon dioxide from the environment into organic

carbon. Rubisco is the most abundant enzyme on the planet; roughly 50-60% of plant

leaf protein is made up of this enzyme, and its presence is theorized to be one of the

reasons that human life on earth is able to exist (Kung, 1976). In addition, approximately

half of land plants’ total cellular nitrogen is dedicated to the generation of this protein;

1 this may be due to Rubisco being a very poor catalyst. The average turnover for most

Rubisco enzymes is anywhere from 3-5 sec-1, which would explain the abundance of this

protein in plants in order to maintain the amount of carbon dioxide fixed to support growth. Up to 1 x 1011 metric tons of carbon dioxide is fixed into organic carbon every year and this is largely possible because of the many plant and microbial sources of

Rubisco found in the biosphere. With ever increasing amounts of carbon dioxide released into the atmosphere, attempts at engineering a more efficient Rubisco enzyme in organisms could help to offset the buildup of greenhouse gases in our environment. In

addition, a more efficient Rubisco could lead to plants that are able to grow in harsher

environments, thus increasing their yield and energy production.

Figure 1.1. The Calvin-Benson-Bassham (CBB) reductive pentose phosphate pathway. Enzymes involved in the pathway are: Rubisco, ribulose 1,5-bisphosphate carboxylase/oxygenase; PRK, phosphoribulokinase; G3PDH, glyceraldehyde 3- phosphate dehydrogenase; PGK, 3-phosphoglycerate kinase; F1,6BPase, fructose 1,6- bisphosphatase; Aldolase; Triose Phosphate Isomerase; Transaldolase; Transketolase; R5P Isomerase, ribose 5-phosphate isomerase. Substrates in this pathway are: Ru5P, ribulose 5-phosphate; RuBP, ribulose 1,5-bisphosphate; 3-PGA, 3-phosphoglycerate; 1,3- bisPGA, 1,3-bisphosphoglycerate; 3-PGAL, glyceraldehyde 3-phosphate; DHAP, dihydroxyacetone phosphate; F1,6-bisP, fructose 1,6-bisphosphate; F6P, fructose 6- phosphate; E4P, erythrose 4-phosphate; Xu5P, xylulose 5-phosphate; R5P, ribose 5- phosphate; S7P, sedoheptulose 7-phosphate; S1,7-bisP, sedoheptulose 1,7-bisphosphate.

4 Rubisco is arguably the most abundant protein on earth (Tabita, 1999; Ellis, 1979;

Kung, 1976). Many gene sequences representing Rubisco have been submitted to gene

bank databases. The protein that these genes encode plays an important role in the

photosynthetic tissue of higher forms of eukaryotic life such as terrestrial plants, but may

also be found in lower phototrophic eukaryotes such as green, red and brown algae and in

prokaryotic cyanobacteria and phototrophic eubacteria. Moreover, large numbers of

chemoautotrophic prokaryotes, which grow in the absence of light and use dark chemical

reactions to provide the energy to support growth, depend on Rubisco and the CBB CO2 fixation pathway. Rubisco molecules from these sources have been previously categorized into two groups, termed form I and form II, based on their amino acid sequence (Figure 1.2). The form I enzymes are found in virtually all eukaryotic photosynthetic organisms and also all cyanobacteria and most eubacteria that use the

CBB pathway to assimilate CO2. Form I enzymes are divided into two groups termed the

“green” group and “red” group. The green group is subdivided into subgroup A, which includes proteobacterial and some marine cyanobacterial enzymes, and subgroup B, which includes green plastids and the bulk of cyanobacterial enzymes. The red group is further divided into subgroup C, which includes various proteobacterial enzymes, and subgroup D, which includes chromophytic and rhodophytic algae enzyme (Figure 1.2)

(Tabita, 1999).

Figure 1.2. Unrooted neighbor joining tree of Rubisco large subunit sequences found in the biosphere. Currently there are 4 forms of Rubisco (form I – in red and green, form II – in purple, form III – bronze) and the RLP sequences (form IV – in cyan), forms I-III are bona fide Rubiscos capable of RuBP-dependent CO2 fixation reactions. Form IV are Rubisco-like proteins (RLP) and have some sequence and structure similarities to the other bona fide forms of Rubisco but are incapable of RuBP-dependent CO2 fixation reactions. Bootstrap values at the nodes indicate the percentage of times that particular node was present in 1000 trials. Only nodes with less than 70% bootstrap support are indicated. The scale bar represents 0.1 substitutions per site (Hanson and Tabita, 2003).

6 Form I enzymes have a complex structure made up of two different subunits, a

large subunit of 55 kDa and a small subunit of 15 kDa. In land plants and green algae,

the rbcL gene that encodes the large subunit is located and transcribed in the chloroplast while the rbcS gene that encodes the small subunit is located in the nucleus. A precursor

RbcS protein is transported from the cytoplasm and processed in the chloroplast, where the mature RbcS protein assembles with large subunits to form the holoenzyme.

Together, the large and small subunits form a sixteen subunit quarternary structure

comprised of eight large and eight small subunits, denoted as L8S8 or more specifically,

(L2)4(S4)2. Two large subunits dimerize to form the active site and four dimers of large subunits form an octamer. The small subunits reside on the top and the bottom of this octamer in a tetrameric fashion (Figure 1.3A).

Form II Rubiscos are found mainly in photosynthetic bacteria and the discovery of these enzymes and their crystal structures reveal a much simpler quaternary structure

(Tabita and McFadden, 1974a, b; Schneider et al., 1990; Tabita, 1999). Form II proteins are found in various purple photosynthetic bacteria, but also are found in some chemoautotrophic bacteria and eukaryotic marine dinoflagellates (Figure 1.2) (Tabita,

1999). These enzymes are comprised solely of large subunits of relatively the same size

as form I large subunits, about 50 kDa per monomer, in an (L2)x configuration that dimerize to form the active site and do not contain small subunits (Figure 1.3B). The large subunits of form II Rubisco proteins share 25-30% amino acid identity with the form I large subunits. Form II enzymes can exist as a simple dimer but in many cases the quaternary structure involves an oligimerization of dimers, such as the form II Rubisco

7 from Rhodopseudomonas palustris whose large subunits come together to form a hexameric quarternary structure (unpublished results).

Figure 1.3. Representative crystal structures of the three forms of bona fide Rubiscos. Form I Rubisco (A) from Spinacia oleracea (8RUC) has an (L2)4(S2)4 configuration containing both large and small subunits. Form II Rubisco (B) from Rhodospirillum rubrum (5RUB) has a much simpler quaternary structure and is of a dimer of large subunits; however other form II enzymes can be of an (L2)x configuration. Form III archaeal enzymes are also made up of only large subunits that form dimers, or in some cases, have a more complex quaternary structure of (L2)x, as seen (C) with the Rubisco from T. kodakaraensis (1GEH) which is a pentamer of dimers.

8 In the mid-1980’s, many crystal structures of Rubisco from various organisms

became available, both with and without the necessary substrates that “activate” the

enzyme as well as structures that contained a transition state analog for RuBP, 2’

carboxyarabinitol 1,5-bisphophate (CABP). Along with these solved structures,

mutagenic studies using site-directed mutagenesis targeting active site residues have

allowed researchers to dissect the mechanism by which all Rubiscos function. Firstly,

Rubisco must be “activated” by the formation of a carbamate moiety with the addition of

an exogenous carbon dioxide molecule at the ε-amino group of a highly conserved lysine

residue located within the active site pocket of the enzyme. The pKa of this residue, about

7.5, is lower than lysine residues found elsewhere, and allows for a positive charge to exist on the ε-amino group for carbamylation. Besides this, the active site has a positive electrostatic field which is altered by the carbamylation step (Lu et al., 1992). Next, a divalent metal cation is linked to carbamate formation and is stabilized by neighboring negatively charged and highly conserved, polar amino acids that will also assist in catalysis. Again, the electrostatic potential at the active site is affected and is more

positive than the unactivated enzyme and creates a favorable environment for the addition

of the RuBP substrate. The lysine residue involved in carbamate formation will thus act

as the general base in the reaction mechanism (Figure 1.4). The divalent metal cation is

typically a magnesium atom, although other divalent metals such as manganese, zinc, and

copper have been shown to be able to support activity. Even though these other metals

support activity, optimal activity is lost and some metals, such as manganese, can favor

the oxygenase activity, more than the more desired carboxylase activity. When the

enzyme is in the activated state it is ready to accept the five carbon substrate, ribulose

9 1,5-bisphosphate (RuBP), which will lead to a conformational change in the enzyme that

causes the enzyme to assume a “closed” conformation involving a flexible loop,

commonly denoted as Loop 6. Loop 6 is on a strand located between the sixth alpha-

helix and sixth beta-strand of the C-terminal portion of the enzyme and in the “closed”

conformation folds over the active site and interacts with amino acids within the active

site as well as the N-terminal portion of an associating monomeric subunit. Additionally,

the C-terminal strand of the subunit moves into position over the Loop 6 thus “locking”

the substrate in the active site (Cleland et al., 1998; Gutteridge and Gatenby, 1995; Lu et al., 1992; Shneider et al., 1990).

10 CABP (RuBP transition state analog)

Glutamic acid

Magnesium ion Aspartic acid

Carbamylated Lysine

Figure 1.4. The active site of Rubisco. The magnesium ion is shown at the center in green. The magnesium ion is held tightly by three highly conserved amino acids, a carbamylated form of lysine and two polar amino acids, aspartic acid and glutamic acid. Above the magnesium ion is the transition state analog, CABP, similar to the RuBP substrate used in the Rubisco reaction. The side of the magnesium ion opposite of the amino acids is then free to bind to both RuBP, holding onto two oxygen atoms, and the carbon dioxide molecule that will be attached to the sugar (http://www.palaeos.com/Eukarya/Lists/EuGlossary/Images/Rubisco.gif).

Stabilizing the substrate in the active site allows for the initiation of the first step of the reaction. Specifically, the first step occurs as a proton is abstracted from the C2 position of the RuBP substrate, this will cause the formation of a very unstable transition state enediol intermediate with a double bond between the second and third alpha carbons 11 of RuBP. The next step, which is also the rate-limiting step of the reaction, is an

irreversible neucleophilic attack of a carbon dioxide molecule, independent of the carbon

dioxide molecule used for carbamate formation, at the double bond between the C2 and

C3 carbons forming a six carbon intermediate that is lysed to form two identical

molecules of 3-phosphoglyceric acid (3-PGA). The carbon dioxide molecule added to

the enediol intermediate competes with an oxygen molecule for the same site. When

oxygen is used as the gaseous substrate, a five carbon intermediate is produced that splits

to form two unidentical molecules, one 3-PGA and one 2-phosphoglycolic acid (2-PG)

(Figure 1.5). The production of two 3-PGA molecules by the carboxylase reaction is

more favored since the two molecules are further utilized in the CBB cycle, resulting in

the formation of organic metabolites that will be used as basic building blocks to support

cellular growth. In addition, some of the metabolites may be recycled to form the RuBP

substrate needed for additional CO2 assimilation to occur. Unfortunately, the addition of

oxygen, to create one 3-PGA and one 2-PG, may be a metabolically wasteful process

since the creation of 2-PG is further utilized during aerobic metabolism which consumes

ATP and eventually gives results in the release of carbon in the form of carbon dioxide at

the completion of the cycle. Consequently, the oxygenase reaction results in up to 50%

less carbon assimilated under normal aerobic conditions (Cleland et al., 1998; Gutteridge

and Gatenby, 1995).

The ability of a Rubisco from a specific organism to differentiate between the two

gaseous substrates, either carbon dioxide or oxygen, is defined by the enzyme’s substrate

specificity factor (Ω). The specificity factor (Ω), is defined as υc/υo =

(KcVo/KoVc)([CO2]/[O2]), where υ is the initial velocities for the carboxylase (υc) or

12 oxygenase (υo) reactions, K is the Michaelis-Menten constant for the carboxylase (Kc) or oxygenase (Ko) reactions, and V is the maximal velocities for the carboxylase (Vc) or oxygenase (Vo) reactions. Ω describes the enzyme’s catalytic efficiency for either the

carboxylase or oxygenase reactions at any particular gaseous substrate concentration. In

addition, Ω = VcKo/VoKc, where (Vmax/Km) reflect the catalytic efficiencies for the carboxylase (Vo/Kc) and oxygenase (Vo/Ko) reactions, respectively. Thus, the higher the

specificity value, the greater the enzyme favors the carboxylase reaction. The substrate

specificity value has been determined for many Rubiscos, each having a distinct value

among the different forms of Rubisco. Form I Rubiscos have the highest values as well

as the largest range, with values that range from 20-240. Form II enzymes have much

lower Ω values and range between 10-15 (Tabita, 1999). There are very few calculated

specificity values for the form III enzymes; however, a value of 310 has been reported for

the hyperthermophile Thermococcus kodakaraensis although this value is controversial

due to the lack of experimental details reported to obtain this value (Ezaki et al., 1999).

13 2- CH2OPO3 CH2OPO32- CH2OPO32- H C COO H C COO H2 H C H C O H C O COOH CO2 H C O H C O 3-Phosphoglycerate (2) 2- CH2OPO32- CH2OPO3 2- CH2OPO32- CH2OPO3 2-Carboxy-3-keto Hydrated C O C O arabinitol-1,5-bisphosphate intermediate

H C O C O H C O H C O 2- 2- CH2OPO3 CH2OPO32- CH2OPO3 2,3-enediol(ate) Ribulose-1,5- CH2OPO32- COOH bisphosphate 2-Phosphoglycolate HO C O O- O 2 H2O + C O COO- H C OH H C OH CH2OPO32- 2- 2-Hydroperoxy-3-keto CH2OPO3 arabinitol-1,5-bisphosphate 3-Phosphoglycerate

Figure 1.5. Carboxylase and oxygenase reaction mechanism of Rubisco.

Rubiscos from aerobic organisms such as terrestrial plants are able to function in

the presence of atmospheric oxygen although the carboxylase function is forced to

compete with oxygen for the enediol intermediate at the active site. Many efforts to

engineer a more efficient Rubisco enzyme that will increase the carboxylase reaction or

decrease the oxygenase reaction or somehow alter the relative affinity for either CO2 or

O2 would be beneficial in that such a protein could potentially increase crop yields as well as decrease the amount of carbon dioxide in the atmosphere, with the potential to decrease atmospheric greenhouse gas concentrations. In order to succeed at engineering this kind of Rubisco enzyme, it is very important to understand the molecular mechanism by which the enzyme differentiates between the two competing gaseous substrates,

14 carbon dioxide or oxygen, for the enediol intermediate. In addition, it will be important to understand the factors that mitigate and influence the different kinetic properties of the enzyme. The important kinetic constants of diverse forms of Rubisco, particularly Ω and

Kc, differ considerably even among closely related and structurally similar proteins, but there is little molecular understanding as to the basis for this variation (Jordan and Ogren,

1981; Horken and Tabita, 1999). Although all Rubisco proteins have conserved key residues for known mechanistically important functions, it is apparent that different non- conserved residues and regions of the protein must influence the specificity of the substrates carbon dioxide and oxygen as well as other kinetic factors. If this relationship between kinetic factors and residues that influence catalysis can be determined, it will uncover the “holy grail” of all Rubisco enzymes and could have a significant impact on the environment for reasons previously mentioned (Cleland et al., 1998; Spreitzer, 1999;

Tabita, 1999).

Genome sequencing projects of organisms from the third domain of life, the archaea, revealed the presence of Rubisco enzymes in these organisms. The function of these enzymes remained obscure, however, because other enzymes necessary to complete the CBB pathway, particularly phosphoribulokinase (PRK), the enzyme that is necessary for the production of RuBP and thus is required to supply substrate for Rubisco, is not present in these organisms (Bult et al., 1996; Galagan et al., 2002; Klenk et al., 1997). In addition, other means for fixing carbon dioxide that is necessary for autotrophic growth was discovered in some of these organisms, such as the acetyl CoA pathway. To date, there is no evidence that the CBB reductive pentose phosphate pathway provides a major means by which these organisms assimilate CO2 (Deppenmeier et al., 2002; Ezaki et

15 al.,1999; Galagan et al., 2002; Klenk et al., 1997; Watson et al., 1999; Finn and Tabita,

2003). This brings up the perplexing question as to what exactly Rubisco is doing in

these organisms. To date, there are 12 Rubisco sequences that are derived from archaeal

organisms and are grouped as form III Rubiscos (Sato et al., 2007). Of the bona fide

Rubiscos from archaea, many of them have very different features not found in other

form I or form II Rubiscos, such as activity at high salt concentrations, activity at extreme

temperatures as high as 100°C, and activity achieved only under strict anaerobic

conditions. While their sequences are highly divergent from forms I and II, and contain

unique characteristics not found in the other forms, archaeal Rubiscos have been grouped

together in their own group, form III (Figure 1.2). Even though there are differences in

primary sequence, structure, and requirements for optimal activity, the form III enzymes

retain the ability to carry out carboxylation as a consequence of conservation of the key

residues implicated in catalysis, as discussed above (Watson et al., 1999). Interestingly,

with all these varied features, archaeal form III Rubiscos still maintain the simple large

subunit monomer that dimerizes to form the active site, and do not have small subunits

associated with them. Again, most of the archaeal Rubiscos are simple dimers, such as

previously studied Rubiscos from Methanocaldococcus jannaschii, Methanosarcina

acetivorans, and Archaeoglobus fulgidus. But, at least in one instance, there is a higher

order pentamer of dimers for the Rubisco from Thermococcus kodakaraensis KOD1

(Figure 1.3C). All the archaeal Rubiscos large subunits are slightly smaller in size compared to the form I and II Rubiscos; i.e., about 45 to 48 kDa (Kitano et al., 2001).

Since the discovery of Rubisco enzymes in archaea in the mid-1990’s, efforts to

study the properties of these enzymes have increased because they may yield important

16 clues that will help further our understanding of substrate specificity in Rubisco.

Previous studies indicated that RuBP may be formed by a novel means via the conversion

of ribose 1,5-bisphosphate to RuBP (Finn and Tabita, 2004). Various metabolites

including 5-phosphoribose-D-1-pyrophosphate (PRPP) led to the formation of RuBP.

More recent studies indicated that adenosine monophosphate (AMP) (presumably derived from PRPP) leads to the formation of ribose 1,5-bisphosphate, catalyzed by an AMP phosphorylase, and then RuBP via a specific isomerase. This pathway, along with

Rubisco, provides a means to produce 3-PGA that can then be used to produce ATP

(Figure 1.6) (Sato et al., 2007).

AMP phosphorylase Ribose 1,5-bisphosphate isomerase

Tk-DeoA Tk-E2b2

Pi adenine

AMP Ribose 1,5- Ribulose 1,5- bisphosphate bisphosphate

Figure 1.6. Metabolic pathway of RuBP production in archaeal organisms. Tk-DeoA, the adenine monophosphate (AMP) phosphorylase, catalyzes conversion of AMP to ribose 1,5-bisphosphate (R1,5P). Tk-E2b2 then catalyzes an isomerization reaction to form ribulose 1,5-bisphosphate (RuBP), the substrate for Rubisco, which will then lead to the formation of two molecules of 3-PGA.

Rubisco enzymes from archaeal sources have many unusual properties not associated with form I and form II enzymes. The ability of archaeal Rubiscos to sustain activity in unusual conditions such as higher salt concentrations or higher temperatures are just a few of the properties not found in other Rubiscos. One of the most important characteristics of archaeal Rubiscos is their apparent sensitivity to molecular oxygen.

Many of the archaeal Rubiscos studied to date are sensitive to oxygen; they lose carboxylase activity when exposed to and assayed in the presence of oxygen, even in the presence of high levels of CO2 which normally swamp out the effects of oxygen (Watson

18 and Tabita, 1999; Finn and Tabita, 2003). Previous studies in our lab have shown that optimal carboxylase activity is obtained under strictly anaerobic conditions for Rubisco enzymes from the archaeal organisms A. fulgidus, M. jannaschii, and M. acetivorans which were obtained both in the host organisms as well as recombinant proteins from

Escherichia coli (Watson et al., 1999). When exposed to molecular oxygen and assayed in the presence of air, these Rubiscos exhibit a substantial loss of activity as compared to activity obtained when assayed under strict anaerobic conditions. Interestingly, when the oxygen exposed Rubisco enzymes from A. fulgidus and M. jannaschii are returned to an anaerobic environment, partial but not full activity is obtained. Unfortunately, for the M. acetivorans Rubisco, carboxylase activity was not recovered when the oxygen-exposed enzyme was returned to an anaerobic atmosphere like that seen for other archaeal

Rubiscos (Finn and Tabita, 2003). Sensitivity to oxygen to this extreme is not observed in form I and form II Rubiscos (Jordan and Ogren, 1981). It was suggested that intensive investigations of the structure-function relationship of these form III archaeal Rubisco enzymes might lead to a better understanding of the basis for competing carbon dioxide and oxygen selectivity for the enediol intermediate.

Although form III archaeal Rubiscos are the least understood of all bona fide forms of Rubisco and detailed study of such enzymes from strictly anaerobic organisms has not thus far been accomplished. For such a study, characterization of Archaeoglobus fulgidus RbcL2 was thought to show promise with respect to shedding light on unusual properties found for this group of Rubisco enzymes, particularly the molecular basis for the interesting oxygen sensitivity. The ability to easily produce recombinant proteins suggested that a mutagenesis approach might be useful, with the aim of directing such

19 studies towards CO2/O2 substrate specificity and O2 sensitivity. Such studies might then eventually provide clues as to determining how all forms of Rubisco select between the two gaseous substrates. In addition, experiments were designed that utilize a Rubisco knockout strain from Rhodobacter capsulatus as a surrogate host to assist in demonstrating and understanding how wild-type and mutant forms of the A. fulgidus enzyme affect the organism’s ability to grow under phototrophic and aerobic chemoautotrophic conditions.

CHAPTER 2

KINETIC PROPERTIES OF ARCHEAOGLOBUS FULGIDUS RBCL2; SITE-

DIRECTED MUTAGENESIS STUDIES TO PROBE THE BASIS FOR

OXYGEN SENSITIVITY

INTRODUCTION

The organism Archaeoglobus fulgidus is a sulfate reducing organism that is

predominantly found in hydrothermal vents located at the bottom of the ocean floor; this

organism has also been discovered in other location such as sub-surface oil fields. A. fulgidus is an extreme thermophile whose discovery linked two distant forms of archaea, those that exhibit an anaerobic thermophilic sulphur-based metabolism and organisms that undergo methanogenesis. A. fulgidus is an obligate anaerobe that uses sulfate, thiosulfate or sulfite as electron acceptors. Growth with carbon sources from lactate and pyruvate may be achieved in the presence of sulfate or thiosulfate, although other carbon sources such as formate, fumarate and H2/CO2 may be utilized but are less effective and require the presence of thiosulfate. A. fulgidus grows between 60 and 90°C with the

optimal temperature between 75 and 80°C (Achenbach-Richter et al., 1987; Zellner et al.,

1987). In the mid-1990’s, the organism’s genome was fully sequenced and the presence

21 of two putative Rubisco genes was noted that shared sequence similarity to form I and II

Rubisco enzymes. The deduced amino acid sequence of the rbcL1 gene was later shown

to resemble form IV Rubisco or the Rubisco-like protein (Watson et al., 1999), while the rbcL2 gene appeared to encode a bona fide form III Rubisco (Finn and Tabita, 2003).

For the latter protein (RbcL2), the presence of highly conserved residues that are important for catalysis and RuBP substrate binding, was noted (Watson et al., 1999; Finn and Tabita, 2003) although the presence of other genes that encode proteins for the CBB pathway were not found (Klenk et al., 1997). Additionally, carbon fixation apparently was via the acetyl CoA pathway; thus the role of Rubisco in this organism at that time was unknown. Known key active site residues found in Rubisco enzymes were completely conserved in A. fulgidus rbcL2 with the exception of a valine residue at position 158 rather than the usual threonine residue found in form I enzymes and an isoleucine in form II enzymes (Figure 2.1).

22 C I Syn PCC 6301 RbcL MPKTQSAAGYKAGVKDYKLTYYTPDYTP-KDTDLLAAFPVSPQPGVPADEAGAAIAAESS 59 II R.rubrum CbbM ------MDQSSRYVNLALKEEDLIAGGEHVLCAYIMKPKAGYGYVATAAHFAAESS 50 III A.fulgidus RbcL2 ------MAEFE-IYREYVDKSYEP-QKDDIVAVFRITPAEGFTIEDAAGAVAAESS 48 .::. :. * . .*.*** R I Syn PCC 6301 RbcL TGTWTTVWTDLLTDMDR-YKGKCYHIEPVQGEENS---YFAFIAYPLDLFEE------GS 109 II R.rubrum CbbM TG--TNVEVCTTDDFTRGVDALVYEVDEA------RELTKIAYPVALFHRNITDGKAM 100 III A.fulgidus RbcL2 TGTWTSLHPWYDEERVKGLSAKAYDFVDL-GDGS----SIVRIAYPSELFEP------HN 97 * *.: : *.. : **** *. C I Syn PCC 6301 RbcL VTNILTSIVGNVFGFKAIRSLRLEDIRFPVALVKTFQGPPHGIQVERDLLNKYGRP---M 166 II R.rubrum CbbM IASFLTLTMGNNQGMGDVEYAKMHDFYVPEAYRALFDGPSVNISALWKVLGRPEVDGGLV 160 III A.fulgidus RbcL2 MPGLLASIAGNVFGMKRVKGLRLEDLQLPKSFLKDFKGPSKGKEGVKKIFGVADRP---I 154 :...*: ** *: . :: *: .* :.** . . . : : C C C RubisCO motif:GXDFXKXDE I Syn PCC 6301 RbcL LGCTIKPKLGLSAKNYGRAVYECLRGGLDFTKDDENINSQPFQRWRDRFLFVADAIHKSQ 226 II R.rubrum CbbM VGTIIKPKLGLRPKPFAEACHAFWL-GGDFIKNDEPQGNQPFAPLRDTIALVADAMRRAQ 219 III A.fulgidus RbcL2 VGTVPKPKVGYSAEEVEKLAYELLSGGMDYIKDDENLTSPAYCRFEERAERIMKVIEKVE 214 * ***:* .: . : .* * *:** . : .: . : :

I Syn PCC 6301 RbcL AETGEIKGHYLNVTAPTCEEMMKR-----AEFAKELGMPIIMHDFLTAGFTANTTLAKWC 281 II R.rubrum CbbM DETGEAKLFSANITADDPFEIIARGEYVLETFGENASHVALLVDGYVAGAAAITTARRRF 279 III A.fulgidus RbcL2 AETGEKKSWFANITAD-VREMERR-----LKLVAELGNPHVMVDVVITGWGALEYIRDLA 268 **** * *:** *: * . : . : * * * CR R C I Syn PCC 6301 RbcL RDNGVLLHIHRAMHAVIDR-QRNHGIHFRVLAKCLRLSGGDHLHSGTV-VGKLEGDKAST 339 II R.rubrum CbbM PDN--FLHYHRAGHGAVTSPQSKRGYTAFVHCKMARLQGASGIHTGTMGFGKMEGES--- 334 III A.fulgidus RbcL2 EDYDLAIHGHRAMHAAFTR-NAKHGISMFVLAKLYRIIGIDQLHIGTAGAGKLEGQKWDT 327 * :* *** *... . * . .* *: * . :* ** **:** . R R I Syn PCC 6301 RbcL LGF-VDLMREDHIERDRSRGVFFTQDWASMPGVLPVASGGIHVWHMPALVEIFG-DDSVL 397 II R.rubrum CbbM ----SDRAIAYMLTQDEAQGPFYRQSWGGMKACTPIISGGMNALRMPGFFENLGNANVIL 390 III A.fulgidus RbcL2 VQN-ARIFSEVEYTPDEGDAFHLSQNFHHIKPAMPVSSGGLHPGNLEPVIDALG-KEIVI 385 :. *.: : * ***:: : ..: :* : :: RR I Syn PCC 6301 RbcL QFGGGTLGHPWGNAPGATANRVALEACVQARNEGRDLYREGGDILREAGKWSPELAAALD 457 II R.rubrum CbbM TAGGGAFGHIDGPVAGARSLRQAWQAWRDGVP------VLDYAREHKELARAFE 438 III A.fulgidus RbcL2 QVGGGVLGHPMGAKAGAKAVRQALDAIISAIP------LEEHAKQHPELQAALE 433 ***. ** * .** : * * :* .. : : .. ** *::

I Syn PCC 6301 RbcL LWKEIKFEFETMDKL------472 II R.rubrum CbbM SFPGDADQIYPGWRKALGVEDTRSALPA 466 III A.fulgidus RbcL2 KWGRVTPI------441 :

Figure 2.1. Amino acid sequence alignment of Form III archaeal A. fulgidus RbcL2 and representative examples of form I (Synechococcus sp. strain PCC 6301) and form II (R. rubrum) Rubiscos. Multiple sequence alignments were performed by using ClustalW (Thompson et al., 1994). The accession numbers for each deduced large subunit sequence are as follows: A. fulgidus rbcL2, O28635; Synechococcus PCC 6301, P00880; and R. rubrum, P04718. Residue identities are marked with an asterisk, conserved substitutions are marked with a colon, and semiconserved substitutions are marked with a period. Known active-site residues determined to be within 3.3 Å of the bound transition state analog CABP in the Synechococcus PCC 6301 enzyme are labeled C for catalytic and R for RuBP binding properties. Where these residues are identical in all three sequences, they are colored red, with the exception of Synechococcus PCC 6301 T170 which is colored blue. The characteristic Rubisco motif sequence,GXDFXKXDE, is underlined.

23 Based on sequence homologies, archaeal Rubiscos represent a special class of

enzyme, termed form III (Tabita, 1999; Watson and Tabita, 1997) to distinguish these

enzymes from previously characterized form I and form II Rubiscos. Even with these

considerable differences in primary sequence, the form III enzymes retain many features

characteristic of all forms of Rubisco, mainly the ability to carry out carboxylation as a

consequence of conservation of the key residues implicated in catalysis (Watson et al.,

1999), as discussed above. Other than the high conservation of catalytic residues, A. fulgidus RbcL2 has only 33-37% sequence identity with the form I enzymes and only 31-

37% sequence identity with form II enzymes. Within the form III archaeal enzymes, it is only 44% identical with methanogen Rubiscos but has a 72% identity with the pyrococci enzymes such as Thermococcus kodakaraensis KOD1. Although there is no evidence that the CBB reductive pentose phosphate pathway provides a major means by which these organisms assimilate CO2, Rubisco and a novel pathway to synthesize RuBP, appear to be used for a type of purine/pyrimidine salvage pathway in most archaeal organisms that possess Rubisco genes (Finn and Tabita, 2003; Sato et al., 2007). Our laboratory has determined that the archaeal Rubisco genes from Archaeoglobus fulgidus,

Methanocaldococcus jannaschii and Methanosarcina acetivorans encode bona fide

Rubiscos, capable of catalyzing substantial activity, both in the native organisms as well as within Escherichia coli (Watson et al., 1999; Finn and Tabita, 2003). Moreover, the archaeal genes can be expressed in a phototrophic eubacterial host in which the native

Rubisco genes had been inactivated, such that the archaeal genes may complement the organism to allow CO2-dependent growth (Finn and Tabita, 2003).

24 The A. fulgidus rbcL2 gene encodes a protein (RbcL2) of 441 amino acids with a

monomer molecular weight of 48.5 kDa. This form III Rubisco exists as a homodimer, as

does the enzyme from M. jannaschii and M. acetivorans (Watson et al., 1999; Finn and

Tabita, 2003). In this respect, the quartenary structure of the three form III archaeal enzymes resembles the bacterial form II Rubisco from Rhodospirillum rubrum (Tabita,

1974). However, A. fulgidus RbcL2 exhibits unique properties not found in other forms of Rubisco such as optimal activity at temperatures exceeding 80°C. In addition, this protein is highly sensitive yet reversibly inhibited upon exposure to air-levels of oxygen, necessitating that the enzyme be prepared under strictly anaerobic conditions, in order to obtain optimal activity. This enzyme is derived from a very strict anaerobe, and like other Rubiscos, there are no motifs that would suggest oxygen involvement in stability.

In the current study, we have focused on this unusual reversible inhibition by low concentrations of oxygen and examine the possible involvement with a unique residue,

Met-295, that appears to be located at an influential site within the structure of the protein. Other residues in a hydrophobic pocket region also appear to contribute to oxygen sensitivity. The unique oxygen sensitivity of the form III archaeal Rubiscos may provide clues as to how the active site of this enzyme has evolved to become more stable in the presence of oxygen in more evolutionarily advanced form I and form II Rubisco proteins.

MATERIALS AND METHODS

Plasmids, bacterial strains and growth conditions. In vitro studies employed

RbcL2 from Archaeoglobus fulgidus using purified recombinant protein obtained from

Escherichia coli after over-expression of the rbcL2 gene in this organism. All cloning steps were performed in E. coli JM109 (Yanisch-Perron et al., 1985) prior to transformation into E. coli BL-21(DE3) (Stratagene, La Jolla, California) for over expression of the rbcL2 gene. E. coli cultures were grown in Luria-Bertani (LB) media containing 1% tryptone, 0.5% yeast extract, and 1% NaCl (w/v). A. fulgidus rbcL2 (Af

1638, accession number NC_000917) was cloned directly from genomic DNA. Primers designed with an NdeI (5'GGAATTCCATATGGCGGAGTTTGAGATTTACAGA3') restriction site at the N terminus and a BamHI (5'ATTTTAGATTGGCGTAACCCTG3') restriction site at the C terminus were used to amplify the rbcL2 gene from A. fulgidus genomic DNA using Pfu polymerase. The gene was ligated into pCR-Script (Stratagene) and sequenced to assess whether there were PCR-induced mutations. Using the NdeI and

BamHI sites in that vector, the gene was subcloned into pET11a (Novagen).

Site-directed mutagenesis. Site-directed mutagenesis was performed using

QuikChange site-directed mutagenesis kit from Stratagene (Papworth et al., 1996). The

Met-295 ATG sequence within the A. fulgidus rbcL2 gene was replaced with GCG,

TTC, and GAC to obtain alanine, phenylalanine, and aspartate, respectively. Automated

26 sequencing was performed to confirm the sequences of mutant genes using 3730 DNA

Analyzer system (Applied Biosystems) at the OSU Plant-Microbe Genomics Facility.

The mutant genes were inserted into fresh pET11a plasmid after digestion with NdeI and

BamHI.

Over-expression of the rbcL2 gene and purification of recombinant RbcL2

protein. E. coli BL-21(DE3) cells with transformed pET11a vector containing A.

fulgidus rbcL2 were grown using 2.8 l broad bottom flasks containing 2 l of LB media at

37°C at 120 rpm to minimize aeration to an OD600 of 0.4. The temperature of the media containing the cultures was then raised to 42°C for 30 min for expression of E. coli heat shock proteins, produced from the genes dnaJ, dnaK and groEL, to increase the amount of soluble recombinant protein. The media was then allowed to cool to room temperature before inducing with 0.1 mM Isopropyl-ß-D-thiogalactopyranoside (IPTG) and allowed to shake at 120 rpm for 16 h. Cells were harvested to remove LB media and then washed with anaerobic wash buffer (100 mM N,N-bis(2-Hydroxyethyl) glycine (Bicine), pH 8.3,

10 mM MgCl2, 1 mM EDTA) stored in an anaerobic chamber. Cells were centrifuged again in centrifuge bottles containing screw caps with rubber seals. Cell pellets were recovered in the chamber and were then stored at -70°C before further protein purification with column chromatography.

All preparation and manipulation to the cell material was performed in an

anaerobic chamber. Prior to column chromatography, cells were resuspended in wash

buffer supplemented with 10 mM phenylmethylsulfonyl fluoride (PMSF) and 50 µg/ml deoxyribonucleic acid I (DNase I) and were disrupted using a pressurized French pressure cell (at 110,000 kPa) flowing directly into a sealed anaerobic serum vial sparged

27 with argon gas. The lysed cells were then centrifuged at 16,000 x g at 4°C for 20 min in

screw cap centrifuge tubes with rubber sealed caps. The supernatant was decanted into a

serum vial and placed in an 80°C water bath for twenty minutes and then allowed to cool

on ice for 1 h. The heat stable extract was transferred to a fresh screw cap centrifuge tube

with rubber sealed caps and centrifuged at 30,000 x g at 4°C for 30 min. The heat stable

supernatant was syringe filtered using 0.22 µm filters before loading onto columns for

further purification.

Column chromatography was performed under anaerobic conditions using a Bio-

Rad BioLogic HR Workstation. Heat stable extract was loaded onto a Q-Sepharose

strong anion exchange column equilibrated with wash buffer supplemented with 50 mM

NaHCO3 and 10 mM β-mercaptoethanol (Buffer A). Samples were eluted in 2 M NaCl in Buffer A, recombinant A. fulgidus RbcL2 elutes at ~0.4 M NaCl. Fractions were

monitored for activity using a modified protocol of the standard Rubisco assay under

anaerobic conditions (Tabita et al., 1978). Fractions with activity were pooled and

concentrated with a Millipore 30,000 MWCO concentrator and loaded onto a 110 ml

Superose-12 gel filtration column. Fractions with activity were pooled and further

purified based on hydrophobic interaction using a Phenyl-sepharose column. Samples

were eluted with decreasing salt starting with 2 M (NH4)2SO4, recombinant A. fulgidus

RbcL2 elutes at ~0.4 M (NH4)2SO4. Fractions with activity were pooled and concentrated with a 30,000 MWCO Millipore concentrator using a centrifuge and then loaded onto a 1 ml G-25 desalting column to remove any remaining (NH4)2SO4. Purified protein was stored in 20% glycerol at -70°C in anaerobically sealed serum vials.

28 Radiometric Rubisco Assays. A. fulgidus RbcL2 was assayed for activity under

a strict anaerobic atmosphere unless otherwise noted. The previously described assay was used and modified to optimize carboxylase activity (Tabita et al., 1978). Buffers and substrates were bubbled with argon gas in sealed glass serum vials prior to use. In an anaerobic chamber, enzyme was prepared in glass serum vials in 100 mM Bicine-NaOH, pH 8.3, 10 mM MgCl2, 1 mM EDTA, and 0.4 M NaCl. Vials were sealed in the chamber

and then placed in a React Therm III™ heating/stirring module (Pierce, Rockford, IL) set

at 83°C after the addition of 50 mM NaHCO3 in Bicine buffer, containing a final

14 concentration of 2 µCi NaH[ C]O3. Reactions were initiated by the addition of

anaerobic RuBP and stopped with the addition of aerobic propionic acid. Vials were

unsealed and dried overnight in a vacuum oven at 65°C. Samples were resuspended in

200 µl 2 N HCl and counted in 3 ml scintillation cocktail using a Tri-Carb 2100TR

Liquid Scintillation Analyzer (Packard Instrument, Meriden, CT). The Bradford method

was used to determine protein concentrations, BSA was used as the standard (Bradford,

1976).

Kinetic measurements. Purified enzymes were used for all kinetic

measurements of kcat, KCO2, KO2, KRuBP, and Ω. The KCO2 was determined under strict anaerobic conditions using sealed vials as previously described with few modifications

14 (Smith and Tabita, 2003). Dilutions of NaH[ C]O3 were prepared in 100 mM Bicine–

NaOH buffer with 10 mM MgCl2. The pH of the buffer was usually around 8.3, and the exact pH was recorded for each assay. Assays were performed at 83°C, initiated by the addition of anaerobic RuBP, and terminated after 30 sec by addition of 100 µl of aerobic

propionic acid. Vials were unsealed and dried overnight in a vacuum oven at 65°C.

29 Products were resuspended in 2 N HCl and counted in scintillation cocktail. Results were

plotted using Sigma Plot 2002 v8.0, deriving the KCO2 and KO2 by fitting values to a hyperbolic curve and double reciprocal plot. The concentration of CO2 was derived using the pH and the Henderson–Hasselbach relationship. Solubility of CO2 at 83°C was calculated from published values to obtain an equation that was extrapolated to 83°C

(Dean, 1985). After determining the average volume of the glass vials (2.2 ml), various concentrations of oxygen were introduced into the vials by removing a certain percent of the anaerobic headspace and replacing it with the same amount of oxygen from a sealed serum vial sparged with ultrapure oxygen. This percentage of oxygen introduced to the vial was then used to determine how much oxygen (in µM) was present in the vial and then the solubility of oxygen was determined using solubility charts available from

Unisys®.

The KRuBP was measured similarly to the KCO2, determined under strict anaerobic conditions in sealed serum vials at 83°C. Various concentrations of RuBP were prepared and sparged with argon gas. Assays were initiated with the addition of RuBP to the assay vials containing activated enzyme and ran for 30 sec then stopped by the addition of aerobic propionic acid. Samples were dried overnight, resuspended and counted in scintillation cocktail. Results were plotted using Sigma Plot.

Specificity was measured under conditions of saturating O2 (1.23 mM) with 200

mM NaHCO3 in 100 mM Bicine buffer (pH 8.3), 10 mM MgCl2. The concentration of

CO2 was calculated from the Henderson–Hasselbach relationship, as described above for

3 KCO2. Reactions were initiated by addition of [1- H]RuBP, and incubated at 83°C for 2 h.

The reaction was halted by the addition of 200 mM NaBH4 and incubated at room 30 temperature for 15 min. Excess NaBH4 was consumed by the addition of 400 mM

Glucose and incubated for an additional 15 min at room temperature. Samples were

diluted with distilled water and products formed were separated from the enzyme by

centrifugation in a Millipore 30,000 MWCO concentrator. The samples were frozen at

-70°C until further use. Reaction products were separated with a MonoQ resin using a

Dionex DX500 chromatography system (Dionex Corporation, Sunnyvale, CA) and

detected with an in-line scintillation counter (IN/US b-Ram, Tampa, FL), as described

(Harpel et al., 1993).

Molecular modeling of A. fulgidus RbcL2. Modeling of the A. fulgidus RbcL2 was performed using Deep View Swiss PDB Viewer, spdbv 3.7 (Guex and Peitsch,

1997). The template used to model the dimer form of the enzyme was Thermococcus kodakaraensis KOD1 crystal structure (PDB, 1GEH), the closest related Rubisco large subunit based on sequence identity, 72%, with A. fulgidus RbcL2 (Berman et al., 2000).

RESULTS

In this study, the focus was on the unusual reversible inhibition by low concentrations of oxygen. The possible involvement with a unique residue, Met-295, that appears to be located at an influential site within the structure of the protein, was examined. It was surmised that the unique oxygen sensitivity of the form III archaeal

Rubiscos might provide clues as to how the active site of this enzyme has evolved to become more stable in the presence of oxygen in more evolutionarily advanced form I and form II Rubisco proteins.

The A. fulgidus rbcL2 gene was over expressed in E. coli BL-21(DE3) and the resultant recombinant protein purified under strictly anaerobic conditions to virtual homogeneity (Figure 2.2).

Figure 2.2. Coomassie-stained SDS-PAGE of samples from Rubisco purification. The A. fulgidus rbcL2 gene was expressed in E. coli and samples obtained from: uninduced E. coli cells (lane 1); soluble extract of French Press disrupted E. coli cells after induction (lane 2); supernatant obtained after centrifuging the heat-treated extract for 20 min at 80°C (lane 3); Q-Sepharose anion exchange chromatography (lane 4); Superose-12 gel filtration (lane 5); phenyl-Sepharose hydrophobic chromatography (lane 6).

By means of nondenaturing polyacrylamide gel electrophoresis and gel filtration

chromatography, the purified A. fulgidus RbcL2 appeard to exist as a dimer of large

subunits, similar to the previously determined holoenzyme structure of the form III M.

jannaschii RbcL (Watson et al., 1999; Finn and Tabita, 2003). Like the M. jannaschii

and T. kodakaraensis archaeal Rubiscos, the A. fulgidus enzyme was highly active at

temperatures up to 93°C, with a temperature optimum for activity of 83°-93°C (Finn and

Tabita, 2003; Ezaki et al., 1999). Interestingly, activity was detected as low as 23°C

(Figure 2.3), quite different from the M. jannaschii enzyme (Finn and Tabita, 2003).

) 12

l/min 10 o m 8 its (n n

U 6

0 23 33 43 53 63 73 83 93

Temperature (°C)

Figure 2.3. Carboxylase Activity for A. fulgidus RbcL2 at varying temperatures. Recombinant, purified enzyme from E. coli was assayed at different temperatures under strict anaerobic conditions. At each temperature, 1 µg of enzyme was used. The assay was initiated with the addition of RuBP and allowed to proceed for 1 min before the addition of aerobic propionic acid to stop the reaction.

Optimum activity for the A. fulgidus RbcL2 was also achieved in the presence of

0.3 M NaCl, when compared to a range of KCl and (NH4)2SO4 concentrations (Figure

2.4). Under strictly anaerobic conditions at 83°C, the expected product, [14C]3-PGA, was produced in stoichiometric amounts by the A. fulgidus RbcL2 catalyzed reaction, as

14 determined using both a CO2 incorporation assay, or via a nonradioactive coupled PGA enzymatic assay (Watson et al., 1999). These results demonstrated that this enzyme is a bona fide Rubisco. Furthermore, A. fulgidus RbcL2 was found to possess an unusually

34 -1 high kcat of about 23 sec at 83°C (specific activity of 28.9 µmol/min/mg), compared to

-1 the characteristically low kcat values of 3-5 sec that have been reported for other forms of Rubisco, assayed at their optimum temperatures (Hartman and Harpel, 1993; Hartman and Harpel 1994).

25 ) in

l/m 20

o KCl

m NaCl

(n (NH ) SO s 15 4 2 4 t i n U 10

0 0 200 400 600 800 1000

[Salt] (mM)

Figure 2.4. Requirement and Inhibitory Effects of Salt for Carboxylase Activity of A. fulgidus RbcL2. Recombinant, purified enzyme over expressed in E. coli was assayed with different salts at varying concentrations under strict anaerobic conditions. The salts that were tested were KCl, NaCl, and (NH4)2SO4. For each assay, 1 µg of enzyme was used. The assay was initiated with the addition of RuBP and allowed to proceed for 1 minute before the addition of aerobic propionic acid to stop the reaction.

Aside from its extreme thermostability and high intrinsic activity, the A. fulgidus

RbcL2, unlike its T. kodakaraensis homolog (Ezaki et al., 1999), was found to be 35 sensitive to molecular oxygen. Thus, substantial activity loss was obtained in

preparations exposed to oxygen and/or assayed in the presence of oxygen, even in the

presence of overwhelming excesses of bicarbonate. Current experiments, as well as

previous studies, indicated that the activity lost after exposure of A. fulgidus RbcL2 to molecular oxygen could be recovered (Watson et al., 1999; Finn and Tabita, 2003).

When the enzyme was exposed to molecular oxygen for 30 min and assayed in the presence of air, some 10-15% of the overall carboxylase activity was obtained compared to enzyme kept fully anaerobic and assayed under strict anaerobic conditions. When the oxygen-exposed enzyme was subsequently injected into an anaerobic vial and assayed under strict anaerobic conditions, 65% of the carboxylase activity was obtained compared to enzyme preparations maintained and assayed under anaerobic conditions (Figure 2.5).

Several repetitions of these experiments with different enzyme preparations indicated that levels of recovered activity ranged from 65% to nearly full recovery. The fact that the level of activity recovery varied suggested that perhaps differing amounts of oxygen

might be carried over from the oxygen incubation vials to the anaerobic assay tubes.

Assays performed at a lower temperature (23°C) comparing specific activities of

complete anaerobic samples and oxygen exposed samples resulted in the same amount of

activity lost after oxygen exposure compared to the samples assayed at 83°C, the major

difference was the much lower amount of specific activity obtained at the lower

temperature, as previously mentioned.

25000

20000

15000

(nmol/min/mg)

vity ti 10000 Ac ific 5000 Spec

0 Anaerobic Oxygen Exposed Recovered

Figure 2.5. Recovery of carboxylase activity of A. fulgidus RbcL2 upon oxygen exposure. Wild-type was assayed initially under strict anaerobic conditions (Anaerobic). The enzyme was then exposed to molecular oxygen for 30 min and assayed in the presence of air (Oxygen Exposed). An aliquot was then taken from the vial containing oxygen exposed enzyme and injected into an anaerobic vial and assayed (Recovered). No O2 scavenging system was present in the “recovered” vials. Assay conditions were the same in all cases using a strict anaerobic Rubisco assay with the exception of the “oxygen exposed” samples in which the assay vials were not sealed with rubber septas and crimped with aluminum caps.

37 In order to alleviate this problem, experiments were designed to scrub out all

vestiges of oxygen from enzyme preparations that had been exposed to oxygen and then

transferred and diluted into the anaerobic assay vials. Scrubbing was accomplished by

incorporating commercially available protocatechuic acid (PCA) 3,4-dioxygenase (PCD)

and its substrate, PCA, into the Rubisco assay mixture to “fix” any oxygen that remained

in solution. Assays were set up similar to the oxygen exposure assay however after 30

minutes of oxygen exposure, enzyme was injected into a vial containing 0.5 units PCD

and 2 mM PCA. Briefly, the PCD enzyme catalyzes the benzene ring cleavage of

protocatechuic acid via the addition of one molecular oxygen molecule. The three and

four carbon position of the ring accept a single oxygen atom each resulting in the

disruption of the ring and creating β-carboxymuconic acid (Fujisawa, 1968; Watson et al., 1999). To optimize this reaction, the sample vials containing the oxygen exposed A. fulgidus RbcL2, PCA and PCD was incubated at 37°C for 10 minutes. After that time, the vials were placed in the 83°C heating block for the Rubisco assay. The PCA dioxygenase-scrubbing system was highly effective, resulting in the recovery of all available carboxylase activity (Figure 2.6). O2-mediated inhibition was thus fully

reversible; partial (65%) restoration of activity obtained in vials lacking the oxygen

scrubbing system was clearly attributable to oxygen carried over from the initial

incubation.

38 16000

14000

12000

10000

8000 ty (nmol/min/mg)

6000 Activi

4000 Specific 2000

0 0 10203040506070 Time (minutes)

Figure 2.6. Reversibility of oxygen inhibition of A. fulgidus RbcL2. Assays were performed in the absence (●) or presence (○) of molecular oxygen, followed by removal of molecular oxygen and replacement with an anaerobic atmosphere at 20 min and the addition of an O2 scavenging system containing protocatechuate (PCA) 3,4-dioxygenase (20) at 30 min (indicated by the arrow). In all cases, the enzyme was dialyzed in a CO2- free, O2-free buffer of 100 mM Bicine-NaOH (pH 8.3) containing 10 mM MgCl2, 1 mM EDTA, and 0.4 M NaCl.

Inhibition by low amounts of oxygen, especially in the presence of a large excess

of bicarbonate (CO2) in the otherwise anaerobic assay, is something that is not seen with

the highly studied form I and form II enzymes, simply because CO2 and O2 compete for the same enediolate-enzyme complex. Thus 20-50 mM bicarbonate in an assay typically 39 simply swamps out any O2 that might otherwise inhibit the carboxylation reaction. For

the A. fulgidus RbcL2 enzyme, a viable assumption is that oxygen binding must be quite

efficient since simply diluting oxygen-exposed enzyme into an anaerobic assay with high

levels of bicarbonate was not sufficient to fully reactivate the enzyme. Furthermore, full

activity was restored only after the addition of the O2 scavenging system. Thus,

experiments were initiated to measure the affinity of the A. fulgidus enzyme for its

substrates CO2, O2, and RuBP. The usual anaerobic methods were employed as described in the experimental procedures and the KCO2, KO2, and KRuBP were determined at 83°C. The KCO2 value was determined to be 51 ± 8 µM. To calculate the KO2, various fixed concentrations of pure oxygen were injected into vials that were assayed with varying amounts of CO2, as for the KCO2 determination. Double reciprocal plots clearly

showed that O2 was a competitive inhibitor with respect to CO2 (Figure 2.7). In addition, replots of the data the double reciprocal plots allowed the KO2 or Ki for O2 to be determined (5 ± 1 µM) (Figure 2.8). The KRuBP was determined to be 20 ± 5 µM (Table

1). The KCO2 and KRuBP values fall within the range of reported values for other form I

and form II Rubiscos (Tabita, 1999). The most notable difference was the extremely low

KO2 value for A. fulgidus RbcL2. Form I and form II Rubiscos typically have KO2 values ranging from 500 to 1000 µM; thus it is apparent that the unusual sensitivity of A. fulgidus RbcL2 activity to oxygen may be attributed to the extreme high affinity of this archaeal form III enzyme for oxygen.

0.0006

0.0005

mg) n/ i

l/m 0.0004

mo (n y t

i 0.0003

Activ c 0.0002 ecifi

1/Sp 0.0001

.04 -0.02 0.00 0.02 0.04 0.06

1/[CO2] (µM)

Figure 2.7. Double reciprocal plot of RuBP carboxylase activity (in nmol/min/mg) of wild-type A. fulgidus RbcL2 as a function of varying concentrations (µM) of CO2 at fixed levels (●, 0 µM; ○, 4.2 µM; ■, 21.0 µM; □, 42.1 µM) of O2. The assay conditions were as described in the experimental procedures. Vmax and Km values and lines were drawn using Sigma Plot v8.0.

0.025 ) -1 0.020 [CO2] ∆ / -1

tivity 0.015 c A

ific 0.010 Spec ∆

e (

Slop 0.005

0 1020304050

[O2] (µM)

Figure 2.8. Secondary replot of the slopes from the double reciprocal plot used to calculate the Ki for O2 for wild-type A. fulgidus RbcL2. Lines were drawn using Sigma Plot v8.0.

The low residual carboxylase activity seen with the A. fulgidus enzyme in the presence of low concentrations of oxygen was also observed for other archaeal form III enzymes from M. jannaschii and M. acetivorans (Watson et al., 1999; Finn and Tabita,

2003). Moreover, as a result of various genome sequencing projects, there are now

several available archaeal Rubisco sequences in the data base. These archaeal sequences,

along with many additional form I and form II Rubisco sequences available, primarily

from well studied Rubiscos including those with solved crystal structures, prompted a

bioinformatics analysis of all available form I, form II, and form III sequences to

determine if there might be unique residues within structurally significant regions of the

enzyme that might perhaps influence the unusual kinetic properties of the archaeal

Rubisco.

42 Eight form III Rubiscos from archaea, A. fulgidus RbcL2, T. kodakaraensis

(AB018555), M. jannaschii (AAB99239), M. acetivorans (AAM07894), Methanosarcina

mazei (AAM30945), Pyrococcus abyssi (CAB50122), Pyrococcus furiosus (AAL81280), and Pyrococcus horikoshii (BAA30036) were aligned using ClustalW (Thompson et al.,

1994). Of the 441 amino acids in A. fulgidus RbcL2, there are 107 amino acids that were identical among all eight archaeal Rubiscos. To determine the uniqueness of these 107 amino acids, all eight form III Rubisco sequences were then compared to a large representative group of form I and form II Rubiscos. By process of elimination of the

107 amino acids, 55 of these differed such that the form I and form II proteins did not contain the amino acid found in the form III Rubiscos. The remaining 52 amino acids had sequence identity with either the form I, the form II, or both forms of Rubisco. These

55 amino acids were then examined with respect to their position within the crystal

structures of representative form I (Synechococcus PCC 6301), and form II (R. rubrum)

Rubiscos, as well as to a homology model of the structure of the dimeric A. fulgidus

RbcL2 (Figure 2.9), with the focus on residues within 3-4 Å of the 17 conserved active

site residues, either upstream or downstream of these active site residues in the peptide

backbone or surrounding them, particularly the active site pocket, on neighboring peptide

backbone loops, α-helices, or β-strands with the hope of finding nearby residues that

could lead to clues influencing the oxygen insensitivity. Of these 55 amino acid

differences, two residues, Met-295 and Gln-309 of the protein were of particular interest

due to their position within the A. fulgidus RbcL2 model structure as well because of changes of these residues in form I and form II Rubiscos. Form I Rubiscos have a highly conserved phenylalanine residue and form II Rubiscos have a highly conserved alanine

43 residue at the same position as Met-295 in A. fulgidus RbcL2. The residue at this position in all three forms of Rubisco is approximately 3 Å from a highly conserved arginine residue, Arg-279 in A. fulgidus RbcL2, which has been previously implicated in

other Rubisco enzymes as being necessary for the binding and stabilization of the RuBP

substrate during the enzyme’s catalysis. A histidine residue in the form I and a glycine

residue in the form II Rubiscos are located at the same position as the Gln-309 residue in

the A. fulgidus RbcL2. Again, the residue at this position in all three forms of Rubisco is

approximately 3 Å from a highly conserved histidine residue, His-311 in A. fulgidus

RbcL2, which has also been previously implicated in other Rubisco enzymes as being

necessary for the binding and stabilization of the RuBP substrate during the enzyme’s

catalysis.

Figure 2.9. Quarternary structure prediction of A fulgidus RbcL2. The predicted structure of the A. fulgidus RbcL2 protein was modeled to the known structure of the previously crystallized T. kodakaraensis KOD1 RbcL (1GEH), using Deep View Swiss PDB Viewer, spdbv 3.7. The ribbon models of the two monomers are shaded cyan and orange, respectively. Within each monomer, the main features are the catalytically important residues, colored red, predicted to be within 3.3 Å of a bound five carbon transition-state molecule, shown to be as positioned in the same areas as other Rubisco enzymes. The first nine amino acids were not predicted and thus are not part of the model structure.

45 Met-295 of the A. fulgidus rbcL2 gene was altered according to established site- directed mutagenesis protocols, with the M295F recombinant A. fulgidus protein synthesized and purified to mimic the form I enzymes at this position and the M295A protein produced to mimic the form II enzymes. Finally, an M295D recombinant enzyme was synthesized and purified in order to introduce a charged amino acid that is not present at this position in any of the three forms of Rubisco. Gln-309 of the A. fulgidus rbcL2 gene was also altered at its position to mimic the residues found in form I (His) and form II (Gly) enzymes. No additional mutant was made at this position to introduce a different charged environment not encountered in the form I and form II enzymes. All five mutant proteins, M295A, M295D, M295F, Q309G and Q309H were initially analyzed in extracts prepared from small scale cultures (25 ml) along with the wild-type enzyme, produced from cells grown under exactly the same conditions. Large scale growths were also performed to obtain greater amounts of protein when required using the exact same procedure as the one used for over expression of the wild-type A. fulgidus

RbcL2 Rubisco. For the initial assays using the small scale growths, anaerobically prepared heat-treated supernatant fractions were used as the source of enzyme since significant amounts of heat labile E. coli proteins could be removed (Figure 2.2). Each sample was assayed under strictly anaerobic conditions to monitor activity; then an aliquot of the enzyme preparation was exposed to molecular oxygen and reassayed. The wild-type, M295A, M295F and both mutants of Q309 enzymes retained 11-26 percent activity compared to anaerobic controls. Surprisingly, the M295D enzyme showed the least sensitivity to oxygen exposure and retained roughly 60 percent of its activity after the same oxygen exposure regimen (Figure 2.10 and 2.11).

7000

6000 g

m 5000 n/ mi l/ o

m 4000 n

( Anaerobic y t i v

i Oxygen Exposed t

c 3000 A ic if

pec 2000 S

1000

0 Wild-type M295A M295D M295F Q309G Q309H Heat Stable Extract Samples

Figure 2.10. Comparison of carboxylase activity of A. fulgidus RbcL2 wild-type and mutant enzymes upon oxygen exposure. Heat stable extracts from 25 ml cultures were assayed initially under strict anaerobic conditions (Anaerobic). The enzyme was then exposed to molecular oxygen for 30 min and assayed in the presence of air (Oxygen Exposed). Assay conditions were the same in all cases using a strict anaerobic Rubisco assay with the exception of the oxygen exposed samples in which the assay vials were not sealed with rubber septas and crimped with aluminum caps.

47 70

60 e n i

a 50 t e R

y t

i 40 v ti

Ac 30 t n e c r 20 Pe 10

0 Wild-type M295A M295D M295F Q309G Q309H Heat Stable Extract Samples

Figure 2.11. Percent activity retained of A. fulgidus RbcL2 wild-type and mutant Rubiscos. Heat stable extracts from 25 ml cultures were assayed initially under strict anaerobic conditions. The enzyme was then exposed to molecular oxygen for 30 min and assayed in the presence of air. The percentage of activity retained is the amount of carboxylase activity measured in the oxygen exposed sample compared to the amount of carboxylase activity measured in the anaerobic sample.

These results prompted further studies of the M295D enzyme; accordingly large scale growths allowed significant amounts of purified recombinant protein to be prepared, much like the wild-type recombinant enzyme (Figure 2.2). Purified wild-type and M295D proteins were assayed both anaerobically and aerobically after 30 min exposure to oxygen. While the purified wild-type enzyme again showed 85-90% loss of activity upon oxygen exposure and assay in the presence of air (Figure 2.5), the M295D enzyme lost only 60-65% of its activity (Figure 2.12), similar to results obtained with the partially purified M295D protein from heat stable extracts.

83 rubber septasandcrim exception ofthe“oxygenexposed”sam conditions werethesam Figure 2.12.Recoveryof each ofthesubstrates(K deleterious effectsofoxygen. molecular oxygenwaschanged;thisenzym (Recovered). NoO containing oxygenexposedenzymeandinjected in thepresenceofair(OxygenExposed).An (Anaerobic). Theenzymewasthenexposedtomolecularoxygenfor30m exposure. Mutanten

° Specific Activity (nmol/min/mg) C. Therewaslittlechange in

10 12 14 16 2 4 6 8 000 000 000 000 000 000 000 000 Clearly, theM295Denzym 0 2 zy A scavengingsy ped withalum naer m CO e e carboxylaseactivityof inallcasesusing M295Dwasassaye obi 2 , K c Oxy O As withthewild-typeenzym 2 , andK th e wasalteredin e K inum g stem en CO p 49 RuBP les inwh Ex waspresentinthe“recovered 2 caps. andK posed ) weredeterm d e a strictanaerobicRubiscoassaywiththe

appearedm initiallyunderstrictan aliquotwasthentakenfrom A. fulgidus RuBP ich theassayvialswerenotsealedwith such awaythatth intoananaerobicvialandassayed Reco fortheM295Denzym v e ined fortheM r ed RbcL2 M295Duponoxygen u e ch lesssusceptibletothe , thekineticconstantsfor e norm aerobic con 295D enzym ” i vials.As n andassayed a e l responseto . However, thevial d itions e say at

in agreement with the recovery experiment, there was an approximate 5-fold increase in

the KO2, from 5 ± 1 µM for the wild-type enzyme to 24 ± 7 µM determined for the

M295D protein (Table 2.1). While double reciprocal plots at varying CO2 concentrations at several fixed levels of O2 for the M295D enzyme (Figure 2.13) showed more scatter than the wild-type enzyme (Figure 2.7), replots of the data gave a linear response such that accurate and reproducible kinetic constants could be determined (Figure 2.14).

Again, the M295D enzyme showed the expected competitive inhibition by O2 with respect to CO2. Furthermore, the M295D enzyme retained significantly more activity than the wild-type enzyme when both enzymes were incubated with concentrations of

oxygen ranging from 1% (4.21 µM) to 20% (84.2 µM) in the gas phase (Figure 2.15).

Levels of oxygen beyond 20% (up to 100%) yielded the same difference in the percent of

activity retained. Like the wild-type enzyme, the M295D enzyme recovered fully after oxygen-exposed enzyme was incubated with PCD and its substrate PCA in order to scrub out any oxygen carried over.

50 a a a a Enzymes Kc Ko KRuBP kcat Ω (VcKo/VoKc)

µM µM µM s-1

Wild-type 51 ± 8 5 ± 1 20 ± 5 23.1 4 ± 0.6

M295D 58 ± 11 24 ± 7 21 ± 3 17.7 13 ± 1

a Average of at least three independent assays.

Table 2.1. Kinetic properties of purified, recombinant wild-type and mutant M295D Rubisco from A. fulgidus RbcL2.

0.0006

) 0.0005

l/min/mg 0.0004

(nmo

0.0003

0.0002

1/Specific Activity 0.0001

.04 -0.02 0.00 0.02 0.04 0.06 1/[CO ] (µM) 2

Figure 2.13. Double reciprocal plot of RuBP carboxylase activity (in nmol/min/mg) of M295D A. fulgidus RbcL2 as a function of varying concentrations (µM) of CO2 at fixed levels (●, 0 µM; ○, 4.2 µM; ■, 21.0 µM; □, 42.1 µM) of O2. The assay conditions were as described in the experimental procedures. Vmax and Km values and lines were drawn using Sigma Plot v8.0.

51 0.025

) -1 ] 2

O 0.020

[C ∆ /

-1 vity 0.015

Specific Acti 0.010

∆ ope (

l S 0.005

0 1020304050 [O ] (µM) 2

Figure 2.14. Secondary replot of the slopes from the double reciprocal plot used to calculate the

Ki for O2 M295D A. fulgidus RbcL2. Lines were drawn using Sigma Plot v8.0.

100

d Wild-type e M295D in 60 a t Re y vit i t 40 Ac %

0 0 153045607590 [O ] (µ M) 2

Figure 2.15. Retention of carboxylase activity in the presence of oxygen. The wild-type and M295D enzyme were exposed to varying amount of oxygen and assayed for carboxylase activity. The percent activity retained is the difference in activity between the anaerobic sample compared to the oxygen exposed sample. The M295D enzyme retained significantly more activity than the wild-type enzyme when both enzymes were incubated with concentrations of oxygen ranging from 1% (4.21 µM) to 20% (84.2 µM) in the gas phase. 52 The low residual activity that is retained by the wild-type A. fulgidus enzyme

upon exposure to oxygen suggested that it would be feasible to determine the CO2/O2 specificity of the enzyme; i.e., the specificity factor or Ω value. Thus, experiments were

performed with both the wild-type and M295D enzymes, using established methods to

separate and quantitate the specific carboxylase and oxygenase reaction products ([3H]3-

PGA and [3H]2-PG, respectively) obtained from a reaction mixture containing [1-

3 H]RuBP and both gaseous substrates CO2 and O2 (Harpel et al., 1993). The results of

this experiment indicated that the wild-type A. fulgidus RbcL2 enzyme catalyzed, albeit

weakly and over a long time period, oxygen-dependent formation of [3H]2-PG (Figure

2.16A), which was not formed in the absence of oxygen (Figure 2.16C). Clearly, comparisons of the chromatographic profiles in the presence and absence of oxygen indicated that the level of [3H]3-PGA produced is greatly reduced under an oxygen

14 atmosphere, in agreement with the [ C]O2 incorporation assays showing inhibition of carboxylase activity in the presence of oxygen. In addition, it was apparent that the

M295D enzyme produced significantly more [3H]2-PG than the wild-type enzyme

(Figure 2.16B). Presumably, the increase in the KO2 or Ki for O2 for the M295D enzyme impinges on the fact that the carboxylation reaction is less inhibited in the presence of oxygen than the wild-type enzyme. The levels of [3H]3-PGA and [3H]2-PG produced, at the concentrations of O2 and CO2 utilized in this reaction, allowed a calculation to be

made of the relative CO2/O2 substrate specificity factor (Ω) for this archaeal enzyme and the M295D mutant protein (Table 2.1). The consequences of enhanced carboxylase activity in the presence of oxygen for the M295D enzyme, and the increase in the KO2, caused about a 3-fold increase in the Ω value compared to the wild-type protein.

Figure 2.16. Anion exchange chromatographic separation of Rubisco reaction products. [3H]3-PGA and [3H]2-PG generated from a completed reaction mixture containing [1- 3H]RuBP after 2 h reaction at 83°C. (A) Wild-type and (B) M295D A. fulgidus Rubisco 3 were incubated in the presence of both molecular oxygen and CO2 to generate [ H]3- 3 PGA and [ H]2-PG. (C) Wild-type A. fulgidus RbcL2 was incubated in the absence of O2 3 in the presence of CO2 for 2 h at 83°C to allow the generation of only [ H]3-PGA. Peaks at the beginnings of the chromatographic profiles represent degraded RuBP produced in this reaction mixture at high temperatures. Peaks at the ends of the profiles represent RuBP reduction products produced after the addition of NaBH4 to quench the reaction in the absence of enzyme (Harpel et al., 1993).

A 250 [3H]3-PGA [3H]2-PG 200

150 ) m (cp

ts 100 Coun 50

0 10203 040506070 Time (min) B 250 3 [ H]3-PGA [3H]2-PG 200

150 ) m (cp s

t 100

Coun 50

0 10203040506070 Time (min)

C 250 [3H]3-PGA 200

150 pm) c

s ( 100 unt

Co 50

0 10203040506070 Time (min)

55 To assess whether the role of the methionine to aspartic acid substitution at

position 295 in the enzyme might be unique, substitutions were considered at other

positions in close proximity to the active site. Similar to how Met-295 is positioned in

regards to the active site, Ser-363, roughly 10 Ǻ from Met-295 according to the modeled

structure, is situated in what appears to be a hydrophobic pocket that surrounds one side

of the active site. In addition, the model structure shows an ionic interaction of the side

chain of Ser-363 with highly conserved and catalytically important residues Gly-313 and

Thr-314 of A. fulgidus RbcL2. Gly-313 and Thr-314, found in all forms of Rubisco,

show no ionic interactions with the amino acid residue equivalent to Ser-363 of RbcL2 in

form I and form II enzymes. This unique interaction and positioning of Ser-363 in a key

hydrophobic pocket of RbcL2, similar to Met-295, thus suggested that Ser-363 of RbcL2

might be a likely candidate for further investigation by site-directed mutagenesis.

Sequence alignments of form I, II and III enzymes show that alanine is uniquely

conserved at this position in the form I enzymes and isoleucine is uniquely conserved at

this same position in the form II enzymes. In the form III enzymes, serine is present at

this position for A. fulgidus, T. kodakaraensis, and M. jannaschii, while the remaining methanogens and all of the pyrococci have an alanine present in this position. Because of the residues found in this position in the above enzymes, Ser-363 was changed to an alanine, isoleucine, and valine in A. fulgidus RbcL2. Initial analysis of heat stable

extracts as well as purified S363I and S363V enzymes indicated that recombinant RbcL2

mutated in this position exhibited less oxygen sensitivity compared to the wild-type

enzyme. However, the S363A enzyme did not show a change in oxygen sensitivity

compared to the wild-type enzyme. The S363I and S363V enzymes retained 48% and

56 42% activity when exposed to oxygen compared to the enzyme assayed under anaerobic

conditions (Figure 2.17). Clearly, these two mutations have a similar effect on oxygen

sensitivity as the M295D enzyme, perhaps indicating that substitutions at Met-295 are not

merely eliminating some chemical modification or oxidation effect of the methionine side

chain. To further probe the importance of the mutations at Met-295 and Ser-363, double

mutations were constructed and analyzed. When exposed to oxygen, the M295D/S363I

and M295D/S363V enzymes retained even higher levels of activity when compared to

the single mutations; i.e., 82% activity was retained for the M295D/S363I enzyme and

86% activity for the M295D/S363V enzyme (Figure 2.17). However, as a consequence

of changing these two residues near the active site, the absolute activity levels (specific

activities or kcat) of these two double mutations were much lower than the wild-type or single mutation enzymes.

100

n 70

ini 60 ma

Re 50

tivity 40

Ac 30

% 20

Wild-type M295D S363I S363V M295D/ M295D/ S363I S363V Anaerobica,b 10396 ± 637 12732 ± 376 15854 ± 692 15468 ± 919 615 ± 23 2715 ± 170

a,b O2-Exposed 1255 ± 18 6019 ± 664 7699 ± 855 6617 ± 170 505 ± 89 2361 ± 124

Figure 2.17. Wild-type and mutant A. fulgidus RbcL2 carboxylase activity remaining after exposure to oxygen. Wild-type, M295D, S363I, S363V, M295D/S363I, and M295D/S363V recombinant purified enzymes were assayed for activity under strictly anaerobic conditions. These preparations were then exposed to pure oxygen as described in Experimental Procedures and assayed in the presence of air. The percent activity retained is the amount of carboxylase activity obtained under anaerobic conditions compared to the amount of carboxylase activity obtained after oxygen exposure and assay in the presence of air. Absolute specific activities (nmol CO2 fixed/min/mg) obtained under both conditions are listed below each sample.

DISCUSSION

14 In previous studies, it was demonstrated via [ C]O2 radiometric assays and

Western immunoblots that crude cell extracts of A. fulgidus do indeed contain functional

Rubisco (Finn and Tabita, 2003). Since A. fulgidus is a thermophilic strict anaerobe

isolated from the bottom of the ocean near hydrothermal vents, it is not surprising that

Rubisco from this organism is adapted to function under similar extreme conditions in

vitro. Indeed this form III homodimer catalyzes a reaction with a kcat that is 4 to 5 fold higher than other forms of Rubisco. Of considerable interest, and unlike other well studied Rubiscos, is the substantial loss in carboxylase activity of the A. fulgidus enzyme in the presence of molecular oxygen, even when CO2 levels are in great excess. This was

found to be a reversible effect and in this study it was shown that full activity could be recovered so long as all vestiges of oxygen were removed form the reaction mixture.

Thus far, this response to oxygen has been observed only for certain form III archaeal

Rubiscos, including the enzymes from both mesophilic and thermophilic methanogenic archaea such as M. jannaschii and M. acetivorans (Finn and Tabita, 2003). The ability to isolate substantial amounts of purified recombinant A. fulgidus protein, coupled with the recent determination of the structure of the related Rubisco from T. kodakaraensis

(Maeda et al., 1999), suggested that it might be feasible to design experiments to elucidate the molecular basis for the unusual properties exhibited by the A. fulgidus

59 enzyme. Of particular interest is the extremely high kcat and oxygen sensitivity of this

enzyme. The response to molecular oxygen was clearly shown to be a classic

competition with CO2 for the enediolate intermediate of the enzyme, as observed for all

Rubisco proteins. However, what distinguished the A. fulgidus enzyme from other sources of Rubisco, was the extremely high affinity this enzyme showed to molecular oxygen, with Ki values (of about 5 µM), that were nearly 3 orders of magnitude lower than typical form I or form II enzymes. Clearly, this high affinity for molecular oxygen underscores why inhibition of carboxylase activity was initially obtained even in reaction mixtures that contained levels of CO2 that normally swamp out the inhibitory effects of oxygen for form I and form II Rubiscos.

Obviously, Rubisco from organisms like A. fulgidus is not ever expected to

encounter molecular oxygen. With the interesting in vitro response of this enzyme to

oxygen noted in this study, we proceeded to further analyze this protein with expectations

that such studies might eventually provide clues as to how the active site of Rubisco

evolved in more evolutionary advanced organisms. In analyzing the linear sequence of

the A. fulgidus (and other archaeal) Rubiscos compared to other well-studied form I and

form II enzymes, Met-295 was singled out for further attention. After altering this

residue by site-directed mutagenesis and preparing recombinant M295A, M295D, and

M295F proteins, it was apparent that the M295D enzyme showed substantially less

sensitivity to molecular oxygen than the wild-type protein. A homology model of the

homodimeric structure of the A. fulgidus RbcL2 was constructed (Figure 2.9), based on

the solved structure of the highly homologous (72 % amino acid identity) T.

kodakaraensis enzyme (Maeda et al., 1999). Like the large subunits of all Rubiscos,

60 known residues necessary for catalysis (Cleland et al., 1998) are conserved and are positioned within the A. fulgidus structure in the same locale as in other Rubisco structures. As for Met-295, it was found to be situated in a hydrophobic pocket created by residues along the active site and in close proximity to a highly conserved residue,

Arg-279 (Figure 2.1 and 2.18), found in all other forms of Rubisco and shown to be necessary for substrate (RuBP) binding (Zhang et al., 1994). In A. fulgidus RbcL2, there is no hydrogen bond to the Arg-279 residue (Figure 2.18A). However, there is definite hydrogen-bonding to the equivalent Arg residue in all other form I and form II Rubisco structures; e.g., originating from the oxygen atom of the carbonyl group of His-324 from the peptide backbone of the Synechococcus PCC6301 enzyme. In the A. fulgidus RbcL2 model, the distance between the corresponding arginine residue (Arg-279) to the carbonyl group of the equivalent histidine (His-311) of the peptide backbone is ~3.70 Å. The model structure suggests that a mutation to an aspartate residue at the Met-295 position would allow for an ionic interaction between one of the hydroxyl side chains of the aspartate residue with one of the side chain nitrogen atoms of Arg-279 (Figure 2.18B).

All the other amino acid mutations made at position 295 suggested either unfavorable conformations or no ionic interactions with Arg-279. In addition, many rotamers were available for the aspartic acid substitution at the methionine position; the rotamers with the lowest score, thus the most favorable conformation, all had hydrogen bonding interactions with Arg-279. Although this occurrence is seen within the model structure, perhaps stabilization of the Arg-279 residue is necessary for the carboxylation activity to function in the presence of oxygen, or perhaps there is some significance to the presence of hydrophobic pockets surrounding the active site. Thus, the amino acid side chain

61 situated in these pockets might play a role in the enzyme’s overall carboxylation activity in the presence of oxygen, as demonstrated for Met-295. This could perhaps be the reason why hydrogen bonding was observed in a different position in other solved crystal structures but not observed in the model of A. fulgidus RbcL2.

Figure 2.18. Predicted side-chain interactions with Met-295 in wild-type A. fulgidus RbcL2 and the mutant M295D enzyme. Side chains shown are amino acids Met-295 (A) and Asp-295 (B), as well as conserved amino acids found in all other forms of Rubisco. In A. fulgidus RbcL2 and the mutant M295D enzyme, His-278, Arg-279, and His-311, are illustrated as they are necessary for catalysis and binding of the five carbon substrate, RuBP. The model structure predicts no ionic interactions between Arg-279 and Met-295 in the wild-type form of the enzyme (A). In the M295D mutant, the model predicts an ionic interaction between the hydroxyl group of the Asp-295 residue and the amino group of the Arg-279 residue (dashed purple line).

63 A

64 Further investigation into this localized structural change led us to another amino

acid, Ser-363, which we predicted might have a similar affect on oxygen sensitivity. A

possible alteration to the hydrogen bond interactions with the highly conserved Arg-279,

independent of the suggested hydrogen bonding between M295D and Arg-279, might be

influenced by mutations to this Ser-363 residue. Again, the model structure indicates that

this amino acid is on β-strand 6, situated in a hydrophobic pocket adjacent to the active site. Alanine is strictly conserved at this same position in the form I enzymes and isoleucine in the form II enzymes. Form III enzymes have either a serine or an alanine.

For A. fulgidus RbcL2, the S363I and S363V mutant enzymes showed increased retention

of activity when enzymes were assayed in the presence of oxygen, while the S363A enzyme mimicked the wild-type enzyme and lost activity when assayed in the presence of oxygen. These results were quite reminiscent to what was found for enzymes containing substitutions at Met-295. Additionally, recombinant M295D/S363I and

M295D/S363V double mutations retained very high levels of activity when assayed in the presence of oxygen, suggesting an additive effect of mutations at two influential residues in hydrophobic pockets situated close to the active site. The mutations at Ser-

363 and the double mutations of M295D with the Ser-363 mutants will prompt further investigations of these interactions and how they influence the activity of the enzyme and more importantly the interaction with molecular oxygen. In addition, the potential to biologically select double mutants that retain high absolute activity is something that might be quite feasible and most revealing (Smith and Tabita, 2003).

CHAPTER 3

FURTHER STUDIES OF MUTANT RIBULOSE 1,5-BISPHOSPHATE

CARBOXYLASE/OXYGENASE PROTEINS FROM ARCHAEOGLOBUS FULGIDUS

AND A RELATED ARCHAEON.

INTRODUCTION

Previously, RuBP-dependent fixation of CO2 was detected in recombinant A.

fulgidus RbcL2 Rubisco synthesized in E. coli (Chapter 2, Kreel and Tabita, 2007). The

-1 wild-type enzyme has a very high specific activity (28.9 µmol/min/mg) and kcat (23 sec ) and is active at temperatures ranging from 23°C to 93°C, with optimal activity at 83°C.

The enzyme was also found to be extremely sensitive to oxygen, making it necessary to purify and assay the enzyme under strict anaerobic conditions. Additionally, incubation with low levels of oxygen resulted in loss of overall carboxylase activity, even in the presence of saturating amounts of CO2 (supplied as 20-50 mM sodium bicarbonate).

Typically, with other (form I and form II) Rubiscos, such high levels of CO2 overcomes the inhibitory effects of oxygen. However, activity of the A. fulgidus enzyme was found to recover when all traces of oxygen were removed from solution using an enzyme-based oxygen-scrubbing system. Experiments further revealed that this enzyme possesses an extremely low Ko (Km for oxygen). The value determined for the A. fulgidus enzyme (5 66 µM) is the lowest determined for any Rubisco; e.g., form I and II Rubiscos have values ranging anywhere from 500 – 1000 µM. It is thus apparent that the unusual sensitivity of

A. fulgidus RbcL2 to oxygen may be attributed to the fact that this form III archaeal enzyme has an extremely high affinity for oxygen.

Bioinformatic studies of all forms of Rubisco, in which primary sequences and solved crystal structures along with homology models were compared and revealed potential important residue differences within key regions of the form III archaeal enzymes. Amino acid substitutions of some of these residues in A. fulgidus RbcL2 were thus prepared and were shown to confer altered sensitivities to oxygen. Met-295, a residue that is conserved among all form III enzymes, was changed at the equivalent position to residues typically found in form I and form II proteins. In addition, Met-295 was also converted to a more polar, positively charged residue, aspartic acid. While Met-

295 alterations that mimic residues typically found in form I and II Rubisco did not produce any discernible changes in the enzyme’s sensitivity to oxygen, the M295D mutant A. fulgidus RbcL2 enzyme showed a significant change in its sensitivity to oxygen compared to the wild-type form of the enzyme; a nearly five-fold increase in the

Ko for oxygen (24 µM) was obtained for the M295D enzyme compared to the very low value (5 µM) for the wild-type enzyme.

In this chapter, additional mutations were made to further test the uniqueness of the kinetic changes observed in the M295D enzyme. The model structure of A. fulgidus

RbcL2 suggests that hydrogen bonding interactions exists between the side chain of

M295D with the side chain of Arg-279. Arg-279 is a highly conserved residue among all bona fide forms of Rubisco and is catalytically significant since it is essential for the

67 binding of the five carbon substrate, RuBP. In crystal structures of form I

(Synechococcus PCC6301 RbcL, 1RBL) and form II (R. rubrum CbbM, 5RUB) the equivalent arginine residue interacts via hydrogen bonding, with another highly conserved His residue that also is important for RuBP binding. However, in the model structure for A. fulgidus RbcL2, the interaction between these two residues Arg-279 and

His-311, does not appear to occur. The construction of mutant proteins to create this interaction was thus investigated and as a result of these considerations one major difference found in the A. fulgidus RbcL2 model structure compared to the form I and II

Rubiscos was noted. This is residue Ser-363. Interestingly, this residue resides in a hydrophobic pocket in which many of the constituent hydrophobic residues are conserved among form I, II and III Rubiscos. However, only in the form III archaeal Rubisco is there a serine positioned in the pocket; the residues found in form I and II enzymes are hydrophobic. Again, alterations in the A. fulgidus RbcL2 protein were made to mimic form I (Ala) and form II (Ile) enzymes; also an additional hydrophobic residue (Val) was changed at this position since prior studies with an A375V mutant of the form I

Synechococcus PCC6301 enzyme displayed interesting effects with respect to its ability to interact with oxygen (Satagopan, Scott, Smith and Tabita, submitted). The S363A

RbcL2 enzyme did not exhibit any significant changes in oxygen sensitivity compared to the wild-type enzyme, however the S363I and S363V proteins gave comparable oxygen sensitivity results to M295D. Moreover, the oxygen insensitivity of double mutants

(M295D/S363I and M295D/S363V) appeared to be additive, providing additional indications of the importance of both M295D and Ser-363 on the oxygen sensitivity of the enzyme.

68 The above mutant proteins were further investigated and additional kinetic and some structural studies (circular dichroism, CD) were conducted on M295D, S363I,

S363V, M295D/S363I and M295D/S363V. Furthermore, potential in vivo physiological effects of these mutant proteins was examined after complementing a Rubisco knockout strain of Rhodobacter capsulatus (strain SBI/II-), with genes that encode the various mutant A. fulgidus enzymes. These studies allow one to investigate the ability of these enzymes to confer some defined phenotypic change that can be related back to the properties of the enzyme. In addition, other residues in the A. fulgidus RbcL2 enzyme were constructed to further validate residues, Met-295 and Ser-363 as well as the double mutants M295D/S363I and M295D/S363V. To test if the effects on oxygen insensitivity observed for A. fulgidus RbcL2 may be observed with other form III arachaeal Rubiscos, similar mutations were made at equivalent sites in the Thermococcus kodakaraensis

RbcL. In T. kodakaraensis RbcL, these residues are Met-298 and Ser-366, which are located in a similar hydrophobic pocket as Met-295 and Ser-363 of the A. fulgidus enzyme.

MATERIALS AND METHODS

Plasmids, bacterial strains and growth conditions. Plasmids and bacterial

strains used are summarized (Table 3.1). In vitro studies of Archaeoglobus fulgidus

RbcL2 and Thermococcus kodakaraensis RbcL were performed with recombinant protein prepared from Escherichia coli. All cloning steps were performed in E. coli JM109

(Yanisch-Perron et al., 1985) prior to transformation into E. coli BL-21(DE3)

(Stratagene, La Jolla, California) for over expression of the A. fulgidus rbcL2 and T.

kodakaraensis rbcL genes. E. coli cultures were grown in Luria-Bertani (LB) media

containing 1% Tryptone, 0.5% Yeast Extract, and 1% NaCl (w/v). T. kodakaraensis rbcL

(Tk 2290, accession number NC_006624) was cloned directly from genomic DNA.

Primers designed with an NdeI

(5'GCATATGATGGTTGAGAAGTTTGATACGATATACGACTACTATGTTGACAA

GGGCTACG3') restriction site at the N terminus and a BamHI

(5'GCGGATCCTCAGACTGGAGTAACGTGACCCCACTTCTCCAGGG3') restriction

site at the C terminus were used to amplify the rbcL gene from T. kodakaraensis genomic

DNA using Pfu polymerase. The gene was ligated into pCR2.1-TOPO vector (Invitrogen)

and sequenced to determine if there were any PCR-incorporated mutations. Using the

NdeI and BamHI sites in that vector, the gene was subcloned into pET11a (Novagen). A.

fulgidus rbcL2 (Af 1638, accession number NC_000917) was cloned directly from

genomic DNA as previously described (Chapter 2; Kreel and Tabita, 2007). 70 In vivo studies of Archaeoglobus fulgidus RbcL2 were performed with after

complementation of the genes in Rhodobacter capsulatus SBI/II-. Construction of the

Rubisco deletion strain of R. capsulatus SBI/II-, from wild-type strain SB1003, has been

described (Paoli, Vichivanives and Tabita, 1998). All growth on plates and in liquid

media for R. capsulatus was at 30°C. R. capsulatus was grown aerobically on Peptone

yeast extract (PYE) plates or in a liquid SOC media containing Ormerod’s basal salts as previously described and both were supplemented with 1 mg/ml of nicotinic acid and 1 mg/ml of thiamine-hydrochloride (Smith and Tabita, 2003; Paoli et al., 1998; Falcone and Tabita, 1991; Ormerod et al., 1961). Antibiotics in PYE plates or in SOC media were used at the following concentrations: 100 mg/ml of rifampicin, 2 mg/ml of tetracycline, 10 mg/ml of spectinomycin, and 5 mg/ml of kanamycin. DL-Malate was added to 0.4% (w/v) for photoheterotrophic growth on plates and in liquid media.

Minimal medium plates for photoautotrophic growth, and minimal malate medium plates

(0.4% malate) for photoheterotrophic growth, were incubated in jars containing a

CO2/H2-generating system (5–6% CO2, BBL GasPak system, Becton Dickson

Microbiology Systems, Cockeysville, MD). In some cases, photoautotrophic plates were grown under conditions where jars were flushed for 15 min with premixed 20% CO2/80%

H2. All phototrophic jars contained a palladium catalyst to remove O2 from the

atmosphere, and all jars were incubated in water-baths in front of lights. Growth curves

for R. capsulatus SBI/II- liquid cultures were generated by obtaining absorbance readings at 660 nm at intervals of six to 12 h. Minimal media for photoautotrophic growth and minimal malate media (0.4% malate) for photoheterotrophic growth were set up in the anaerobic chamber in 25 ml sealed tubes containing 10 ml media. The tubes were capped

71 with rubber stoppers and crimped inside the chamber. For photoautotrophic growth, the

headspace was exchanged by vacuuming and sparging with premixed 20% CO2/80% H2

at 1 min intervals for three cycles, performed every 24 h. When cultures reached an A660 between 1.2 and 1.5, a turbidity range known to yield maximum Rubisco specific activity

(data not shown), the cells were harvested by centrifugation and then washed with 100 mM Bicine–NaOH (pH 8.3), 10 mM MgCl2, and the resultant cell pellets stored at -80°C.

Plasmid pRPS-MCS3 was constructed specifically for the complementation

system, as previously described (Smith and Tabita, 2003). This broad host range

expression plasmid contains was derived from pBBR1-MCS3 and contains a tetracycline-

resistance gene, a mobilization site, a lacZα gene with a multiple cloning site (MCS) as

well as the promoter region for the cbbM gene (pcbbM) from R. rubrum, including its

cognate transcriptional activator gene, cbbR, to help drive expression of genes cloned

into the multiple cloning site (Kovach et al., 1995). A pET11a clone of rbcL2 from A.

fulgidus that would facilitate directional cloning into pRPS-MCS3 was constructed by

amplification of rbcL2. Primers designed with a KpnI

(5'CGGGTACCGTTGAAGATAAAACTTCTATCCCCC3') restriction site at the N

terminus and a SacI (5'GCGGAGCTCTTAGATTGGCGTAACCCTGCCC3') restriction

site at the C terminus were used to amplify the rbcL2 gene from the pET11a plasmid

containing the gene using Taq polymerase. The gene was ligated into pRPS-MCS3 and

sequenced for PCR-incorporated mutations. The ligated plasmid was transformed in the

E. coli JM109 host strain. A. fulgidus rbcL2 wild-type and mutant genes were directionally cloned into the plasmid using blue-white screening to facilitate isolation of desired insertions.

72 The pRPS-MCS3 plasmid harboring A. fulgidus rbcL2 wild-type or mutant genes

was mobilized into R. capsulatus in trans. 10 ml R. capsulatus SBI/II- was grown chemoheterotrophically in SOC at 30°C to logarithmic stage (~2 days) prior to mating.

E. coli JM109, with transformed pRPS-MCS3 containing A. fulgidus rbcL2, was grown in 1 ml LB to logarithmic stage prior to mating. Helper strain E. coli HB101, which harbors the transfer genes required for conjugation on plasmid pRK2013, was grown in 1 ml LB to logarithmic stage (5-6 hours) (Paoli et al., 1998; Falcone and Tabita, 1991;

Falcone and Tabita, 1993; Figurski and Helinski, 1979). R. capsulatus SBI/II- was

harvested by centrifugation at 7,000 rpm for 10 min at 4°C. Cells were washed with 5

mls of phosphate buffer and centrifuged again as before. The cell pellet was resuspended

in 1 ml phosphate buffer. E. coli JM109 and HB101 were harvested at 16,000 rpm for 2

min in an Eppendorf 5415D bench top centrifuge. The cell pellets were washed with 1

ml phosphate buffer and centrifuge again as before. The 1 ml of R. capsulatus SBI/II- was used to resuspend each E. coli cell pellet. The mixture of cells was then centrifuged for 5 min in an Eppendorf 5415D bench top centrifuge. Approximately half of the supernatant was poured off and the cell pellet was resuspended by vortexing. The mating mixture was plated as concentrated as possible onto PYE and incubated for 36 h at 30°C.

Afterwards, an ample amount of the mating mixture was streaked out onto a PYE agar plate containing rifampicin and tetracycline (PYErif-tet). These plates were incubated for

three days before transconjugates were streaked for individual colonies onto a replicate

PYErif-tet plate.

Site-directed mutagenesis. Site-directed mutagenesis was performed using the

QuikChange site-directed mutagenesis kit from Stratagene (Papworth et al., 1996). The

73 Ser-363 TCA sequence within the A. fulgidus rbcL2 gene was replaced with GCA, ATC,

and GTA to obtain alanine, isoleucine, and valine, respectively. The Ile-312 ATA sequence within the A. fulgidus rbcL2 gene was replaced with GCA, TCA, and ACA to obtain alanine, serine, and threonine, respectively. The Met-298 ATG sequence within the T. kodakaraensis rbcL gene was replaced with GAC to obtain aspartate. The Ser-366

AGC sequence within the T. kodakaraensis rbcL gene was replaced with ATC, and GTC to obtain isoleucine and valine residues, respectively. Automated sequencing was performed to confirm the sequences of mutant genes using a 3730 DNA Analyzer system

(Applied Biosystems) at the OSU Plant-Microbe Genomics Facility. The mutant genes were inserted into fresh pET11a plasmid after digestion with NdeI and BamHI.

Over-expression of the A. fulgidus rbcL2 and T. kodakaraensis rbcL wild-type and mutant genes in small scale cultures for initial analysis. E. coli BL-21(DE3) cells with transformed pET11a vector containing A. fulgidus rbcL2 or T. kodakaraensis rbcL were grown to an OD600 of 0.4 using a 50 ml Erlenmeyer flask containing 25 ml LB

media at 37°C at 120 rpm to minimize aeration. The temperature of the media containing

the cultures was then raised to 42°C by placing the flasks in a water bath for 30 min to

facilitate expression of E. coli heat shock (chaperone) proteins encoded by the dnaJ,

dnaK and groEL genes. This serves to increase the amount of soluble recombinant protein. The cultures were then allowed to cool to room temperature before inducing with 0.1 mM Isopropyl-ß-D-thiogalactopyranoside (IPTG); the cultures were then shaken

at 120 rpm for 16 h at room temperature. Cells were harvested to remove LB media and

then washed with anaerobic wash buffer, either 100 mM N,N-bis(2-Hydroxyethyl) glycine (Bicine), pH 8.3, 10 mM MgCl2, 1 mM EDTA flushed with argon and stored in

74 an anaerobic chamber. Cells were centrifuged again in anaerobic centrifuge bottles

containing screw caps with rubber seals. Cell pellets were recovered in the chamber and

were then stored at -70°C before further protein purification by column chromatography.

Cells were resuspended in wash buffer and were disrupted using a Retsch MM200 cell

mill. Resuspended samples were mixed with glass beads (0.10 – 0.25 µm) in a 1:1 (v/w)

ratio and were placed in a 2 ml centrifuge tube with a screw cap containing rubber seals.

The samples were then placed in the cell mill and allowed to process for 9 min at a

frequency of 30 sec-1 at 4°C. The samples were then centrifuged at 13,100 x g for 10 min

at 4°C in an Eppendorf 5414R bench top centrifuge. Further processing of the samples

was performed for the thermostable enzymes from A. fulgidus and T. kodakaraensis. The

supernatant was collected in the anaerobic chamber and placed in 2.7 ml gas-tight and crimped glass serum vials and then taken outside of the anaerobic chamber and heat treated in an 80°C water bath for 15 min to precipitate labile E. coli proteins. After the heat step, the samples were placed on ice for 30 min. The samples were then taken back into the chamber and placed into fresh centrifuge tubes with rubber seals and centrifuged at 16,100 x g under the same conditions as previously mentioned. The heat stable supernatant was collected and used for further experiments.

Over-expression of the A. fulgidus rbcL2 and T. kodakaraensis rbcL gene and purification of recombinant proteins RbcL2 and RbcL. E. coli BL-21(DE3) cells with transformed pET11a vector containing A. fulgidus rbcL2 or T. kodakaraensis rbcL were grown using 2.8 l broad bottom flasks containing 2 l of LB media to an OD600 of 0.4 at 37°C and shaken at 120 rpm to minimize aeration. Growth conditions were identical to the small scale growth as previously described.

75 All preparation and manipulation of cell material was performed in an anaerobic chamber. Prior to column chromatography, cells were resuspended in wash buffer

supplemented with 10 mM phenylmethylsulfonyl fluoride (PMSF) and 50 µg/ml deoxyribonucleic acid I (DNase I) and were disrupted using a pressurized French pressure cell (at 110,000 kPa) flowing directly into a sealed anaerobic serum vial sparged with argon gas. The lysed cells were then centrifuged at 16,000 x g at 4°C for 20 min in

screw cap centrifuge tubes with rubber sealed caps. The supernatant was decanted into a

serum vial and placed in an 80°C water bath for 20 min and then allowed to cool on ice

for 1 h. The heat stable extract was transferred to a fresh screw cap centrifuge tube with

rubber sealed caps and centrifuged at 30,000 x g at 4°C for 30 min. Supernatant from

either the thermostable A. fulgidus RbcL2 or T. kodakaraensis RbcL-containing extracts

were syringe filtered using 0.22 µm filters before loading onto columns for further

purification via column chromatography.

Column chromatography was performed in the anaerobic hood using a Bio-Rad

BioLogic HR Workstation. Purifications were similar for A. fulgidus RbcL2 and T.

kodakaraensis RbcL. For A. fulgidus RbcL-2 and T. kodakaraensis RbcL, syringe

filtered heat-stable extract was loaded onto a Q-Sepharose strong anion exchange column

equilibrated with wash buffer supplemented with 50 mM NaHCO3 and 10 mM β-

mercaptoethanol (Buffer A). Samples were eluted using a gradient of 0 - 2 M NaCl in

Buffer A; recombinant Rubisco enzyme from both organisms elute at ~0.4 M NaCl.

Fractions were monitored for activity using a modified protocol of the standard Rubisco

assay under anaerobic conditions (Tabita et al., 1978). Fractions with activity were

pooled and concentrated with a Millipore 30,000 MWCO concentrator and loaded onto a

76 110 ml Superose-12 gel filtration column. Fractions with activity were pooled and

further purified based on hydrophobic interaction using a Phenyl-sepharose column.

Samples were eluted with decreasing salt starting with 2 M (NH4)2SO4 and again, both

recombinant elute at ~0.4 M (NH4)2SO4. Fractions with activity were pooled and concentrated with a 30,000 MWCO Millipore concentrator using a centrifuge and then loaded onto a 1 ml G-25 desalting column to remove any remaining (NH4)2SO4. Purified protein was stored in 20% glycerol at -70°C in anaerobically sealed serum vials.

Radiometric Rubisco Assays. Purified recombinant enzyme of the two organisms was assayed for activity under a strict anaerobic atmosphere unless otherwise noted. The previously described assay was used and modified to optimize carboxylase activity (Tabita et al., 1978). Buffers and substrates were bubbled with argon gas in sealed glass serum vials prior to use. In an anaerobic chamber, enzyme was prepared in glass serum vials in 100 mM Bicine-NaOH, pH 8.3, 10 mM MgCl2, 1 mM EDTA, and

0.4 M NaCl. Vials were sealed in the chamber and then placed in a React Therm III™ heating/stirring module (Pierce, Rockford, IL) set at 83°C for form III A. fulgidus RbcL2 or T. kodakaraensis RbcL after the addition of 50 mM NaHCO3 in buffer, containing a

14 final concentration of 2 µCi NaH[ C]O3. Reactions were initiated by the addition of anaerobic RuBP and stopped with the addition of aerobic propionic acid. Vials were unsealed and dried overnight in a vacuum oven at 65°C. Samples were resuspended in

200 µl 2 N HCl and counted in 3 ml scintillation cocktail using a Tri-Carb 2100TR

Liquid Scintillation Analyzer (Packard Instrument, Meriden, CT). The Bradford method was used to determine protein concentrations, BSA was used as the standard (Bradford,

1976).

77 Kinetic measurements. Purified enzymes were used for all kinetic measurements of kcat, KCO2, KO2, KRuBP, and Ω. The KCO2 was determined under strict anaerobic conditions using sealed vials as previously described with few modifications

14 (Smith and Tabita, 2003). Dilutions of NaH[ C]O3 were prepared in 100 mM Bicine–

propionic acid. Vials were unsealed and dried overnight in a vacuum oven at 65°C.

Products were resuspended in 2 N HCl and counted in scintillation cocktail. Results were

Unisys®.

The KRuBP was measured similarly to the KCO2, determined under strict anaerobic conditions in sealed serum vials at 83°C. Various concentrations of RuBP were prepared and sparged with argon gas. Assays were initiated with the addition of RuBP to the assay 78 vials containing activated enzyme and ran for 30 seconds then stopped by the addition of

aerobic propionic acid. Samples were dried overnight, resuspended and counted in

scintillation cocktail. Results were plotted using Sigma Plot 2002 v8.0, deriving the

KRuBP by fitting values to a hyperbolic curve.

Specificity was measured under conditions of saturating O2 (1.23 mM) with 200

mM NaHCO3 in 100 mM Bicine-NaOH (pH 8.3), 10 mM MgCl2. The concentration of

CO2 was calculated from the Henderson–Hasselbach relationship, as described above for

3 KCO2. Reactions were initiated by addition of [1- H]RuBP, and incubated at 83°C for 2 h.

The reaction was halted by the addition of 200 mM NaBH4 and incubated at room temperature for 15 min. Excess NaBH4 was consumed by the addition of 400 mM

Glucose and incubated for an additional 15 minutes at room temperature. Samples were

diluted with distilled water and products formed were separated from the enzyme by

centrifugation in a Millipore 10,000 MWCO concentrator. The samples were frozen at

-70°C until further use. Reaction products were separated with a MonoQ resin using a

Dionex DX500 chromatography system (Dionex Corporation, Sunnyvale, CA) and

detected with an in-line scintillation counter (IN/US b-Ram, Tampa, FL), as described

(Harpel et al., 1993).

Molecular modeling of A. fulgidus RbcL2. Modeling of the A. fulgidus RbcL2 was performed using Deep View Swiss PDB Viewer, spdbv 3.7 (Guex and Peitsch,

1997). The template used to model the dimer form of the enzyme was the Thermococcus kodakaraensis KOD1 crystal structure (PDB, 1GEH), the closest related Rubisco large subunit based which possesses 72% amino acid sequence identity, with A. fulgidus RbcL-

2 (Berman et al., 2000).

79 Western immunoblots using polyclonal antibodies to archaeal Rubisco

proteins. Antiserum directed against purified A. fulgidus Rubisco was prepared in

rabbits by Cocalico Biologicals, Inc. (Reamstown, PA) and Western immunoblots were used to test the specificity of the antiserum. Proteins resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (Laemmli, 1970) were transferred to polyvinylidene difluoride membranes (Immobilon-P; Millipore, Bedford, MA) according to directions supplied by the manufacturer using a BioRad Transblot semi-dry transfer cell (BioRad, Hercules, CA). Washes and incubations with antibodies were carried out as described using antibodies directed against the archaeal Rubisco that was used, at a dilution of 1:3000 (Towbin et al., 1979). Immunoblots were developed with the

Attophos detection reagent according to the manufacturer’s instructions (Amersham,

Buckinghamshire, England) and visualized with a Molecular Dynamics Storm 840 imaging system (Molecular Dynamics, Sunnyvale, CA).

Circular Dichroism (CD). Far-UV CD measurements were taken using an Aviv model 62A DS spectrometer, scanning from 190 to 260 nm at 1 nm intervals with a bandwidth of 1 nm and a 2 sec signaling average; three scans were averaged for each sample. Data were collected using 1 cm quartz cuvettes with screw caps containing rubber septas with temperature controller set to 83°C. All samples were prepared in the anaerobic chamber and protein concentrations were ~1 mg/ml in 20 mM Tris–HCl buffer, pH 8.3. Samples were allowed to equilibrate at 83°C for 30 min. Data was smoothed by applying a negative exponential 10 degree polynomial regression with SigmaPlot v8.0.

80 Plasmid or Strain Relevant Characteristics Reference

Plasmids

pET11a Dubendorff and pBR322 derivative containing T7 autogenes Studier, 1991

pET11a-AfulRbcL2 Finn and Tabita, pET11a with NdeI/BamHI - Af 1638 2003

pET11a-TkodRbcL Kreel and Tabita, pET11a with NdeI/BamHI - Tk 2290 2008

pRPS-MCS3 Broad host range vector derivative from Smith and Tabita, pBBR1-MCS3 containing pcbbM and cbbR 2003

pRPS-MCS3-AfulRbcL2 Kreel and Tabita, pRPS-MCS3 with KpnI/SacI - Af 1638 2008

pRPS-MCS3-MaceRbcL Finn and Tabita, pRPS-MCS3 with KpnI/SacI - Ma 4555 2003

pCR-2.1-TOPO- Kreel and Tabita, TkodRbcL pCR-2.1-TOPO with PCR product Tk 2290 2008

pRK2013 Fergurski et al., Harbors transfer genes required for conjugation 1979

Bacterial Strains

E. coli JM109 Appropriate strain used for routine cloning Yanisch-Perron and high quality miniprep DNA applications et al., 1985

E. coli BL21(DE3) High-level expression by IPTG induction of T7 Weiner et al., RNA polymerase from lacUV5 promoter 1994

Possesses phenotype which increases transformation E. coli XL-10 Gold Jerpseth et al., efficiency of DNA used in site-directed mutagenesis. 1997

E. coli HB101 Host strain harboring pRK2013 Hanahan, 1983

R. capsulatus SB1003 Yen and Marrs, Wild-type 1976

R. capsulatus SBI/II- cbbLS/ cbbM Paoli et al., 1998

Table 3.1. Plasmids and strains used in this study.

RESULTS

In this study, the focus is on additional residues involved in the reversible inhibition by low concentrations of oxygen. Ser-363, much like Met-295, appears to be located at an influential site within the structure of the protein; these two residues may interact with catalytically important residues to alter kinetic parameters in the enzyme.

Similar to how Met-295 is positioned in regards to the active site, Ser-363, roughly 10 Ǻ from Met-295 according to the modeled structure, is situated in what appears to be a hydrophobic pocket that surrounds one side of the active site. The hydrophobic pocket appears to be a highly conserved region among Rubisco enzymes; residues that surround this amino acid, Ser-363, are either conserved or identical hydrophobic residues (Table

3.2). In addition, the model structure shows an ionic interaction of the side chain of Ser-

363 with highly conserved and catalytically important residues Gly-313 and Thr-314 of

A. fulgidus RbcL2. Gly-313 and Thr-314, found in all forms of Rubisco, show no ionic interactions with the amino acid residue equivalent to Ser-363 of RbcL2 in form I and form II enzymes. This unique interaction and positioning of Ser-363 in a key hydrophobic pocket of A. fulgidus RbcL2, similar to Met-295, thus suggested that Ser-

363 of A. fulgidus RbcL2 might be a likely candidate for further investigation by site- directed mutagenesis.

82 Syn. PCC6301 CbbL R. rubrum CbbM A. fulgidus RbcL2 (form I) (form II) (form III)

Gly-326 Gly-323 Gly-313

Thr-327 Thr-324 Thr-314

Lys-331a Lys-329 a Lys-319 a

Leu-332 Leu-330 Leu-320

Val-373 Ile-365 Val-361

Pro-374 Pro-366 Pro-362

Ala-375 Ile-367 Ser-363

Ser-376 a Ser-368 a Ser-364 a

Gly-378 a Gly-370 a Gly-366 a

Ile-379 Met-371 Leu-367

Phe-391 Leu-383 Leu379

Gly-392 Gly-384 Gly-380

Val-396 Ile-390 Val-384

Leu-397 Leu-391 Ile-385

aCatalytic Residue

Table 3.2. Amino acid residues involved in the formation of a hydrophobic pocket in a specific region of the Rubisco enzyme and sequence comparison of these residues between forms I, II and III Rubiscos.

Further characterization and determination of kinetic parameters of mutants

of A. fulgidus RbcL2. In Chapter 2, initial characterization of single and double mutant recombinant proteins indicated that Ser-363 is a key residue in the hydrophobic pocket,

83 and this residue, along with Met-295, appeared to influence the response to oxygen

exposure. To further illustrate the differential effects of molecular oxygen, each of the

purified mutant proteins, along with wild-type RbcL2, was exposed to increasing levels of molecular oxygen. The mutant enzymes retained significantly more activity than the wild-type enzyme when all enzymes were incubated with concentrations of oxygen ranging from 10% (42.1 µM) to 100% (421 µM) in the gas phase (Figure 3.1). Clearly,

these enzymes were altered in such a way that the normal response to molecular oxygen

was changed; the mutant enzymes, especially the M295D/S363I protein, appeared much

less susceptible to the deleterious effects of oxygen.

As with the wild-type and M295D enzymes, the kinetic constants for each of the

substrates (KCO2, KO2, and KRuBP and Ω) were determined for the S363I, S363V,

M295D/S363I and M295D/S363V enzymes at 83°C. The results showed that there was

little change in the KCO2 for the S363I, S363V, M295D/S363I and M295D/S363V enzymes (Table 3.3). In agreement with the recovery experiment, there was an approximate 3-fold increase in the KO2, from 5 ± 1 µM for the wild-type enzyme to 18 ±

2 µM or 16 ± 1 determined for the S363I and S363V proteins, respectively (Table 3.3).

Surprisingly, there was a substantial increase in the KO2 for the double mutants

M295D/S363I and M295D/S363V with calculated values of 427 ± 96 and 91 ± 9,

respectively. These values are comparable to the calculated KO2 values for form I and II enzymes which are able to maintain carboxylase activity in the presence of oxygen

(Jordan and Ogren, 1981). Much like the initial studies with the M295D enzyme

(Chapter 2), replots of the data gave a linear response such that accurate and reproducible kinetic constants could be determined. Again, these mutant enzymes showed the 84 expected competitive inhibition by O2 with respect to CO2. Additional experiments were initiated to measure the affinity of the A. fulgidus enzyme for the RuBP substrate. The usual anaerobic methods were employed as described in Materials and Methods and the

KRuBP was determined at 83°C. The KRuBP values determined for the S363I, S363V,

M295D/S363I and M295D/S363V mutants were significantly higher than the wild-type

enzyme (Table 3.3). These KRuBP values are also higher than the range of reported values for other form I and form II Rubiscos (Tabita, 1999). The most notable difference was the extremely high KRuBP value for the double mutants; M295D/S363I and

M295D/S363V have values of 1646 ± 310 and 1381 ± 88, respectively. Thus it is

apparent that the low specific activity in these two mutants compared to the wild-type

enzyme is attributed to the change in the KRuBP. In addition, these extremely high values in comparison to the single mutants suggest that there is an additive effect on the KRuBP.

Finally, it is notable that the specificity for all the mutant enzymes increased relative to the wild-type protein, undoubtedly as a consequence of the increase in the Ko/Kc ratio

(decreased Kc/Ko).

100

ining 60

Rema

ty i 40

% Activ

0 0 100 200 300 400 500

[Oxygen] (µM)

Figure 3.1. Retention of carboxylase activity in the presence of oxygen. Wild-type (●), M295D (○), S363I (▼), S363V (∆), M295D/S363I (■) and M295D/S363V (□) enzymes were exposed to varying amount of oxygen and assayed for carboxylase activity. The percent activity retained is difference in activity between the anaerobic samples compared to the oxygen exposed sample. The M295D/S363I and M295D/S363V enzymes retained significantly more activity than all the other enzymes when all enzymes were incubated with concentrations of oxygen ranging from 10% (42.1 µM) to 100% (421 µM) in the gas phase.

86 a a a a Enzymes Kc Ko Kc/Ko KRuBP kcat Ω

(VcKo/VoKc) µM µM µM s-1

Wild-type 51 ± 8 5 ± 1 10.2 20 ± 5 23.1 4 ± 0.6

M295D 58 ± 11 24 ± 7 2.4 21 ± 3 17.7 13 ± 1

S363I 79 ± 3 18 ± 2 4.3 570 ± 94 9.5 9 ± 2

S363V 88 ± 7 16 ± 1 5.5 118 ± 6 12.4 8 ± 0.5

M295D/S363I 74 ± 5 427 ± 96 0.2 1646 ± 310 0.7 9 ± 1

M295D/S363V 62 ± 17 91 ± 9 0.7 1381 ± 88 4.3 9 ± 0.7

a Average of at least three independent assays.

Table 3.3. Kinetic properties of purified, recombinant wild-type and mutant Rubiscos from A. fulgidus RbcL2 assayed at 83°C.

Circular Dichroism (CD scans) of wild-type and mutant of A. fulgidus RbcL2

enzymes. In order to test whether or not large conformational changes accompanied the

observed activity changes and lower sensitivity to molecular oxygen, most notably in

double mutants M295D/S363I and M295D/S363V, CD scans were performed. Clearly, it

was important to determine if there was some facile way to test whether the wild-type

and mutant A. fulgidus enzyme’s structural integrity was affected under anaerobic versus

oxygen exposed conditions at 83°C. The enzyme was first dialyzed into 20 mM Tris-

HCl, pH 8.3 to remove the normal storage and assay buffer (100 mM Bicine pH 8.3, 10

mM MgCl2, 1 mM EDTA, 50 mM NaHCO3) since interference occurs at lower wavelengths (less than 190 nm) with Bicine buffer. The dialyzed enzymes were then assayed and it was verified that the 20 mM Tris-HCl buffer did not affect enzyme activity

87 (data not shown). CD scans were collected as described in Materials and Methods. All

samples were prepared under strict anaerobic conditions in the anaerobic chamber and

were placed in a 1 cm path length quartz cuvette with screw cap lid and rubber septa.

Data for the anaerobic samples were collected first (Figure 3.2). Variation in the molar

ellipticity between samples can be attributed to the deviation in protein concentration.

Clearly, there were no detectable significant alterations to secondary structures, α-helix,

β-sheets and loops. The cuvettes were then flushed with pure oxygen through the rubber septa for 30 minutes at room temperature and afterwards data was collected similar to the anaerobic samples at 83°C. Anaerobic versus oxygen exposed samples were compared within each sample (Figure 3.3).

Wavelength (nm) 210 220 230 240 250 0.0

/dmol) -2.0 2 cm

(deg -4.0 3 Wild-type M295D ty x 10 i S363I -6.0 S363V M295D/S363I M295D/S363V

Molar Elliptic -8.0

Figure 3.2. Far UV CD spectra of A. fulgidus RbcL2 wild-type and various mutant proteins under strict anaerobic conditions. Measurements were performed at 83°C as described in Materials and Methods. Wild-type (●), M295D (○), S363I (▼), S363V (∆), M295D/S363I (■) and M295D/S363V (□) samples were analyzed in 20 mM Tris-HCl at a protein concentration of ~1 mg/ml.

Wavelength (nm)

210 220 230 240 250 0.0

mol)

/d -2.0 2 cm x

(de -4.0

3 10

ty x i c -6.0 Anaerobic Oxygen Exposed

Molar Ellipti -8.0

Figure 3.3. Far UV CD spectra of wild-type A. fulgidus RbcL2 under anaerobic and oxygen exposed conditions. Measurements were performed at 83°C as described in Materials and Methods section. Anaerobic wild-type enzymes (●) was prepared anaerobically and measured in a quartz cuvette with a screw cap containing a rubber septa. The cuvette was then sparged with 100% oxygen and scanned (○). Samples were analyzed in 20 mM Tris-HCl at a protein concentration of ~1 mg/ml.

90 To supplement the CD results obtained with the wild-type and mutant A. fulgidus RbcL2 proteins, discontinuous nondenaturing PAGE (13%) gels were utilized to assess whether or not the mutant enzymes migrated in the same manner as the wild-type form. Electrophoretic migration of mutant Rubisco proteins is often altered and may be clearly observed with such gels (Read and Tabita, 1992; Ramage et al., 1998). Before loading onto the gel, samples were prepared fresh from -80°C glycerol stocks and assayed at 83°C to ascertain that the enzymes were still active. Like the CD studies, the initial results with gels performed in air on the laboratory bench indicated that changes in activity and sensitivity to molecular oxygen was not due to major changes in the structure of either the wild-type or mutant forms of this enzyme (Figure 3.4). Furthermore, PAGE gels were run under anaerobic conditions in the anaerobic chamber. All buffers and solutions needed to prepare the gels, as well as the samples, were prepared under anaerobic conditions by sparging the buffers and solutions with argon gas with the exception of the stock of 30:0.8 (%:%) acrylamide:bis-acrylamide. Results for aerobic nondenaturing PAGE gel electrophoresis were identical to gels performed anaerobically in the anaerobic chamber (data not shown).

Figure 3.4. Coomassie-stained discontinuous native PAGE of samples from A. fulgidus RbcL2 wild-type and mutant Rubiscos. The A. fulgidus rbcL2 gene was expressed in E. coli and samples were obtained either through partially purified heat stable extracts (where indicated) or FPLC column chromatography purification. 5 µg of each sample was loaded per lane as follows: A. fulgidus wild-type (lane 1); M295D (lane 2); S363I (lane 3); heat stable extract I312S (lane 4); M295D/S363I (lane 5); heat stable extract M295D/I312S/S363I (lane 6); Native protein standard (lane 7).

It was noted that heat treatment of crude E. coli extracts removed substantial amounts of extraneous protein (Figure 2.2, lane 3; Figure 3.4, lanes 4-6); moreover increasing the temperature from 83°C to 90°C, allowed even more unwanted proteins to be removed. The end-result was a fairly highly purified, yet not completely homogeneous, preparation of this thermophilic Rubisco. Inasmuch as there was a desire to examine the properties of multiple mutant proteins with a facile purification regimen, it was deemed essential to verify that the partially purified preparations behaved similar to

92 completely purified preparations. Thus, five mutant recombinant proteins, S363A,

S363I, S363V, M295D/S363I and M295D/S363V, along with the wild-type enzyme,

were initially prepared from small scale cultures (25 ml) grown under exactly the same

conditions. For these preliminary assays, anaerobically prepared heat-treated supernatant fractions were used as the source of enzyme. To monitor activity, each sample was assayed under strictly anaerobic conditions; then an aliquot of this enzyme preparation was exposed to molecular oxygen and reassayed. The wild-type and mutant S363A enzymes retained 10-17 percent activity compared to anaerobic controls. Interestingly, the S363I and S363V mutant enzymes showed the same relative insensitivity to oxygen exposure as the M295D enzyme and retained roughly 50-80 percent of its activity after the same oxygen exposure regimen. The double mutants M295D/S363I and

M295D/S363V showed even less sensitivity to oxygen than the single mutants and wild- type enzyme, retaining 85-98 percent activity after exposure to oxygen, although the overall catalytic activity of these double mutants was greatly affected. Results from anaerobic and oxygen exposed experiments with heat stable extract wild-type and mutant enzymes are summarized (Table 3.4). It was apparent that these assays indicated close coincidence to results obtained with purified preparations of the same proteins (Chapter

2, Table 2.17). Thus, it was concluded that heat-treated partially purified recombinant preparations could safely be used for screening, at least initially, the effects of several different mutations.

93 Anaerobic carboxylase O -exposed carboxylase % activity Enzymes 2 activitya activitya retained Wild-type 6484 645 10

M295D 6935 3682 53

S363A 7105 1230 17

S363I 4888 3813 78

S363V 8511 4219 50

M295D/S363I 267 227 85

M295D/S363V 1184 1167 98

anmol/min/mg

Table 3.4. Carboxylase activity at 83°C of heat stable extracts from A. fulgidus RbcL2 wild-type and mutant enzymes under anaerobic and oxygen-exposed conditions.

Additional mutations in the hydrophobic pocket. To assess whether the

interactions of the methionine to aspartic acid substitution at position 295 and serine to

isoleucine or valine at position 363 in the enzyme might be unique, substitutions were

considered at other positions in close proximity to these two residues. The objective was

to determine if a mutation close to these two residue positions would suppress the altered

changes observed in the mutant forms of the enzyme, either single or double, and revert

back to the kinetics observed in the wild-type form of the enzyme. In the model structure

of A. fulgidus RbcL2, Ile-312 resides on β-strand 6, most notably between His-311 and

Gly-313. Not only is His-311 catalytically important for the binding of RuBP during catalysis, there are implications in solved crystal structures of form I and II enzymes that the carbonyl group on the peptide backbone of this amino acid interacts and perhaps stabilizes a neighboring, highly conserved and catalytically important arginine residue

94 (Arg-279 in A. fulgidus RbcL2). However, according to the model structure of A.

fulgidus RbcL2, there are no indications that such an interaction exists (Kreel and Tabita,

2007). By contrast, Gly-313, another highly conserved residue that is involved in

catalysis, interacts with the neighboring, and highly conserved Thr-314, as well as with

Ser-363, according to the model structure of A. fulgidus RbcL2 (Figure 3.5). Comparing

the A. fulgidus RbcL2 model structure to existing form I and form II crystal structures

shows that Ser-363 interacts with both Gly-313 and Thr-314 but no interaction occurs

with residues equivalent to Ser-363 in form I and form II enzymes. Based on these

observations and the unique interactions observed in A. fulgidus RbcL2, Ile-312 was

deemed a good candidate for further studies. Ile-312 is directed away from the active site

and towards α-helix 6. In addition, there appears to be no interactions with surrounding

residues. Compared to form I and form II enzymes, the residues at this position are

alanine, serine or threonine; thus a less bulky hydrophobic or a charged residue could

have some impact in this localized region of the A. fulgidus enzyme (Figure 3.5). With these considerations in mind, Ile-312 was mutated to alanine, serine and threonine to mimic the residues found at the same position in form I and II enzymes. Single mutations as well as a combination of double mutants and triple mutants with M295D,

S363I or S363V were made. Again, small scale growth and heat stable recombinant proteins were prepared, however in this instance cells were disrupted with a cell mill instead of using the French Press. The cell mill provided a much easier means to process multiple samples under anaerobic conditions and was very reproducible and convenient for these studies. Heat stable enzymes were obtained from extracts after exposure to

90°C under anaerobic conditions. Initial analyses of heat stable recombinant extracts of

95 the single mutant enzymes I312A, I312S and I312T indicated that they all exhibited

about the same oxygen sensitivity compared to the wild-type enzyme (Table 3.5). The

I312A, I312S and I312T enzymes retained 30%, 16% and 15% of activity when exposed to oxygen compared to the enzyme assayed under anaerobic conditions. The double

mutants, M295D/I312A., M295D/I312S and M295D/I312T showed roughly the same

response to oxygen sensitivity, 40%, 46% and 45%, respectively, much like the

previously studied single mutant, M295D. Also, much like the absolute activity levels

(specific activities or kcat) from heat stable extracts of the M295D mutant being at or higher than the wild-type enzyme, the two double mutations also had the same levels of activity as the wild-type enzyme, unlike the results seen with the low levels of activity with the double mutants M295D/S363I and M295D/S363V enzymes (Table 3.5) (Kreel

and Tabita, 2007). Double mutations of I312A/S363I and I312S/S363I were constructed

and the oxygen sensitivity results (75 and 78% activity remaining, respectively, after

exposure of anaerobic preparations to oxygen) were similar to that obtained with a single

mutation to Ser-363, but not to the single Ile-312 mutant enzyme. This level of recovery

after oxygen exposure is a bit less than what was routinely obtained with the other double

mutant enzymes, M295D/S363I and M295D/S363V. To further probe the effect of

changes in residues Met-295, Ile-312 and Ser-363, triple mutations were constructed at

all three positions and analyzed. When assayed anaerobically, the M295D/I312A/S363I,

M295D/I312S/S363I, M295D/I312A/S363V and M295D/I312S/S363V enzymes all had substantially lower levels of activity when compared to the single as well as double mutations, even when assays were performed in the presence of extremely high concentrations of RuBP to eliminate the possibility of a severely altered KRuBP. The

96 specific activities of the M295D/I312A/S363I and M295D/I312S/S363I enzymes were

particularly low, 51 and 46 nmol/min/mg specific activity, respectively, while the

specific activities of the M295D/I312A/S363V and M295D/I312S/S363V enzymes was

somewhat higher, 146 and 140 nmol/min/mg, respectively. In summary, it appears that as a consequence of changing these three residues near the active site, absolute activity levels (specific activities or kcat) of single mutations were similar or slightly lower than the wild-type enzyme; double mutations, with the exception of M295D/S363I and

M295D/S363V were also similar to the wild-type enzyme, while the activity for the triple mutations were significantly lower than the wild-type, single and double mutant enzymes

(Table 3.5).

Gly-313 Ile-312 Thr-314

Ser-363

His-311

Figure 3.5. The hydrophobic pocket of Archaeoglobus fulgidus RbcL2 Rubisco showing interactions of Ser-363 with conserved residues Gly-313 and Thr-314. Ile-312 is situated on β–strand 6 between a highly conserved residue, His-311, necessary for the binding of RuBP and the Gly-313 residue that interacts with Ser-363 and is directed away from the active site towards the α–helix 6. The loop 6 structure that is important by folding over the active site during catalysis is in between the β–strand 6 and α–helix 6 and colored in red.

98 a a a Enzymes Anaerobic O2-exposed % activity retained Wild-type 1234 251 20 M295D 1825 796 44 S363I 1769 1345 76 S363V 2543 1158 46 I312A 957 284 30 I312S 1323 208 16 I312T 1503 230 15 M295D/S363I 87 81 93 M295D/S363V 248 224 90 M295D/I312A 1297 523 40 M295D/I312S 1109 507 46 M295D/I312T 2246 1003 45 I312A/S363I 1488 1117 75 I312S/S363I 1197 943 78 M295D/I312A/S363I 51 46 89 M295D/I312S/S363I 46 40 87 M295D/I312A/S363V 146 103 70 M295D/I312S/S363V 140 104 74 anmol/min/mg

Table 3.5. Carboxylase activity at 83°C of heat stable extracts from A. fulgidus RbcL2 wild-type and mutant enzymes under anaerobic and oxygen exposed conditions.

Complementation of bacterial Rubisco deletion mutants with the A. fulgidus

rbcL2 gene. A Rubisco deletion mutant (strain SBI/II-) of R. capsulatus cannot grow

under photoheterotrophic or photoautotrophic conditions. Previous studies in this

laboratory indicated that introduction of form III mesophilic archaeal Rubisco genes

allowed growth of strain SBI/II-. In particular, Methanococcus acetivorans rbcL was able to complement growth under photoheterotrophic and photoautotrophic conditions so long as strict measures were taken to insure anaerobicity. Unfortunately, Methanocaldococcus jannaschii rbcL failed to complement growth under either of these growth conditions.

The lack of complementation for the latter archaeal gene is most likely due to the fact R. capsulatus grows optimally at 30°C while the M. jannaschii Rubisco shows little or no

99 activity at temperatures at or below 65°C (Finn and Tabita, 2003). Recently, growth complementation with the form III archaeal Rubisco gene from Thermococcus kodakaraensis was accomplished with a strain of Rhodopseudomonas palustris in which the three endogenous sets of Rubisco genes were knocked out (Yoshida et al., 2006).

Since A. fulgidus RbcL2 has a temperature range of activity from 23°C to 93°C, much

like its homolog T. kodakaraensis Rubisco, attempts were undertaken to complement R.

capsulatus SBI/II- with the A. fulgidus rbcL2 gene. Photoheterotrophic (malate minimal media) and growth tubes used in these experiments were allowed to equilibrate in the anaerobic chamber; in addition all inoculation into the culture tubes was performed in a strictly anaerobic environment. Under such conditions, it was found that the A. fulgidus

- rbcL2 gene was expressed in R. capsulatus strain SBI/II and that this recombinant archaeal enzyme could complement growth under anaerobic photoheterotrophic conditions (Figure 3.6). In addition, SDS-PAGE and Western immunoblot experiments provided further evidence that the A. fulgidus gene was expressed to low levels (as expected) in the R. capsulatus Rubisco deletion strain under anaerobic photoheterotrophic growth conditions (Figure 3.7). Furthermore, radiometric Rubisco assays performed on crude extracts from photoheterotrophically grown cells indicated that Rubisco activity levels were low under these conditions and was strictly dependent on anaerobiosis. The levels of activity were substantially enhanced when assays were performed at 83°C, in keeping with the observed properties of the A. fulgidus enzyme

(Table 3.6). Since Rubisco knockout strains of Rhodobacter do eventually develop adaptive mutations over long time intervals that enable photoheterotrophic growth in the absence of the CBB pathway, we determined whether A. fulgidus rbcL2 expression would

100 allow photoautotrophic growth of Rhodobacter Rubisco knockout strains (Wang et al.,

1993). The CBB pathway and a functional Rubisco are absolutely required when these organisms are grown photoautotrophically under a H2/CO2 environment (Paoli et al.,

1995; Tabita, 1999). Again, complementation to photoautotrophic growth depends on strict anaerobiosis; sealed and crimped tubes prepared in the anaerobic chamber were used in these experiments. A. fulgidus RbcL2 Rubisco supported growth in the R. capsulatus Rubisco knockout strain under photoautotrophic conditions under a gas phase of 20% CO2 balanced with hydrogen (Figure 3.8). What is most striking is the extensive lag in growth observed for the R. capsulatus SBI/II- complemented with the wild-type A.

fulgidus RbcL2 enzyme. This was not observed during photoheterotrophic growth.

Experiments using the “oxygen-scrubbing” assay with protocatechuate 3,4-dioxygenase

(PCD) were conducted to test whether or not there were any vestiges of air present in the

growth tubes, particularly since the wild-type A. fulgidus RbcL2 enzyme is much more

oxygen sensitive than the mutant enzymes. Thus far, oxygen was not detected and there

was no indication for the formation of the product of the oxygenation of protocatechuic

acid (3,4-dihydroxybenzoic acid), namely β-carboxymuconic acid, catalyzed by PCD. In

addition, SDS-PAGE and Western immunoblot experiments provided further evidence

that the A. fulgidus gene was expressed in the R. capsulatus Rubisco deletion strain under

anaerobic photoautotrophic growth conditions (Figure 3.9). Again, radiometric Rubisco

assays were performed on crude extracts from photoautotrophically grown cells similar to

those performed with crude extracts from photoheterotrophically grown cells (Table 3.6).

Radiometric Rubisco assays proved the complemented strains had Rubisco activity under

the phototrophic conditions that was greatly enhanced at 83°C. After both

101 photoheterotrophic and photoautotrophic growth of the complemented strains, plasmid

pRPS-MCS3 (containing the A. fulgidus rbcL2 gene) was re-isolated from R. capsulatus

SBI/II- using a plasmid miniprep kit and sequenced to determine if there were any point mutations that may have been selected under these growth conditions. No mutations

- were detected. Subsequent reintroduction of the plasmid into a new SBI/II background

again resulted in photoheterotrophic and photoautotrophic growth, eliminating the

- possibility that mutations in strain SBI/II somehow allowed complementation.

102

2.5

2.0

1.5

nm)

1.0 OD (660

0.5

0.0 0 50 100 150 200 250 300 350 Time (hrs)

Figure 3.6. Complementation and growth of R. capsulatus cbbLS/cbbM knockout strain SBI/II- using the A. fulgidus wild-type and mutant Rubisco (rbcL2) genes delivered with plasmid pRPS-MCS3MA. Photoheterotrophic growth using malate as the carbon source was performed in glass tubes prepared under anaerobic conditions. Wild-type R. capsulatus SB1003 (●); R. capsulatus strain SBI/II- (○); R. capsulatus strain SBI/II- containing plasmid pRPS-MCS3 and wild-type A. fulgidus rbcL2 (▼) or mutants M295D ( ); S363I (■); S363V (□); M295D/S363I (♦); M295D/S363V (◊); M295D/I312A/S363V (▲); M295D/I312S/S363V (∆).

103

RbcL2

1 2 3 4 5 6 7 8 9 10 11 12 13

Figure 3.7. Coomassie-stained SDS-PAGE (top) and Western immunoblot (bottom) of extracts of R. capsulatus SBI/II- complemented with plasmid pRPS-MCS3-AfulRbcL2 (containing A. fulgidus rbcL2). The immunoblot was tested using antibodies directed against purified recombinant A. fulgidus RbcL2 Rubisco. All lanes contained soluble crude extract prepared from photoheterotrophically-grown stationary phase cultures grown photoheterotrophically from the following: wild-type R. capsulatus strain SB1003 (lane 2); wild-type R. capsulatus SBI/II- containing pRPS-MCS3 with no insert (lane 3); R. capsulatus SBI/II- complemented with plasmid pRPS-MCS3-MaceRbcL (containing M. acetivorans rbcL) (lane 4); R. capsulatus SBI/II- complemented with plasmid pRPS- MCS3-AfulRbcL2 (containing A. fulgidus rbcL2) (lane 5); R. capsulatus SBI/II- complemented with plasmid pRPS-MCS3-AfulRbcL2 mutated to M295D (lane 6); S363I (lane 7); S363V (lane 8); M295D/S363I (lane 9); M295D/S363V (lane 10); M295D/I312A/S363V (lane 11); M295D/I312S/S363V (lane 12); and purified recombinant A. fulgidus Rubisco (lane 13). Each lane received approximately 2 µg of protein. BioRad Low Range Molecular Weight Standard was used as the marker in lane 1.

104

1.8

1.6

1.4

1.2

nm) 1.0

0.8

OD (660 0.6

0.4

0.2

0.0 0 100 200 300 400 500

Time (hrs)

Figure 3.8. Complementation and growth of R. capsulatus cbbLS/cbbM knockout strain - SBI/II using A. fulgidus wild-type and mutant Rubisco (rbcL2) genes in plasmid pRPS- MCS3MA. Photoautotrophic growth was performed under an atmosphere of 20% CO2/80% H2 in glass tubes prepared under anaerobic conditions. Wild-type R. capsulatus SB1003 (●); R. capsulatus strain SBI-II- containing plasmid pRPS-MCS3 and A. fulgidus rbcL2 wild type (○) or mutants M295D (▼); S363I (■); and S363V ( ).

105

1 2 3 4 5 6

Figure 3.9. Coomassie-stained SDS-PAGE(top) and Western immunoblot (bottom) of extracts of R. capsulatus SBI/II- complemented with plasmid pRPS-MCS3-AfulRbcL2 (containing A. fulgidus rbcL2). The immunoblot was tested using antibodies directed against purified recombinant A. fulgidus RbcL2 Rubisco. All lanes contained soluble crude extract prepared from stationary phase cultures grown photoautotrophically from the following: wild-type R. capsulatus strain SB1003 (lane 2); R. capsulatus SBI/II- complemented with plasmid pRPS-MCS3-AfulRbcL2 (containing A. fulgidus rbcL2) (lane 3); R. capsulatus SBI/II- complemented with plasmid pRPS-MCS3-AfulRbcL2 mutated to M295D (lane 4); S363I (lane 5); S363V (lane 6). Each lane received approximately 2 µg of protein. BioRad Low Range Molecular Weight Standard was used as the marker in lane 1.

106

Enzymes Photoheterotrophic Photoautotrophic 30°Ca 83°Ca 30°Ca 83°Ca Wild-type 14 353 226 3797 M295D 10 218 129 1805 S363I 0.2 22 86 1237 S363V 9 566 93 1896 M295D/S363I 0.5 60 NG NG M295D/S363V 1.2 268 NG NG M295D/I312A/S363V ND 11 NG NG M295D/I312S/S363V 0.2 20 NG NG anmol/min/mg

Table 3.6. Specific activity of crude soluble wild-type and mutant A. fulgidus RbcL2 obtained from R. capsulatus SBI/II- grown photoheterotrophically and photoautotrophically. ND represents assays where no activity was detected. NG represents samples that were unable to grow under the specific growth condition thus assays were unable to be performed.

Initial characterization of the form III Thermococcus kodakaraensis RbcL

Rubisco. The next experiment was to determine if the residues that affect the response to molecular oxygen for the A. fulgidus RbcL2 play a similar role in other Rubiscos. T. kodakaraensis RbcL is a form III archaeal Rubisco that is a close homolog to A. fulgidus

RbcL2 (72% sequence identity). Moreover, the crystal structure of the T. kodakaraensis

RbcL has been solved and of course this protein has served as the template for the molecular modeling of A. fulgidus RbcL2. When aligned, Met-298 and Ser-366 in T. kodakaraensis RbcL corresponds to Met-295 and Ser-363, respectively of A. fulgidus

RbcL2 (Figure 3.10). Using site-directed mutagenesis protocols described in Materials and Methods, Met-298 was changed to an aspartic acid residue and Ser-366 was altered to either an isoleucine or valine residue. Single mutations of M298D, S366I and S366V, double mutations of M298D/S366I and M298D/S366V and the wild-type form of the T.

107 kodakaraensis enzyme were produced and small scale growth using E. coli BL-21(DE3) and preparation of recombinant protein were performed as previously described in

Materials and Methods.

CR R III A.fulgidus RbcL2 IHGHRAMHAAFTR-NAKHGISMFVLAKLYRIIGIDQLHIGTA 315 III T.kodakaraensis RbcL IHGHRAMHAAFTR-NPYHGISMFVLAKLYRLIGIDQLHVGTA 318 :* **. *.. . * .* *: * . :* **

R R RR III A.fulgidus RbcL2 IKPAMPVSSGGLHPGNLEPVIDALG-KEIVIQVGGGVLGHPMG 397 III T.kodakaraensis RbcL IKAAFPTSSGGLHPGNIQPVIEALG-TDIVLQLGGGTLGHPDG 400 : * ***:: : ..: :* : :: ***. ** *

Figure 3.10. Partial amino acid sequence alignment of A. fulgidus and T. kodakaraensis archaeal form III Rubiscos. Multiple sequence alignments were performed by using ClustalW (Thompson et al., 1994). Residue identities are marked with an asterisk, conserved substitutions are marked with a colon, and semiconserved substitutions are marked with a period. Known active-site and highly conserved residues are labeled C for catalytic and R for RuBP binding properties and colored red. Amino acids colored blue are identical to the position of either Met-295 or Ser-363 residues in the model structure of A. fulgidus RbcL2.

108 The initial analyses of these mutated recombinant T. kodakaraensis Rubiscos was

carried out using heat-treated extracts obtained from small scale E. coli cultures. Cell

pellets retrieved from 25 ml growth cultures were disrupted using a cell mill as described

in Materials and Methods. Since T. kodakaraensis is a hyperthermophilic organism the

RbcL enzyme had the same basic heat stability characteristics properties as A. fulgidus

RbcL2. Thus, the enzyme was highly purified after simply heating crude extracts to

90°C for 15 min followed by immersion into an ice bucket and subsequent removal of

denatured protein by centrifugation. Assays indicated that such preparations of wild-type

T. kodakaraensis RbcL possessed high specific activity at high temperatures (83°C)

under strict anaerobic conditions similar to activity observed with heat stable extracts

obtained with the wild-type enzyme from A. fulgidus RbcL2. Likewise, after exposure to

oxygen, activity decreased 37% compared to enzyme kept anaerobic (Table 3.8). This

loss of activity upon oxygen exposure was not nearly as severe as that obtained with the

wild-type A. fulgidus RbcL2 enzyme (10-15% activity remaining) under the same conditions (Table 3.8, Chapter 2, Kreel and Tabita, 2007). These initial results prompted further studies on the effects of molecular oxygen with purified wild-type and mutant T. kodakaraensis RbcL. Thus, over-expression of the various T. kodakaraensis rbcL genes was performed with large scale cultures and the resultant heat stable recombinant RbcL proteins purified to homogeneity by column chromatography as described in Materials and Methods. These preparations were examined by SDS and nondenaturing PAGE to check for purity and the basic migration properties of the native enzymes. SDS-PAGE indicated that the enzyme was of high purity and most contaminating enzymes were removed (Figure 3.11). The native PAGE gel indicated that the T. kodakaraensis RbcL

109 enzyme migrated much slower than the A. fulgidus enzyme, which is a dimer (Figure

3.12). The electrophoretic migration of the T. kodakaraensis RbcL is consistent with its reported decamer structure; i.e., a pentamer of dimers (Ezaki et al., 1999; Kitano et al.,

2001).

Figure 3.11. Coomassie-stained SDS-PAGE of samples from T. kodakaraensis RbcL Rubisco purification. The T. kodakaraensis rbcL gene was expressed in E. coli and samples obtained from: uninduced E. coli cells (lane 2); soluble extract of French Press disrupted E. coli cells after induction (lane 3); supernatant obtained after centrifuging the heat-treated (90°C) extract for 20 min at 80°C (lane 4); Q-Sepharose anion exchange chromatography (lane 5); Superose-12 gel filtration (lane 6); phenyl-Sepharose hydrophobic chromatography (lane 7). Lane 1 contains the low range SDS protein standards.

110

Figure 3.12. Coomassie-stained native PAGE of samples from A. fulgidus RbcL2 and T. kodakaraensis RbcL Rubiscos. Native PAGE protein standard (lane 1); A. fulgidus RbcL2 (lane 2); T. kodakaraensis RbcL (lane 3).

Under strictly anaerobic conditions with assays performed at 83°C, a high specific activity of 20.0 µmol/min/mg for the purified wild-type T. kodakaraensis RbcL enzyme was achieved. After exposure to molecular oxygen via the usual protocols, the specific activity of the purified enzyme was 49% of that compared to enzyme maintained under anaerobic conditions. Although the difference in activity of anaerobic and oxygen exposed samples is 3-5 times higher than the differences observed with the wild-type A. fulgidus RbcL2, the T. kodakaraensis RbcL is clearly oxygen sensitive. In part these studies were undertaken because of the rather surprising CO2/O2 substrate specificity values of 290-310 at 80 to 90°C previously reported (Ezaki et al., 1999). Such results would indicate a stupendously highly favorable carboxylase reaction that is not inhibited by the presence of oxygen. Using precisely defined conditions (Materials and Methods)

111 CO2/O2 specificity (Ω) values of the purified wild-type T. kodakaraensis RbcL enzyme was determined. The Ω was found to be 6 ± 0.2 (Table 3.7, Figure 3.13A). Attempts to

assay the enzyme under the high CO2 concentration conditions described by Ezaki et al. were performed at 83°C; however the correct specificity value (Ω) could not be accurately calculated because these authors did not specify the precise concentrations of gaseous CO2 and O2 substrate concentrations used in their assays. Thus, even though the resulting chromatogram displays formation of the characteristic separated product peaks

(3-PGA and 2-PG), and the expectedly large 3-PGA peak (Figure 3.13B and C), it was not possible to accurately calculate the CO2/O2 specificity (Ω) value for both T. kodakaraensis RbcL and A. fulgidus RbcL2 under these conditions, simply because the concentrations of the gaseous substrates were not specified.

112

Figure 3.13. Anion exchange chromatographic separation of Rubisco reaction products. [3H]3-PGA and [3H]2-PG generated from a completed reaction mixture containing [1- 3H]RuBP after 2 h reaction at 83°C. T.kodakaraensis Rubisco was incubated in the 3 3 presence of both molecular oxygen and CO2 to generate [ H]3-PGA and [ H]2-PG under (A) defined conditions as described in Materials and Methods and (B) under conditions previously described (Ezaki et al., 1999). In (C), wild-type A. fulgidus RbcL2 was assayed under the same conditions previously described (Ezaki et al., 1999). Peaks at the beginnings of the chromatographic profiles represent degraded RuBP produced in this reaction mixture at high temperatures.

113 A 250

[3H]3-PGA [3H]2-PG 200

150 ) pm c

( s

t 100 un

o C 50

0 102030405060 TIme (min)

B 250

200 [3H]3-PGA [3H]2-PG 150

)

pm (c

ts 100 n Cou 50

0 102030405060 Time (min)

C 250

200 3 3 [ H]3-PGA [ H]2-PG 150

(cpm) s 100 Count 50

0 102030405060 Time (min)

114 Since the specificity value (Ω) for the wild-type T. kodakaraensis RbcL Rubisco

at 83°C could be determined under defined conditions, it was desirable to determine the

key kinetic constants for each of the substrates (KCO2, KO2, and KRuBP). The usual anaerobic methods were employed at 83°C. The KCO2 value was determined to be 79 ± 5

µM (Table 3.7). To calculate the KO2, one fixed concentration of pure oxygen (100%,

421 µM at 83°C) was injected into vials that were then assayed with varying amounts of

CO2. This experiment also yielded the KCO2 value. Similar to kinetic assays performed

for the determination of KO2 for A. fulgidus RbcL2 (Figure 2.8, Chapter 2), double

reciprocal plots clearly showed that O2 was a competitive inhibitor with respect to CO2.

In addition, replots of the data from the double reciprocal plots indicated that the KO2 was

43 ± 2 µM (Table 3.7). Likewise, the KRuBP was determined to be 14 ± 2 µM (Table 3.7).

Unlike previous experiments with this enzyme (Ezaki et al., 1999), these results clearly demonstrate that the basic kinetic properties of T. kodakaraensis RbcL are very similar to those recently determined for A. fulgidus RbcL2. In retrospect, this does not appear to be unusual due to the fact that these two enzymes are so closely related (i.e., 72% sequence identity at the amino acid level). Moreover, previously reported specificity values of 310 at 90°C would indicate that this enzyme would be virtually insensitive to oxygen and would represent an enzyme that retained only carboxylase activity with the highest reported specificity value for any Rubisco. While this remains a dream of Rubisco biochemists, the data reported here does not support this claim (Ezaki et al., 1999).

115 a a a a Enzyme Kc Ko Kc/Ko KRuBP kcat Ω (VcKo/VoKc)

µM µM µM s-1

T.kodakaraensis 79 ± 5 43 ± 2 1.8 14 ± 2 16.6 6 ± 0.2 RbcL a Average of at least three independent assays. Values are reported in nmol/min/mg

Table 3.7. Kinetic properties of purified, recombinant wild-type Rubisco from T. kodakaraensis RbcL assayed at 83°C.

The effect of mutations in T. kodakaraensis RbcL. To assess if the amino acid

residues previously found to influence the response to oxygen for A. fulgidus RbcL2

might have similar or different effects on T. kodakaraensis RbcL, mutations were made

at positions Met-298 and Ser-366 in T. kodakaraensis RbcL. The residue changes are

equivalent to single and double mutations made in A. fulgidus RbcL2. The pET11a

plasmids harboring the T. kodakaraensis rbcL mutant genes M298D, S366I, S366V,

M298D/S366I and M298D/S366V, were transformed into E. coli BL21(DE3) and the

genes over-expressed in small scale cultures as described previously. Heat stable extracts were prepared and samples assayed at 83°C under strict anaerobic conditions as well as after exposure to molecular oxygen. The results (Table 3.7) were similar to what had previously been obtained for heat stable extracts of wild-type A. fulgidus RbcL2 (Table

3.4). The M298D, S366I and S366V enzymes had slightly lower activities compared to wild-type RbcL, however these samples retained higher activities after oxygen exposure, averaging 51, 75, and 71%, respectively, compared to the wild-type enzyme. The double mutants of M298D/S366I and M298D/S366V lost a significant amount of activity compared to the wild-type enzyme, however they still maintained high levels of activity after oxygen exposure, 78 and 75%, respectively (Table 3.7).

116 Anaerobic carboxylase O -exposed carboxylase % activity Enzymes 2 activitya activitya retained Wild-type 6125 2326 37

M298D 4900 2345 51

S366I 827 594 75

S366V 4184 2846 71

M298D/S366I 4.1 3.2 78

M298D/S366V 28 21 75 anmol/min/mg

Table 3.8. Carboxylase activity at 83°C of heat stable extracts from T. kodakaraensis RbcL wild-type and mutant enzymes under anaerobic and oxygen exposed conditions.

117

DISCUSSION

14 In previous studies, it was demonstrated via [ C]O2 radiometric assays that purified A. fulgidus RbcL2 Rubisco is functional under strict anaerobic conditions and has a high sensitivity to oxygen (Chapter 2) (Kreel and Tabita, 2007). Since A. fulgidus is a thermophilic strict anaerobe isolated from the bottom of the ocean near hydrothermal vents, it is not surprising that Rubisco from this organism is adapted to function under similar extreme conditions in vitro. Of considerable interest is the substantial loss in carboxylase activity of the A. fulgidus enzyme in the presence of molecular oxygen, even when CO2 levels are in great excess. Mutations to this form III Rubisco homodimer at positions Met-295 and Ser-363 catalyzes a reaction with a kcat that is 4 to 5 fold higher

than other forms of Rubisco much like the wild-type enzyme however these residue

changes result in the formation of an enzyme that is much less sensitive to molecular

oxygen. Again, this effect of oxygen was found to be a reversible effect and in previous

studies it was shown that full activity could be recovered so long as all vestiges of oxygen were removed form the reaction mixture (Chapter 2) (Kreel and Tabita, 2007). Thus far, it appears that this response to oxygen has been observed only for form III archaeal

Rubiscos, including the enzymes from both mesophilic and thermophilic archaea such as

M. acetivorans, M. jannaschii, A. fulgidus and T. kodakaraensis (Kreel and Tabita, 2007;

Finn and Tabita, 2003; this study). Previous experiments have shown that substantial amounts of highly purified recombinant A. fulgidus protein may be obtained. Coupled

118 with the solved crystal structure of the related Rubisco from T. kodakaraensis (Maeda et al., 1999), it was deemed feasible to investigate the molecular basis for the unusual properties exhibited by the A. fulgidus enzyme (Finn and Tabita, 2003; Kreel and Tabita,

2007). Earlier studies had shown that one of the unusual properties is the high sensitivity to oxygen. The response to molecular oxygen was clearly shown to be a classic competition with CO2 for the enediolate intermediate of the enzyme, as observed for all

Rubisco proteins. However, what distinguished the A. fulgidus enzyme from other

sources of Rubisco was the extremely high affinity this enzyme showed to molecular

oxygen, with Ki values (of about 5 µM). This is nearly 3 orders of magnitude lower than

typical form I or form II enzymes. Clearly, this high affinity for molecular oxygen

underscores why inhibition of carboxylase activity was initially obtained even in reaction

mixtures that contained levels of CO2 that normally swamp out the inhibitory effects of oxygen for form I and form II Rubiscos. In previous studies, the effect of a methionine to aspartate substitution at position 295 was analyzed (Chapter 2) (Kreel and Tabita, 2007).

In addition to Met-295, analysis of the linear sequence of RbcL2 from A. fulgidus and other archaeal Rubiscos compared to other well-studied form I and form II enzymes brought attention to another residue, Ser-363. Ser-363 is in close proximity to the active site, near Met-295, and this residue is positioned in a distinct hydrophobic pocket. After altering this residue by site-directed mutagenesis and preparing recombinant S363I and

S363V proteins, it was apparent that these enzymes showed substantially less sensitivity to molecular oxygen than the wild-type protein, much like the M295D mutant. Further investigation of this amino acid using homology modeling of the homodimeric structure of the A. fulgidus RbcL2 indicated that this amino acid is on β-strand 6, situated in a

119 hydrophobic pocket adjacent to the active site (Figure 3.14). Alanine is strictly

conserved at this same position in the form I enzymes and isoleucine in the form II

enzymes. Form III enzymes have either a serine or an alanine. According to the model

structure, ionic interactions of the side chain of Ser-363 occur with the highly conserved

and catalytically important residues Gly-313 and Thr-314 in the model structure of A.

fulgidus RbcL2. In the form I and form II enzymes there is an absence of similar ionic

interactions with the amino acids present in the same position, strongly suggesting that this serine residue is affects the catalytic activity. Furthermore, many of the amino acids that surround this serine residue are highly conserved in form I, II and III Rubiscos

(Table 3.2). For A. fulgidus RbcL2, the increase in the Michaelis constant (Ko) for oxygen for the S363I and S363V mutant enzymes suggests that this amino acid plays an important role in the enzymes ability to retain activity in the presence of oxygen, similar to M295D. This is supported by previous findings of increased activity retention when enzymes were assayed in the presence of oxygen (Chapter 2) (Kreel and Tabita, 2007).

Perhaps this effect is caused by either a disruption of the hydrogen bonding interactions between Gly-313, Thr-314 and Ser-363 or due to localized structural changes in the area of this hydrophobic pocket, or both. Additionally, recombinant M295D/S363I and

M295D/S363V double mutations and their apparent additive effect in the retention of activity after exposure (Chapter 2) (Kreel and Tabita, 2007) was further supported by the greatly increased Ko values determined for these enzymes. Not only were the observed

Ko values 84-fold and 18-fold, respectively, higher than the wild-type enzyme, but the Ko of 427 ± 96 µM for the M295D/S363I enzyme falls within the range of Ko values observed for many form I and form II Rubiscos (Jordan and Ogren, 1981).

120 Unfortunately, the higher observed Ko’s for the double mutants M295D/S363I and

M295D/S363V of A. fulgidus RbcL2 also result in a significantly lower kcat value.

Figure 3.14. Model structure of monomer subunit of A. fulgidus RbcL2. The predicted structure of the A. fulgidus RbcL2 protein was modeled to the known structure of the T. kodakaraensis KOD1 RbcL (1GEH), using Deep View Swiss PDB Viewer, spdbv 3.7. Within the monomer, the highlighted features are the residues that form the hydrophobic pocket, colored red and cyan, surrounding the Ser-363 residue, colored yellow. Met-295 is highlighted in purple. Highly conserved and catalytically important residues that are predicted to interact or be influenced by Met-295 or Ser-363 are highlighted in white. The gold colored ribbons of the monomer represent the remaining highly conserved active site residues and are shaded orange.

121 With respect to the low activity observed for the two double mutants, initial

assays were performed at what was later determined to be limiting levels of RuBP

substrate (data not shown). Indeed, the Km for RuBP (KRuBP) was much higher than the wild-type enzyme (Table 3.2). Thus, substantially increasing the concentration of RuBP in normal assays resulted in significantly higher activity for both these enzymes, however still not at the level obtained for the wild-type and single mutants. Furthermore, the

S363I and S363V single mutants show a higher KRuBP than both the wild-type and the

M295D enzyme; the dramatic increase in the KRuBP for double mutants might suggest an

additive effect that is perhaps due to some form of non-cooperative binding between

these two mutated residues since the mutated positions, Met-295 and Ser-363 are in close

proximity to highly conserved active site residues necessary for the binding of RuBP.

Since these two mutations have such a profound effect on reducing oxygen

sensitivity, attempts were made to recover the high activity observed for the wild-type

and single mutations form A. fulgidus RbcL2, while maintaining the extremely high Ko values. Using the model structure of the A. fulgidus RbcL2 enzyme and comparing its hydrogen bonding network to previously solved crystal structures of form I

(Synechococcus PCC6301) and form II (R. rubrum CbbM) enzymes, revealed that Ile-

312 in A. fulgidus RbcL2 may be an important residue for such considerations.

Molecular modeling showed that Ile-312 resides between Met-295 and Ser-363 in A. fulgidus RbcL2 (Figure 3.5). In form I Rubiscos there is either a serine, threonine, or alanine in this position and for form II Rubiscos there is either a valine or threonine.

Also, since the crystal structure of the form I Synechococcus PCC6301 enzyme (PDB,

1RBL) is in the closed conformation, this structure could be used to give some insight

122 into the folded structure of all Rubiscos; i.e. provide some indications as to how this

amino acid position may cause disruptions if it is too bulky. The A. fulgidus RbcL2 wild-

type enzyme contains Ser-363 and Ile-312 (Figure 3.5) however with mutations of S363I

and S363V plus double mutations of M295D/S363I and M295D/S363V, this region may

be overwhelmed with bulky residues that cause distortions to the active site that influence

optimal enzymatic activity. Site-directed mutagenesis of Ile-312 residue to a smaller, less

bulky residue such as that found in the form I and form II enzymes might perhaps cause some sort of reversion to the wild-type properties from the characteristics elicited by the mutant enzymes, particularly double mutants M295D/S363I and M295D/S363V.

However, the results obtained after site-directed mutagenesis revealed that such a

reversion did not occur.

The site-directed mutagenesis studies suggested that there was a strong influence

of mutations at Ser-363 and this region of the enzyme (Table 3.5). Much like the wild-

type enzyme, single mutations to Ile-312 had the same response to molecular oxygen.

However, double mutants that contained the M295D change along with the alterations at

Ser-363 showed a significant (additive) increase in the percent activity retained after

oxygen exposure. This appeared also to be true, in at least one instance, with mutants

that had changes in Ile-312 and Ser-363. Although the goal to determine if it might be

possible to change the low kcat and extremely high Km for RuBP observed for the

M295D/S363I and M295D/S363V mutant enzymes was not achieved, the results clearly

indicated that Ser-363 is an influential site for reducing the A. fulgidus RbcL2 enzyme’s

sensitivity to oxygen.

123 These results were further supported by growth complementation studies with

these mutant enzymes under phototrophic conditions in R. capsulatus SBI/II-. Under photoheterotrophic conditions, A. fulgidus RbcL2 wild-type and all of the mutant enzymes (cloned into plasmid vector pRPS-MCS3) were all able to support growth.

However, the double and triple mutants took a considerably longer time to grow compared to the wild-type and single mutant enzymes. Such results could be attributed to the extremely low kcat and/or poor binding of RuBP by these enzymes. Nonetheless,

mutant and wild-type enzymes were synthesized, as verified through SDS-PAGE gel,

Western immunoblot analysis, and enzymatic activity assays at optimal temperatures for

the A. fulgidus RbcL2 enzyme (83°C) after cultures were grown at the optimal growth

temperature for R. capsulatus (30°C). In addition, it was possible to reisolate plasmid

DNA and verify (by sequencing) that the expected genes were present in the cultures.

Under photoautotrophic conditions, only the wild-type and single mutant forms of the

enzyme were able to complement growth; the presence of these enzymes were verified as

discussed above for photoheterotrophic growth. The higher levels of enzymatic activity

observed at both assay temperatures under photoautotrophic conditions compared to

photoheterotrophic conditions was undoubtedly due to increased Rubisco synthesis

directed by the vector promoter under the former growth conditions. Ultimately, the need

to complement growth under aerobic chemoautotrophic growth conditions in the dark will provide some physiological significance for the single mutants’ (M295D, S363I and

S363V) ability to retain activity in vitro in the presence of oxygen. In addition, random mutagenesis of the A. fulgidus rbcL2 gene and subsequent complementation under these growth conditions will allow for positive selection of additional mutants that may be able

124 to function and complement growth in the presence of molecular oxygen. Moreover, it

will be entirely feasible to use the double mutants of either (M295D/S363I or

M295D/S363V) as a template for selecting for additional mutant proteins that possess high kcat values, while also maintaining insensitivity to oxygen and a very high Ko value.

T. kodakaraensis RbcL shares 72% amino acid sequence identity with A. fulgidus

RbcL2. Because these two highly related form III Rubisco enzymes share many similar features, it was difficult to accept that their reported CO2/O2 substrate specificity (Ω)

values would be at such polar extremes of reported specificity values (Ezaki et al., 1999;

Kreel and Tabita, 2007). Recombinant T. kodakaraensis RbcL was thus prepared and specificity values were determined under rigorously defined conditions with known concentrations of CO2 and O2. Our results indicate that the CO2/O2 substrate specificity value for T. kodakaraensis RbcL was 6 ± 0.2, much lower than the previously determined

value of 310 (Ezaki et al., 1999) and more close to the value of 4 reported for the A.

fulgidus enzyme (Kreel and Tabita, 2007). Moreover, the T. kodakaraensis RbcL

Rubisco is also oxygen sensitive, although not so sensitive as the very closely related A.

fulgidus RbcL2. Since the model structure of A. fulgidus RbcL2 is based on the solved

structure of the highly homologous T. kodakaraensis enzyme, it is not surprising that many of the ionic bonding interactions that are suggested in the model structure of A. fulgidus RbcL2 appear in the solved structure of T. kodakaraensis RbcL (Kitano et al.,

2001). Like the large subunits of all Rubiscos, known residues necessary for catalysis are conserved and are positioned within the T. kodakaraensis structure in the same locale as in other Rubisco structures (Cleland et al., 1998). Similar to the model structure for A. fulgidus RbcL2, Met-298 in T. kodakaraensis RbcL was found to be situated on α–helix 5

125 positioned next to β–strand 5, adjacent to the active site. Met-298 was also found to be in

close proximity to a highly conserved residue, Arg-282, found in all other forms of

Rubisco and known to be necessary for substrate (RuBP) binding (Zhang et al., 1994). In

T. kodakaraensis RbcL, there is no hydrogen bond to the Arg-282 residue, while there is definite hydrogen-bonding to the equivalent arginine residue in all other form I and form

II Rubisco structures; e.g., originating from the oxygen atom of the carbonyl group of

His-324 from the peptide backbone of the Synechococcus PCC6301 enzyme (Figure

3.15). The distance between the corresponding arginine residue (Arg-282) to the carbonyl group of the equivalent histidine (His-311) of the peptide backbone is ~3.61 Å.

A mutation at position Met-298 to an aspartate residue suggests an ionic interaction between one of the hydroxyl side chains of the aspartate residue with one of the side chain nitrogen atoms of Arg-282 (Figure 3.16). All the other amino acid mutations made at position 298 suggested either unfavorable conformations or no ionic interactions with

Arg-282. In addition, many rotamers were available for the aspartic acid substitution at the methionine position; the rotamers with the lowest score, thus the most favorable conformation, all had hydrogen bonding interactions with Arg-282. Further investigation led us to another amino acid, Ser-366, which we predicted might have a similar affect on oxygen sensitivity since identical mutations were made in A. fulgidus RbcL2. Again, the structure indicated that this amino acid is on β-strand 6, situated in a hydrophobic pocket adjacent to the active site. Interestingly, the same results occurred after exposure to molecular oxygen with the single mutations of S366I and S366V, very much like the results with mutations made at the same site as A. fulgidus RbcL2. Also, the double mutations of M298D/S366I and M298D/S366V showed an extremely low kcat

126 reminiscent of the results shown for A. fulgidus RbcL2. As previously explained, the

model structure for A. fulgidus RbcL2 suggests that at this positions, Ser-363, there is a

hydrogen bonding interaction with both Gly-313 and Thr-314 that is essentially lost when either isoleucine or valine is introduced into the site. The solved structure for T. kodakaraensis RbcL shows a slightly different interaction. Compared to the A. fulgidus

RbcL2 model, there is an interaction with the side chain of Ser-366 with the peptide

backbone of Gly-316 however there is no interaction with the side chain of Thr-317.

Since the results with both A. fulgidus and T. kodakaraensis Rubisco indicate changes in

sensitivity to oxygen (as well as substantially loss in activity for the T. kodakaraensis

S366I enzyme) compared to the wild-type form of the enzyme, perhaps the interaction

that is most important is the interaction between the serine residue situated in the

hydrophobic pocket and the highly conserved glycine residue. Perhaps the threonine

residue does not affect the oxygen sensitivity as much, even though the model structure

of A. fulgidus RbcL2 suggests that it does (Figure 3.17).

Although it is not precisely understood what exactly the role is of these amino

acids at these positions in the form III Rubisco enzymes, ultimately it should become a

priority to crystallize many of these enzymes, both wild-type and mutant forms, so that

more direct observations can be made towards which elucidating which changes occur to

induce the rather substantial alterations in kinetic parameters that define this enzyme.

Such studies could also eventually lead to a suggested mechanism as to how the two

gaseous substrates, carbon dioxide or oxygen, are differentiated in the overall mechanism

in these Rubisco enzymes.

127

Figure 3.15. Comparison of side-chain interactions with Arg-295 in (A) form I Synechococcus PCC6301 and Arg-282 and in (B) form III T. kodakaraensis Rubisco enzymes. Highly conserved amino acids necessary for the binding of the five carbon substrate, RuBP, in Synechococcus PCC6301/T. kodakaraensis enzymes include Arg- 295/Arg-282, His-298/His-285, and His-327/His-314 and are shown in ball and stick figures off of the ribbon structure. Phe-311 in Synechococcus PCC6301 and Met-298 in T. kodakaraensis are at the same position in sequence alignments, situated on α-helix 5. The carbonyl of the peptide backbone of His-327 in Synechococcus PCC6301 Rubisco forms an ionic interaction, depicted by a dashed purple line) with the side chain of Arg- 295 (A) whereas this interaction is not observed between the corresponding His-314 and Arg-282 in T. kodakaraensis Rubisco (B).

128 A

129

Figure 3.16. Predicted side-chain interactions with Met-298 in wild-type T. kodakaraensis RbcL and the mutant M298D enzyme. Side chains shown are amino acids Met-298 (A) and Asp-298 (B), as well as conserved amino acids found in all other forms of Rubisco. In T. kodakaraensis RbcL and the mutant M298D enzyme, His-285, Arg- 282, and His-314, are illustrated as they are necessary for catalysis and binding of the five carbon substrate, RuBP. The solved crystal structure shows no ionic interactions between Arg-282 and Met-298 in the wild-type form of the enzyme (A). In the M298D mutant, the model predicts an ionic interaction between the hydroxyl group of the Asp- 298 residue and the amino group of the Arg-282 residue (dashed purple line).

130 A

131

Figure 3.17. Comparison of side chain interactions of Ser-366 with Gly-316 and Thr-317 in the solved crystal structure of T. kodakaraensis RbcL with corresponding residues in the model structure of A. fulgidus RbcL2. Ser-366 in T. kodakaraensis RbcL (A) is situated on β-strand 6 pointing away from the active site, situated in a highly conserved hydrophobic pocket, and interacts with highly a highly conserved residue, Gly-316, depicted by dashed purple lines. Ser-366 does not interact with the otherr highly conserved residue, Thr-317. Conversely, the model structure of A. fulgidus RbcL2 (B) suggests that the identical residues in this region, Ser-363, Gly-313 and Thr-314 all interact to form hydrogen bonds (dashed purple lines).

132 A

133

CHAPTER 4

SUMMARY OF WORK PERFORMED

Summary of work performed. Forms I, II, and III ribulose 1,5-bisphosphate

(RuBP) carboxylase/oxygenase (Rubisco) are all capable of catalyzing carboxylation and/or oxygenation of the five carbon sugar substrate ribulose 1,5-bisphosphate (RuBP).

Two identical molecules of 3-phosphoglycerate are obtained after carboxylation while one molecule of 3-PGA and one molecule of 2-phosphoglycolate are obtained as a result of the oxygenolysis of RuBP. Rubisco enzymes may be classified into three distinct forms (form I, II and III) based on sequence homologies (Tabita, 1999; Watson et al.,

1999). The discovery that Rubisco may be present in representatives of the third domain of life, the Archaea, underline, for the first time, that Rubisco may be found in organisms that thrive only in strictly anaerobic environments. Not only has the presence of Rubisco in archaea stimulated investigations as to why this enzyme is even present in such organisms, since carbon dioxide is not fixed using the Calvin-Benson-Bassham pathway, but more importantly, this finding has triggered studies on potential unique structural adaptations compared to other bona fide Rubisco molecules. Since archaeal Rubiscos appeared to show extreme oxygen sensitivity, it was thus considered that the archaeal enzyme might provide clues as to how the active sight might have evolved to allow this

134 enzyme to function in an aerobic environment. Recent studies suggest that archaeal

Rubisco functions in purine/pyrimidine salvage/recycling pathways in which compounds

such as 5-phosphoribose-D-1-pyrophosphate (PRPP) and AMP lead to the formation of

RuBP via a novel precursor, ribose-1,5-bisphosphate. Thus, in these organisms, Rubisco

catalysis and the subsequent formation of PGA appears to be important for allowing such

compounds to enter central carbon metabolism (Figure 1.6) (Finn and Tabita, 2004; Sato

et al., 2007).

Investigating the basis for oxygen sensitivity of form III archaeal Rubisco could

give key clues relative to structure and function relationships and CO2/O2 substrate specificity. Such information might then lead to an understanding as to why structurally related proteins often possess vastly different kinetic properties and especially quite different gaseous substrate selectivity. Sequence alignments, along with consideration of

various available structural models (and homology modeling of the form III

Archaeoglobus fulgidus RbcL2 Rubisco) have offered much information relative to

unique features conserved in archaeal Rubiscos. Bioinformatic studies identified two

specific residues, Met-295 and Ser-363, that subsequent site-directed mutagenesis and

expression studies resulted in recombinant proteins with altered kinetic properties,

especially reduced sensitivity to oxygen. While other amino acid positions were

considered for site-directed mutagenesis as well as other amino acid changes at positions

295 and 363, the M295D, S363I and S363V mutant proteins retained the most

carboxylase activity after oxygen exposure compared to the wild-type enzyme. Similar

responses to molecular oxygen were obtained for recombinant Thermococcus

kodakaraensis RbcL Rubisco when identical amino acid alterations were made. In

135 addition, mutant A. fulgidus RbcL2 proteins exhibited strikingly different responses in

vivo when such enzyme molecules were used to complement photoheterotrophic and

photoautotrophic growth of a double Rubisco knockout strain (SBI/II-) of Rhodobacter capsulatus. Growth rates appear to be consistent with in vitro kinetic experiments, especially double mutants M295D/S363I and M295D/S363V, whose significantly lower kcat in vitro appears to correlate with much slower growth under both photoheterotrophic and photoautotrophic conditions compared to the wild-type and single mutants of A. fulgidus RbcL2. Interestingly, when R. capsulatus SBI/II- was complemented with wild- type A. fulgidus rbcL2 and mutant rbcL2 genes, cultures containing the wild-type gene

showed a considerable lag in growth when cultured under photoautotrophic conditions

with minimal media (Figure 3.8). Similar results were obtained when such

- complemented SBI/II strains were cultured under aerobic conditions in a CO2/H2/O2 atmosphere on PYE plates. It is not well understood why this lag in growth occurred when the wild-type A. fulgidus rbcL2 gene was employed in these complementation experiments. One idea is that the presence of low, contaminating levels of oxygen were present in these “anaerobic” cultures and that this caused inhibition of growth, especially cultures containing the more oxygen-sensitive (low Ko) wild-type enzyme. Thus far, using an oxygen scrubbing assay with protocatechuate 3,4-dioxygenase, we have not yet been able to detect any adventitious levels of oxygen in these cultures (data not shown).

Of course it is recognized that minute amounts of oxygen may still be present, but undetected with this assay, or there may be other reasons why the wild-type enzyme seems to specifically confer this lag in growth. Note, after this lag in growth is overcome, all cultures grow at about the same rate and have similar doubling times.

136 Hypotheses to explain the results obtained. An important aspect of these

studies is that the residues identified, particular Ser-363, reside in a hydrophobic pocket

that is highly conserved in all three forms of Rubisco according to solved crystal and

model structures. Recent studies in our laboratory have revealed that a mutation in Ala-

375 of the form I Synechococcus PCC6301 Rubisco (the equivalent site to Ser-363 of the

A. fulgidus enzyme) show interesting results. When alanine at position 375 was changed

to a valine of the Synechococcus PCC6301 enzyme, the resultant A375V protein was

almost completely oxygen insensitive enzyme and possessed a higher Ko compared to the wild-type enzyme (Dr. Sriram Satagopan, personal communication). Clearly, this localized hydrophobic pocket region of Rubisco influences oxygen binding to the enediol intermediate of RuBP in the active site. This has now been confirmed with three different enzymes, the form I Synechococcus PCC6301 enzyme and the two archaeal form III Archaeoglobus fulgidus RbcL2 and Thermococcus kodakaraensis RbcL enzymes

studied in this investigation. Upon further analysis, it is apparent that this hydrophobic

pocket is located in a position that could influence the catalytically important loop 6

structure. As previously mentioned in Chapter 1, loop 6 folds over the active site after

the insertion of RuBP into the active site; then a highly conserved lysine residue on loop

6 interacts with the N-terminal portion of the opposing large subunit. Alterations to the

hydrophobic pocket, specifically residues that reside within the pocket, such as Ser-363

in A. fulgidus RbcL2, Ser-366 in T. kodakaraensis RbcL, or Ala-375 in Synechococcus

PCC6301 CbbL could conceivably influence the ability of loop 6 to fold over the active site, interact with the N-terminal portion of the opposing subunit, and subsequently allow catalysis to occur. It is still unknown, but should continue to be investigated further,

137 whether substitutions of either serine or alanine residues with bulky amino acids in the

hydrophobic pocket of form III or form I enzymes, respectively, alter or distort the

hydrophobic pocket as hypothesized. In addition, whether such changes lead to altering

loop 6 folding also needs to be investigated further. Finally, it is conceivable that there

might be more important amino acid residues within this hydrophobic pocket structure;

i.e., residues that surround the serine or alanine residues directly in the pocket. Such

residues could then be mutated in order to determine whether such changes influence

gaseous substrate specificity and Ko values.

Observations of space filling models from crystal structures of form I enzymes

with bound CABP molecules indicate that there exists only one channel for the carbon

dioxide and oxygen gaseous substrates to move into the active site for subsequent

nucleophilic attack at the C2 position of the enediol RuBP intermediate. This is the

opening created by loop 6 that interacts and binds to the specific region at the N-terminus

of an opposing large subunit of a dimeric pair. More specifically, it has been suggested

that after the RuBP molecule moves into the active site, the enzyme undergoes a

conformational change that includes folding of loop 6, followed by interaction with the

N-terminal portion of the opposing subunit to form a “locked” position. At this point, the

2,3-enediol intermediate of RuBP is formed and becomes associated with the enzyme.

For gaseous substrate entry, loop 6 and the N-terminal region must subsequently become

“unlocked” and this involves the reopening of the loop and after either CO2 or O2 enter the active site, loop 6 and the N-terminal region “lock” again (Dr. Sriram Satagopan, personal communication). The suggestion for the existence of only one channel for substrate entry is highly supported by crystal structures and site-directed mutations made

138 in residues at the loop 6 region, as well as mutations in the N-terminal region that result

in altered Michaelis constants. However, at this time there is still no absolutely proven

precise route or location within the protein for gaseous substrate entry into the active site.

Nonetheless, the positioning of the carboxylic group at the C2 position of the transition state analog CABP in crystal structures of form I enzymes suggests that the region comprised of loop 6 of one subunit and the N-terminal region of the second subunit is the most likely area for gaseous substrate entry. Thus alterations of residues that are in or part of the hydrophobic pocket may not directly affect the active site where RuBP binds.

Rather, such residues may have a substantial impact as to how loop 6 is able to fold over the active site and “lock” into position to allow catalysis to occur.

It is unclear then why recombinant proteins that have single mutations at either

Met-295 or Ser-363 in A. fulgidus RbcL2 yield similar kinetic values. Moreover, the basis for the additive effect observed with double mutations that result in major changes in kinetic values, particularly KRuBP and Ko, is also very curious. While Ser-363 and the region of the hydrophobic pocket has a more direct effect on the folding of loop 6 over the active site, due to this residue’s very close association with loop 6, perhaps the region where Met-295 resides seemingly more indirectly influences how loop 6 properly folds over the active site. Perhaps, potential ionic interactions between Asp-295 of the M295D enzyme and highly conserved and catalytically important Arg-279 of A. fulgidus RbcL2

(as proposed in Chapter 2) may be quite important. Thus, such an interaction with Arg-

295 may affect catalysis by dramatically altering enediol formation and subsequent nucleophilic attack by either carbon dioxide or oxygen, this is supported by the fact that the M295D enzymes possess altered Ko values, presumably via such specific interactions

139 with Arg-279. This may explain why there appears to be an additive effect upon

recovery from oxygen inactivation by the double mutant enzymes compared to the single

mutants (i.e., the M295D/S363I and M295D/S363V enzymes compared to either the

M295D or S363I/S363V mutant enzymes).

It was considered that perhaps other forms of Rubisco might provide evidence to

support the hypothesis for the universal importance of disrupted ionic interactions with

highly conserved and catalytically important arginine residues, such as Arg-279 of the A.

fulgidus enzyme. In Rhodospirillum rubrum form II Rubisco, Ala-305 is the equivalent

residue to Met-295 of A. fulgidus RbcL2, and Ile-367 of form II Rubisco is equivalent to

Ser-363 of the A. fulgidus enzyme. Site-directed mutagenesis produced recombinant

A305D, I367S or I367V R. rubrum CbbM enzymes. Assays performed at 30°C of crude soluble extracts of the Ile-367 mutants indicated that about the same level of activity was maintained compared to the wild-type enzyme. More interesting, there was a loss of activity, by two orders of magnitude, with single and double mutations that contained

A305D substitutions (A305D, A305D/I367S and A305D/I367V) (Table 4.1). Even with higher amounts of RuBP than standard amounts used, to test for altered KRuBP, the specific activity was unchanged. SDS-PAGE as well as Western immunoblot analyses under denaturing conditions exhibit the same level of expression for all the mutant enzymes (data not shown). Via the criteria of normal migration in nondenaturing gels, both the wild-type and all mutant forms of R. rubrum CbbM enzyme folded properly;

Western immunoblot analyses of the nondenaturing gels confirmed proper folding of both the wild-type and all mutant forms (Figure 4.1). As mentioned in Chapter 2, the crystal structure of R. rubrum Rubisco shows an ionic interaction originating from the

140 oxygen atom of the carbonyl group of His-321 from the peptide backbone to the side

chain of the conserved arginine residue, Arg-287. Since model structures of M295D of

A. fulgidus RbcL2 suggests an ionic interaction with the aspartate side chain and the Arg-

279 residue (an interaction not found in the wild-type enzyme), perhaps the same

interactions occur in A305D of R. rubrum CbbM. Thus, the A305D R. rubrum enzyme could severely alter the ability of RuBP to bind efficiently to the active site via the introduction of additional interactions. Based on the best rotameric score from the

SwissProt PDB modeling program, the configuration does not suggest additional binding interaction between A305D and Arg-287 (Figure 4.2). This may indicate that disruption of the hydrophobic region where Ala-305 resides occurs when a bulkier, anionic charged amino acid is substituted at this position. Whatever the precise disruption in the R. rubrum form II enzyme, activity is severely curtailed by reasons different from that postulated for the M295D A. fulgidus enzyme.

141

Figure 4.1. Coomassie-stained SDS-PAGE (top) and Western immunoblot (bottom) of crude soluble extracts of R. rubrum CbbM. The immunoblot was tested using antibodies directed against purified recombinant R. rubrum CbbM Rubisco. All lanes contained soluble crude extracts containing recombinant enzyme prepared from E. coli: wild-type R. rubrum CbbM (lane 2); A305D (lane 3); I367S (lane 4); I367V (lane 5); A305D/I367S (lane 6); A305D/I367V (lane 7); purified wild-type R. rubrum CbbM (lane 8); purified A. fulgidus RbcL2 (lane 9). Each lane received approximately 5 µg of protein. BioRad Low Range Molecular Weight Standard was used as the marker in lane 1.

A305D/ A305D/ Enzymes Wild-type A305D I367S I367V I367S I367V Specific Activity 1713a 23a 1316a 1860a 4a 11a (nmol/min/mg) aAverage of two independent assays.

Table 4.1. Carboxylase activity at 30°C of crude soluble extracts from R. rubrum CbbM wild-type and mutant enzymes.

142

Figure 4.2. Comparison of side-chain interactions with Ala-305 in the solved crystal structure from wild-type R. rubrum CbbM and the predicted interaction of mutant A305D. Side chains shown are amino acids Ala-305 (A) and Asp-305 (B), as well as conserved amino acids found in all other forms of Rubisco. In R. rubrum CbbM and the mutant A305D enzyme, His-291, Arg-287, and His-321, are illustrated as they are necessary for catalysis and binding of the five carbon substrate, RuBP. The solved crystal structure shows ionic interactions between Arg-287, His-291, and His-321 in the wild-type form of the enzyme (A). In the A305D mutant, the model predicts no ionic interaction between the Asp-305 residue and the amino group of the Arg-287 residue compared to observations at the same position of form III A. fulgidus RbcL2 and T. kodakaraensis RbcL.

143 A

144

SUGGESTIONS FOR FUTURE EXPERIMENTS

Random mutagenesis with A. fulgidus rbcL2 that alters kcat: complementation of R. capsulatus strain SBI/II- under photoheterotrophic and photoautotrophic

conditions and selection under aerobic chemoautotrophic conditions. Using A.

fulgidus rbcL2 as the template, PCR reactions containing small amounts of MnCl2 will

induce random errors during amplification of the gene. Sub-cloning the error filled gene

into the host vector pRPS-MCS3 followed by tri-parental mating the vector into the

double Rubisco knockout strain R. capsulatus SBI/II- will allow for the selection of

functional Rubisco enzymes. Positive selection for mutated A. fulgidus rbcL2 that supports growth under aerobic chemoautotrophic conditions on minimal media where no growth is observed for the host strain containing the wild-type gene, would be extremely valuable. Such mutated genes would presumably encode for A. fulgidus RbcL2 proteins that have increased insensitivity to molecular oxygen. Since in vitro studies with the double mutants M295D/S363I and M295D/S363V have shown a dramatic increase in the

Ko compared to wild-type enzyme, yet possess a severely altered kcat, presumably due to the altered KRuBP, the use of these double mutants as the template during random mutagenic PCR reactions and subsequent positive selection for growth under photoautotrophic and chemoautotrophic conditions could be very useful. Such selections might not only provide mutations that “rescue” the enzyme’s kcat, but in the case of

145 positive growth under aerobic chemoautotrophic conditions, would maintain the high Ko observed from in vitro kinetic studies.

Site-directed mutagenesis in the hydrophobic pocket region of A. fulgidus

RbcL2. Continuation of the site-directed mutagenesis studies and further in vitro analyses of recombinant proteins is needed in order to further understand the importance of this hydrophobic region and all of the amino acids involved. Although there is a high conservation and high similarity among all three forms of bona fide Rubiscos of the amino acids in this region (Table 3.2), where non-identical residues exist in this region need to be investigated. In particular, mutations to A. fulgidus RbcL2 Leu-367 should be made to mimic form I (Ile) and form II (Met) and tested for oxygen sensitivity. One approach that could be taken to further study the details and formulate a more detailed mechanism is to develop a random mutagenesis with error prone PCR that is more directed, that is directly target this exact region or stretch of the gene, and select for positive (or negative) growth using R. capsulatus SBI/II- under varying growth

conditions. This “directed random” mutagenesis approach would allow for only this

region to be mutagenized and no other region of the enzyme to be altered. In addition,

other amino acids in this hydrophobic pocket region may come about and be of some

interest that other wise would not have been introduced to the enzyme after sequence

analysis, computational model studies, or site-directed mutagenesis.

Further analysis of equivalent mutations of A. fulgidus M295D, S363I and

S363V made in form II Rhodospirillum rubrum CbbM. It has already been shown that

mutations at Met-295 and Ser-363 in A. fulgidus RbcL2 can be applied to the very closely

related T. kodakaraensis RbcL, with strikingly similar results. To a small degree,

146 applications of mutations at equivalent sites have been applied to form I (Synechococcus

PCC6301 RbcL – A375V) and form II (R. rubrum CbbM – A305D, I367S and I367V).

Continuing application of mutations discovered in the highly oxygen sensitive form III A.

fulgidus RbcL2 to other forms of Rubisco will eventually lead to a better understanding

of the structure and function relationship between Rubisco and the two gaseous

substrates, carbon dioxide and oxygen, that compete for the same enediol intermediate of

RuBP in the active site of the enzyme.

In addition, it is vital to the study of this enzyme as well as to all Rubisco

enzymes, that this enzyme and a selection of the mutants made in this work, such as

S363I and M295D/S363I, have their structures solved so that a better understanding as to

how and why the kinetic parameters, such as the dramatically different Ko and KRuBP, become altered. Such studies will be critical to develop Rubisco molecules that support maximum CO2 fixation even in the presence of oxygen.

147

LIST OF REFERENCES

Achenbach-Richter, L., K. O. Stetter, and C. R. Woese. 1987. A possible biochemical missing link among archaebacteria. Nature. 327:348-349.

Berman, H. M., J. Westbrook, Z. Feng et al. 2000. The protein data bank. Nucleic Acids Res. 28:235-242.

Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254.

Bult, C. J., O. White, G. J. Olsen et al. 1996. Complete genome sequence of the methanogenic archeon, Methanococcus jannaschii. Science. 273:1058-1073.

Cleland, W. W., T. J. Andrews, S. Gutteridge, F. C. Hartman, and G. H. Lorimer. 1998. Mechanism of Rubisco: the carbamate as general base. Chem. Rev. 98:549-562.

Dean, J. A., 1985. Lange’s Handbook of Chemistry. McGraw-Hill, New York, NY.

Deppenmeier, U., A. Johann, T. Hartsch., R. Merkl, R. A. Schmitz et al. 2002. The genome of Methanosarcina mazei: evidence for lateral gene transfer between bacteria and archaea. J. Mol. Microbiol. Biotechnol. 4:453-461.

Dubendorff, J. W., and F. W. Studier. 1991. Controlling basal expression in an inducible T7 expression system by blocking the target T7 promoter with lac repressor. J. Mol Biol. 219:45-59.

Ellis, R. J., 1979. The most abundant protein in the world. Trends Biochem. Science. 4:241-244.

Ezaki, S., N. Maeda, T. Kishimoto, H. Atomi, and T. Imanaka. 1999. Presence of a structurally novel type ribulose-bisphosphate carboxylase/oxygenase in the hyperthermophilic archaeon, Pyrococcus kodakaraensis KOD1. J. Biol. Chem. 274:5078-5082.

148 Falcone, D. L. and F. R. Tabita. 1991. Expression of endogenous and foreign ribulose 1,5-bisphosphate carboxylase-oxygenase (Rubisco) genes in a Rubisco deletion mutant of Rhodobacter sphaeroides. J. Bacteriol. 173:2099-2108.

Falcone, D. L. and F. R. Tabita. 1993. Complementation analysis and regulation of CO2 fixation gene expression in a ribulose 1,5 bisphosphate carboxylase-oxygenase deletion strain of Rhodospirillum rubrum. J. Bacteriol. 175:5066-5077.

Figurski, D. and D. R. Helinski. 1979. Replication of an origin-containing derivative of the plasmid RK2 dependent on a plasmid function provided in trans. Proc. Natl. Acad. Sci. USA. 76:1648-1652.

Finn M. W., and F. R. Tabita. 2003. Synthesis of catalytically active form III ribulose 1,5-bisphosphate carboxylase/oxygenase in archaea. J. Bacteriol. 185:3049- 3059.

Finn M. W., and F. R. Tabita. 2004. Modified pathway to synthesize ribulose 1,5- bisphosphate in methanogenic archaea. J. Bacteriol. 186:6360-6366.

Fujisawa, H. and O. Hayaishi. 1968. Protocatechuate 3,4-dioxygenase. I. Crystallization and characterization. J. Biol Chem. 243:2673-2681.

Galagan, J. E., C. Nusbaum, A. Roy, M. G. Endrizzi et al. 2002. The genome of M. acetivorans reveals extensive metabolic and physiological diversity. Genome Res. 12:532-542.

Guex, N. and M. C. Peitsch. 1997. SWISS-MODEL and the Swiss-PdbViewer:an environment for comparative protein modeling. Electrophoresis. 18:2714-2723.

Gutteridge, S. and A. A Gatenby. 1995. Rubisco synthesis, assembly, mechanism, and regulation. The Plant Cell. 7:809-819.

Hanahan, D. 1983. Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166:557-580.

Hanson, T. E. and F. R. Tabita. 2003. Insights into the stress response and sulfur metabolism revealed by proteome analysis of a Chlorobium tepidum mutant lacking the Rubisco-like protein. Photosyn. Res. 78:231-248.

Harpel, M. R., E. H. Lee, and F. C. Hartman. 1993. Anion-exchange analysis of ribulose-bisphosphate carboxylase/oxygenase reaction: CO2/O2 specificity determination and identification of side products. Anal. Biochem. 209:367-374.

149 Hartman, F. C. and M. R. Harpel. 1993. Chemical and genetic probes of the active- site of D-ribulose-1,5-bisphosphate carboxylase oxygenase – a retrospective based on the 3-dimensional structure. Adv. Enzymol. 6:71-75.

Hartman, F. C. and M. R. Harpel. 1994. Structure, function, regulation, and assembly of D-ribulose-1,5-bisphosphate carboxylase/oxygenase. Annu. Rev. Biochem. 63:197-232.

Horken, K. M. and F. R. Tabita. 1999. Closely related form I ribulose bisphosphate carboxylase/oxygenase molecules that possess different CO2/O2 substrate specificities. Arch. Biochem. Biophys. 361:183-194.

Jerpseth, B., M. Callahan, and A. Greener. New Competent Cells for Highest Transformation Efficiencies. 1997. Strategies. 10:37–38.

Jordan, D. B. and W. L. Ogren. 1981. Species variation in the specificity of ribulose bisphosphate carboxylase/oxygenase. Structure. 291:513-515.

Kitano, K., N. Maeda, T. Fukui, H. Atomi, T. Imanaka, and K. Miki. 2001. Crystal structure of a novel-type archaeal Rubisco with pentagonal symmetry. Structure. 9:473-491.

Klenk, H. P., R. A. Clayton, J. F. Tomb et al. 1997. The complete sequence of the hyperthermophilic sulfate-reducing archaeon Archaeoglobus fulgidus. Nature. 390:364-370.

Kovach, M. E., P. H. Elzer, D. S. Hill, G. T. Robertson, M. A. R. Farris, M. Roop, and K. M. Peterson. 1995. Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying different antibiotic-resistance cassettes. Gene. 166:175-176.

Kreel, N. E., and F. R. Tabita. 2007. Substitutions at Methionine 295 of Archaeoglobus fulgidus ribulose-1,5-bisphosphate carboxylase/oxygenase affect oxygen binding and CO2/O2 specificity. J. Biol. Chem. 282:1341-1351.

Kung, S. 1976. Tobacco fraction I protein: a unique genetic marker. Science. 191:429- 434.

Laemmli, U. K. 1970. Cleaving of structural proteins during the assembly of the head of bacteriophage T4. Nature. 227:680-685.

Lu, B. Y. Lindqvist, and G. Schneider. 1992. Electrostatic fields at the active site of ribulose-1,5-bisphosphate carboxylase. PROTEINS: Structure, Function, and Genetics. 12:117-127.

150 Maeda, N., K. Kitano, T. Fukui, S. Ezaki, and T. Imanaka. 1999. Ribulose bisphosphate carboxylase/oxygenase from the hyperthermophilic archaeon Pyrococcus kodakaraensis KOD1 is composed solely of large subunits and forms a pentagonal structure. J. Mol. Biol. 293:57-66.

Ormerod, J. G., K. S. Ormerod, and H. Gest. 1961. Light-dependent utilization of organic compounds and photoproduction of molecular hydrogen by the photosynthetic bacteria; relationships with nitrogen metabolism. Arch. Biochem. Biophys. 94:449-463.

Paoli, G. C., N. S. Morgan, F. R. Tabita, and J. M. Shively. 1995. Expression of the structural cbbLcbbS and cbbM genes and distinct organization of the cbb Calvin cycle structural genes of Rhodobacter capsulatus. Arch. Microbiol. 164:396-405.

Paoli, G. C., P. Vichivanives, and F. R. Tabita. 1998. Physiological control and regulation of the Rhodobacter capsulatus cbb operons. J. Bacteriol. 180:4258-4269.

Papworth, C., J. C. Bauer, J. Braman, and D. A. Wright. 1996. Site-directed mutagenesis in one day with greater than 80% efficiency. Strategies. 9:3-4.

Ramage, R. T., B. A. Read, and F. R. Tabita. 1998. Alteration of the alpha helix region of cyanobacterial ribulose 1,5-bisphosphate carboxylase/oxygenase to reflect sequences found in high substrate specificity enzymes. Arch. Biochem. Biophys. 349:81- 88.

Read, B. A., and F. R. Tabita. 1992. Amino acid substitutions in the small subunit of ribulose 1,5-bisphosphate carboxylase/oxygenase that influence catalytic activity of the holoenzyme. Biochemistry. 31:519-525.

Sato, T., H. Atomi, and T. Imanaka. 2007. Archael type III rubiscos function in a pathway for AMP metabolism. Science. 315:1003-1006.

Schneider, G., Y. Lindquist, and T. Lundqvist. 1990. Crystallographic refinement and structure of ribulose-1,5-bisphosphate carboxylase from Rhodospirillum rubrum at 1.7Å resolution. J. Mol. Biol. 4:989-1008.

Smith, S. and F. R. Tabita. 2003. Positive and negative selection of mutant forms of prokaryotic (cyanobacterial) ribulose-1,5-bisphosphate carboxylase/oxygenase. J. Mol. Biol. 331:557-569.

Spreitzer, R. J. 1999. Questions about the complexity of chloroplast ribulose-1,5- bisphosphate carboxylase/oxygenase. Photosyn. Res. 60:29-42.

Tabita, F. R. 1999. Microbial ribulose 1,5-bisphosphate carboxylase/oxygenase: A different perspective. Photosyn. Res. 60:1-28.

151

Tabita, F. R., P. Caruso, and W. Whitman. 1978. Facile assay of enzyme unique to the Calvin cycle in intact cells, with special reference to ribulose 1,5-bisphosphate carboxylase. Anal. Biochem. 84:462-472.

Tabita, F. R. and B. A. McFadden. 1974a. D-Ribulose 1,5-diphosphate carboxylase from Rhodospirillum rubrum. I. Levels, purification, and effects of metallic ions. J. Biol. Chem. 249:3453-3458.

Tabita, F. R. and B. A. McFadden. 1974b. D-Ribulose 1,5-diphosphate carboxylase from Rhodospirillum rubrum. II. Quarternary structure, composition, catalytic, and immunological properties. J. Biol. Chem. 249:3459-3464.

Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673- 4680.

Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA. 76:4350-4354.

Wang, X., D. L. Falcone, and F. R. Tabita. 1993. Reductive pentose phosphate independent CO2 fixation in Rhodobacter sphaeroides and evidence that ribulose bisphosphate carboxylase/oxygenase activity serves to maintain the redox balance of the cell. J. Bacteriol. 175:3372-3379.

Watson, G. M. F., and F. R. Tabita. 1997. Microbial ribulose 1,5-bisphosphate carboxylase/oxygenase: A molecule for phylogenetic and enzymological investigation. FEMS. Lett. 146:13-22.

Watson, G. M. F., J. P. Yu, and F. R. Tabita. 1999. Unusual ribulose1,5- bisphosphate carboxylase/oxygenase of anoxic archaea. J. Bacteriol. 181:1569-1575.

Weiner, M. P., C. Anderson, B. Jerpseth, S. Wells, B. Johnson-Browne, et al. 1994. Studier pET system vectors and hosts. Strategies. 7:41–43.

Yanisch-Perron, C., J. Vierira, and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M14mp18 and pUC19 vectors. Gene. 33:103-119.

Yen, H. C. and B. Marrs. 1976. Maps of genes for carotenoid and bacteriochlorophyll biosynthesis in Rhodopseudomonas capsulatus. J. Bacteriol. 126:619-629.

152 Yoshida, S., M. Inui, H. Yukawa, T. Kanao, K. Tomizawa, H. Atomi, and T. Imanaka. 2006. Phototrophic growth of a Rubisco-deficient mesophilic purple nonsulfur bacterium harboring a type III Rubisco from a hyperthermophilic archaeon. J. Biotechnol. 124:532-544.

Zellner, G., E. Stackebrandt, H. Kneifel, P. Messner, W. E. B. Sletyr, E. C. De Marcario, H. P. Zabel, K. O. Stetter, and J. Winter. 1987. Isolation and characterization of a thermophilic, sulfate reducing archaebacterium, Archaeoglobus fulgidus strain Z. System. Appl. Microbiol. 11:151-160.

Zhang, K. Y. J., D. Cascio, and D. Eisenberg. 1994. Crystal structure of the unactivated ribulose 1,5-bisphosphate carboxylase/oxygenase complexed with a transition state analog, 2-carboxy-D-arabinitol 1,5-bisphosphate. Protein Sci. 3:64-69.

153