B12 METABOLISM IN HUMANS By NICOLE AURORA LEAL A ...

B12 METABOLISM IN HUMANS
By
NICOLE AURORA LEAL
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2004

Copyright 2004
by
Nicole Aurora Leal

The work in this dissertation is dedicated to my family, especially my parents, for
their endless love and support. I would also like to dedicate this dissertation to my
fiancé, Tearlach Bigger, whose love, patience, and encouragement guided me through
this enlightening experience.

ACKNOWLEDGMENTS
I would sincerely like to thank my mentor, Dr. Thomas A. Bobik. I find myself
continually amazed by his keen intellectual insight and unceasing dedication to research
and the furthering of science. I would like to thank the rest of my committee,
Dr. Terence R. Flotte, Dr. Nemat O. Keyhani, Dr. Julie A. Maupin-Furlow, and
Dr. Keelnatham T. Shanmugam. Their help and guidance along the way were greatly
appreciated.
I owe a debt of gratitude to Dr. Ruma Banerjee and Horatiu Olteanu, who provided
purified human methionine synthase reductase, which was instrumental in the MSR-ATR
studies. Special thanks go to Dr. Peter E. Kima for his constant optimism and
much-needed expertise in the area of cell biology. Additional thanks go to the entire
faculty, staff, and graduate students, especially Stephanie Havemann, at the Microbiology
and Cell Science Department.
I would also like to extend my appreciation to Dr. Gregory Havemann, Celeste
Johnson, Dorothy Park, Edith Sampson, and all the other members of the Bobik lab, for
making it a friendly and supportive place to work.
The text of Chapter 2 in this dissertation, in part or in full, is a reprint of the
material as it appears in the Journal of Biological Chemistry (volume 278, pp. 9227-9234,
2003). The text of Chapter 4 in this dissertation, in part or in full, is a reprint of the
material as it appears in Archives of Microbiology (volume 180, pp. 353-361, 2003).
iv

TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ................................................................................................. iv
LIST OF TABLES............................................................................................................. ix
LIST OF FIGURES .............................................................................................................x
LIST OF ABBREVIATIONS........................................................................................... xii
ABSTRACT.......................................................................................................................xv
CHAPTER
1 INTRODUCTION ........................................................................................................1
History of Cobalamin ...................................................................................................1
Structure of Cobalamin.................................................................................................2
Biosynthesis of Cobalamin...........................................................................................3
Cobalamin-Dependent Enzymes ..................................................................................4
Cobalamin in Humans ..................................................................................................6
Cobalamin-Dependent Enzymes in Humans.........................................................6
Methylmalonyl-CoA mutase..........................................................................7
Methionine synthase.......................................................................................9
Cobalamin Absorption and Transport .................................................................11
Intracellular Cobalamin Metabolism...................................................................12
β-ligand transferase ......................................................................................12
Cob(III)alamin reductase..............................................................................13
Cob(II)alamin reductase...............................................................................14
Adenosyltransferase .....................................................................................15
Formation of methylcobalamin ....................................................................17
Inherited Disorders of Cobalamin Metabolism ..........................................................18
Combined Methylmalonic Aciduria and Homocystinuria ..................................19
Methylmalonic Aciduria without Homocystinuria..............................................21
Homocystinuria without Methylmalonic Aciduria..............................................23
Salmonella as a Model Organism for the Study of Cobalamin Metabolism ..............25
Research Overview.....................................................................................................26
v

2 IDENTIFICATION OF THE HUMAN AND BOVINE ATP:COB(I)ALAMIN
ADENOSYLTRANSFERASE CDNAS BASED ON COMPLEMENTATION OF
A BACTERIAL MUTANT........................................................................................36
Introduction.................................................................................................................36
Materials and Methods ...............................................................................................38
Chemicals and Reagents......................................................................................38
Bacterial Strains and Growth Media ...................................................................39
General Protein and Molecular Methods and ATR Assays.................................39
P22 Transductions ...............................................................................................39
Screening the Bovine and Human Liver cDNA Libraries...................................40
Cloning the Bovine Adenosyltransferase Coding Sequence
for High-Level Expression...............................................................................40
Cloning the Human Adenosyltransferase Coding Sequence
for High-Level Expression...............................................................................41
Cloning the Human Adenosyltransferase Coding Sequence
for Complementation Studies ..........................................................................42
Growth of Adenosyltransferase Expression Strains and Preparation
of Cell Extracts ................................................................................................43
Growth Curves.....................................................................................................43
Western Blots ......................................................................................................44
DNA Sequencing and Analysis...........................................................................45
Results.........................................................................................................................45
Screening of the Bovine and Human Liver cDNA Libraries for Clones that
Express Adenosyltransferase Activity .............................................................45
Identification of a Human Gene Related to the Bovine Adenosyltransferase
cDNA ...............................................................................................................47
Putative Subcellular Localization of Adenosyltransferase Enzymes ..................48
High-Level Expression of the Bovine and Human Adenosyltransferases
in E. coli...........................................................................................................48
Assay of the Bovine and Human Adenosyltransferase Fusion Proteins
for ATR Activity..............................................................................................49
Complementation of an S. enterica Mutant Deficient in ATR Activity by a
Human Adenosyltransferase cDNA Clone ......................................................50
Level of Human Adenosyltransferase Enzyme in Normal and cblB Mutant
Human Skin Fibroblasts...................................................................................51
Conserved Amino Acids and Distribution of Adenosyltransferase Enzymes.....52
Discussion...................................................................................................................53
3 PURIFICATION AND INITIAL CHARACTERIZATION OF THE HUMAN
ATP:COB(I)ALAMIN ADENOSYLTRANSFERASE AND ITS INTERACTION
WITH METHIONINE SYNTHASE REDUCTASE .................................................62
Introduction.................................................................................................................62
Materials and Methods ...............................................................................................64
Chemicals and Reagents......................................................................................64
Bacterial Strains and Growth Media ...................................................................65
vi

General Molecular and Protein Methods.............................................................65
ATP:cob(I)alamin Adenosyltransferase Assays..................................................65
Construction of ATR Expression Strains ............................................................66
Purification of the Human ATR Variants............................................................66
MSR-ATR Assay.................................................................................................68
Measurement of Cob(I)alamin with Iodoacetate.................................................68
Separation and Quantification of Cobalamins by HPLC ....................................68
Effect of Ionic Strength on the MSR-ATR System.............................................69
DNA Sequencing.................................................................................................69
Results.........................................................................................................................70
Purification of the Human ATR Variants............................................................70
Linearity of the ATR reaction .............................................................................71
ATR Reaction Requirements...............................................................................71
Alternative Nucleotide Donors............................................................................71
Km and Vmax Values for ATR 239K and 239M ...................................................71
MSR Reduces Cob(II)alamin to Cob(I)alamin for AdoCbl Synthesis................72
Cob(I)alamin is Sequestered by the MSR-ATR System .....................................73
MSR Produces little Cob(I)alamin in the Absence of the ATR ..........................75
Stoichiometry of the MSR-ATR system .............................................................75
Ionic Strength Dependence of Cobalamin Reduction and Adenosylation
by the MSR-ATR System ................................................................................75
Discussion...................................................................................................................76
4 PDUP IS A COENZYME-A-ACYLATING PROPIONALDEHYDE
DEHYDROGENASE ASSOCIATED WITH THE POLYHEDRAL BODIES
INVOLVED IN B12-DEPENDENT 1,2-PROPANEDIOL DEGRADATION BY
SALMONELLA ENTERICA SEROVAR TYPHIMURIUM LT2...............................92
Introduction.................................................................................................................92
Materials and Methods ...............................................................................................95
Bacterial Strains, Media, and Growth Conditions...............................................95
General Molecular Methods ................................................................................95
General Protein Methods.....................................................................................95
P22 Transduction.................................................................................................96
Cloning pduP into pLAC22.................................................................................96
Construction of His8-PduP ..................................................................................97
Purification of His8-PduP under Denaturing and Nondenaturing Conditions.....97
Propionaldehyde Dehydrogenase Assays............................................................98
Identification of Propionyl-CoA a Product of the Propionaldehyde
Dehydrogenase Reaction .................................................................................99
Construction of a Nonpolar pduP Deletion.......................................................100
Antibody Preparation.........................................................................................101
Western Blots ....................................................................................................101
Electron Microscopy .........................................................................................102
DNA Sequencing and Analysis.........................................................................102
Chemicals and Reagents....................................................................................102
Results.......................................................................................................................103
vii

Effect of a Precise pduP Deletion on the Growth of S. enterica on Minimal
1,2-Propanediol Medium ...............................................................................103
Propionaldehyde Dehydrogenase Activity in Wild-type S. enterica
and a ∆pduP Mutant.......................................................................................105
High-level Production of the PduP Protein .......................................................105
Purification of Recombinant His8-PduP Protein ...............................................106
Propionyl-CoA is a Product of the PduP Reaction............................................107
Preparation and Specificity of the Anti-PduP Antiserum..................................108
Localization of PduP Immunoelectron Microscopy..........................................109
Propionaldehyde Dehydrogenase Activity is associated
with Purified pdu Bodies ...............................................................................110
Discussion.................................................................................................................110
5 CONCLUSIONS ......................................................................................................119
Identification of the Bovine and Human Adenosyltransferase.................................119
Biochemical Characterization of the Human Adenosyltransferase..........................120
Methionine Synthase Reductase is a Cob(II)alamin Reductase
for the Human Adenosyltransferase.....................................................................121
Future Experimentation ............................................................................................123
LIST OF REFERENCES.................................................................................................125
BIOGRAPHICAL SKETCH ...........................................................................................139
viii

LIST OF TABLES
Table page
1-1. Adenosylcobalamin- and methylcobalamin-dependent reactions .............................30
2-1. Bacterial strains .........................................................................................................56
2-2. Specific activities of bovine and human ATP:cob(I)alamin ATRs...........................59
3-1. ATP:cob(I)alamin adenosyltransferase activity during ATR 239K and 239M
purification ...............................................................................................................82
3-2. Use of alternative nucleotide donors by the human ATR .........................................84
3-3. The MSR-ATR system sequesters cob(I)alamin.......................................................85
4-1. Bacterial strains .......................................................................................................113
4-2. Propionaldehyde dehydrogenase activity in extracts from wild-type S. enterica
and strain BE191 (∆pduP)......................................................................................114
4-3. Propionaldehyde dehydrogenase activity during polyhedral body purification......115
ix

LIST OF FIGURES
Figure page
1-1. Structure of cobalamins.............................................................................................29
1-2. Different classes of cobalamin-dependent enzymes..................................................31
1-3. Pathways of propionyl-CoA catabolism and cobalamin metabolism........................32
1-4. Pathways involving methionine synthase and the reductive activation of
methionine synthase by methionine synthase reductase ..........................................33
1-5. Cobalamin metabolism in mammalian cells, and disorders associated with
deficiencies in this pathway .....................................................................................34
1-6. Proposed pathway of AdoCbl-dependent 1,2-propanediol degradation
in S. enterica.............................................................................................................35
2-1. Multiple sequence alignment of adenosyltransferase enzymes.................................57
2-2. SDS-PAGE analysis of cell extracts from bovine and human ATR
expression strains .....................................................................................................58
2-3. Complementation of an ATR-deficient bacterial mutant for AdoCbl-dependent
growth on 1,2-propanediol by a plasmid that expresses the human ATR................60
2-4. Western blot analysis of ATR expression by normal and cblB mutant cell lines .....61
3-1. Propionyl-CoA metabolism, methionine synthesis, and intracellular cobalamin
metabolism ...............................................................................................................81
3-2. Purification of the human ATR .................................................................................83
3-3. Determination of kinetic constants for ATR 239K and ATR 239M .........................86
3-4. Absorbance spectra of an MSR-ATR assay ..............................................................87
3-5. Two schemes depicting the conversion of cob(II)alamin to AdoCbl by the
MSR-ATR system....................................................................................................88
3-6. MSR produces little cob(I)alamin in the absence of the ATR enzyme.....................89
x

3-7. Stoichiometry of the MSR-ATR system ...................................................................90
3-8. Effect of ionic strength on the MSR-ATR system ....................................................91
4-1. SDS-PAGE analysis of His8-PduP ..........................................................................116
4-2. Western blot using the anti-PduP antiserum............................................................117
4-3. Immunogold localization of the PduP enzyme........................................................118
xi

A adenine
LIST OF ABBREVIATIONS
ADP adenosine diphosphate
AdoCbl adenosylcobalamin
AIM aldehyde indicator medium
Amp ampicillin
ATP adenosine triphosphate
ATR adenosyltransferase
BSA bovine serum albumin
C cytosine
°C degree centigrade
Cam chloramphenicol
cbl cobalamin
cDNA complementary deoxyribonucleic acid
CH3Cbl methylcobalamin
CH3THF methyl tetrahydrofolate
CNCbl cyanocobalamin, Vitamin B12
CMCbl carboxymethylcobalamin
ddH20 distilled deionized water
DMB dimethylbenzimidazole
DNA deoxyribonucleic acid
xii

DTT dithiothreitol
E. coli Escherichia coli
FAD flavin adenine dinucleotide
FldA flavodoxin
FMN flavin mononucleotide
Fre flavin oxidoreductase
Fpr flavodoxin reductase
G guanine
GSCbl glutathionylcobalamin
HOCbl hydroxycobalamin
IF intrinsic factor
IFCR intrinsic factor cobalamin receptor
IPTG isopropyl-β-D-thiogalactopyranoside
Kan kanamycin
kDa kiloDaltons
Km Michaelis Constant for enzyme activity
LB Luria Bertani medium
MCM methylmalonyl-CoA mutase
MTS mitochondrial targeting sequence
min minutes
mg milligram
mM millimolar
MS methionine synthase
xiii

MSR methionine synthase reductase
NADPH nicotinamide adenine dinucleotide phosphate – reduced form
NCE no carbon E medium
OD optical density
PAGE polyacrylamide gel electrophoresis
PCR polymerase chain reaction
SAM S-adenosylmethionine
S. enterica Salmonella enterica
SDS sodium dodecyl sulfate
T thymine
TCII transcobalamin II
TCII-R transcobalamin II receptor
TCA tricarboxylic acid
THF tetrahydrofolate
Tris tris hydroxymethyl aminomethane
tRNA transfer ribonucleic acid
U uracil
V volt
Vmax maximal rate of enzyme activity
XCbl inactive cobalamin derivatives
X-Gal 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside
xiv

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
B12 METABOLISM IN HUMANS
By
Nicole Aurora Leal
August 2004
Chair: Thomas A. Bobik
Major Department: Microbiology and Cell Science
In humans, the B12 coenzymes, adenosylcobalamin and methylcobalamin, are
required cofactors for propionate metabolism and methionine biosynthesis, respectively.
Humans are incapable of de novo synthesis of the B12 coenzymes and require complex
precursors such as vitamin B12 in their diet. The metabolism of vitamin B12 to
adenosylcobalamin requires two successive reductions and an adenosylation reaction;
however, many aspects of the genetics and biochemistry of this process have not been
elucidated. This dissertation focuses on the isolation and characterization of the human
genes and enzymes involved in the formation of adenosylcobalamin from vitamin B12.
Deficiencies in this process result in methylmalonic aciduria, a rare disorder that is often
fatal in newborns.
Because of the sophisticated genetic methods that are available, Salmonella
enterica was used as a model system. By complementation of an S. enterica mutant, a
bovine adenosyltransferase cDNAs was isolated and subsequent sequence similarity
xv

searches identified a homologous human cDNA. The bovine and human cDNAs were
independently cloned, expressed in Escherichia coli, and the encoded proteins were
shown to have adenosyltransferase activities similar to previously studied bacterial
adenosyltransferases. Additional studies, using Western blots, showed that
adenosyltransferase expression was altered in cell lines derived from patients with cblB
methylmalonic aciduria when compared to cell lines from normal individuals.
Two common human adenosyltransferase polymorphic variants were expressed in
E. coli, purified, and biochemically characterized. Both had high specificity for ATP and
comparable Km and Vmax values. An adenosyltransferase-linked cob(II)alamin reductase
assay was developed and used to show that the human methionine synthase reductase
functions as a cob(II)alamin reductase for adenosylcobalamin formation. A linked assay
was also performed in the presence of a cob(I)alamin trapping agent, and results from
these experiments suggested that human adenosyltransferase and methionine synthase
reductase interact. Importantly, results from this work provide information that will
allow improvements in diagnosis and treatment of methylmalonic aciduria, a devastating
childhood disorder.
xvi

CHAPTER 1
INTRODUCTION
History of Cobalamin
In 1925, Whipple and colleagues discovered that a diet of liver aided in the
formation of blood cells in iron-deficient dogs. The next year, Minot and Murphy (Minot
and Murphy 1926) were the first to document that patients with severe pernicious anemia
(a disease that affects the formation of red blood cells) were successfully treated with a
special diet consisting mainly of liver. Eight years later (in 1934) Whipple, Minot, and
Murphy were awarded the Nobel Prize in Medicine and Physiology for their work on
liver therapy against anemias.
Following these studies, Castle (Kass 1978) concluded that the treatment of
pernicious anemia by this specialized diet was dependent on the absorption of an
extrinsic anti-pernicious anemia factor by an intrinsic protein in gastric juice. It is now
known that Castle’s extrinsic anti-pernicious anemia factor is vitamin B12 and the
intrinsic factor is a specific cobalamin transporter (synthesized by the stomach) that was
shown to enhance the curing effect by vitamin B12 (Kass 1978). In 1948, two research
groups led by Folkers and Smith independently isolated the red crystalline anti-pernicious
anemia factor from bovine liver, and proposed the name vitamin B12 (Rickes et al. 1948,
Smith and Parker 1948). The isolation of vitamin B12 inspired collaborative efforts
between two laboratories (led by Alexander Todd and Dorothy Hodgkin) to elucidate its
chemical structure. After 6 years, the crystal structure was resolved and finalized in the
1

2
laboratory of Dorothy Hodgkin (Hodgkin et al. 1955), a feat for which she was awarded
the 1964 Nobel Prize in Chemistry.
Today we know that two active cobalamin derivatives are required for enzyme
catalysis. One is the adenosyl derivative (adenosylcobalamin) which was identified in
1958, when Barker and coworkers reported the isolation of a light-sensitive cobalamin
compound that was a required cofactor for glutamate mutase (Barker et al. 1958).
Shortly after the discovery of this compound, its crystal structure was resolved by
Lenhert and Hodgkin (Lenhert and Hodgkin 1961). The other biologically active
cobalamin derivative was discovered later as methylcobalamin, a required cofactor for
methionine synthase (Guest et al. 1962a, Guest et al. 1962b). The structure of
methylcobalamin was resolved decades later by Jenny Gluskers laboratory (Rossi et al.
1985).
Structure of Cobalamin
Cobalamin is the largest and most complex cofactor known to man. It is a
water-soluble molecule composed of three general structural units: a central ring, an
upper ligand, and a lower ligand (Figure 1-1). At the core of cobalamin is a ring structure
designated as corrin, which is synthesized from uroporphyrinogen III, the last common
intermediate of heme, siroheme, chlorophyll, and cobalamin synthesis. The corrin ring
consists of four reduced pyrrole rings (A, B, C, and D). Unlike heme and chlorophyll (in
which the pyrrole rings are linked by four methylene bridges), cobalamin has a direct
bond between rings A and D. In cobalamin, the four nitrogen atoms of the pyrrole rings
form coordinate bonds with a central cobalt atom; whereas the central atom of the
tetrapyrrole ring is iron and magnesium, in heme and chlorophyll, respectively. Below
the plane of the corrin ring is the α-ligand of the cobalt atom, dimethylbenzimidazole

3
(DMB). The DMB moiety is covalently attached to the corrin ring by a ribose phosphate
bonded to the aminopropanol side chain on ring D and is also coordinated to the central
cobalt atom. The exact function of the lower ligand is not completely understood, but has
been suggested that it may have a role in coenzyme binding and catalysis (Banerjee and
Ragsdale 2003). Above the plane of the corrin ring is the β-ligand of the cobalt atom.
The β-ligand varies in the different forms of cobalamins and includes cyano, hydroxy,
glutathionyl, methyl, and adenosyl moieties. Cyanocobalamin (vitamin B12, CNCbl) is
an artifact obtained during isolation of cobalamins from natural sources and is the most
common pharmaceutical form (Rickes et al. 1948). CNCbl together with
hydroxycobalamin (HOCbl), and glutathionylcobalamin (GSCbl) are precursors for
synthesis of the biologically active coenzymes, methylcobalamin (CH3Cbl) and
adenosylcobalamin (AdoCbl).
Biosynthesis of Cobalamin
Cobalamin or closely related corrinoid compounds are required in all kingdoms of
life with perhaps the exception of plants and fungi. De novo synthesis of B12 is restricted
to some Bacteria and Archaea (Roth et al. 1996), requiring all other life forms that use
B12 to obtain complex precursors from their diet. Characterized as the most complex
nonpolymeric natural substance known, cobalamin is synthesized de novo by a multi-step
enzymatic process consisting of uroporphyrinogen III synthesis, side chain modification,
cobalt insertion, aminopropanol addition, DMB biosynthesis, and addition of methyl or
adenosyl moieties (Battersby 1994). The microbial biosynthesis of cobalamin can
proceed along two different pathways: one aerobic and the other anaerobic (Scott 2003).
These pathways differ both in their requirement for molecular oxygen and the timing of

cobalt insertion. Pseudomonas denitrificans and Bacillus megaterium use the aerobic
pathway of cobalamin synthesis whereby molecular oxygen is used to assist with ring
contraction and cobalt is inserted late in the synthesis. In contrast, cobalamin synthesis
by the anaerobic pathway (used by Salmonella enterica and Propionibacterium
4
freundenreichii) is prevented by the presence of molecular oxygen, and cobalt is inserted
early during synthesis. Unlike prokaryotes, humans are unable to synthesize cobalamin
de novo and are dependent on a dietary source of a complex precursor such as vitamin
B12 from which AdoCbl and CH3Cbl are synthesized (Baker and Mathan 1981).
Cobalamin-Dependent Enzymes
AdoCbl and CH3Cbl share structural and chemical similarities; however, they
catalyze very different biochemical reactions. The weak covalent carbon-cobalt bond of
cobalamin, which is highly reactive and crucial for enzyme catalysis, is a shared trait of
both cofactors (Frey and Reed 2000). In AdoCbl-dependent reactions, carbon-based free
radicals are generated by the homolytic cleavage of the carbon-cobalt bond (Halpern
1985); while in methyltransferase reactions, the carbon-cobalt bond of CH3Cbl is
heterolytically cleaved, releasing a methyl group and cob(I)alamin (Banerjee 1997).
AdoCbl-dependent enzymes can be classified into four types: eliminases (including
dehydratases and deaminases), amino mutases, carbon skeleton mutases, and
ribonucleotide reductases. The first three types are isomerases that catalyze
1,2-rearrangement reactions (Table 1-1). Their main use is in the fermentation of small
molecules and they use the same basic catalytic mechanism (Banerjee 1999). The
reaction is triggered by substrate binding, causing homolytic cleavage of the
carbon-cobalt bond, thereby generating an adenosyl radical and cob(II)alamin. The
adenosyl radical abstracts a hydrogen atom from the substrate, forming a substrate radical

5
and deoxyadenosine. After hydrogen abstraction, rearrangement of the substrate radical
results in product radical formation. The product radical then abstracts a hydrogen atom
from deoxyadenosine, forming both product and an adenosyl radical. Finally, the
adenosyl radical recombines with cob(II)alamin, reforming the coenzyme (which can be
used in additional catalytic cycles). The last type of AdoCbl-dependent enzymes is
ribonucleotide reductases, which generate deoxyribonucleotides for DNA synthesis.
There are three classes of ribonucleotide reductases. Class I ribonucleotide reductases
are strict aerobic enzymes and are divided into two subclasses (Class Ia and Ib). Class Ia
ribonucleotide reductases are found in eukaryotes, plants, viruses, and some prokaryotes.
These enzymes contain three conserved cysteines, a tyrosyl radical, and a nonheme diiron
center which are essential for catalysis (Aberg et al. 1989, Nordlund and Eklund 1993).
Class Ib ribonucleotide reductases are found in prokaryotes and vary from class Ia in their
lack of allosteric regulation and the use of different electron carriers for ribonucleotide
reduction (Jordan et al. 1994). The Class II ribonucleotide reductases require AdoCbl,
are found in bacteria and archaea and are active both aerobically and anaerobically.
Their reaction mechanism is similar to that of the isomerase reaction, differing where the
adenosyl radical does not interact directly with the substrate, but with a cysteinal radical
in the active site of the enzyme (Booker et al. 1994). Class III ribonucleotide reductases
are oxygen sensitive and are found in some prokaryotes and methanogens (Reichard
1993). Although this class has not been well characterized it is thought that a glycyl
radical, an iron sulfur center and S-adenosylmethionine (SAM) are required for catalysis
(Harder et al. 1992).

6
CH3Cbl-dependent enzymes are involved in methyl transfer reactions (Table 1-1).
Only the reaction mechanisms of methionine synthase and the acetyl-CoA synthase have
been studied extensively (Ljungdahl and Wood 1982, Jarrett et al. 1998); however, all of
the methyltransferases are thought to follow the same general mechanism. The reaction
involves three steps, and begins with the methylation of enzyme-bound cob(I)alamin by a
methyl-group donor, forming enzyme-bound CH3Cbl. The carbon-cobalt bond of
CH3Cbl is then heterolytically cleaved, resulting in enzyme-bound cob(I)alamin and a
methyl cation, which is transferred to the substrate forming the methylated product.
Methyltransferases are essential for methionine synthesis and are also involved in acetate
and methane synthesis (Schneider and Stroinski 1987). Figure 1-2 shows the different
types of B12-dependent enzymes and the reactions they catalyze.
Cobalamin in Humans
Humans are incapable of de novo synthesis of cobalamin, and must take up
complex cobalamin precursors from the diet. Cobalamin is a water-soluble vitamin that
cannot freely diffuse across the intestinal wall because of its large size and requires
specific carriers and transporters for uptake and delivery to cells. Once in the cells,
cobalamin is then metabolized to CH3Cbl in the cytosol and AdoCbl in the mitochondria,
by a series of enzyme-mediated reactions.
Cobalamin-Dependent Enzymes in Humans
In humans, AdoCbl and CH3Cbl are required for two enzymes that are essential in
metabolism (Kolhouse and Allen 1977). One of these enzymes is methylmalonyl-CoA
mutase (MCM), an AdoCbl-dependent enzyme used to metabolize propionyl-CoA
resulting from the breakdown of odd-chain fatty acids, cholesterol, and some amino
acids. The other is methionine synthase (MS), a CH3Cbl-dependent enzyme used for

7
homocysteine recycling and methionine synthesis. Deficiencies in MCM or MS can lead
to methylmalonic aciduria or homocystinuria (rare but often lethal childhood disorders).
Methylmalonyl-CoA mutase
MCM is the only AdoCbl-dependent enzyme that is present in both bacterial and
mammalian systems. In humans, it resides in the mitochondrial matrix, and is used for
the metabolism of propionyl-CoA to succinyl-CoA. In this metabolic pathway,
propionyl-CoA is carboxylated to (2S)-methylmalonyl-CoA, isomerized to
(2R)-methylmalonyl-CoA, and lastly rearranged to succinyl-CoA by MCM (Figure 1-3).
In contrast, in some bacteria including Propionibacterium shermanii, MCM is used in the
opposite metabolic flow to produce propionate from succinate (Allen et al. 1964).
The gene encoding MCM has been sequenced and cloned from many organisms
including human, P. shermanii, mouse, Streptomyces cinnamonensis, Porphyromonas
gingivalis, and Sinorhizobium meliloti (Banerjee and Chowdhury 1999). In humans, the
location of the MCM gene was mapped to chromosome 6p12-21.2, and the cDNA was
shown to have a mitochondrial targeting sequence (MTS) (Ledley et al. 1988).
In native conformation, the human and bacterial MCMs are 150 kiloDaltons (kDa),
and are dimeric proteins. The crystal structure of P. shermanii MCM has been
determined, and is an αβ heterodimer that binds one mole of AdoCbl (Marsh et al. 1989).
The human MCM is a homodimer of two α subunits that binds two moles of AdoCbl and
has 60% sequence identity to the α chain of the bacterial MCM (Mancia et al. 1996).
UV, visible, and electron paramagnetic resonance spectroscopic studies have
shown that AdoCbl binding by MCM occurs via “Base-off” (class I) interaction
(Padmakumar and Banerjee 1995). “Base-off” binding occurs by the displacement of the

DMB ligand by a histidine residue on the α-subunit of MCM. DMB is still attached to
the corrin ring through a nucleotide loop; however, in the class I interaction, the
coordinate bond of DMB to cobalt is displaced and substituted by His610 of the
α-subunit. The exact role of the His610 residue is not fully understood, but current
8
studies suggest that its major role is in DMB displacement and organization of the active
site for AdoCbl binding and enzyme catalysis (Vlasie et al. 2002).
The isomerization of methylmalonyl-CoA to succinyl-CoA by MCM requires
homolytic cleavage of the carbon-cobalt bond of enzyme-bound AdoCbl (Padmakumar
and Banerjee 1997). Homolysis of the organometallic bond occurs once
methylmalonyl-CoA binds MCM. This results in the formation of an adenosyl radical
and cob(II)alamin. The carbon-cobalt bond of free AdoCbl is weak and in solution has a
rate of homolysis of 4x10 -10 sec -1 . With the addition of MCM, the rate of homolysis is
increased to higher than 600 sec -1 , suggesting that MCM promotes the homolysis of
AdoCbl for this rearrangement reaction (Padmakumar and Banerjee 1997). The adenosyl
radical generated by homolysis abstracts hydrogen from the methyl-group of
methylmalonyl-CoA, forming a primary-substrate radical and 5’-deoxyadenosine. The
primary-substrate radical undergoes an intramolecular 1,2-rearrangement resulting in the
formation of a secondary product radical that abstracts hydrogen from deoxyadenosine.
Lastly, there is recombination of the adenosyl radical and cob(II)alamin, regenerating
AdoCbl and releasing succinyl-CoA that can flow into the tricarboxylic acid (TCA)
cycle.
In humans, defects in the enzyme MCM cause a block in propionyl-CoA
metabolism, resulting in a buildup of methylmalonyl-CoA that is hydrolyzed to

methylmalonic acid. Methylmalonic acid is excreted into the urine, resulting in
methylmalonic aciduria, an inherited disorder that is often fatal in newborns.
Methionine synthase
MS catalyzes the conversion of CH3THF and homocysteine to THF and
methionine. This enzyme is found in both bacteria and humans. In E. coli, two MS
enzymes occur and they differ in their requirement for CH3Cbl (Drummond and
Matthews 1993). In humans, only CH3Cbl-dependent MS is present. The E. coli and
human genes encoding CH3Cbl-dependent MS have been cloned and encode large
9
monomeric proteins of 136 and 141 kDa, respectively (Banerjee et al. 1989, Leclerc et al.
1996). The human gene was mapped to chromosome 1 at position q43 (Leclerc et al.
1996). The bacterial and the human enzymes share 55% identity in amino acid sequence,
and have established similarities in the mechanism of catalysis (Hall et al. 2000).
However, only the E. coli CH3Cbl-dependent MS has been studied extensively.
Methionine synthase from E. coli is a modular enzyme consisting of four domains,
all of which are crucial for enzyme catalysis (Goulding et al. 1997). The N-terminal
domain (residues 2-353) has been cloned, expressed, and shown to be involved in
homocysteine binding, and methyl group transfer from CH3Cbl to homocysteine
(Goulding et al. 1997). It contains three cysteine residues used to coordinate a zinc ion
essential for homocysteine binding, which are conserved in other MS enzymes including
Homo sapiens, Caenorhabditis elegans, and Synechocystis species (Goulding and
Matthews 1997). The second domain (residues 354-649) was shown to catalyze methyl
transfer from CH3THF to cob(I)alamin (Goulding et al. 1997). The third domain
(residues 650-896) was lacking enzymatic activity; however, it was shown to be essential
for CH3Cbl binding (Banerjee et al. 1989). This fragment was crystallized and shown to

ind CH3Cbl in the class I “Base-off” mode (Drennan et al. 1994). The last domain
10
(residues 897-1227), termed the activation domain, was shown to bind SAM required for
the reductive activation of MS (Dixon et al. 1996, Hall et al. 2000).
The mechanism of MS catalysis and reactivation is shown in Figure 1-4. The
methyltransferase reaction requires heterolytic cleavage of the carbon-cobalt bond of
CH3Cbl resulting in cob(I)alamin and a methyl group. After heterolysis, MS transfers its
methyl group to homocysteine, forming methionine and MS-bound cob(I)alamin. During
catalysis, additional methyl groups are provided by methyltetrahydrofolate (CH3THF),
which regenerates CH3Cbl and releases tetrahydrofolate (THF) (Banerjee and Matthews
1990). Occasionally, the MS-bound cob(I)alamin is oxidized to cob(II)alamin during
catalytic turnover, and reductive activation is required to restore the methylation cycle
(Fujii et al. 1977, Banerjee et al. 1990, Drummond and Matthews 1993). This involves
reduction of cob(II)alamin and methylation by SAM to form CH3Cbl. The reduction of
cob(II)alamin for MS activation differs in E. coli and humans. Reduced flavodoxin
(FldA) provides the needed electron in E. coli (Fujii and Huennekens 1974). However,
humans lack flavodoxin and instead use the dual flavoprotein methionine synthase
reductase (MSR) (Olteanu and Banerjee 2001).
In humans, a block in the methylation cycle caused by defects in MS results in
homocystinuria, a severe and sometimes fatal childhood disease (Fenton and Rosenberg
2000). In addition, partial defects in MS can lead to elevated homocysteine, a major
cardiovascular disease risk factor (Refsum et al. 1998). A deficiency in MS also results
in the trapping of cellular folate as CH3THF, making it unavailable for other

folate-dependent reactions including purine and pyrimidine biosynthesis (Wilson et al.
1999).
Cobalamin Absorption and Transport
11
As mentioned above, humans are incapable of de novo cobalamin synthesis, and
require complex precursors in their diet. Suitable precursors (such as HOCbl or CNCbl)
can be obtained by the consumption of beef, liver, poultry, fish, eggs, dairy products, and
vitamin supplements (Stabler 1999). The cobalt of these cobalamin molecules is in the
+3 oxidation state (cob(III)alamin), the form that is recognized for cobalamin absorption
and transport. Once ingested, cobalamin molecules that are bound to food proteins are
released by the combined action of proteases and acid in the stomach (Del Corral and
Carmel 1990). Haptocorrin (a cobalamin carrier protein) binds released cobalamin and
transports it from the stomach to the small intestines. Haptocorrin has a high specificity
for cobalamin and when bound to cobalamin, protects it from damage by acids in the
stomach, until it reaches the small intestines where cobalamin is liberated from
haptocorrin via digestion by pancreatic enzymes. After being released from haptocorrin,
cobalamin is bound by intrinsic factor (IF), a glycoprotein produced by the parietal cells
(gastric glands lining the stomach). The IF-cobalamin complex is resistant to further
digestion in the small intestines because of the carbohydrates on the IF. The
IF-cobalamin complex recognizes the IF-cobalamin receptor (IFCR) on the epithelial
cells of the distal third small intestines (ileum), and is transported into these cells via
receptor-mediated endocytosis. In the epithelial cell, IF is degraded by the acidic
environment of the lysosome, and cobalamin is released. Transcobalamin II (TCII), a
serum-transport protein in the epithelial cells, binds released cobalamin, and transports it
out of the cell and into the bloodstream until it is taken up by other cells. The

12
TCII-cobalamin complex binds the TCII receptor on the target cell, and gains entry via
endocytosis where it can be used for intracellular cobalamin metabolism (Rosenblatt and
Fenton 1999, Seetharam 1999, Watkins and Rosenblatt 2001).
Intracellular Cobalamin Metabolism
Once the TCII-cobalamin complex is internalized by the target cell, the transport
protein is degraded in the lysosomal vesicle, and cobalamin is released from this
complex. A cobalamin transporter then shuttles free cobalamin from the lysosome to the
cytosol of the cell. Intracellular cobalamin metabolism (Figure 1-5) is thought to be
similar in prokaryotes and eukaryotes (Qureshi et al. 1994, Watkins and Rosenblatt
2001). In the cytosol, cob(III)alamin, in the form of HOCbl or CNCbl, is converted to
GSCbl and then reduced to cob(II)alamin. These steps are proposed to be catalyzed by a
β-ligand transferase and a cob(III)alamin reductase, respectively. After cobalamin
reduction in the cytosol, cob(II)alamin is localized to the mitochondrial matrix and is
reduced to cob(I)alamin by a mitochondrial cob(II)alamin reductase. The final step in
AdoCbl formation is the adenosylation of cob(I)alamin to form AdoCbl by an
ATP:cob(I)alamin adenosyltransferase. The pathways of AdoCbl and CH3Cbl synthesis
are shared, and thought to branch apart after the reduction of cob(III)alamin. For CH3Cbl
metabolism, cytosolic cob(II)alamin associates with MS, forming
MS-bound-cob(II)alamin which is reduced to cob(I)alamin by MSR and methylated by
SAM to form MS-CH3Cbl.
β-ligand transferase
The first step in cobalamin metabolism is removal of the β-ligand. Cobalamin
β-ligand transferase activity has been detected in crude cell extracts of Clostridium

tetanomorphum, P. shermanii, Euglena gracilis, human leukocytes, human skin
fibroblasts, rat liver, and rabbit spleen (Weissbach et al. 1961, Brady et al. 1962,
Watanabe et al. 1987, Pezacka et al. 1990, Pezacka 1993). The enzymes involved and
13
their encoding genes have not been identified. However, it was determined that for the
enzymatic conversion of CNCbl to AdoCbl, cell extracts required supplementation with
adenosine triphosphate (ATP), reduced flavin, and reduced glutathione (Weissbach et al.
1962). Studies in mammalian cells suggest that GSCbl is a product of the β-ligand
transferase reaction (Pezacka et al. 1990). In bacteria, this GSCbl intermediate has not
been found.
Cob(III)alamin reductase
The second enzymatic step in cobalamin metabolism is cob(III)alamin reduction.
Reductase activity has been detected in crude cell extracts of C. tetanomorphum,
P. shermanii, E. gracilis, human, and rat (Weissbach et al. 1961, Brady et al. 1962,
Watanabe et al. 1987, Pezacka et al. 1990, Pezacka 1993). Extracts of C. tetanomorphum
required NADH and either flavin mononucleotide (FMN) or flavin adenine dinucleotide
(FAD) to mediate reductase activity (Walker et al. 1969). The flavin oxidoreductase
(Fre) of S. enterica was purified and shown to mediate the nonenzymatic reduction of
cobalamin in vitro (Fonseca and Escalante-Semerena 2000). Fre reduces flavin
nucleotides, which in turn chemically reduces cob(III)alamin. In E. gracilis,
cob(III)alamin reductase activity was purified from the mitochondrial fraction, and was
ultimately identified as a flavoprotein with an absolute requirement for NADPH but not
FAD or FMN (Watanabe et al. 1987). Further analysis of this protein revealed that it
contained one molecule of either FAD or FMN that remained bound during the

14
purification, explaining why cob(III)alamin reductase activity was retained without the
addition of these flavins. Cell extracts from human fibroblasts and rat liver have
cob(III)alamin reductase activity associated with both microsomes and the mitochondrial
membrane, that was derived from oxidation-reduction reactions by cytochromes and
flavoproteins (Watanabe et al. 1989, Watanabe et al. 1996). There has been no genetic
evidence that the flavoproteins studied in these systems are the physiological
cob(III)alamin reductases. The studies by Watanabe and colleagues also indicate that
both prokaryotic and eukaryotic systems produce redundant enzymes with the ability to
reduce cobalamin; however, this reduction could be a primarily nonenzymatic process.
Cob(II)alamin reductase
Cob(I)alamin is one of the strongest nucleophiles known to exist in aqueous
solution, and is a powerful reductant (E°’ = -0.61 V) (Banerjee et al. 1990). Because of
the nucleophilic nature of cob(I)alamin, cob(II)alamin reduction is an energetically
unfavorable reaction. Thus, it is no surprise that cob(II)alamin reductase activity can
only be studied by coupling the reaction to an alkylation event such as adenosylation or
methylation. Cob(II)alamin reductase activity has been detected in cell-free extracts of
C. tetanomorphum and P. shermanii; however, these enzymes have not been identified,
and the encoding genes are unknown (Brady et al. 1962, Weissbach et al. 1962, Walker et
al. 1969). In humans, cob(II)alamin reductase activity has been detected in mitochondrial
fractions, but the enzyme was never identified, and the gene is unknown (Fenton and
Rosenberg 1978b).
Purification of the cob(II)alamin reductase from cell extracts of C. tetanomorphum
did not produce a single enzyme, but produced a complex containing the adenosylating

enzyme. This suggests that cob(II)alamin reductase and the adenosyltransferase may
15
exist as a structural complex under certain conditions (Walker et al. 1969). Advantages
of this complex would decrease the probability of unfavorable cob(I)alamin side
reactions and instead allow for adenosylation because of the close proximity of
cob(I)alamin to the adenosyltransferase.
The E. coli FldA protein has been shown to function as a cob(II)alamin reductase
for the formation of AdoCbl (Fonseca and Escalante-Semerena 2001). In vitro,
flavodoxin reductase (Fpr), flavodoxin (FldA), and purified CobA (an
adenosyltransferase) catalyze AdoCbl synthesis from cob(III)alamin (Fonseca and
Escalante-Semerena 2001). In this coupled assay, it is proposed that Fpr uses NADPH to
reduce FldA, which then reduces cob(III)alamin to cob(I)alamin while bound to CobA
(Fonseca and Escalante-Semerena 2001). To date, there are no reports showing an
interaction between FldA and CobA. Although this reducing system is functional in
vitro, there is no genetic evidence that FldA is the physiological cob(II)alamin reductase
(Fonseca and Escalante-Semerena 2001).
Adenosyltransferase
Adenosylation is the terminal step in AdoCbl metabolism. ATP:cob(I)alamin
adenosyltransferase (ATR) activity was detected in crude cell extracts of human
fibroblast cells, but the enzyme involved and the encoding gene were never identified
(Fenton and Rosenberg 1981). ATRs from P shermanii, C. tetanomorphum,
P. denitrificans, Thermoplasma acidophilum, and S. enterica have been purified and
partially characterized (Brady et al. 1962, Vitols et al. 1966, Debussche et al. 1991, Suh
and Escalante-Semerena 1995, Johnson et al. 2001, Saridakis et al. 2004). The genes
encoding these enzymes have been identified from P. denitrificans, T. acidophilum, and

16
S. enterica (Crouzet et al. 1991, Suh and Escalante-Semerena 1993, Johnson et al. 2001,
Saridakis et al. 2004).
To date, three families of adenosyltransferases have been identified: EutT-type,
PduO-type, and CobA-type. These families are classified based on amino acid sequence
similarity and examples of each are found in S. enterica. EutT is thought to be involved
in the conversion of CNCbl to AdoCbl for ethanolamine utilization (Kofoid et al. 1999).
PduO adenosylates cob(I)alamin for AdoCbl synthesis which is required for
B12-dependent 1,2-propanediol metabolism (Johnson et al. 2001). CobA can substitute
for PduO in cob(I)alamin adenosylation for 1,2-propanediol metabolism, but its primary
role is the adenosylation of cobalamin intermediates for de novo B12 biosynthesis (Suh
and Escalante-Semerena 1995, Johnson et al. 2001). The CobA-type and PduO-type
ATRs have been extensively studied and well characterized biochemically.
CobA ATR from S. enterica has been overexpressed and purified (Suh and
Escalante-Semerena 1995). This enzyme was shown to have a broad specificity for
reduced corrinoid substrates, which could be due to its role in de novo cobalamin
synthesis (Suh and Escalante-Semerena 1995). The X-ray crystal structure of CobA in its
free state, complexed with ATP and HOCbl has been resolved (Bauer et al. 2001). It is a
homodimer, which is consistent with the subunit composition of CobO ATR from
P. denitrificans (Debussche et al. 1991, Bauer et al. 2001). The N-terminal P-loop
(GNGKGKT) of CobA coordinates the α-, β-, and γ-phosphates of ATP, however the
crystal structure shows that this interaction occurs in the opposite orientation when
compared to other P-loop dependent nucleotide hydrolases (Bauer et al. 2001). This
unique mode of ATP binding positions the 2’-OH group to allow for nucleophilic attack

at the 5’-C of ATP by cob(I)alamin resulting in adenosylation (Fonseca et al. 2002).
Additionally, triphosphate was identified as a reaction byproduct of adenosylation and
was also found to be a strong inhibitor of the reaction (Fonseca et al. 2002).
17
The archaeal T. acidophilum PduO-type ATR has been purified, crystallized, and
partially characterized (Saridakis et al. 2004). T. acidophilum ATR has 27% amino acid
identity to PduO. This ATR is specific for ATP and deoxy-ATP. Additionally, the
crystal structure revealed a trimeric configuration. Each of the subunits is composed of
five alpha helical domains that aid in subunit assembly due to hydrophobic, ionic, and
polar interactions depending on the domain. In contrast to CobA, there is no
recognizable P-loop motif suggesting a different binding method for ATP.
Formation of methylcobalamin
The synthesis CH3Cbl is coupled to cob(II)alamin reduction (Huennekens et al.
1982). Once cob(II)alamin is reduced, methylation is dependent on SAM as the methyl
donor for the formation of CH3Cbl. The E. coli cob(II)alamin reductase requires FldA
for the reductive activation of MS (Fujii and Huennekens 1974). Humans lack FldA,
requiring an alternate reductase for CH3Cbl formation.
Because of the role of FldA in bacteria, it was proposed that the human reductase
would contain binding sites for FMN and FAD (Leclerc et al. 1998). Using homology
based PCR, a cDNA with consensus sequences to predicted binding sites for FMN, FAD,
and NADPH was cloned and proposed to encode methionine synthase reductase (MSR),
the cob(II)alamin reductase required for MS activation (Leclerc et al. 1998). The MSR
gene (MTRR) was mapped to chromosome 5 at position p15.2 (Leclerc et al. 1999).
Analysis of the gene revealed that the N-terminal sequence is consistent with cytosolic
targeting, but alternative splicing at the 5’ end of MTRR mRNA generates a

mitochondrial targeting sequence (Leclerc et al. 1999). There have been no studies
18
documenting the role of MSR in the mitochondria. Banerjee and Olteanu (2001) showed
that MSR is a monomeric protein of 77 kDa, that when purified contains stoichiometric
amounts of FAD and FMN. In an NADPH-dependent reaction, purified MSR reduced
cob(II)alamin to cob(I)alamin for activation of MS in vitro (Olteanu and Banerjee 2001).
In these same studies, it was found that the reduction of cobalamin bound to MS by MSR
is dependent on electrostatic interactions.
Inherited Disorders of Cobalamin Metabolism
In humans, inherited disorders in the metabolic pathways of both AdoCbl and
CH3Cbl synthesis result in methylmalonic aciduria and homocystinuria. Methylmalonic
aciduria is caused by a block in the pathway of propionyl-CoA metabolism which causes
an accumulation methylmalonyl-CoA that is hydrolyzed to methylmalonic acid, and
excreted into the blood and urine (Rosenblatt and Cooper 1990). Clinically,
methylmalonic aciduria presents itself by lethargy, recurrent vomiting, dehydration,
respiratory distress, feeding difficulties, failure to thrive, hypotonia, and death (Ciani et
al. 2000). Homocystinuria is the result of a block in the methylation cycle for methionine
synthesis, and results in an accumulation of homocysteine, which is excreted into the
blood and urine (Rosenblatt and Cooper 1990). Homocystinuria presents itself by
megaloblastic anemia, delayed development, neurological disorders, cardiovascular
disease, and death (Kapadia 1995). All of the disorders involved in cobalamin
metabolism are inherited as autosomal recessive traits (Rosenblatt and Fenton 1999).
There have been no reported heterozygotes with the presentation of these diseases.
Inborn errors of cobalamin metabolism have been classified into eight distinct
complementation groups, cblA to cblH (Figure 1-5). This classification is based on

iochemical studies and complementation analysis of human cultured fibroblast cells
(Fenton and Rosenberg 1978a). Complementation analysis examines the uptake and
19
conversion of labeled precursors (propionate, CH3THF) by fused and unfused fibroblast
cell lines derived from two patients with cobalamin disorders. In these studies, fused
cells that stimulated the incorporation of labeled precursors complemented each other and
were assigned to different complementation groups. If the fused cells were unable to
stimulate incorporation of the precursors they were assigned to the same
complementation group (Gravel et al. 1975). An assay that has helped to determine if
defects are due to decreased synthesis of one or both cobalamin coenzymes involves
incubating the cultured fibroblasts with labeled HOCbl and measuring the cofactors
formed against control cell lines.
Combined Methylmalonic Aciduria and Homocystinuria
Defects in cellular cobalamin metabolism resulting in both methylmalonic aciduria
and homocystinuria are defined by the complementation groups cblF, cblC, and cblD.
Cultured fibroblasts from patients in these groups were analyzed and shown to have a
deficiency in the conversion of CNCbl or HOCbl to AdoCbl or CH3Cbl cofactors
(Qureshi et al. 1994). The absence of these cofactors renders both MCM and MS
inactive. There are over 100 patients with combined methylmalonic aciduria and
homocystinuria. The cblC complementation group has more than 100 patients, the cblD
group has only two patients who are siblings, and the cblF group has six patients
(Watkins and Rosenblatt 2001).
Individuals with the cblF disorder have impaired MCM and MS activities. The
total intracellular pools of AdoCbl and CH3Cbl are reduced when compared to control
cells. Cultured cells incubated with [ 57 Co]-labeled CNCbl showed an accumulation of

20
unmetabolized, non-protein bound CNCbl contained within the lysosomes (Qureshi et al.
1994). There was no synthesis of either AdoCbl or CH3Cbl in these cells. This disorder
is defined by a deficiency in the release of cobalamin from the lysosomes after
endocytosis making cobalamin unavailable to intracellular enzymes (Kapadia 1995). The
gene encoding cobalamin lysosomal transport has not been identified. Other clinical
presentations of this disorder include hypotonia, stomatitis, facial abnormalities,
developmental retardation, and failure to thrive (Fenton and Rosenberg 2000).
Individuals diagnosed with the cblC and cblD disorders have very similar
biochemical characteristics. Both of these groups have decreased total cobalamin content
in the liver, kidney, and fibroblasts (Kapadia 1995). Cultured cells from these patients
incubated with [ 57 Co]-labeled CNCbl or HOCbl have reduced conversion to labeled
AdoCbl or CH3Cbl when compared to controls (Mahoney et al. 1971). Because of the
reduced levels of these cofactors, MCM and MS activities are deficient in cultured cells
even though the enzymes are normal (Mellman et al. 1978). In light of these results, it is
proposed that cblC and cblD complementation groups result from a defect in cytosolic
cob(III)alamin reductase activity (Watkins and Rosenblatt 2001). Based on
complementation studies, the cblC and cblD groups are classified as mutations occurring
at unique chromosomal locations; however, the specific role of these two groups has not
been determined and their corresponding genes remain unidentified.
The clinical presentation of the cblC disorder is more severe than cblD. In addition
to methylmalonic aciduria and homocystinuria, the cblC group usually presents itself in
the neonatal stage (early-onset) resulting in megaloblastic anemia, congenital
malformations, lethargy, failure to thrive, poor feeding, and death (Rosenblatt et al.

1997). Both cases of the cblD disorders did not develop problems until later on in life
21
(late-onset) resulting in megaloblastic marrow, neuromuscular problems, methylmalonic
aciduria, homocystinuria, and mental retardation (Rosenblatt and Cooper 1990). Since
the cblD group is biochemically similar to cblC but presents itself as less severe, it is
suggested that cblD is a leaky form of cblC, or that it may be affecting an unidentified
step before or after the reduction of cobalamin (Qureshi et al. 1994). The cblC, cblD, and
cblF groups respond to pharmacological doses of HOCbl which reduces methylmalonic
aciduria and homocystinuria but does not eliminate them (Fenton and Rosenberg 2000).
Methylmalonic Aciduria without Homocystinuria
Defects in the cellular metabolism of cobalamin resulting in methylmalonic
aciduria without homocystinuria are defined by cblA, cblB, and cblH complementation
groups. Complementation analysis of fibroblast cells from these patients has shown that
the cblA, cblB, and cblH defects occur at three distinct chromosomal locations (Qureshi
et al. 1994, Watkins et al. 2000). Patients with these disorders have a deficiency in
intracellular AdoCbl levels and have normal CH3Cbl levels (Watkins and Rosenblatt
2001). There are at least 45 patients diagnosed with cblA, 36 individuals with cblB, and
only one reported individual with cblH disorder (Watkins and Rosenblatt 2001).
The cblA and cblH complementation groups are deficient in mitochondrial AdoCbl
synthesis. Cultured cells from these patients were unable to convert [ 57 Co]-labeled
HOCbl to labeled AdoCbl but had no abnormality in CH3Cbl synthesis (Fenton and
Rosenberg 2000). Cell extracts from cblA patients required a reducing system for the
synthesis of AdoCbl; however, intact mitochondria from these patients were unable to
synthesize AdoCbl even in the presence of a reducing system. On the other hand, control
intact mitochondria can synthesize AdoCbl from HOCbl without the addition of reducing

22
agents (Fenton and Rosenberg 1978b), suggesting that the cblA disorder could be due to a
deficiency in cobalamin reduction or abnormalities of mitochondrial cobalamin binding
and transport (Mahoney et al. 1975).
The gene encoding the cblA disorder was identified (MMAA) and shown to localize
to chromosome 4q31.1-2 (Dobson et al. 2002). It appears that this gene product has
mitochondrial functionality due to the presence of an N-terminal MTS. Dobson and
colleagues suggested that the MMAA protein does not encode a mitochondrial cobalamin
reductase because it lacks NADPH, and flavin binding motifs that would be essential for
reductase activity (Watanabe et al. 1987), (Dobson et al. 2002). The MMAA protein has
61% similarity to ArgK in E. coli, an accessory protein common to lysine (L), arginine
(A), ornithine (O), and LAO transport systems (Dobson et al. 2002). It is thought that
ArgK could have some function in B12 transport in E. coli and has been proposed that
MMAA could encode a mitochondrial cobalamin transporter (Dobson et al. 2002).
Recently, MeaB, a homolog of the human MMAA, from Methylobacterium
extorquens AM1 was identified (Korotkova et al. 2002). Homology searches (using
MeaB and MCM) identified a fusion protein from Burkholderia fungorum with an
N-terminal MeaB domain and a C-terminal MCM domain (Korotkova and Lidstrom
2004). The MCM and MeaB homologue fusion protein in B. fungorum suggests that
these two proteins would interact in systems where the genes are expressed separately
(Korotkova and Lidstrom 2004). It has been shown that the M. extorquens MeaB exists
as a complex with MCM (Korotkova and Lidstrom 2004). It is now thought that the cblA
group (MMAA) encodes a protein that would interact with MCM and protect it from
inactivation (Korotkova and Lidstrom 2004).

23
The cblB complementation group is characterized by a deficiency in the conversion
of cobalamin to AdoCbl. Cell extracts derived from patients with cblB disorder were
unable to synthesize AdoCbl even when provided with a reducing system, labeled
HOCbl, and ATP (Fenton and Rosenberg 1981). Patients with this disorder are defective
in ATP:cob(I)alamin ATR activity. At the start of this study, the gene encoding the cblB
defect was unknown and the protein had not been isolated.
The clinical presentation of cblA, cblB, and cblH disorders are very similar and
include methylmalonic aciduria, lethargy, failure to thrive, vomiting, dehydration,
respiratory distress, and anemia (Fenton and Rosenberg 2000). These disorders have
been treated by a diet restricted in amino acid precursors of methylmalonate and
cobalamin supplementation (Kapadia 1995).
Homocystinuria without Methylmalonic Aciduria
Inborn errors of cellular cobalamin metabolism resulting in homocystinuria without
methylmalonic aciduria are defined by cblE and cblG complementation groups. There
has been only one patient in the cblE group that had combined homocystinuria and mild
methylmalonic aciduria (Wilson et al. 1999). Cultured fibroblasts from these patients
show a deficiency in CH3Cbl synthesis and decreased methionine synthase activity
(Watkins and Rosenblatt 1989). Complementation analysis has shown that cblE and
cblG defects occur at distinct chromosomal locations (Watkins and Rosenblatt 1988).
Fibroblasts from patients with cblE and cblG disorders showed normal incorporation of
labeled propionate into macromolecules; however, the incorporation of labeled CH3THF
was reduced when compared to controls suggesting an abnormality in the methionine
synthase pathway (Watkins and Rosenblatt 1989). To date there are 12 patients

24
diagnosed with the cblE disorder and 28 individuals with the cblG disorder (Fenton and
Rosenberg 2000).
Members of the cblE complementation group are deficient in methionine synthase
activity due to a block in CH3Cbl formation. Fibroblast extracts from patients with this
disorder had normal methionine synthase activity when high concentrations of reducing
agent were added to the assay, but activity was reduced or absent when the reducing
agent was used at lower concentrations (Watkins and Rosenblatt 1988). From these
studies it was shown that the cblE group is defective in cobalamin reduction. The gene
corresponding to this group was identified (MTRR) and localized to chromosome
5p15.2-p15.3 (Leclerc et al. 1998). The MTRR protein or MSR is required for the
reductive activation of methionine synthase and was reviewed above (Wilson et al. 1999).
The cblG complementation group is also deficient in methionine synthase activity
but has normal levels of CH3Cbl. Fibroblast extracts from these patients were assayed
and showed no MS activity under optimal conditions (Watkins and Rosenblatt 1989).
This group is defined by defects in MS. Human MS is a modular enzyme with domains
containing binding sites for homocysteine, CH3THF, SAM, and cob(II)alamin (Banerjee
1997). The cDNAs from cblG mutants have been analyzed and shown to contain
missense and nonsense mutations (Leclerc et al. 1996). It appears that a majority of
deficiencies in the cblG group are due to the lack of MS protein (caused by a premature
stop codon due to nonsense mutations) and not in cobalamin binding. The clinical
presentations of the cblE and cblG groups include homocystinuria without methylmalonic
aciduria, developmental delay, megaloblastic anemia, and neurological disorders

(Watkins et al. 2002). Some patients respond to treatment with HOCbl (Watkins and
Rosenblatt 2001).
25
Salmonella as a Model Organism for the Study of Cobalamin Metabolism
Salmonella enterica is an important model organism for studies of B12-dependent
processes (Schneider and Stroinski 1987, Roth et al. 1996). More is known about
cobalamin metabolism and physiology in S. enterica than in any other single organism.
Cobalamin transport, cobalamin biosynthesis, intramolecular rearrangements, and methyl
transfer reactions have been investigated in S. enterica (Roth et al. 1996). In addition,
S. enterica has a well-defined genetic system that includes methods for transformation,
transduction, gene expression, directed, chemical, and transposon mutagenesis and a
known genome sequence.
One of the best-studied AdoCbl dependent enzymes in S. enterica is diol
dehydratase. This enzyme is crucial for the metabolism of 1,2-propanediol. The genes
required for growth of S. enterica on 1,2-propanediol are organized at the propanediol
utilization (pdu) locus. Determination of the DNA sequence indicated that this locus has
23 genes (Bobik et al. 1997, Bobik et al. 1999): six pdu genes are thought to encode
enzymes needed for the 1,2-propanediol degradative pathway (pduCDEPQW) (Bobik et
al. 1997); two are involved in transport and regulation (pduF and pocR) (Bobik et al.
1992, Chen et al. 1994); two are probably involved in diol dehydratase reactivation
(pduGH) (Bobik et al. 1999); one is needed for the conversion of cobalamin to AdoCbl
(pduO) (Johnson et al. 2001); five are of unknown function (pduLMSVX); and seven
share similarities to genes involved in the formation of carboxysomes (pduABJKNTU),
polyhedral bodies found in certain cyanobacteria and chemoautotrophs (Shively and
English 1991, Shively et al. 1998). The large number of genes required for the

26
degradation of propanediol suggests that this compound serves an important purpose in
the lifestyle of Salmonella (Roth et al. 1996).
Salmonella uses propanediol as a carbon and energy source in a B12-dependent
manner. The proposed pathway of propanediol degradation was determined based on
biochemical studies (Figure 1-6). The pathway begins with the conversion of
1,2-propanediol to propionaldehyde, a reaction catalyzed by the AdoCbl-dependent diol
dehydratase (Toraya et al. 1979, Obradors et al. 1988). Propionaldehyde is then
disproportionated to propanol and propionic acid by a reaction series thought to involve
propanol dehydrogenase, phosphotransacylase, and propionate kinase (Toraya et al. 1979,
Obradors et al. 1988). Aerobic growth of S. enterica on 1,2-propanediol requires
addition of CNCbl (or other complex cobalamin precursors) to growth media.
Research Overview
Coenzyme B12-dependent processes are essential for human health.
Methylmalonyl-CoA mutase and methionine synthase, the only two reported
B12-dependent enzymes in humans, are crucial for propionate metabolism and methionine
synthesis. Inborn errors of cobalamin metabolism can block these pathways resulting in
methylmalonic aciduria and homocystinuria, serious diseases that are often fatal in
newborns. Cobalamin metabolism has been studied in both bacterial and eukaryotic
systems, and has been found to be quite similar. Progress in identifying human genes
involved in B12 metabolism has been slow due to difficulties in purifying the
corresponding enzymes and the lack of facile genetic techniques. To circumvent these
problems, S. enterica was used as a model system for studies of B12 metabolism in
humans.

Chapter 2 of this dissertation reports the identification of the human
27
ATP:cob(I)alamin adenosyltransferase involved in AdoCbl metabolism. An S. enterica
mutant deficient in ATR activity allowed for the isolation of a bovine ATR cDNA.
Subsequent sequence similarity searching was used to identify a homologous human
cDNA. Both the bovine and human cDNAs were independently cloned and
overexpressed in E. coli. Enzyme assays showed that both the bovine and human
enzymes had ATR activity in vitro. Subsequent studies showed that the human cDNA
clone complemented an ATR-deficient S. enterica strain for AdoCbl-dependent growth
on 1,2-propanediol in vivo. In addition, Western blots were used to show that ATR
expression is altered in cell lines derived from cblB methylmalonic aciduria patients
compared with cell lines from normal individuals.
Chapter 3 of this dissertation reports the biochemical properties of the human ATR
and its interaction with methionine synthase reductase. Two common polymorphic
variants of the ATR, which are found in normal individuals, were expressed in E. coli and
purified to apparent homogeneity. Purified ATR variants were used for kinetic studies
and the catalytic properties of both ATR variants including the Vmax and the Km for ATP
and cob(I)alamin were determined. Investigations also showed that the purified
methionine synthase reductase in combination with purified ATR can convert
cob(II)alamin to AdoCbl in vitro. In this system, MSR reduced cob(II)alamin to
cob(I)alamin which was adenosylated to AdoCbl by the ATR enzyme and results
indicated that MSR and ATR interact in such a way that the highly reactive intermediate
(cob(I)alamin) was sequestered. The finding that MSR can reduce cob(II)alamin to
cob(I)alamin for AdoCbl synthesis (in conjunction with the prior finding that MSR

educed cob(II)alamin for the activation of methionine synthase) indicates a dual
physiological role for the MSR enzyme.
28
Chapter 4 of this dissertation is not directly related to B12 metabolism in humans,
but was the research performed during my first two years in Dr. Bobik’s Laboratory.
This work focused on the identification of PduP, a coenzyme-A-acylating
propionaldehyde dehydrogenase associated with polyhedral bodies involved in
AdoCbl-dependent 1,2-propanediol degradation by S. enterica. A PCR based method
was used to construct a precise nonpolar chromosomal deletion of the gene pduP. The
resulting pduP deletion strain grew poorly on 1,2-propanediol minimal media and
expressed 105-fold less propionaldehyde dehydrogenase activity than did wild-type
S. enterica grown under similar conditions. An E. coli strain was constructed for
high-level production of His8-PduP, which was purified by nickel-affinity
chromatography and was shown to have propionaldehyde dehydrogenase activity.
Reverse-phase High Performance Liquid Chromatography (HPLC) and mass
spectrometry were used to verify the product of the PduP reaction. Immunogold electron
microscopy was used to show that PduP is associated with a polyhedral organelle
involved in AdoCbl-dependent 1,2-propanediol degradation.

Corrin
Ring
Dimethyl
benzimidazole
H H2NC 2NC
O H O 3C H3C O
H H2NC 2NC
Aminopropanol
H H3C 3C
H H3C 3C
H H3C 3C
C
C
H H3C 3C
H H2NC 2NC
CH
C
C N N
Co 3+
Co 3+
C
C N N
O
HN
A
O
C
D C
C
C C
C
C
C
H H3C 3C
H
C
CH 3
R
C
CH 3
N
N
OH
29
R= upper β-ligand
H H3C 3C CNH 2
C
C
CNH 2
O
O
O
B
O
O
C
C
P O
O
CH
CH CH2OH 2OH
CH 3
CH 3
O
CNH 2
R= Active
Adenosyl
AdoCbl
Methyl
CH CH3Cbl 3Cbl
R= Inactive
Cyano
CNCbl,
vitamin B B12 12
Hydroxy
HOCbl
Glutathionyl
GSCbl
N
NH 2
N
OH
CH 2
R
R
O
N
N
OH
R CH 3
C N
R OH
R GS
Figure 1-1. Structure of cobalamins. The major parts of cobalamin are labeled: upper
β-ligand, corrin ring, aminopropanol, and dimethylbenzimidazole. The four
pyrrole rings of the corrin are labeled with capital letters. The upper ligand
can be any of the groups listed in the legend. Active and inactive forms of
cobalamin are dependent on the upper ligand.

Table 1-1. Adenosylcobalamin- and methylcobalamin-dependent reactions
AdoCbl-dependent Isomerases MeCbl-dependent Methyltransferases
Diol dehydratase:
Methionine synthase:
1,2-propanediol propionaldehyde + homocysteine CH3THF methionine +
H20
THF
Glycerol dehydratase:
glycerol β-hydroxypropionaldehyde
+ H20
Ethanolamine ammonia lyase:
ethanolamine acetaldehyde + H20
L-β-Lysine aminomutase:
L-β-lysine
L-erythro-3,5-diaminohexanoic acid
D-α-Lysine aminomutase:
D-α-lysine 2,5-diaminohexanoic acid
D-Ornithine aminomutase:
D-ornithine 2,4-diaminovaleric acid
Methylmalonyl-CoA mutase:
methylmalonyl-CoA succinyl-CoA
Glutamate mutase:
glutamate β-methylaspartate
Isobutyryl-CoA mutase:
isobutyryl-CoA butyryl-CoA
Methyleneglutarate mutase:
2-methylene glutarate
3-methylitaconate
Ribonucleotide reductase:
NTP dNTP
30
Acetyl -CoA synthesis via Corrinoid/Fe
protein:
(a) CH3THF + corrinoid/FeS protein
THF + CH3-corrinoid Fe/S protein
(b) CH3-corrinoid Fe/S protein + carbon
monoxide corrinoid Fe/S protein +
acetyl-CoA
Methyltetrahydromethanopterin:coenzyme
M methyltransferase:
CH3-H4MPT + coenzyme M
H4MPT + CH3-coenzyme M
Methanol:2-mercaptoethanesulfonic acid
methyltransferase:
CH3OH + coenzyme M
CH3-coenzyme M + H20

Classes
Eliminases
Dehydratases
OH
Aminomutases
OH
1,2-propanediol
H 2N COH
NH 2
Carbon skeleton mutases
HOC
Ribonucleotide reductases
(P)PPO
ethanolamine
methylmalonyl-CoA
H
H
OH
O
NDP, NTP
Methyltransferases
O
Deaminases
H 2N
O
H
C
CH 3
NH 2
H
OH
H
OH
O
BASE
31
AdoCbl
diol dehydratase
ethanolamine ammonia lyase
L-β-lysine 5,6 aminomutase
O
propionaldehyde
H 3C
CO SCoA
AdoCbl
methylmalonyl CoA-mutase
HOC C
cob(I)alamin
Reactions
AdoCbl
AdoCbl
CH 3Cbl
O
CH
acetaldehyde
β-lysine L-erythro 3,5 diaminohexanoic acid
AdoCbl
ribonucleotide reductase
H 2
C
H 2
+
+
succinyl-CoA
HOC CH CH2 CH2 SH
CH3THF THF
HOC CH CH2 CH2 homocysteine
methionine synthase
methionine
O
H 3C
O
(P)PPO
NH 2
NH 2
H
H
OH
NH 2
O
dNDP, dNTP
H 2O
NH 3
O
COH
CO SCoA
H
H
H
BASE
S CH 3
Figure 1-2. Different classes of cobalamin-dependent enzymes. The enzymes involved
and the reactions they catalyze are highlighted.

H3C C CO SCoA
H 2
propionyl-CoA
carboxylase
HOOC
propionyl-CoA
CH 3
CH CO
Biotin, HCO 3 2- ,
ATP, Mg 2+
SCoA
(2S)-methylmalonyl-CoA
methylmalonyl-CoA
epimerase
HOOC
(2R)-methylmalonyl-CoA
methylmalonyl-CoA
mutase
HOOC C C
H 2
H
C CO SCoA
CH 3
H 2
succinyl-CoA
CENTRAL
METABOLISM
adenosylcobalamin
CO SCoA
32
Valine
Isoleucine
Methionine
Threonine
Odd chain fatty acids
Thymine
Cholesterol
HOOC
adenosyltransferase
CH 3
CH COOH
methylmalonic acid
cob(I)alamin
cob(II)alamin
cob(II)alamin
reductase
cob(III)alamin
reductase
GSCbl
β-ligand
transferase
HOCbl
Figure 1-3. Pathways of propionyl-CoA catabolism and cobalamin metabolism.
Propionyl-CoA carboxylase catalyzes the formation of
(2S)-methylmalonyl-CoA. Methylmalonyl-CoA epimerase catalyzes the
conversion of (2S)-methylmalonyl-CoA to the (2R) isomer.
AdoCbl-dependent methylmalonyl-CoA mutase catalyzes the conversion of
(2R)-methylmalonyl-CoA to succinyl-CoA. In humans, a deficiency in
MCM or in any of the metabolic steps needed for AdoCbl synthesis leads to
methylmalonic aciduria.

A)
B)
CH 3THF
33
MS-cob(I)alamin
Methionine
MS-CH3Cbl Synthase
MS-cob(II)alamin
methionine
Methionine
Synthase
Reductase
THF
homocysteine
MS-CH 3Cbl
SAM SAH
Figure 1-4. Pathways involving methionine synthase and the reductive activation of
methionine synthase by methionine synthase reductase. A) Proposed
pathway of methionine synthesis. In this pathway, CH3Cbl-dependent MS
transfers a methyl from CH3THF to homocysteine forming methionine. B)
Reductive activation of MS-bound cob(II)alamin. MSR reduces cobalamin,
and SAM provides a methyl for the reactivation of MS which reenters the
methylation cycle.

TCII
HOCbl
β-ligand
transferase
Lysosome
TCII
HOCbl
TCII
HOCbl
HOCbl
GSCbl
cblF
Cytoplasm
34
Methylmalonyl-CoA
Methylmalonyl-CoA
Mutase mut
Adenosylcobalamin
Succinyl-CoA
cblC, cblD
Mitochondrion
Cob(III)alamin Reductase
cblA
Adenosyltransferase
Cob(II)alamin
Reductase
MS Reductase
SAM cblE
MS-CH MS-CH3Cbl 3Cbl
cblB
Cob(I)alamin
cblH
Cob(II)alamin
Cob(II)alamin
MS-Cob(II)alamin
CH CH3THF 3THF
MS-Cob(I)alamin
Homocysteine
cblG
Methionine Synthase
Methionine
Figure 1-5. Cobalamin metabolism in mammalian cells, and disorders associated with
deficiencies in this pathway. The pathways by which inactive cobalamin
precursors are taken up and metabolized to active cofactors are shown. The
cytoplasmic and mitochondrial compartments are labeled. The specific steps
affected by inborn errors of cobalamin metabolism (mut, cblA-cblH) are
labeled.

propanol
dehydrogenase
(PduQ)
propanol
OH
OH
diol dehydratase
(PduCDE)
NAD +
35
1,2-propanediol
OH
O
propionaldehyde
NADH
+ H +
adenosylcobalamin
NAD +
NADH
+ H +
adenosyltransferase
(PduO)
CoA-dependent
propionaldehyde dehydrogenase
(PduP)
O
SCoA
propionyl-CoA
O
PO 3 2-
propionyl-phosphate
ADP
propionate
Figure 1-6. Proposed pathway of AdoCbl-dependent 1,2-propanediol degradation in
S. enterica. AdoCbl-dependent diol dehydratase catalyzes the conversion of
1,2-propanediol to propionaldehyde. Propionaldehyde is then
disproportionated into propanol and propionate presumably by aldehyde
dehydrogenase (PduP), phosphotransacylase, propionate kinase (PduW) and
propanol dehydrogenase (PduQ). When grown aerobically, S. enterica is
unable to synthesize cobalamin de novo and requires cobalamin precursors
(XCbl) for the metabolism to AdoCbl.
ATP
phosphotransacylase
propionate kinase
(PduW)
O
O
XCbl

CHAPTER 2
IDENTIFICATION OF THE HUMAN AND BOVINE ATP:COB(I)ALAMIN
ADENOSYLTRANSFERASE CDNAS BASED ON COMPLEMENTATION OF A
BACTERIAL MUTANT
Introduction
Enzymes dependent on the vitamin B12 coenzymes, AdoCbl and CH3Cbl have a
broad but uneven distribution among the three domains of life (Dolphin 1982, Schneider
and Stroinski 1987, Banerjee 1999). In higher animals, there are two known
cobalamin-dependent enzymes. CH3Cbl dependent MS is needed for the methylation of
homocysteine to methionine (Cauthen et al. 1966, Drummond and Matthews 1993), and
AdoCbl-dependent MCM plays an essential role in the conversion of propionyl-CoA to
the TCA cycle intermediate, succinyl-CoA (Banerjee and Chowdhury 1999, Matthews
1999). This latter process occurs in three steps: propionyl-CoA is carboxylated to
(2S)-methylmalonyl-CoA, isomerized to (2R)-methylmalonyl-CoA, and finally
rearranged to succinyl-CoA in a reaction catalyzed by AdoCbl-dependent MCM
(Figure 1-3). In higher animals, propionyl-CoA is produced from the breakdown of the
amino acids valine, isoleucine, methionine, and threonine, as well as thymine, cholesterol
and odd-chain fatty acids; hence, MCM is essential for the complete catabolism of each
of these compounds (Banerjee and Chowdhury 1999).
In humans, inherited defects that impair the activity of AdoCbl-dependent MCM
lead to methylmalonic aciduria, a rare but severe disease that is often fatal in the first year
of life (Rosenblatt and Fenton 1999, Fenton et al. 2000, Olteanu and Banerjee 2001).
Such inherited deficiencies can result from mutations in the MCM structural gene (mut)
36

or from mutations that impair the acquisition of the required cofactor, AdoCbl
(Rosenblatt and Fenton 1999, Watkins and Rosenblatt 2001). Higher animals are
incapable of de novo synthesis and, hence, must obtain AdoCbl by synthesis from
37
complex precursors taken up from their diet (Banerjee 1999). Suitable precursors include
vitamin B12 (CNCbl) and other cobalamins with various β-ligands (XCbls) (Banerjee
1999). The pathway by which XCbls are converted to AdoCbl has been studied in
several organisms and is thought to be similar in both prokaryotes and eukaryotes
(Figure 1-3): XCbl is converted to glutathionyl-cobalamin (GSCbl), reduced to
cob(II)alamin, further reduced to cob(I)alamin and finally adenosylated to AdoCbl
(Friedmann 1975, Huennekens et al. 1982, Pezacka et al. 1990, Pezacka 1993, Watanabe
et al. 1996). In humans, four complementation groups (cblABCD), associated with
methylmalonic aciduria, are thought to correspond to genes involved in the conversion of
XCbls to AdoCbl (Figure 1-5). Enzymatic assays of fibroblast extracts have indicated
that the cblC and cblD complementation groups encode cytoplasmic enzyme(s) needed
for the conversion of XCbls to cob(II)alamin (Mellman et al. 1979). Similar studies have
indicated that the cblA and cblB complementation groups correspond to a mitochondrial
cob(II)alamin reductase and an ATP:cob(I)alamin ATR enzymes, respectively (Mahoney
et al. 1975, Fenton and Rosenberg 1981). To date, the human genes that correspond to
the cblABCD complementation groups have not been identified. Progress in this area has
been slow due to the difficulties in purifying the relevant proteins (Rosenblatt and Cooper
1990).
We recently identified the PduO ATR of Salmonella, and showed that this enzyme
has partial functional redundancy with the CobA enzyme (Johnson et al. 2001). In these

studies, it was also shown that a Salmonella pduO cobA double mutant lacked
38
AdoCbl-dependent diol dehydratase activity due to the ATR deficiency. Furthermore, it
was found that a plasmid-encoded source of ATR enzyme restored diol dehydratase
activity to a cobA pduO double mutant (Johnson et al. 2001). Because this enzymatic
activity can be readily detected on aldehyde indicator medium (AIM), we expected that
the ATR-deficient Salmonella strain could be used to screen expression libraries for
cDNAs that encode ATR enzymes. Such an approach would circumvent the difficulties
associated with ATR purification.
Here, we report the isolation of an ATR cDNA, from a bovine liver library, by
complementation of an ATR-deficient Salmonella mutant for color formation on
aldehyde indicator medium. Subsequently, sequence similarity searching identified the
homologous human cDNA and its corresponding gene. Both the human and bovine
cDNAs are shown to express ATR activity and complement an ATR-deficient bacterial
mutant. In addition, Western blots were used to show that expression of the human ATR
was altered in cell lines derived from three cblB patients compared with a cell line
derived from a normal individual. We propose that the human gene identified here
coincides with the cblB complementation group, defects in which lead to methylmalonic
aciduria. The identification of genes involved in methylmalonic aciduria is important for
the development of improved methods for the diagnoses and treatment of this devastating
disorder.
Chemicals and Reagents
Materials and Methods
CNCbl and HOCbl were purchased from Sigma Chemical Company (St. Louis,
MO). Titanium (III) citrate was prepared as previously described (Bobik and Wolfe

1989). 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-Gal) and
isopropyl-β-D-thiogalactopyranoside (IPTG) were from Diagnostic Chemicals Ltd.
39
(Charlottetown, Prince Edward Island, Canada). Restriction enzymes and T4 DNA ligase
were from New England Biolabs (Beverly, MA). Other chemicals were from Fisher
Scientific (Norcross, GA).
Bacterial Strains and Growth Media
Bacterial strains used in this study are listed in Table 2-1. The minimal medium
used was NCE (Vogel and Bonner 1956, Berkowitz et al. 1968) supplemented with 0.4%
1,2-propanediol, 200 ng/ml of CNCbl, 1 mM MgSO4, 0.3 mM each of valine, isoleucine,
leucine, and threonine. LB medium was the rich medium used (Difco Laboratories,
Detroit, MI) (Miller 1972). MacConkey and aldehyde indicator media were
supplemented with 1,2-propanediol and CNCbl and prepared as previously described
(Johnson et al. 2001).
General Protein and Molecular Methods and ATR Assays
Bacterial transformation, polymerase chain reaction (PCR), restriction enzyme
digests, and other standard molecular and protein methods were performed as previously
described (Maniatis et al. 1982, Johnson et al. 2001). ATR assays were also performed
as previously reported (Johnson et al. 2001).
P22 Transductions
Transductional crosses were performed as described previously (Davis et al. 1980).
In preparing lysates of galE mutant strains, overnight cultures were grown on LB
medium supplemented with 0.2% galactose and 0.2% glucose with the addition of
appropriate antibiotics. A P22 HT105/1 int-201 phage was used for transductional
crosses at a concentration of 2x10 8 phage/ml (Schmieger 1971).

40
Screening the Bovine and Human Liver cDNA Libraries
Uni-ZAP XR bovine and human liver cDNA libraries were from Stratagene (La
Jolla, CA). Titering, amplification, and mass excision of the cDNA carried on
pBlueScript SK(+/-) were done according to the Manufacturer’s protocol except that the
titering procedure used F-top agar with 0.2 mM thymine, but no NaCl (Miller 1972)
instead of NZY top agar. Following mass excision, the cDNA expression plasmids were
purified using a QIAprep spin mini prep kit (Qiagen, Chatsworth, CA). A portion of the
resulting expression library was used to transform S. enterica strain BE253 by
electroporation. Strain BE253 is ATR-deficient due to pduO and cobA mutations, and
carries pSJS1240 which provides rare tRNAs that enhance expression of heterologous
genes in S. enterica. Following electroporation, transformation mixtures were suspended
in molten aldehyde indicator medium that had been supplemented with 1,2-propanediol,
HOCbl and 100 µg/ml ampicillin (Amp), and cooled to 45 ºC. The molten medium was
poured into 100 x 15 mm sterile Petri dishes, allowed to solidify, and incubated at 37 ºC
in the dark for ~12 h. Resultant colonies were screened for red/brown color formation.
This procedure allowed screening of about 5,000 transformants per plate.
Cloning the Bovine Adenosyltransferase Coding Sequence for High-Level
Expression
PCR was used to amplify the bovine ATR coding sequence. Plasmid pNL121,
isolated in this study from a bovine cDNA library, provided the template DNA. The
primers used for amplification of the full-length coding sequence were
5’-GCCGCCGGTACCGATGACGACGACAAGTTCGGCACGAGCCCGGGAGGT-3’
(forward) and 5’-GCCGCCAAGCTTGCTTGGTTCCTCGATGAAGCA-3’ (reverse).
To eliminate the predicted mitochondrial targeting sequence (MTS), forward primer

41
5’-GCCGCCGGTACCGATGACGACGACAAGCCCCAGGGCGTGGAAGACGGG-3’
was used in conjunction with the reverse primer described above. These primers
introduced KpnI and HindIII restriction sites into the PCR products and were designed
such that following cloning, the bovine ATR coding sequence would be fused to both
N-terminal glutathione S-transferase and His6 tags. PCR products obtained using the
primers described above were restricted with KpnI and HindIII, and ligated to the
pET-41a expression vector (Novagen, Cambridge, MA) that had been similarly digested
(Maniatis et al. 1982). Ligation mixtures were used to transform E. coli DH5α by
electroporation and transformants were selected by plating on LB-kanamycin (Kan)
medium. Pure cultures were prepared from selected transformants and plasmid DNA
isolated from these strains was analyzed by restriction digestion and DNA sequencing.
Clones having the expected DNA sequences were transformed into E. coli strain BL21
(DE3) RIL (Stratagene) for high-level expression.
Cloning the Human Adenosyltransferase Coding Sequence for High-Level
Expression
The human ATR coding sequence was cloned via PCR using a strategy similar to
that described above for the bovine enzyme. IMAGE cDNA clone 2822202 provided the
template DNA (Incyte Genomics, Palo Alto, CA). For amplification of the full-length
coding sequence, the following primers were used: forward
5’-GCCGCCAGATCTGGATGACGACGACAAGATGGCTGTGTGCGGCCTGG-3’
and reverse 5’-GCCGCCAAGCTTTCAGAGTCCCTCAGACTCGGCCG-3’. To
eliminate the putative MTS, primer 5’-GCCGCCAGATCTGGATGACGACGACAAGC
CTCAGGGCGTGGAAGACGGG-3’ was used as the forward primer. The primers
described above, introduced BglII and HindIII restriction sites that were used for cloning

42
into the pET-41a expression vector. The BglII site was positioned such that the resulting
clones would express the human ATRs as fusion proteins with N-terminal GST and His6
tags. Ligation, transformation and analysis of clones were performed as described above
for the bovine expression clones. Clones with the expected DNA sequence were
transformed into E. coli strain BL21 (DE3) RIL for high-level expression.
Cloning the Human Adenosyltransferase Coding Sequence for Complementation
Studies
For complementation studies, the human ATR coding sequence was amplified by
PCR and cloned into plasmid pLAC22 which allows IPTG-inducible expression in
S. enterica (Warren et al. 2000). The primers used for amplification were forward,
5’-GCCGCCAGATCTTATGCCTCAGGGCGTGGAAGACGGG-3’ and reverse,
5’-GCCGCCAGCTTTCAGAGTCCCTCAGACTCGGCCG-3’. These primers eliminate
the putative MTS and provide the needed ATG start triplet such that the first five amino
acids of the expressed protein would be MPQGV. The PCR product was restricted with
BamHI and HindIII and ligated into pLAC22 that had been digested with BglII and
HindIII. Ligation mixtures were used to transform S. enterica strain BE253
(ATR-deficient) by electroporation. Transformants were selected on AIM supplemented
with 1,2-propanediol and Amp. Use of this medium allowed the identification of clones
expressing ATR activity. Pure cultures were prepared from selected transformants and
plasmid DNA isolated from these strains was analyzed by restriction digestion and DNA
sequencing. Clones having the expected DNA sequences were moved into strain BE266
(pduO cobA) via P22 transduction. Pure cultures prepared from the resultant colonies
were determined to be phage-free by cross-streaking against P22 H5 and used for
complementation studies.

43
Growth of Adenosyltransferase Expression Strains and Preparation of Cell Extracts
The E. coli strains used for expression of the bovine and human ATRs were grown
on LB supplemented with 25 µg/ml Kan at 37°C with shaking at 275 rpm in a New
Brunswick Scientific shaker incubator. Cells were grown to an absorbance of 0.6-0.8 at
600 nm and protein expression was induced by the addition of 1 mM IPTG. Cells were
incubated at 37°C with shaking at 275 rpm for an additional 3 hours. Cells were removed
from the incubator, held on ice for 10 minutes, and then harvested by centrifugation at
6,690 x gmax for 10 minutes using a Beckman JLA-10.500 rotor. The cells were
resuspended in 3 ml of 50 mM sodium phosphate, 300 mM NaCl, pH 7, and broken using
French Pressure Cell (SLM Aminco, Urbana, IL) at 20,000 psi. Phenylmethylsulfonyl
fluoride was added to the cell extract to a concentration of 100 µg/ml to inhibit proteases.
The crude cell extract was centrifuged at 16,000 rpm (31,000 x gmax) for 30 minutes using
a Beckman JA20 rotor to separate the soluble and insoluble fractions. The supernatant
was decanted (soluble fraction), and the pellet was resuspended in 1 ml of 50 mM sodium
phosphate, 300 mM NaCl, pH 7 (inclusion body fraction). Both fractions were used for
the ATR assays.
Growth Curves
Cells to be used as inocula for growth curves were grown overnight at 37°C in 2 ml
LB medium or LB medium supplemented with Amp at 100 µg/ml for strains carrying
pLAC22. A portion of the LB culture (1.5 ml) was pelleted by centrifugation, and
resuspended in 1 ml of minimal medium. Then resuspended cells (125 µl) were used as
the inoculum. Cells were grown in 16 x 150-mm culture tubes containing 5 ml of
minimal medium and incubated at 37°C in a New Brunswick model C-24 water bath with

44
a shaking speed of 250 rpm. Cell growth was determined by measuring the absorbance at
600 nm using a Milton Roy model Spectronic 20D + spectrophotometer.
Western Blots
Human skin fibroblasts from one normal individual (MCH45) and three patients
diagnosed with cobalamin b (cblB) disorder (WG1680, WG1879, and WG2127) were
obtained from the Repository for Mutant Human Cell Strains at Montreal Children’s
Hospital. These cell lines were grown in Eagle’s minimum essential medium
supplemented with Earle’s salts, L-glutamine, 1% nonessential amino acids, 10%
heat-inactivated fetal bovine serum, 100 IU/ml penicillin, 100 µg/ml streptomycin, and
0.2 µg/ml HOCbl and incubated at 37°C with 5% CO2, 95% air mixture. For lysate
preparation, each of the cell lines were cultured in 3, T150 cm 2 tissue culture flasks and
grown to confluence. Harvested fibroblasts were washed twice with phosphate-buffered
saline and lysed in 50 mM Tris, pH 7.3, 100 mM KCl, 5 mM MgCl2, 1.6% Triton-X-100
and a protease inhibitor mixture (Roche Diagnostics, Manheim, Germany). Insoluble
cellular matter was removed by centrifugation at 12,000 x g at 4 °C for 15 minutes.
Protein concentration of the soluble extract was determined using a Micro BCA assay
(Pierce, Rockford, IL). Cell Extracts (220 µg protein) were fractionated by 10% sodium
dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a
Hybond-P, polyvinylidene difluoride membrane (Amersham Biosciences). The
membrane was incubated at room temperature for one hour with rocking in 10 mM
Tris-HCl, pH 8, 150 mM NaCl, 0.05% Tween 20 (TBST) containing 3% bovine serum
albumin and 3% mouse serum. The primary antibody used was affinity-purified rabbit
IgG anti-human-ATR diluted 1:1000. The secondary antibody was monoclonal

anti-rabbit IgG (γ-chain specific) biotin conjugate (Sigma) and was used at a 1:25,000
45
dilution. All membrane incubations were carried out at room temperature for one hour
with rocking in TBST. The membrane was prepared for detection using streptavidin
horseradish peroxidase conjugate (Caltag Laboratories, Burlingame, CA) diluted
1:10,000, and Immun-Star HRP Chemiluminescent Kit (Bio-Rad, Hercules, CA)
following the Manufacturer’s protocol. Membranes were exposed to X-ray film from
Research Products International (Mt. Prospect, IL).
DNA Sequencing and Analysis
DNA sequencing was carried out by University of Florida, Interdisciplinary Center
for Biotechnology Research, DNA Sequencing Core facility, and University of Florida,
Department of Microbiology and Cell Science, DNA Sequencing Facility as described
previously (Bobik et al. 1999). Sequence similarity searches were carried out using the
Blast family of sequence analysis software (Altschul et al. 1990, Altschul et al. 1997).
Multiple sequence alignments were done with ClustalX using default parameters except
as noted below (Thompson et al. 1997). The slow and accurate method was used for
pairwise alignments with a gap opening and extension penalties of 35 and 0.75,
respectively. Multiple sequence alignment parameters were gap opening and extension
penalties of 15 and 0.3, respectively, and delay divergent sequences were set at 25%.
Results
Screening of the Bovine and Human Liver cDNA Libraries for Clones that Express
Adenosyltransferase Activity
Our previous studies showed that S. enterica strain BE253 forms red/brown
colonies on AIM supplemented with 1,2-propanediol and CNCbl when it is transformed
with an ATR expression plasmid (Johnson et al. 2001). The basis of the color formation

46
is that ATR activity restores the ability of strain BE253 to convert CNCbl to AdoCbl, the
required cofactor for diol dehydratase. Given AdoCbl, diol dehydratase converts
1,2-propanediol to propionaldehyde which reacts with components in the aldehyde
indicator medium to form a red/brown precipitate (Johnson et al. 2001). Hence, we
expected that BE253 would be useful for screening expression libraries for ATR activity.
Accordingly, we attempted to identify a human ATR cDNA by transforming strain
BE253 with a human liver cDNA expression library and screening the resultant colonies
for red/brown color formation on AIM. Approximately 250,000 human cDNAs were
screened, but no red/brown colonies were found. Subsequently, a bovine liver cDNA
library was similarly screened and four positive clones were isolated. The DNA
sequences of the four clones were determined and found to be identical. The clone was
1058 bp in length, and included a poly(A) tail indicating that its 3' end was complete.
However, analyses indicated that the 5' end of the clone was incomplete. The clone
included a 228-amino acid open reading frame that would be expressed as an N-terminal
β-galactosidase fusion protein (as was expected given the cloning method employed), but
it lacked the expected ATG triplet. Hence, the bovine clone apparently lacked a portion
of its 5' end.
After the DNA sequence of the bovine cDNA clones was determined, Blastp
searches of the NCBI nonredundant database were conducted, and these searches showed
that the 228-amino acid protein encoded by the bovine cDNA clone was 29% identical
(Expect = 8 x 10 -15 ) to the N-terminal domain (165 amino acids) of the PduO ATR of
S. enterica (Johnson et al. 2001). The findings that the bovine cDNA isolated both
complemented an ATR-deficient mutant of S. enterica and encoded a protein with

sequence similarity to a known ATR enzyme suggested that the bovine clone also
encoded an ATR enzyme.
47
Identification of a Human Gene Related to the Bovine Adenosyltransferase cDNA
Blastp searches showed that the putative bovine ATR enzyme was 88% identical to
a human protein encoded by IMAGE cDNA clone 2822202 (accession number
BC005054). The human cDNA clone was 1128 bp in length and included an appropriate
ATG triplet and a poly(A) tail indicating that it was a full-length clone. Further sequence
analyses showed that the predicted protein encoded by the human cDNA was 250 amino
acids in length and was homologous over its entire length to the putative bovine ATR
except that it included 19 additional N-terminal amino acids (Figure 2-1). This suggested
that the bovine clone was nearly full-length, lacking only a portion of its N-terminus,
probably a small region of its MTS (see below). Perhaps more importantly, however, the
high identity between the human and bovine sequences indicated that both encoded
proteins with similar functions. Furthermore, sequence similarity searches showed
human cDNA BC005054 encoded a protein with 26% amino acid identity the PduO ATR
of S. enterica. Thus, these results tentatively identified sequence BC005054 as the
human ATR cDNA.
Subsequent sequence analyses showed human cDNA BC005054 was 99% identical
to nine regions of human chromosome XII. The IMAGE clone BC005054 encoded a
protein with 2 amino acid substitutions compared with the putative protein encoded by
chromosome XII (Arg-19 to Gln and Met-239 M to Lys). These changes may represent
natural polymorphisms or may have resulted from mutations introduced during cDNA
preparation. The nine regions on chromosome XII were the only regions of the human
genome that showed significant homology to human cDNA BC005054. Hence, it seems

48
likely that these regions correspond to nine exons that encode the human ATR enzyme.
The exon structure of the human ATR is indicated in Figure 2-1.
Putative Subcellular Localization of Adenosyltransferase Enzymes
The amino acid sequences of the human and bovine ATR enzymes were aligned
with 20 related sequences obtained from GenBank (Figure 2-1). This alignment
showed that the human, bovine, and mouse ATR sequences included about 50-60
additional N-terminal amino acids compared with their prokaryotic homologues. In
higher organisms, ATRs localize to mitochondria (Banerjee and Chowdhury 1999);
hence, the additional N-terminal amino acids of the eukaryotic sequences might represent
MTSs. Prediction of the subcellular localization of the human and mouse proteins using
Predotar (Small et al. 2004) and MitoProt II (Claros and Vincens 1996) software
indicated mitochondrial localization: human, MitoProt and Predotar scores both = 0.89;
and mouse MitoProt and Predotar scores = 0.97 and 0.99, respectively. Because the
bovine clones were apparently partial sequences, reliable prediction of subcellular
localization of the presumptive bovine ATR enzyme was not possible (Claros and
Vincens 1996, Small et al. 2004).
High-Level Expression of the Bovine and Human Adenosyltransferases in E. coli
E. coli strains were constructed for high-level expression of the presumptive bovine
and human ATR enzymes. Clones were constructed to allow expression of these
enzymes with and without sequences presumed to be involved in mitochondrial targeting.
Our concern was that the MTS sequences might interfere with enzyme activity as such
sequences are normally removed during mitochondrial localization (Paschen and Neupert
2001). The arrows in Figure 2-1 indicate the regions of the bovine and human ATRs that
were expressed by clones that eliminated their MTSs. Furthermore, in all cases, the

ATRs were produced as fusion proteins with N-terminal glutathione S-transferase and
His6 tags that could be removed by enterokinase cleavage if desired.
49
Production of the presumptive bovine and human ATR fusion proteins was
monitored by SDS-PAGE (Figure 2-2). Both the soluble (lanes 2 thru3) and insoluble
fractions (lanes 5-7) were analyzed. Strains BE255 and BE256 produced large amounts
of protein with molecular masses of 55 and 52 kDa, respectively (Figure 2-2 A, lanes 3,
4, 6, and 7). These values are near the predicted masses (58 and 54 kDa) for the bovine
ATR fusion proteins with and without the presumptive MTS and including the expression
tags. In contrast, a strain carrying the expression plasmid without insert (BE237)
produced little protein near these molecular masses (Figure 2-2 A, lanes 2 and 5)
indicating that BE255 and BE256 were indeed producing large amounts of bovine ATR
fusion proteins with and without the presumptive MTS.
Expression of the human ATR fusion proteins was also monitored by SDS-PAGE
(Figure 2-2 B). The predicted molecular masses of the human ATR fusion proteins with
the expression tags, with and without the predicted MTS are 58 and 55 kDa, respectively.
As shown in Figure 2-2 B, large amounts of proteins with calculated molecular masses of
56 and 52 kDa were produced by expression strains BE257 (ATR + MTS) and BE258
(ATR without MTS), (Figure 2-2 B, lanes 3, 4, 6, and 7) but not by the control strain
BE237 (lanes 2 and 5) which carried that expression plasmid without insert.
Assay of the Bovine and Human Adenosyltransferase Fusion Proteins for ATR
Activity
The cell extracts analyzed by SDS-PAGE (Figure 2-2) were tested for
ATP:cob(I)alamin ATR activity (Table 2-2). Substantial activity was found in both
soluble and inclusion body extracts from strains BE255 (bovine ATR + MTS), BE256

(bovine ATR), BE257 (human ATR + MTS), and BE258 (human ATR). Of the four
50
bovine ATR extracts tested, the highest specific activity measured was 85.7 nmol min -1
mg -1 found in the inclusion body fraction from strain BE255 (ATR + MTS). The highest
specific activity found in the human ATR extracts tested was 98 nmol min -1 mg -1 in the
inclusion body fraction containing human ATR without its presumptive MTS. For each
ATR fusion protein, the majority of the activity was found in the soluble fraction
(Table 2-2, column 4). For both the human and bovine cell extracts, ATR activity was
linear with protein concentration (data not show). Furthermore, when the substrates,
ATP and/or cob(I)alamin were excluded from the assay mixture, no activity was detected.
Also, as expected, cell extracts from strain BE237 (T7 expression vector without insert)
expressed no measurable ATR activity.
Complementation of an S. enterica Mutant Deficient in ATR Activity by a Human
Adenosyltransferase cDNA Clone
We previously showed that an S. enterica strain deficient for both the CobA and
PduO ATRs (cobA pduO) was unable to grow on 1,2-propanediol due to an ATR
deficiency (Johnson et al. 2001). To test whether the human ATR cDNA could
complement this defect, we compared the growth of the wild-type strain, BE265 (pduO
cobA /pLAC22-human ATR) and BE263 (pduO cobA /pLAC22-no insert) on
1,2-propanediol minimal medium (Figure 2-3). The doubling times of the wild-type and
strain BE265 (pduO cobA /pLAC22-human ATR cDNA) were found to be comparable;
they were 8.4 and 8.9 h, respectively. Control experiments showed that strain BE263,
which carried the pLAC22 vector without insert, grew minimally on 1,2-propanediol, and
that growth of strain BE265 (pduO cobA/pNL135-human ATR) required the addition of
IPTG as was expected since expression of the human ATR in this strain is under control

51
of the lacI repressor (not shown). Thus, clearly, the human ATR cDNA complemented
the ATR-deficient bacterial mutant for AdoCbl-dependent growth on 1,2-propanediol.
These results provided further evidence that human cDNA BC005054 encodes an ATR
enzyme, and also showed that this enzyme can function as an ATR under physiological
conditions albeit in a heterologous host.
Level of Human Adenosyltransferase Enzyme in Normal and cblB Mutant Human
Skin Fibroblasts
Above, we presented biochemical and genetic evidence that human cDNA
BC005054 encodes an ATR enzyme. If defects in this enzyme underlie cblB
methylmalonic aciduria, its expression or stability might be altered in cblB mutant cell
lines. To examine the expression of this enzyme in normal and cblB mutant human skin
fibroblasts, Western blots were performed on lysates from these cells using antibody
against recombinant human ATR described above. Figure 2-4 shows a representative
experiment. Lane 5 contained recombinant human ATR lacking both the mitochondrial
targeting sequence and the expression tags. In this lane, a single protein band with an
approximate molecular mass of 25 kDa was recognized by the anti-ATR antibody as
expected. In lane 1, which contained cell extracts from normal skin fibroblasts, the
anti-human ATR antibodies recognized two bands. One of these bands was similar in
molecular mass to the recombinant human ATR (approximately 25 kDa) and the second
band was of lower mass (approximately 23 kDa). Neither the 25 kDa nor the 23 kDa
band was observed in blots performed with pre-immune serum (data not shown)
indicating that these bands most likely represent two processed forms of the human ATR.
Interestingly, ATR-reactive polypeptides of these molecular masses were differentially
expressed in the three cblB cell lines tested (lanes 2-3). The cblB mutants WG1680 and

52
WG2127 did not express detectable levels of the 25 kDa form of the ATR and expressed
reduced levels of the 23 kDa form (lanes 2 and 4). For cblB mutant WG1879, the 25 kDa
form was reduced, and the 23 kDa form was not detectable. These results indicated that
expression of the ATR was significantly altered in cblB mutant human skin fibroblasts.
These results provide direct evidence that defects in the ATR identified in this study
underlie cblB methylmalonic aciduria.
Conserved Amino Acids and Distribution of Adenosyltransferase Enzymes
The human and bovine ATR enzymes were aligned with 20 related sequences
obtained from GenBank (Figure 2-1). Of the sequences shown, the function of the
H. sapiens, B. taurus, and S. enterica ATRs is supported by biochemical evidence
presented here and previously (Johnson et al. 2001). To obtain additional information
about the function of the remaining prokaryotic ATRs, we examined their genomic
context. Genes encoding ATR homologues from S. enterica, L. innocua,
L. monocytogenes, B. halodurans, C. perfringes, C. pasteurianum, and K. pneumonia
were found proximal to either AdoCbl-dependent enzymes or AdoCbl biosynthetic genes.
Since bacteria frequently cluster genes of related function, these finding suggest that the
genes from these organisms do indeed encode ATR enzymes. It is also notable that there
are several regions of high conservation in all the ATRs aligned suggesting that many are
indeed ATR enzymes (Figure 2-1). Thus, PduO-type ATR enzymes appear to have broad
phylogenetic distribution, being found in mammals, gram positive and gram negative
bacteria, and the archaea.
The highly conserved regions found in the ATRs also indicate amino acids likely to
be essential for enzymatic activity, and may represent sites directly involved in catalysis
or in binding of substrates, ATP and cob(I)alamin. There are at least two different ways

in which proteins bind B12, the "base-on" and "base-off" modes. Base-off binding
53
involves displacement of the lower axial ligand of B12 (dimethylbenzimidazole) with an
imidazole side-chain of a histidine residue (Drennan et al. 1994, Shibata et al. 1999,
Marsh and Drennan 2001). The sequence containing the coordinating histidine
(D-X-H-X-X-G), which is conserved among proteins that bind B12 in the “base-off”
mode, is absent from PduO-type ATRs examined in this study. This suggests that the
ATR enzymes described in this study bind B12 in the “base-on” mode.
Discussion
In this report we identified human and bovine cDNAs that encode ATR enzymes.
The highest specific activities measured in cell extracts from expression strains were 85.7
nmol min -1 mg -1 protein for the bovine ATR and 98 nmol min -1 mg -1 protein for the
human ATR. These values are comparable to those previously reported for purified
CobA ATR, and partially purified PduO ATR which were 53 and 312 nmol min -1 mg -1
protein, respectively (Suh and Escalante-Semerena 1993, Suh and Escalante-Semerena
1995). The bovine ATR cDNA was isolated based on its ability to complement an
ATR-deficient S. enterica mutant for color formation on indicator medium, and the
human ATR cDNA was shown to restore the ability of an ATR-deficient mutant to grow
on 1,2-propanediol in an AdoCbl-dependent fashion. This demonstrated that the human
and bovine cDNAs described here encode ATR enzymes that are active under
physiological conditions, although in a heterologous host. Moreover, the bovine and
human ATR cDNAs had 29% and 26% identity to the PduO ATR of S. enterica (Johnson
et al. 2001) and the human enzyme included a presumptive MTS as expected (Fenton et
al. 2000). Thus, genetic, biochemical, and bioinformatic evidence indicate that the
bovine and human cDNAs analyzed here encode ATR enzymes.

54
Prior studies have indicated that ATR defects in humans lead to methylmalonic
aciduria (Fenton and Rosenberg 1981, Banerjee and Chowdhury 1999). MCM is known
to require AdoCbl for activity (Banerjee and Chowdhury 1999) and fibroblast extracts
from methylmalonic aciduria patients with defects in the cblB complementation group
have been shown to lack ATR activity (Fenton and Rosenberg 1981). The ATR cDNA
identified in this study corresponded to a single human gene comprised of nine exons
found on chromosome XII. Furthermore, Western blots indicated that expression of this
ATR was altered in cblB mutant human cell lines compared to normal cell lines. Thus,
we propose that the human gene identified here corresponds to the cblB complementation
group that is involved in methylmalonic aciduria.
The approach used in this study to identify the human ATR cDNA involved
screening mammalian cDNA expression libraries for clones that complemented an
ATR-deficient bacterial mutant. This general strategy may have application to the
identification of additional cDNAs involved in cobalamin metabolism. Human cDNAs
encoding the β-ligand transferase and the cob(III)alamin and cob(II)alamin reductases
(the cblACD complementation groups) have not yet been identified (Rosenblatt and
Fenton 1999, Watkins and Rosenblatt 2001). In this regard, it is notable that in this study
the human ATR cDNA was not directly isolated, but rather the bovine ATR cDNA was
isolated and sequence similarity searching was used to identify the homologous human
gene. The fact that our screen allowed direct isolation of a bovine ATR cDNA, but not a
human ATR cDNA may reflect the relative abundance of ATR mRNA in the human
versus the bovine liver. Ruminant livers have among the highest levels of B12-dependent
enzymes measured in any tissue (Schneider and Stroinski 1987). Thus, ruminant liver

libraries may be particularly useful for isolating mammalian cDNAs involved in
cobalamin metabolism.
55
Importantly, the identification of human genes involved in methylmalonic aciduria
should help with the development of improved methods for diagnosis and treatment of
this rare but devastating disease. Knowledge of the relevant genes will allow DNA-based
methods of diagnosis following amniocentesis or chorionic villi sampling. Such
techniques will allow the identification of the specific genetic lesions involved, and this
may help guide treatment. Some cases of methylmalonic aciduria respond to high-dose
B12 therapy and it seems likely that such cases will be associated with specific mutations
(Ampola et al. 1975, Matsui et al. 1983, Zass et al. 1995). Moreover, as the needed
methodologies become available, knowledge of the genes involved in cobalamin
metabolism may also help with the development of gene or enzyme therapies as
treatments for methylmalonic aciduria.

Table 2-1. Bacterial strains
Species
E. coli
Strain Genotype
BL21 (DE3) RIL
(E. coli B) F - ompT hsdS (rB - mB - ) dcm + Tet r gal
SOLR TM
(Stratagene)
XL1-Blue MRF’
(Stratagene)
BE237
56
λ (DE3) endA Hte (argU ileY leuW Cam r )
e14 - (McrA - ) ∆(mcrCB-hsdSMR-mrr)171 sbcC
recB recJ uvrC umuC::Tn5 (Kan r ) lac gyrA96
relA1 thi-1 endA1 λ R [F’ proAB lacI q Z∆M15] c
Su - (nonsuppressing)
D(mcrA)183 ∆(mcrCB-hsdSMR-mrr)173 endA1
supE44 thi-1 recA1 gyrA96 relA1 lac [F’ proAB
lacI q Z∆M15 Tn10 (Tet r )] c
BL21 (DE3) RIL/pET-41a (T7 expression vector
without insert)
BE255 BL21 (DE3) RIL/pNL128 (bovine ATR+MTS)
BE256
BL21 (DE3) RIL/pNL129 (bovine ATR without
MTS)
BE257 BL21 (DE3) RIL/pNL132 (human ATR+MTS)
BE258
BL21 (DE3) RIL/pNL133 (human ATR without
MTS)
S. enterica serovar Typhimurium LT2
TR6579
metA22 metE551 trpD2 ilv-452 hsdLt6 hsdSA29
HsdB - strA120 GalE - Leu - Pro -
BE253
TR6579 cobA::dcam ∆pduO651
yeex::kan r /pSJS1240 (ileX, argU)
BE254 BE253/pNL121 (pBlueScript with bovine ATR)
BE266
∆pduO651 cobA366::Tn10dCam/pSJS1240
BE263
BE265
(ileX, argU)
∆pduO651
cobA366::Tn10dCam/pLAC22/pSJS1240 (ileX,
argU)
∆pduO651 cobA366::Tn10dCam/pNL135
(human ATR without MTS under plac control)

57
Figure 2-1. Multiple sequence alignment of adenosyltransferase enzymes. ClustalX was
used to align the prokaryotic, eukaryotic, and archaeal homologues of ATR
enzymes. Identical residues are shaded black, and highly conserved residues
are shaded gray. The predicted mitochondrial targeting sequence determined
by MitoProtII software is underlined in the human sequence. Triangles
separate the human ATR into nine regions that correspond to exons on
chromosome XII. Arrows represent the PCR primers used for amplification
of the human ATR without its presumptive MTS.

A.
58
kDa 1 2 3 4 5 6 7 kDa 1 2 3 4 5 6 7
116.25
97.4
97.4
66.2
66.2
45
31
21.5
14.4
45
31
21.5
Figure 2-2. SDS-PAGE analysis of cell extracts from bovine (A) and human (B) ATR
expression strains. Panels A and B lane 1, molecular mass markers. Panel
A, lanes 2 thru 4 are soluble extracts of BE237 (T7 expression vector without
insert), BE255 (bovine ATR + MTS), and BE256 (bovine ATR without
MTS). Panel A, lanes 5 thru 7 are inclusion body fractions from strains
BE237, BE255, and BE256. Panel B, lanes 2 thru 4 are soluble extracts from
BE237 (T7 expression vector without insert), BE257 (human ATR + MTS),
and BE258 (human ATR without MTS), respectively. Lanes 5 thru 7 are
inclusion body fractions from strains BE237, BE257, and BE258.
B.

Table 2-2. Specific activities of bovine and human ATP:cob(I)alamin ATRs
Strain S/I *
59
Total activity
nmol min -1
Total
activity
%
Specific activity
nmol min -1 mg -1
protein
BE255
S 2021 95 47
(bovine ATR+MTS) I 107 5 85.7
BE256
S 3364 84 44.1
(bovine ATR without MTS) I 632 16 30
BE257
S 2445 52 37.5
(human ATR+MTS) I 2296 48 60.8
BE258
S 5481 57 63
(human ATR without MTS) I 4144 43 98
* S=soluble fraction, I=inclusion body fraction

OD 600
1.0
0.5
0.1
60
0 10 20 30 40
Time (hours)
Figure 2-3. Complementation of an ATR-deficient bacterial mutant for
AdoCbl-dependent growth on 1,2-propanediol by a plasmid that expresses
the human ATR. Cells were grown in minimal 1,2-propanediol medium
supplemented with HOCbl and 0.1 mM IPTG. Growth was determined by
measuring the absorbance of cultures at 600 nm (OD600), and data are shown
as a semi-log plot. , Wild-type (S. enterica serovar Typhimurium LT2);
, BE265, ATR-deficient/pNL135(human ATR under plac control); ,
BE263, ATR-deficient /pLAC22 (vector without insert).

61
1 2 3 4 5
25 kDa
Figure 2-4. Western blot analysis of ATR expression by normal and cblB mutant cell
lines. Extracts were prepared from cultured fibroblast lines obtained from a
normal individual and three cblB patients. Lane 1: cell extract from the
MCH45 control cell line. Lanes 2 thru 4: cell extracts from cblB mutant cell
lines WG1680, WG1879 and WG2127, respectively. 220 µg of protein was
loaded into each of lanes 1 thru 4 (It was necessary to load relatively high
amounts of protein because of the low abundance of the ATR enzyme in skin
fibroblasts). Lane 5 contained 100 pg recombinant human ATR lacking the
mitochondrial targeting sequence and affinity tags. The arrow shows the
location of the recombinant ATR enzyme (approximately 25 kDa). The
experiment was performed twice with similar results.

CHAPTER 3
PURIFICATION AND INITIAL CHARACTERIZATION OF THE HUMAN
ATP:COB(I)ALAMIN ADENOSYLTRANSFERASE AND ITS INTERACTION WITH
METHIONINE SYNTHASE REDUCTASE
Introduction
The vitamin B12 coenzymes, adenosylcobalamin (AdoCbl) and methylcobalamin
(CH3Cbl) are required cofactors for at least fifteen different enzymes that have a broad
but uneven distribution among living forms (Schneider and Stroinski 1987, Banerjee
1999). In humans, two B12-dependent enzymes are known (Kolhouse and Allen 1977,
Mellman et al. 1977). AdoCbl -dependent methylmalonyl-CoA mutase (MCM) catalyzes
the reversible rearrangement of methylmalonyl-CoA to succinyl-CoA (Banerjee and
Chowdhury 1999). MCM is found in the mitochondrial matrix and it is required for the
complete catabolism of compounds degraded via propionyl-CoA including
branched-chain amino acids, odd-chain fatty acids, and cholesterol (Figure 3-1) (Banerjee
and Chowdhury 1999). The second B12-dependent enzyme known in humans is
CH3Cbl-dependent methionine synthase (MS). This enzyme is found in the cytoplasm
where it catalyzes the conversion of methyltetrahydrofolate and homocysteine to
tetrahydrofolate and methionine (Matthews 1999). In mammals, methionine is an
essential amino acid, and MS plays a role in recycling the S-adenosylhomocysteine
formed from S-adenosylmethionine-dependent methylation reactions (Weissbach et al.
1963).
Humans are incapable of synthesizing AdoCbl and CH3Cbl de novo and are
dependent on dietary sources of complex cobalamin precursors, such as vitamin B12
62

63
(cyanocobalamin, CNCbl) and hydroxycobalamin (HOCbl) (Stabler 1999). The pathway
by which these precursors are metabolized to the coenzyme forms is thought to be similar
in both prokaryotes and eukaryotes (Figure 3-1). For AdoCbl synthesis, CNCbl is
proposed to be converted to glutathionylcobalamin (GSCbl), reduced successively to
cob(II)alamin and cob(I)alamin respectively and finally, adenosylated to AdoCbl
(Figure 3-1) (Friedmann 1975, Huennekens et al. 1982, Pezacka et al. 1990, Pezacka
1993, Watanabe et al. 1996). For the synthesis of CH3Cbl, cob(II)alamin associated with
MS, is reductively methylated to form CH3Cbl (Figure 3-1) (Drummond and Matthews
1994, Matthews 1999). For this reaction (which is also used for the reductive activation
of MS following adventitious oxidation of the cobalamin cofactor),
S-adenosylmethionine is the methyl-group donor and methionine synthase reductase
(MSR) reduces cob(II)alamin to cob(I)alamin (Leclerc et al. 1998, Olteanu and Banerjee
2001).
In humans, inherited defects in the MCM or MS structural genes or in the genes
needed for the synthesis of B12 coenzymes result in methylmalonic aciduria,
homocystinuria, or combined disease. These rare disorders, which are often fatal in the
first year of life, result from recessive autosomal mutations that fall into nine
complementation groups (mut, cblABCDEFGH) (Qureshi et al. 1994, Kapadia 1995,
Rosenblatt and Fenton 1999, Watkins and Rosenblatt 2001). The cblF, cblC and cblD
defects lead to combined disease. Prior studies indicated that cblF mutations affect
cobalamin transport from the lysosome to the cytoplasm, while cblC and cblD mutations
impair the conversion of complex precursors to cob(II)alamin (Mellman et al. 1979,
Watkins and Rosenblatt 2001). The cblE and cblG complementation groups correspond

64
to the genes for MSR (MTRR) and MS (MTR), respectively, and defects in either gene
results in homocystinuria (Cooper and Rosenblatt 1987, Watkins and Rosenblatt 2001).
The mut group corresponds to the gene that encodes MCM (MUT). The cblA group
(MMAA gene) encodes a protein proposed to protect MCM from inactivation (Korotkova
and Lidstrom 2004) and recent studies have shown that the cblB complementation group
(MMAB gene) encodes a mitochondrial ATP cob(I)alamin adenosyltransferase (ATR)
(Dobson et al. 2002, Leal et al. 2003). Mutations in the cblH complementation group
also result in methylmalonic aciduria, but the nature of the underlying defect and the
corresponding gene are unknown (Watkins and Rosenblatt 2001).
The human enzyme that catalyzes the reduction of cob(II)alamin to cob(I)alamin
for AdoCbl synthesis has not been identified. Recent studies showed that MSR mediates
cob(II)alamin reduction for MS activation (Leclerc et al. 1999, Olteanu and Banerjee
2001), raising the possibility that this enzyme might play a role in AdoCbl synthesis.
Here we describe the purification and initial biochemical characterization of the human
ATR from recombinant E. coli and demonstrate that in vitro human MSR can reduce
cob(II)alamin to cob(I)alamin for AdoCbl synthesis by the ATR enzyme.
Chemicals and Reagents
Materials and Methods
Restriction enzymes and T4 DNA ligase were from New England Biolabs
(Beverly, MA). Titanium III citrate was prepared as previously described (Bobik and
Wolfe 1989). AdoCbl, HOCbl, β-nicotinamide adenine dinucleotide phosphate reduced
form (NADPH), iodoacetic acid (IA), flavin adenine dinucleotide (FAD), and bovine
serum albumin (BSA) were purchased from Sigma Chemical Company (St. Louis, MO).
Isopropyl-ß-D-thiogalactopyranoside (IPTG), ultra-pure ammonium sulfate, adenosine

65
5’- triphosphate (ATP), and dithiothreitol (DTT) were purchased from ICN Biomedicals,
Inc. (Aurora, Ohio). Pefabloc SC PLUS was from Roche Diagnostics (Penzberg,
Germany). All other chemicals and reagents were from Fisher Scientific (Norcross, GA).
Bacterial Strains and Growth Media
The bacterial strains used were E. coli DH5α and BL21 (DE3) RIL (Stratagene, La
Jolla, CA). LB was the rich medium used (Difco, Detroit, MI) (Miller 1972). LB was
supplemented with 25 µg/ml kanamycin and 20 µg/ml chloramphenicol or as indicated.
General Molecular and Protein Methods
Agarose gel electrophoresis and restriction enzyme digests were performed using
standard protocols (Maniatis et al. 1982, Johnson et al. 2001). PCR products and plasmid
DNA were gel purified using QIAquick ® gel extraction kit (QIAGEN, Inc., Valencia,
CA). DNA ligation was accomplished using T4 DNA ligase according to manufacturer’s
instructions. Bacterial transformation and SDS-PAGE were performed as previously
described (Johnson et al. 2001).
ATP:cob(I)alamin Adenosyltransferase Assays
ATR assays were performed as previously reported with some modifications
(Johnson et al. 2001). Reaction mixtures contained: 200 mM Tris-HCl (pH 8.0), 2.8 mM
MgCl2, 10 mM KCl, 0.05 mM HOCbl, 0.4 mM ATP, and 1 mM titanium(III) citrate in a
total volume of 2 ml. Reaction components (except for the ATR) were dispensed into
cuvettes inside an anaerobic chamber (Coy Laboratories, Ann Arbor, MI). The cuvettes
were sealed, removed from the chamber, and incubated at 37°C for 2 minutes. Reactions
were initiated by addition of purified recombinant ATR, and AdoCbl formation was
measured by following the decrease in absorbance at 388 nm (∆ε388 = 24.9 cm -1 mM -1 ).

Construction of ATR Expression Strains
66
Plasmid pNL166 (Leal et al. 2003) was restricted with BglII and HindIII releasing a
645 bp fragment that encodes the ATR enzyme with a lysine at position 239 (ATR
239K). This fragment was gel purified and ligated into a modified pET-41a expression
vector, pTA925 (Johnson et al. 2001). Ligation mixtures were used to transform E. coli
DH5α by electroporation and transformants were selected by plating on LB kanamycin
medium. Pure cultures were obtained from selected transformants and plasmid DNA
isolated from these strains was analyzed by restriction digestion and DNA sequencing.
Clones with the expected DNA sequence were transformed into E. coli strain BL21
(DE3) RIL (Stratagene) for high-level protein production.
PCR was used to construct the ATR polymorphic variant with a methionine at
position 239 (ATR 239M). Plasmid pNL166 (Leal et al. 2003) provided the template
DNA. The primers used for amplification were
5’-GCCGCCAGATCTTATGCCTCAGGGCGTGGAAGACGGG-3’ (forward) and
5’-GCCGCCAAGCTTTCAGAGTCCCTCAGACTCGGCCGATGGGTCATTTTTCAT
GTATATTTTCTCTTGATTCCC-3’ (reverse). These primers introduced BglII and
HindIII restriction sites into the PCR product and were designed to change the lysine at
position 239 to a methionine. Ligation, transformation, and analysis of plasmid DNA
were performed as described above. Clones with the expected DNA sequence were
transformed into E. coli BL21 (DE3) RIL for protein production.
Purification of the Human ATR Variants
Soluble cell extracts of the E. coli expression strains were used for the purification
of both ATR variants. The growth of the cells and the preparation of cell extracts were
carried out as described (Johnson et al. 2001). For ammonium sulfate precipitation, cell

extracts of ATR 239K (160 mg protein) and 239M (180 mg protein) were diluted to 1
mg/ml in 50 mM sodium phosphate (pH 7.0), 300 mM NaCl. Then, ultra-pure
67
ammonium sulfate was added as a fine powder with stirring. Precipitated proteins were
centrifuged for 15 min at 12,000 x gmax using a Beckman JLA-10.500 rotor. The protein
pellets were resuspended in 4 ml of 5 mM potassium phosphate (pH 6.8), and passed
through a 0.45 µm filter. ATR 239K (60 mg) and 239M (35 mg) were applied to a 60 ml
ceramic hydroxyapatite column (Bio-Rad) equilibrated with 5 mM potassium phosphate
(pH 6.8). The column was washed with 60 ml equilibration buffer and eluted with a 600
ml linear gradient of 5 mM to 200 mM potassium phosphate (pH 6.8) at a flow rate of 5
ml/min. Fractions containing ATR of the highest purity were pooled and exchanged into
10 mM potassium phosphate (pH 7.0), 50 mM KCl using a Vivaspin centrifugal
concentrator with a molecular weight cutoff of 10,000 Daltons (Viva Science, Binbrook,
UK). Samples were filtered through a 0.45 µm pore-size membrane and further purified
using a Mono Q HR 10/10 anion exchange column (Amersham Pharmacia Biotech,
Piscataway, NJ). ATR 239K (10 mg) and 239M (9 mg) were applied to the column
which had been equilibrated with 10 mM potassium phosphate (pH 7.0), 50 mM KCl.
Then, the column was washed with 8 ml equilibration buffer and eluted with a 160 ml
linear gradient from 50 to 1000 mM KCl in 10 mM potassium phosphate (pH 7.0) at a
flow rate of 4 ml/min. Protein elution was monitored by following the absorbance of the
column effluent at 280 nm and 8 ml fractions were collected. Purified ATR was
concentrated and stored in 10 mM phosphate buffer (pH 7.0), 130 mM KCl, 50%
glycerol at -20°C.

MSR-ATR Assay
68
The basis of this assay is that MSR converts cob(II)alamin to cob(I)alamin which is
in turn converted to AdoCbl by the ATR enzyme. The conversion of cob(I)alamin to
AdoCbl proceeds with an increase in absorbance at 525 nm that allows quantitation
(∆ε525 = 4.8 cm -1 mM -1 ). Assays contained 200 mM Tris (pH 8.0), 1.6 mM potassium
phosphate, 2.8 mM MgCl2, 100 mM KCl, 0.1 mM cob(II)alamin, 0.4 mM ATP, 1 mM
DTT, 1 mM NADPH, purified MSR and ATR as indicated in the text. The total volume
was 2 ml and anaerobic procedures were used as described for the ATR assay.
Cob(II)alamin was prepared by exposing an anoxic solution of 1 mM AdoCbl to a 150
watt incandescent light at a distance of 15 cm for 30 minutes (Johnson et al. 2001). ATR
239K was purified as described above and MSR was purified as described previously
(Olteanu and Banerjee 2001). Reaction mixtures were incubated at 37°C for 2 min prior
to initiating reactions by the addition of an anoxic solution of NADPH.
Measurement of Cob(I)alamin with Iodoacetate
Iodoacetate reacts rapidly and quantitatively with cob(I)alamin to form
carboxymethylcobalamin (CMCbl). This reaction proceeds with an increase in
absorbance at 525 nm (∆ε525 = 5.1 cm -1 mM -1 ) and hence provides a facile method for
quantitation of cob(I)alamin (Debussche et al. 1991). Anoxic stock solutions of
iodoacetate (40 mM) were prepared and used the same day. The stock solution was
shielded from light using aluminum foil and was added to assay mixtures just prior to the
initiation of reactions.
Separation and Quantification of Cobalamins by HPLC
HOCbl, AdoCbl, and CMCbl were separated and quantified by HPLC using a
NovaPak C18 column (3.9×150 mm) equipped with a C18 Sentry guard column (Waters,

Milford, MA). Samples (200 µl) were loaded onto the column and eluted with 30 ml
69
linear gradient of 10 to 90% methanol in 50 mM sodium acetate (pH 4.6) at a flow rate of
1 ml/min. The absorbance of column effluent was monitored at 365 nm, and analytes
were quantified by comparison of peak areas to a standard curve. The CMCbl standard
was prepared by incubating 2 ml of 200 mM Tris (pH 8.0), 1.6 mM potassium phosphate,
2.8 mM MgCl2, 100 mM KCl, 0.1 mM cob(II)alamin, 1 mM DTT, 0.1 mM FAD, and 0.4
mM iodoacetate at 37°C for 1 hour which resulted in quantitative conversion of
cob(II)alamin to CMCbl (data not shown). Standard curves were prepared using 5, 10,
20, 30, and 40 nmoles of CMCbl, AdoCbl, or HOCbl. Peak areas were integrated using
Waters Breeze software. The standard curves for CMCbl, AdoCbl, and HOCbl, had
regression coefficients (R 2 ) of 0.9985, 0.9993, and 0.9956, respectively. Solutions that
contained CMCbl or AdoCbl were shielded from light to prevent photolysis.
Effect of Ionic Strength on the MSR-ATR System
The ionic strength of MSR-ATR assays was varied by the addition of KCl. Ionic
strength was calculated using the formula I = 1/2Σ(ciz) 2 where I is the ionic strength, ci is
the molar concentration of each type of ion, and z is the charge of each ion (Segel 1976).
For lower values of ionic strength, it was necessary to reduce the Tris-HCl
concentrations.
DNA Sequencing
DNA sequencing was carried out by University of Florida, Interdisciplinary Center
for Biotechnology Research, DNA Sequencing Core facility and University of Florida,
Department of Microbiology and Cell Science, DNA Sequencing facility as described
previously (Bobik et al. 1999).

Purification of the Human ATR Variants
70
Results
The human ATR catalyzes the transfer of a 5'-deoxyadenosyl group from ATP to
the central cobalt atom of cob(I)alamin to form AdoCbl (Chapter 2, Leal et al. 2003).
Two common polymorphic variants of the human ATR having either lysine or
methionine as amino acid 239 (ATR 239M and 239K) were recently shown to be present
in 54% and 46% of 72 control cell lines, respectively (Dobson et al. 2002). We
previously produced ATR 239K as a GST-fusion protein in E. coli (Leal et al. 2003).
Here, subcloning and site-directed mutagenesis were used to construct E. coli expression
strains that produce high levels of ATR 239K and 239M without fusion tags and lacking
their predicted mitochondrial targeting sequences. Both ATR variants were produced at
high levels and purified to apparent homogeneity from soluble cell extracts of
recombinant E. coli. The purification consisted of three steps including ammonium
sulfate precipitation and ceramic-hydroxyapatite and Mono Q chromatography (Table
3-1). Both ATR variants precipitated between 40 and 50% ammonium sulfate saturation
and eluted from the hydroxyapatite and Mono Q columns at 110-135 mM potassium
phosphate and 114-130 mM KCl. ATR 239K and 239M were purified 4.5 and 5.4-fold
with specific activities of 220 and 190 nmol min -1 mg -1 proteins, respectively. The yields
were 14 and 9%. During the purification, inclusion of KCl in chromatography and
storage buffers increased enzyme stability.
SDS-PAGE was used to follow the purification of both ATR variants. Since
purification of both enzymes proceeded similarly, only the gel used to assess the purity of
ATR 239K is depicted (Figure 3-2). Following the final Mono Q chromatography step,
both variants were homogenous following staining with Coomassie Brilliant Blue.

Linearity of the ATR reaction
71
The effect of ATR concentration on activity was determined. Results showed that
the rate of adenosylation was proportional to ATR concentration ranging from 5 µg to
115 µg for both variants and linear regression yielded R 2 values of 0.9995 and 0.9993 for
ATR 239K and 239M, respectively.
ATR Reaction Requirements
To determine the ATR reaction requirements, key assay components were
individually omitted. For both variants, there was no detectable activity in the absence of
ATR, ATP, HOCbl, or titanium (III) citrate (the reducing agent used to produce
cob(I)alamin). In the complete reaction mixture, there was 215 and 194 nmol min -1 mg -1
activity for ATR 239K and 239M, respectively. In the absence of MgCl2, activity
decreased by 20% for both variants. There was no significant difference in activity in the
absence of KCl.
Alternative Nucleotide Donors
The specificity of the ATR variants for ATP, CTP, GTP, UTP, ADP, and AMP was
examined. ATP was found to be the best substrate giving a specific activity of 207 and
197 nmol min -1 mg -1 for ATR 239K and 239M, respectively. These values were set as
100% activity. When CTP, GTP, and UTP were tested as substrates for ATR 239K there
was 9, 16, or 8% activity, respectively (Table 3-2). Similar results were found for ATR
239M (6, 14, or 6% activity with CTP, GTP, and UTP, respectively). For both ATR
variants there was no detectable activity when AMP or ADP was substituted for ATP.
Km and Vmax Values for ATR 239K and 239M
Lineweaver-Burk double-reciprocal plots were used to calculate Km and Vmax values
(Segel 1976). For ATR 239K and ATR 239M, the Km values for ATP were 6.3 µM and

72
6.9 µM, and the Km for cob(I)alamin were 1.2 µM and 1.6 µM, respectively (Figure 3-3).
The Vmax values for ATR 239K and 239M were 250 and 200 nmol min -1 mg -1 ,
respectively. When the Km values for cob(I)alamin were determined, saturating levels of
ATP (500 µM) were added to assay mixtures while varying the concentration of
cob(I)alamin. Similarly, saturating levels of cob(I)alamin (50 µM) were added to assays
when the Km values for ATP were determined. Each value shown in Figure 3-3 is the
average of three measurements of the initial reaction rate at the indicated substrate
concentrations.
MSR Reduces Cob(II)alamin to Cob(I)alamin for AdoCbl Synthesis
The human enzyme that catalyzes the third step of CNCbl assimilation, the
reduction of cob(II)alamin to cob(I)alamin for AdoCbl synthesis by the ATR enzyme, is
unknown. Recently, MSR was shown to catalyze cob(II)alamin reduction for MS
activation (Olteanu and Banerjee 2001). To test whether MSR can also reduce
cob(II)alamin to cob(I)alamin for AdoCbl synthesis, an in vitro assay was used
(MSR-ATR assay). In assays that contained standard components, cob(II)alamin, ATP,
DTT, NADPH, 100 µg purified MSR, and 10 µg purified ATR, AdoCbl was formed at a
rate of 0.96 nmol min -1 . Controls showed that no measurable AdoCbl was formed in the
absence of ATR, MSR, ATP, or cob(II)alamin. When NADPH was excluded from the
reaction mixture, 65% of the total activity remained indicating that DTT can partially
replace NADPH as the electron donor for this reaction. When DTT was omitted, 100%
activity remained with NADPH as the sole electron donor, showing that DTT is not
required for the reaction to occur. In addition, several further tests were used to establish
that AdoCbl was the product of the reaction. The UV-Visible spectrum of completed

73
reactions was characteristic of AdoCbl, and the reaction product was photolyzed by a 30
min exposure to incandescent light with the formation of cob(II)alamin (Figure 3-4).
Moreover, the product of the MSR-ATR reaction co-migrated with authentic AdoCbl by
reverse-phase HPLC following co-injection (not shown). These results show that the
combination of the MSR and ATR enzymes converted cob(II)alamin to AdoCbl. Since
prior studies have shown that the human ATR requires cob(I)alamin for AdoCbl
synthesis and is inactive with cob(II)alamin, these results indicate that MSR reduced
cob(II)alamin to cob(I)alamin for AdoCbl synthesis by ATR.
Cob(I)alamin is Sequestered by the MSR-ATR System
Above, we presented evidence that MSR can reduce cob(II)alamin to cob(I)alamin
for adenosylation by the ATR. In principle, there are two ways in which this might occur
(Figure 3-5). MSR could reduce cob(II)alamin to cob(I)alamin directly, which is
subsequently released into solution and diffuses to ATR. Alternatively, MSR could
reduce cob(II)alamin bound to ATR and the interaction between the two proteins could
sequester cob(I)alamin.
To test whether cob(I)alamin was sequestered or released into solution during
AdoCbl synthesis by the MSR-ATR system, a chemical trap for cob(I)alamin was used.
Iodoacetate (which reacts rapidly and quantitatively with cob(I)alamin to form CMCbl)
was added to an MSR-ATR assay at a 1000-fold excess compared to the ATR (400 µM
compared to 0.4 µM). The assay was allowed to proceed to completion. Then, CMCbl
and AdoCbl were resolved and quantitated by reverse-phase HPLC. Results showed that
35% of the cob(I)alamin formed was converted to CMCbl by reaction with iodoacetate
and that 65% was converted to AdoCbl via the ATR enzyme (Table 3-3, line 1). If "free"

cob(I)alamin had been produced under the conditions used, it is expected that the
majority should have been converted to CMCbl by reaction with the large excess of
iodoacetate prior to diffusion to the ATR for conversion to AdoCbl. Hence, the above
74
results indicated that a significant portion of the cob(I)alamin was sequestered (protected
from reaction with iodoacetate) during conversion of cob(II)alamin to AdoCbl by MSR
and ATR.
As a control, assays were performed that were similar to those described above
except that MSR was replaced with the combination of 1 mM DTT plus 50 µM FAD to
chemically generate cob(I)alamin in situ. Chemical reduction of cob(II)alamin
necessarily produces “free” cob(I)alamin which must then diffuse to the ATR before it
can be converted to AdoCbl. Under these conditions, only 10% of the cob(I)alamin
formed was converted to AdoCbl while 90% was converted to CMCbl (Table 3-3, line 2).
This was in contrast to results obtained when MSR was used to generate cob(I)alamin for
the ATR enzyme where 65% of the cob(I)alamin formed was converted to AdoCbl
(Table 3-3, line1). Hence these results indicate that the MSR-ATR system sequesters
cob(I)alamin during the conversion of cob(II)alamin to AdoCbl.
Interestingly, further studies showed that iodoacetate had relatively little effect on
the total amount of AdoCbl formed by the MSR-ATR system. In the absence or presence
of iodoacetate, 94 or 85 nmol of AdoCbl was produced, respectively (Table 3-3). This
indicated that the majority of the cob(I)alamin produced by the MSR-ATR system was
sequestered.

75
MSR Produces little Cob(I)alamin in the Absence of the ATR
Iodoacetate was also used to test whether MSR can produce “free” cob(I)alamin in
the absence of ATR. Assays were prepared that were similar in composition to the
MSR-ATR assay except that ATR was replaced by iodoacetate. During the first 10 min
of the reaction, cob(I)alamin was formed at a very low rate (0.02 nmol min -1 )
(Figure 3-6). Then, at the 10 min time point, 10 µg of purified ATR was added to the
assay and the rate of cob(I)alamin formation increased approximately 46-fold. These
findings showed that the presence of the ATR greatly enhanced the production of
cob(I)alamin by MSR suggesting a physical interaction between the MSR and ATR.
Control experiments showed that BSA had no effect on the rate of cob(I)alamin
formation when substituted for ATR in the second phase of the reaction shown in
Figure 3-6. Moreover, the controls showed that iodoacetate was working effectively
under our assay conditions. In similar reaction mixtures containing DTT and FAD, the
cob(I)alamin formed was efficiently trapped by iodoacetate (data not shown).
Stoichiometry of the MSR-ATR system
The dependence of ATR activity on the molar ratio of MSR/ATR was determined
at ratios from 0 to 40 (Figure 3-7). Maximal activity occurred at a ratio of approximately
4 moles MSR (which is a monomer) per mol ATR monomer. This ratio is similar to the
optimal stoichiometry for activation of MS by MSR and is a reasonable value for a
physiological process (Olteanu and Banerjee 2001).
Ionic Strength Dependence of Cobalamin Reduction and Adenosylation by the
MSR-ATR System
Variation of the ionic strength from 15 to 720 mM had relatively little effect on the
synthesis of AdoCbl by the MSR-ATR system (Figure 3-8). Within this range, greater

76
than 65% activity was retained. This is in contrast to the relatively narrow ionic strength
dependence of MS activation by MSR (Olteanu and Banerjee 2001).
Discussion
Prior studies identified the human ATR and showed that defects in its encoding
gene (MMAB) underlie cblB methylmalonic aciduria (Dobson et al. 2002, Leal et al.
2003). Investigations also identified two common polymorphic variants of the ATR that
are found in normal individuals (Dobson et al. 2002). Here, both ATR variants were
expressed in E. coli, purified, and found to have similar kinetic properties. The specific
activities of variants 239K and 239M were 250 and 200 nmol min -1 mg -1 , and the Km
values were 6.3 and 6.9 µM for ATP and 1.2 and 1.6 µM for cob(I)alamin, respectively.
These values are roughly similar to those previously reported for bacterial ATR enzymes
where specific activities range from 53 to 619 nmol min -1 mg -1 , and Km values from 2.8 to
110 µM for ATP and from 3 to 5.2 µM for cob(I)alamin (Suh and Escalante-Semerena
1995, Johnson et al. 2001, Saridakis et al. 2004). The Km values for the human ATR are
appropriate to its physiological role since cellular levels of cobalamin are typically 1 µM
or less in higher organisms and ATP concentrations are in the low mM range (Schneider
and Stroinski 1987). Thus, the results reported here show that both common ATR
variants are essentially wild-type and have kinetic properties sufficient to meet cellular
needs for AdoCbl synthesis.
Three classes of ATRs that are unrelated in amino acid sequence have been
identified, the PduO-type, CobA-type, and EutT-type (Johnson et al. 2001). Members of
the CobA-type and the PduO-type have been purified, partially characterized, and their
crystal structures determined (Suh and Escalante-Semerena 1995, Bauer et al. 2001,

Johnson et al. 2001, Saridakis et al. 2004). The CobA enzyme has a relatively broad
specificity for nucleotide donors. It is active with ATP, CTP, GTP, UTP, and ITP and
contains a modified P-loop for nucleotide binding (Bauer et al. 2001). In contrast, the
77
Thermoplasma acidophilum ATR which is a PduO-type enzyme and lacks a recognizable
P-loop, uses only ATP and deoxy-ATP as nucleotide donors (Saridakis et al. 2004).
Thus, different modes of ATP binding are thought to account for the observed differences
in nucleotide specificity by CobA- and PduO-type enzymes. The human ATR is a
PduO-type enzyme, but results presented here show that its nucleotide specificity is
intermediate to that of CobA and T. acidophilum enzymes. In addition, the Km of the
human and T. acidophilum enzymes for ATP is 7 versus 110 µM suggesting some
differences between the two enzymes and the mode of ATP binding which might be
expected since these enzymes are only 32% identical in amino acid sequence.
Results reported here also showed that an in vitro system containing the human
MSR and ATR converted cob(II)alamin to AdoCbl. In this system, MSR reduced
cob(II)alamin to cob(I)alamin which in turn was converted to AdoCbl by ATR. When
0.2 µM purified ATR and 0.6 µM purified MSR was used, the rate of AdoCbl formation
was 0.96 nmol min -1 . It is probable that this level of activity is sufficient to meet
physiological needs. Cells require only very small quantities of AdoCbl. Typical
intracellular levels are about 1 µM whereas other well-known coenzymes are present at
100 to 1000-fold higher concentrations or more (Schneider and Stroinski 1987).
Furthermore, the MSR-ATR system is about 10-fold more active than the analogous
bacterial system which consists of the Fpr, FldA, and CobA proteins (Fonseca and
Escalante-Semerena 2001). Thus, the MSR-ATR system mediates the conversion of

78
cob(II)alamin to AdoCbl at a physiologically relevant rate. To our knowledge, this is the
first reported evidence that MSR can catalyze the reduction of cob(II)alamin to
cob(I)alamin for AdoCbl synthesis.
In addition to its role in AdoCbl synthesis, prior studies showed that MSR also
reduces MS-bound cob(II)alamin to cob(I)alamin for the reductive activation of MS
(Leclerc et al. 1998, Olteanu and Banerjee 2001). It is known that MS activation occurs
in the cytoplasm but that AdoCbl synthesis occurs in the mitochondrion. Therefore, to
carry this dual function MSR must reside in both compartments. Interestingly, recent
studies have indicated that alternative transcript splicing leads to the production of MSR
enzymes with and without mitochondrial targeting sequences providing further evidence
of a dual role of this enzyme (Leclerc et al. 1999).
Findings reported here which indicate dual functionality for MSR bring to light an
interesting parallel between cobalamin reduction in bacteria and in humans. Biochemical
evidence indicated that the E. coli flavodoxin (FldA) catalyzes the reduction of
cob(II)alamin to cob(I)alamin for both the reductive activation of MS (MetH) and the
synthesis of AdoCbl by the bacterial ATR (CobA) (Fujii and Huennekens 1974, Fujii et
al. 1977, Fonseca et al. 2002). Similarly, human MSR reduces cob(II)alamin for both
MS activation (Leclerc et al. 1998, Olteanu and Banerjee 2001) and AdoCbl synthesis by
the human ATR (this study). Thus, although the human MSR and ATR lack significant
sequence similarity to their bacterial counterparts (FldA and CobA), evolution appears to
have driven analogous dual physiological roles for both FldA and MSR enzymes.
In this report, we also conducted studies to determine whether cob(I)alamin was
sequestered or released “free” in solution during the conversion of cob(II)alamin to

AdoCbl by the MSR-ATR system. Experiments in which iodoacetate was used as a
chemical trap indicated that cob(I)alamin was sequestered. This is likely to be
79
physiologically important. Cob(I)alamin is one of the strongest nucleophiles that exists
in aqueous solution and an extremely strong reductant (E°' = -0.61 V) (Schrauzer et al.
1968, Schrauzer and Deutsch 1969, Banerjee et al. 1990). It oxidizes instantaneously in
air and rapidly reduces protons to H2 gas at pH 7 (Schneider 1987). Hence, within the
cells, sequestration of cob(I)alamin would be important by preventing nonspecific or
deleterious side reactions.
The finding that cob(I)alamin was sequestered during the conversion of
cob(II)alamin to AdoCbl also indicates a physical interaction between the MSR and
ATR. This raises the question of whether MSR interacts with both MS and ATR by a
conserved mechanism. Prior studies of the MSR-MS reaction indicated that the optimal
stoichiometry is ~4 MSR/MS and that activity is maximal over a narrow range of ionic
strength indicating that electrostatic interactions are important to the MSR-MS interaction
(Olteanu and Banerjee 2001). Here we found that the optimal ratio for the MSR-ATR
system is also ~4 MSR/ATR, but in contrast to the MSR-MS system, there was no strict
dependence on ionic strength. Hence, although MSR has a conserved function in both
systems (cob(II)alamin reductase), the details of the protein-protein interactions between
MSR and either MS or ATR are apparently somewhat different.
The investigations reported here support prior studies that indicate redundant
systems for mitochondrial cobalamin reduction in mammals. Several lines of evidence
indicated a role for MSR in the reduction of cob(II)alamin to cob(I)alamin for AdoCbl
synthesis: the MSR and ATR enzymes converted cob(II)alamin to AdoCbl at a significant

80
rate; the stoichiometry needed for maximal activity was reasonable (4 MSR/ATR); and
cob(I)alamin was sequestered as is expected for the physiologically relevant system. On
the other hand, prior investigations showed that patients with inherited defects in MSR
(cblE disorder) generally have homocystinuria with the exception of one reported case
resulting in combined homocystinuria and mild methylmalonic aciduria (Wilson et al.
1999). This indicates that in vivo cob(II)alamin reduction for AdoCbl synthesis can occur
independently of MSR. Of the known complementation groups that result in
methylmalonic aciduria, only the cblH and cblA groups have phenotypes consistent with
a role in mitochondrial cobalamin reduction. Recent studies have indicated that the cblA
group functions to protect MCM from inactivation (Korotkova and Lidstrom 2004) and
there has only been one reported case of cblH methylmalonic aciduria (Watkins et al.
2000, Dobson et al. 2002). Thus at this time, there are no strong indications as to the
identity of the alternative reductase(s). Redundant systems for cobalamin metabolism are
known in bacterial systems (Fonseca and Escalante-Semerena 2000, Johnson et al. 2001),
and unpublished results). There, inducible systems provide extra capacity during times of
high demand. Similarly, in humans multiple cobalamin reductases might be required for
an efficient response to physiological stresses of diets that result in higher rates of
propionyl-CoA formation. For example, MSR could be targeted to the mitochondria
under circumstances where the diet is rich in amino acids metabolized via
propionyl-CoA.

propionyl-CoA
PCC
(2S)-methylmalonyl-CoA
MCEE
(2R)-methylmalonyl-CoA
MCM
succinyl-CoA
CENTRAL METABOLISM
breakdown breakdown (AdoCbl (AdoCbl is is unstable unstable in in vivo) vivo)
β-ligand
transferase
81
CNCbl
GSCbl
cob(III)alamin
reductase
cob(II)alamin
MSR
cob(I)alamin
ATR
AdoCbl
Methionine
synthase
[cob(II)alamin]
MSR
CH CH3THF 3THF +
homocysteine
Methionine
synthase
[CH [CH3Cbl] 3Cbl]
inactivation inactivation
THF +
methionine
Figure 3-1. Propionyl-CoA metabolism, methionine synthesis, and intracellular
cobalamin metabolism. In humans, the pathways shown are needed for the
complete catabolism of compounds degraded via propionyl-CoA and for
recycling homocysteine. Abbreviations: PPC, propionyl-CoA carboxylase;
MCEE, methylmalonyl-CoA epimerase; MCM, AdoCbl-dependent
methylmalonyl-CoA mutase; CNCbl, vitamin B12; GSCbl,
glutathionylcobalamin, MSR, methionine synthase reductase; ATR,
ATP:cob(I)alamin adenosyltransferase; THF, tetrahydrofolate.

Table 3-1. ATP:cob(I)alamin adenosyltransferase activity during ATR 239K and 239M purification
Purification step
Total protein
(mg)
Specific activity
(nmol min -1 mg -1 )
Total activity
(nmol min -1 )
Yield
(%)
Purification
(fold)
239K 239M 239K 239M 239K 239M 239K 239M 239K 239M
Cell free extract 160 180 49 35 7840 6300 100 100 1 1
Ammonium
sulfate precipitate
60 35 110 70 6600 2450 84 39 2.2 2
Hydroxyapatite 10 9 118 97 1180 873 15 14 2.4 2.8
Mono Q 5 3 220 190 1100 570 14 9 4.5 5.4
82

kDa
97
66
45
31
21
14
7
83
1 2 3 4
Figure 3-2. Purification of the human ATR. SDS-PAGE was used to assess the
purification of ATR 239K. Lane 1: molecular mass markers, lane 2: 10 µg
of soluble cell free extract from the E. coli expression strain, lane 3: 5 µg of
ammonium sulfate precipitate, lane 4: 4 µg hydroxyapatite eluate, lane 5: 2
µg of Mono Q fraction containing purified ATR. The purification of ATR
239M proceeded similarly (not shown).
5

84
Table 3-2. Use of alternative nucleotide donors by the human ATR
Nucleotide Donor
Activity %
a
ATR 239K ATR 239M
ATP (adenosine-5’-triphosphate) 100 100
ADP (adenosine-5’-diphosphate) ND b ND
AMP (adenosine-5’-monophosphate) ND ND
CTP (cytidine-5’-triphosphate) 9 6
GTP (guanosine-5’-triphosphate) 16 4
UTP (uridine-5’-triphosphate) 8 6
a
ATP is the physiological substrate for the ATR and activity with this nucleotide was set
to 100%.
b ND, none detected

85
Table 3-3. The MSR-ATR system sequesters cob(I)alamin
Variable assay components a
Amount detected (nmol)
AdoCbl CMCbl HOCbl
NADPH, MSR, ATR, and iodoacetate 85 46 45
DTT, FAD, ATR, and iodoacetate 18 165 6
NADPH, MSR, and ATR 94 ND b 81
a Reaction mixtures contained 200 mM Tris (pH 8.0) 1.6 mM KPi, 2,8 mM MgCl2,
100 mM KCl, 0.1 mM (200 nmol) cob(II)alamin, 0.4 mM ATP, and the components
indicated in the table in the following amounts: 10 µg purified ATR, 150 µg purified
MSR, 1 mM NADPH, 1 mM DTT, 50 µM FAD.
Cob(I)alamin was generated enzymatically using NADPH and purified MSR
(lines 1 and 3) or chemically by the combination of DTT and FAD which necessarily
generated “free” cob(I)alamin (line 2).
HOCbl, CMCbl, and AdoCbl were resolved and quantified by HPLC. In the
reverse-phase system used, their retention times were 11.9, 13.8, and 15.3 min,
respectively.
b ND, none detected.

ATR Activity (nmol min-1mg-1 ATR Activity (nmol min ) -1mg-1 )
ATR Activity (nmol min-1mg-1 ATR Activity (nmol min ) -1mg-1 )
A.
200
150
100
1 / Rate
0.012
0.009
0.006
0.003
50
0
0 100
0
0 0.05 0.1 0.15
1 / [ATP]
200 300 400
[ATP], (µM)
0.2
500
150
1 / Rate
0.0120
0.0090
86
250
200
150
100
B.
C. D.
200
200
ATR Activity (nmol min-1mg-1 ATR Activity (nmol min ) -1mg-1 )
ATR Activity (nmol min-1mg-1 ATR Activity (nmol min ) -1mg-1 )
1 / Rate
0.007
0.00525
0.0035
0.00175
0
50
0 0.1 0.2 0.3 0.4 0.5
0
0 10 20
1 / [cob(I)alamin]
30 40 50
[cob(I)alamin], (µM)
100
0.0060
0.0030
120
80
0.006
0.003
50
0
0 100
0
0 0.03 0.05 0.08 0.10 0.13
1 / [ATP]
200 300 400 500
40
0
0 10
0
0
20
0.1 0.2 0.3 0.4
1 / [cob(I)alamin]
30 40
0.5
50
[ATP], (µM) [cob(I)alamin], (µM)
Figure 3-3. Determination of kinetic constants for ATR 239K and ATR 239M. Panels A
and B are for ATR 239K and panels C and D are for ATR 239M. ATR
assays were performed as described in the materials and methods except that
the ATP and cob(I)alamin concentrations were varied as indicated. The
insets are Lineweaver-Burk plots of the data shown in the larger graph.
160
1 / Rate
0.009

Absorbance
0.5
0.4
0.3
0.2
0.1
0.0
400
87
440 480 520 560 600
Wavelength, (nm)
Figure 3-4. Absorbance spectra of an MSR-ATR assay. Prior to the addition of NADPH
the spectrum is that of cob(I)alamin (············); 1 hour after the addition of
NADPH and incubation at 37°C, the spectrum is that of AdoCbl (——); after
photolysis, the spectrum is characteristic of cob(II)alamin (⎯ ⎯ ⎯).

cob(II)alamin
NADPH + H +
MSR
cob(II)alamin
NADPH + H +
MSR
cob(II)alamin
NADPH + H +
MSR
NADP + NADP + NADP +
88
Scheme 1: Cob(I)alamin is released in solution and diffuses to the ATR
cob(I)alamin
ATP
ATR
Scheme 2: MSR reduces ATR-bound cob(II)alamin
ATR + cob(II)alamin
ATR—cob(II)alamin
ATR
MSR
tight tight complex
complex
NADPH NADPH NADPH + H +
+
ATP
AdoCbl
+ PPPi
NADP + NADP + NADP +
AdoCbl
+ PPPi
+ MSR
+ ATR
Figure 3-5. Two schemes depicting the conversion of cob(II)alamin to AdoCbl by the
MSR-ATR system. In scheme 1, MSR reduces cob(II)alamin to
cob(I)alamin and releases it free in solution for diffusion to the ATR. This
scheme seems unlikely due to the high nucleophilic nature of cob(I)alamin.
In scheme 2, MSR reduces cob(II)alamin while bound to the ATR thereby
sequestering the highly reactive cob(I)alamin molecule and protecting it from
oxidation and then adenosylation can occur. MSR, a dual flavoprotein,
mediates the one electron transfer to cob(II)alamin from the two-electron
donor, NADPH.

Wavelength, (525 nm)
0.24
0.23
0.22
0.21
89
ATR addition
0.20
0 5 10 15 20
Time, (min)
Figure 3-6. MSR produces little cob(I)alamin in the absence of the ATR enzyme.
During the first phase of the reaction little cob(II)alamin was reduced to
cob(I)alamin. During this phase, iodoacetate was used to detect
cob(I)alamin. Iodoacetate reacts chemically and quantitatively with
cob(I)alamin to form CMCbl and this reaction proceeds with an increase in
absorbance at 525 nm. At the 10 min time point, 10 µg of ATR was added to
the assay mixture and the rate of cob(I)alamin production increased
approximately 46-fold. This large increase in the rate of cob(I)alamin
formation suggested a physical interaction between the MSR and ATR
enzymes. The assay conditions used were similar to those for the MSR-ATR
assay and controls showed that iodoacetate worked effectively under these
conditions.

Activity, (nmol min-1 Activity, (nmol min ) -1 Activity, (nmol min ) -1 )
2.0
1.6
1.2
0.8
0.4
90
0.0
0 10 20 30 40
MSR/ATR,(mole/mole)
Figure 3-7. Stoichiometry of the MSR-ATR system. The MSR-ATR assay was used for
these studies. 10 µg (0.4 nmol) purified ATR was used with the molar ratio
of MSR indicated. Maximal activity required ~4 MSR/ATR.

Activity, (nmol min-1 Activity, (nmol min ) -1 )
4.0
3.0
2.0
1.0
91
0.0
0 120 240 360 480 600 720
Ionic strength, (mM)
Figure 3-8. Effect of ionic strength on the MSR-ATR system. The MSR-ATR assay
containing 40 µg purified ATR and 28 µg purified MSR was used for these
studies. The ionic strength was varied by the addition of KCl.

CHAPTER 4
PDUP IS A COENZYME-A-ACYLATING PROPIONALDEHYDE
DEHYDROGENASE ASSOCIATED WITH THE POLYHEDRAL BODIES
INVOLVED IN B12-DEPENDENT 1,2-PROPANEDIOL DEGRADATION BY
SALMONELLA ENTERICA SEROVAR TYPHIMURIUM LT2
Introduction
Virtually all Salmonella degrade 1,2-propanediol in a coenzyme-B12-dependent
manner and this ability (which is absent in Escherichia coli) is thought to be an important
aspect of the Salmonella-specific lifestyle (Roth et al. 1996, Price-Carter et al. 2001).
The degradation of 1,2-propanediol may play a role in the interaction of Salmonella with
its host organisms. In Salmonella enterica, in vivo expression technology (IVET) has
indicated that 1,2-propanediol utilization (pdu) genes may be important for growth in
host tissues, and competitive index studies with mice have shown that pdu mutations
confer a virulence defect (Conner et al. 1998, Heithoff et al. 1999). 1,2-Propanediol is
likely to be abundant in anoxic environments, such as the large intestine, since it is
produced by the anaerobic degradation of the common plant sugars rhamnose and fucose.
In addition, 1,2-propanediol degradation by S. enterica provides an important model
system for understanding coenzyme-B12-dependent processes some of which are
important in human physiology, industry, and the environment (Roth et al. 1996).
On the basis of biochemical studies, a pathway for coenzyme-B12-dependent
1,2-propanediol degradation by S. enterica was proposed (Figure 1-6). This pathway
begins with the conversion of 1,2-propanediol to propionaldehyde, a reaction that is
catalyzed by coenzyme-B12-dependent diol dehydratase (Toraya et al. 1979, Obradors et
92

al. 1988). The aldehyde is then disproportionated to propanol and propionic acid by a
reaction series thought to involve propanol dehydrogenase, coenzyme-A
93
(CoA)-dependent propionaldehyde dehydrogenase, phosphotransacylase, and propionate
kinase (Toraya et al. 1979, Obradors et al. 1988, Palacios et al. 2003). This pathway
generates one ATP, an electron sink, and a 3-carbon intermediate (propionyl-CoA),
which can feed into central metabolism via the methyl-citrate pathway (Horswill and
Escalante-Semerena 1997). In S. enterica, the degradation of 1,2-propanediol occurs
aerobically, or anaerobically when tetrathionate is supplied as an electron acceptor (Price-
Carter et al. 2001).
The genes specifically required for growth of S. enterica on 1,2-propanediol are
organized as a single contiguous cluster named the propanediol utilization (pdu) locus.
Determination of the DNA sequence of this locus showed that 1,2-propanediol
degradation was much more complex than prior biochemical studies had suggested.
DNA sequence analyses indicated the pdu locus included 23 genes (Bobik et al. 1997,
Bobik et al. 1999): six pdu genes are thought to encode enzymes needed for the
1,2-propanediol degradative pathway (Bobik et al. 1997); two are involved in transport
and regulation (Bobik et al. 1992, Chen et al. 1994); two are probably involved in diol
dehydratase reactivation (Bobik et al. 1999); one is needed for the conversion of vitamin
B12 (CN-B12) to coenzyme B12 (Johnson et al. 2001); five are of unknown function; and
seven share similarity to genes involved in the formation of carboxysomes, polyhedral
bodies found in certain cyanobacteria and some chemoautotrophs (Shively and English
1991, Shively et al. 1998).

94
The finding that the pdu locus included a number of genes similar to those involved
in carboxysome formation suggested that a related polyhedral body might be involved in
B12-dependent 1,2-propanediol degradation. Indeed, S. enterica was recently shown to
form polyhedral bodies (also called polyhedral organelles) during growth on
1,2-propanediol (Bobik et al. 1999, Havemann et al. 2002). In general appearance, the
pdu polyhedra are similar to carboxysomes. They are 100-150 nm in cross-section and
are composed of a proteinaceous interior covered by a 3- to 4-nm protein shell.
However, carboxysomes and the pdu polyhedra differ in several ways. Carboxysomes,
which contain most of the cell’s ribulose 1,5-bisphosphate carboxylase/oxygenase
(RuBisCO), function to improve autotrophic growth at low CO2 concentrations and are
thought to do so by concentrating CO2 (Shively et al. 2001, Price et al. 2002). By
contrast, the pdu polyhedra of S. enterica do not contain RuBisCO. They consist of a
protein shell composed in part of the PduA protein (Havemann et al. 2002) and are
associated with B12-dependent diol dehydratase and other unidentified proteins (Bobik et
al. 1999). The function of the S. enterica polyhedral bodies is uncertain. Work published
to date suggests that they may serve to minimize aldehyde toxicity by sequestration and
channeling of propionaldehyde and/or by moderating the rate of aldehyde formation
through control of diol dehydratase activity (Chen et al. 1994, Rondon et al. 1995,
Stojiljkovic et al. 1995, Bobik et al. 1999, Havemann et al. 2002, Havemann and Bobik
2003).
If the pdu polyhedra do indeed function to minimize aldehyde toxicity, it might be
expected that aldehyde-degrading enzymes are also associated with these structures. In
this report, we show that the pduP gene encodes a polyhedral-body-associated

CoA-acylating propionaldehyde dehydrogenase important for 1,2-propanediol
degradation.
95
Materials and Methods
Bacterial Strains, Media, and Growth Conditions
The bacterial strains used in this study are listed in Table 4-1. The rich medium
used was LB medium (Difco, Detroit, MI) (Miller 1972). The minimal medium used was
no-carbon-E (NCE) medium (Vogel and Bonner 1956, Berkowitz et al. 1968)
supplemented with 53 mM propanediol, 1 mM MgSO4, 148 nM CNCbl, and 0.3 mM
each valine, isoleucine, leucine, and threonine. In addition, for strains carrying pLAC22
media were also supplemented with 1 mM IPTG, and 100 µg Amp/ml. Cultures were
incubated at 37°C with shaking at 250 rpm as previously described (Leal et al. 2003), and
cell growth was determined by measuring the optical density of cultures at 600 nm.
General Molecular Methods
Agarose gel electrophoresis was done as described previously (Sambrook et al.
1989). Plasmid DNA was purified by the alkaline lysis procedure (Sambrook et al. 1989)
or by using Qiagen products (Qiagen, Chatsworth, CA) according to the manufacturer's
instructions. Following restriction or PCR amplification, DNA was purified using
Qiagen PCR purification or gel extraction kits. Restriction enzymes were used according
to standard protocols (Sambrook et al. 1989). DNA fragments were ligated using T4
DNA ligase according to the manufacturer's directions. Electroporation was carried out
as previously described (Bobik et al. 1999).
General Protein Methods
SDS-PAGE was performed using Bio-Rad Redigels and Bio-Rad Mini-Protean II
electrophoresis units according to the Manufacturer's instructions. Following gel

96
electrophoresis, proteins were stained with Coomassie Brilliant Blue R-250. The protein
concentration of solutions was determined using Bio-Rad protein assay reagent (Bio-Rad)
(Bradford 1976).
P22 Transduction
Transductional crosses were performed as described using P22 HT105/1 int-210
(Davis et al. 1980), a mutant phage that has high transducing ability (Schmieger 1971).
For the preparation of P22 transducing lysates from strains having galE mutations,
overnight cultures were grown on LB-medium supplemented with 11 mM glucose and 11
mM galactose. Transductants were tested for phage contamination and sensitivity by
streaking on green plates against P22 H5.
Cloning pduP into pLAC22
PCR was used to amplify pduP from the template pMGS2 (Havemann et al. 2002).
The primers used for amplification were 5’-GGAATTCGGATCCTATGAATACTTCTG
AACTCGAAAC-3’ (forward) and 5’-GGAATTCAAGCTTCAGTTAGCGAATAGAAA
AGCC-3’ (reverse). These primers introduced BamHI and HindIII restriction sites that
were used for cloning into pLAC22 cut with BglII and HindIII (Warren et al. 2000).
Cloning was carried out by ligation of the complementary cohesive ends formed by
cutting with BamHI and BglII since pduP contains an internal BglII site. Following
ligation, clones were introduced into S. enterica strain TR6579 by electroporation and
transformants were selected by plating on LB-agar supplemented with ampicillin (Amp)
at 100 µg/ml. Pure cultures were prepared from selected transformants, and plasmid
DNA isolated from these strains was cut with NruI, PstI, and HindIII. The DNA
sequence of one clone that released products of the expected sizes following restriction
(2,635, 3,322 and 943 base pairs) was determined and found to be identical to the

97
previously published DNA sequence of pduP (Bobik et al. 1999). This clone (pNL9) as
well as pLAC22 without insert were moved into strain BE191 by P22 transduction and
the resulting strains (BE270 and BE269) were used for complementation studies.
Construction of His8-PduP
PCR was used to fuse an N-terminal His-tag to the PduP protein. Template
pMGS2 was used with the following primers:
5’-GGGGATCCATG(CAT)8ATGAATACTTCTGAACTCGAAAC-3’ (forward); and
5’-GGAATTCAAGCTTCAGTTAGCGA ATAGAAAAGCC-3’ (reverse). These
primers introduced BamHI and HindIII restriction sites that were used for cloning into
pTA925 cut with BglII and HindIII. The ligation mixture was used to transform E. coli
strain DH5α and transformants were selected by plating on LB agar supplemented with
25 µg kanamycin/ml. Plasmid DNA isolated from selected transformants was analyzed
by restriction mapping and DNA sequencing. One plasmid encoding His8-PduP (pNL69)
was introduced into E. coli expression strain BL21 (DE3) RIL by electroporation to form
expression strain BE273.
Purification of His8-PduP under Denaturing and Nondenaturing Conditions
For purification of PduP under denaturing conditions, E. coli strain BE273 was
grown in 500 ml LB Kan (25 µg/ml) broth incubated at 37°C with shaking at 275 rpm in
a 1-liter baffled Erlenmeyer flask. Cells were grown to an optical density of 0.6-0.8 at
600 nm and PduP production was induced by addition of IPTG to 1 mM. Cells were
incubated for an additional 2 hours, and harvested by centrifugation at 6,690 x g for 10
minutes using a Beckman JLA-10.500 rotor. Two grams of cells (wet weight) were
resuspended in 3 ml of 50 mM sodium phosphate, 300 mM NaCl, pH 7 and broken using

98
a French pressure cell at 20,000 psi (SLM Aminco, Urbana, IL). The protease inhibitor
phenylmethylsulfonylflouride was added to the cell extract to a concentration of 100
µg/ml. To separate soluble and insoluble fractions, cell extract was centrifuged at 31,000
x g for 30 minutes using a Beckman JA-20 rotor. Inclusion bodies were purified from the
insoluble fraction using BPER-II according to the manufacturer’s instructions. Purified
inclusion bodies were solubilized overnight in binding buffer (10 mM imidazole, 500
mM NaCl, 20 mM Tris-HCl pH 7.9, and 6 M urea) and filtered using a 0.45-µm filter.
Affinity purification was carried out using 1-ml Amersham Pharmacia Hi-trap chelating
(Ni 2+ ) column according the manufacturer’s instructions with the following
modifications: 6 M urea was added to all buffers, and two wash steps were used, the first
with 20 mM imidazole and the second with 60 mM imidazole. The PduP protein was
then eluted using 150 mM imidazole.
For purification of PduP under nondenaturing conditions, inclusion bodies (isolated
as described above) were resuspended in 50 mM potassium phosphate, pH 7, 25 mM
NaCl and stored at 4 °C for 24 h. Insoluble proteins were pelleted by centrifugation and
PduP that remained in the supernatant was purified by nickel-affinity chromatography
using Ni-NTA resin (Qiagen) according to the manufacturer's instructions. Briefly, the
column was washed with 5 bed volumes of binding buffer (50 mM sodium phosphate,
300 mM NaCl, pH 7) and PduP was eluted with binding buffer supplemented with 40
mM imidazole.
Propionaldehyde Dehydrogenase Assays
Two assays for propionaldehyde dehydrogenase activity were used. Assay one was
carried out as previously described (Walter et al. 1997). Reaction mixtures contained 50

mM CHES (2-[N-cyclohexylamino]ethanesulfonic acid) pH 9.5, 10 mM
propionaldehyde, 1 mM dithiothreitol (DTT), 75 µM NAD + , 100 µM HS-CoA, and an
99
appropriate amount of cell extract. Assays were incubated at 37°C. Enzyme activity was
measured by following the conversion of NAD + to NADH by monitoring the absorbance
of reaction mixtures at 340 nm, and the amount of NADH formed was determined using
∆ε340 = 6.22 mM -1 cm -1 . Assay two was done as described above except that DTT was
omitted and the absorbance of reaction mixtures was followed at 232 nm. An increase in
absorbance at 232 nm occurs when thioester bonds are formed and ∆ε232 ≅ 4.5 mM -1 cm -1
(Dawson et al. 1969).
Identification of Propionyl-CoA a Product of the Propionaldehyde Dehydrogenase
Reaction
Propionaldehyde dehydrogenase reactions were performed as described above
except that CHES buffer was replaced by 50 mM potassium phosphate, pH 7.
Immediately after the reactions reached completion, they were analyzed by reverse-phase
HPLC using conditions previously described (Bobik and Rasche 2003). For the
purification of reaction products, HPLC solvents were buffered with 10 mM ammonium
formate, pH 6.4. Selected reaction products were collected, lyophilized, resuspended in
distilled water at a concentration of 4 µM. For the MALDI-TOF MS (Matrix Assisted
Laser Desorption Ionization-Time of Flight Mass Spectrometry) analyses, samples were
mixed 1:1 with 10 mg/ml α-cyano-4-hydroxycinnamic acid matrix in 0.1% trifluoroacetic
acid in 30% acetonitrile. The sample (1 µl) was spotted on a MALDI plate and analyzed
using an Applied Biosystems Voyager DE-Pro MALDI-TOF MS operated in reflector
mode with a delay of 100 nsec, acc voltage of 20 KV, and a grid voltage of 71.5%. The

100
sample was irradiated with a nitrogen laser (337 nm) and 100 individual laser shots were
collected for each sample.
Construction of a Nonpolar pduP Deletion
Bases 28 to 1,386 of the pduP coding sequence were deleted via a PCR-based
method (Miller and Mekalanos 1988). The deletion was designed to leave the predicted
translational start and stop signals of all pdu genes intact. The following primers were
used for PCR amplification of the flanking regions of the pduP gene: primer 1,
5’-GCTCTAGACCAGGCCAACATCATCCGTGAAGTTAG-3’; primer 2,
5’-TCATCGCGACCTCAGCAGGGTTTCGAGTTCAGAAGTAATCATTG-3’; primer
3, 5’-CGCTAACTGAGGTCGCGATGAATACC-3’; and primer 4,
5’-GAAGAGCTCAATTCTGCGGCGGTACGCTGACCACC-3’. Primers 1 and 2 were
used to amplify 496 DNA bases upstream of the pduP gene, and primers 3 and 4 were
used to amplify 508 DNA bases downstream of pduP gene. The upstream and
downstream amplification products were purified and then fused by a PCR reaction that
included 1 ng/µl of each product and primers 1 and 4. The fused product was cut with
XbaI and SacI (these sites were designed into primers 1 and 4, respectively), and ligated
to suicide vector pCVD442 (Miller and Mekalanos 1988) that had been similarly cut.
The ligation mixture was used to transform E. coli S17.1 by electroporation and
transformants were selected on LB medium supplemented with Amp (100 µg/ml).
Plasmids were extracted from four of the isolated transformants, screened by restriction
analysis, and all released an insert of the expected size (1004 bp). One of these
transformants was used to introduce the pduP deletion into the S. enterica chromosome
using the procedure of Miller and Mekelanos (1988) with the following modification.

101
For the conjugation step, strain BE47 was used as the recipient and exconjugants were
selected by plating on LB agar supplemented with ampicillin (100 µg/ml) and
chloramphenicol (20 µg/ml). Deletion of the pduP coding sequence was verified by PCR
using chromosomal DNA as a template. Lastly, the thr-480 dCAM insertion used for
selection of exconjugants was crossed-off by P22 transduction using a phage lysate
prepared with the wild-type strain and by selecting for prototrophy on NCE glucose
minimal medium.
Antibody Preparation
His8-PduP purified using Ni 2+ -affinity chromatography was resolved on a 12%
Tris-HCl SDS-PAGE gel. The protein band corresponding to His8-PduP was excised and
used as a source of antigen for the preparation of polyclonal antibody in a New Zealand
white rabbit by Cocalico Biologicals (Reamstown, PA). To eliminate antibodies
cross-reacting with E. coli proteins, the antiserum was preadsorbed using acetone powder
prepared from E. coli BL21DE3 RIL containing pTA925 (the expression strain lacking
pduP) as described previously (Harlow and Lane 1988).
Western Blots
Cultures were grown and prepared for SDS-PAGE and electroblotting as
previously described (Havemann et al. 2002). Membranes were probed using adsorbed
anti-PduP antiserum (diluted 1:3,500 in blocking buffer) as the source of the primary
antibody, and goat anti-rabbit conjugated to alkaline phosphatase (diluted 1:3,000 in
blocking buffer) as the secondary antibody. Chromogenic developing agents were used
following the manufacturer’s instructions (Bio-Rad).

Electron Microscopy
102
For electron microscopy, cells were grown in minimal medium supplemented with
37 mM succinate and 26 mM propanediol. Cultures (10 ml) were incubated in 125-ml
Erlenmeyer flasks at 37º C, with shaking at 275 rpm in a New Brunswick C24 Incubator
Shaker.
For immunogold localization of the PduP protein, cells were prepared as previously
described (Bobik et al. 1999). The source of the primary antibody was rabbit polyclonal
anti-PduP antiserum or preimmune serum diluted in 1:100 in PBS. The secondary
antibody used was goat anti-rabbit IgG conjugated to12-nm colloidal gold (Jackson
ImmunoResearch Laboratories, Inc. West Grove, PA) diluted 1:30 in PBS.
DNA Sequencing and Analysis
DNA sequencing was carried out at the University of Florida Interdisciplinary
Center for Biotechnology Research DNA Sequencing Core Facility using Applied
Biosystems automated sequencing equipment (Perkin Elmer, Norwalk, CT) or at the
University of Florida, Department of Microbiology and Cell Science DNA Sequencing
Facility using a LI-COR model 4000L DNA sequencer, automated sequencing
equipment, and Base ImagIR Analysis Software version 04.1h (LI-COR, Lincoln, NE).
The template for DNA sequencing was plasmid DNA purified using Qiagen 100 tips or
Qiagen mini-prep kits. BLAST software was used for sequence similarity searches
(Altschul et al. 1997).
Chemicals and Reagents
Formaldehyde, (R, S)-1,2-propanediol, and antibiotics were from Sigma Chemical
Company (St. Louis, MO). Isopropyl-β-D-thiogalactopyranoside (IPTG) was from
Diagnostic Chemicals Limited (Charlotteville PEI, Canada). Restriction enzymes were

103
from New England Biolabs (Beverly, MA) or Promega (Madison, WI). T4 DNA ligase
was from New England Biolabs. Glutaraldehyde was from Tousimis (Rockville, MD),
and uranyl acetate was from EM Sciences (Washington, PA). LR White resin was from
Ted Pella Inc. (Redding, CA). Acrylamide, agarose, ammonium persulfate, Coomassie
Brilliant Blue R-250, EDTA, ethidium bromide, 2-mercaptoethanol, N,
N-bis-methylene-acrylamide, powdered milk, SDS, and TEMED were from Bio-Rad
(Hercules, CA). Bacterial Protein Extraction Reagent II (BPER-II) was from Pierce
(Rockford, IL). Other chemicals were from Fisher Scientific (Pittsburgh, PA).
Results
Effect of a Precise pduP Deletion on the Growth of S. enterica on Minimal
1,2-Propanediol Medium
Prior biochemical studies indicated that a CoA-acylating propionaldehyde
dehydrogenase was involved in the degradation of 1,2–propanediol by S. enterica
(Toraya et al. 1979, Obradors et al. 1988). Recent DNA sequence analyses of the pdu
operon (a cluster of genes required for 1,2-propanediol degradation by S. enterica)
showed that the PduP protein has sequence similarity to a number of CoA-acylating
aldehyde dehydrogenases and is 32% identical in amino acid sequence to the E. coli
aldehyde/alcohol dehydrogenase, AdhE (Kessler et al. 1991, Bobik et al. 1999). This
suggested that PduP is a CoA-acylating propionaldehyde dehydrogenase needed for the
degradation of 1,2-propanediol by S. enterica.
To examine the role of PduP in 1,2-propanediol degradation, a precise deletion of
the pduP gene was constructed using a PCR-based method, and the effects of this
mutation on the growth of S. enterica on 1,2-propanediol/CN-B12 minimal medium were
examined. The wild-type strain and strain BE191 (∆pduP) grew with generation times of

104
about 7.4 and 21.7 hours, respectively, and reached maximum optical densities at 600 nm
of about 1.38 and 0.49, respectively. This showed the ∆pduP mutant was significantly
impaired for growth on 1,2-propanediol/CNCbl minimal medium.
As a control, we tested whether the growth defect of strain BE191 (∆pduP) could
be corrected by the expression of pduP in trans. The generation times of strains BE270
(∆pduP/pLAC22-pduP) and BE269 (∆pduP/pLAC22-no insert) on
1,2-propanediol/CNCbl minimal medium were 3.2 and 25.4 hours, respectively, and the
maximum optical densities reached by these strains were 1.7 for BE270 and 0.45 for
BE269. Hence, expression of pduP in trans corrected the growth defect of ∆pduP
mutation on 1,2-propanediol/CNCbl minimal medium. This result indicated that the
observed growth defect of BE191 was due to deletion of pduP, but not to polarity or a
mutation inadvertently introduced during strain construction. This was the expected
result since BE191 was constructed by a PCR-based method designed to produce a
precise nonpolar deletion. Strain BE270 was found to grow about 2.2-fold faster than did
the wild-type strain. It appeared that pLAC22 enhanced growth of this strain on
1,2-propanediol minimal medium for unknown reasons, since the wild-type strain also
grew faster when it carried this vector without insert (not shown). Growth tests were
repeated twice with similar results. The fact that BE191 (∆pduP) was impaired for
growth on 1,2-propanediol/CNCbl minimal medium, in conjunction with the control
experiments described above indicates that pduP plays an important role in
1,2-propanediol degradation in vivo.

105
Propionaldehyde Dehydrogenase Activity in Wild-type S. enterica and a ∆pduP
Mutant
To test whether the pduP gene encodes a propionaldehyde dehydrogenase, enzyme
assays were performed on cell extracts of wild-type S. enterica and strain BE191
(∆pduP). The ∆pduP mutant used was the same strain used for the growth studies
described above, and assays were conducted on both the supernatants and the pellets that
resulted from centrifugation of cell extracts at 31,000 x g (Table 4-2). For the wild-type
strain, the majority of the propionaldehyde dehydrogenase activity was found in the
31,000 x g pellet which had a specific activity 1.15 µmol min -1 mg -1 protein. Similar
extracts from the pduP mutant had only 1% of the specific activity of the wild-type strain
(0.011 µmol min -1 mg -1 ). When the substrates, propionaldehyde, NAD + or HS-CoA were
omitted from the assay mixture, no activity was detected. To our knowledge, this was the
first experimental evidence that the pduP gene encodes a CoA-acylating propionaldehyde
dehydrogenase.
It was of interest that the majority of propionaldehyde dehydrogenase activity was
found in the 31,000 x g pellet from the wild-type strain (99%). Prior studies showed that
polyhedral bodies are involved in AdoCbl-dependent 1,2-propanediol degradation by
S. enterica (Bobik et al. 1999, Havemann et al. 2002). Hence, the finding that the PduP
aldehyde dehydrogenase was associated with the 31,000 x g pellet suggested that it might
be associated with the polyhedral bodies which pellet under similar centrifugation
conditions (unpublished results).
High-level Production of the PduP Protein
E. coli strain BE273 (pET41a-His8-pduP) was constructed to produce high levels of
recombinant His8-PduP protein. Protein production by this strain as well as by the

106
control strain BE237 (which is isogenic to BE273 except that it contains the expression
plasmid without insert) was analyzed by SDS-PAGE (Figure 4-1). Both the soluble and
inclusion-body fractions of cell extracts were examined. High amounts of a protein with
a molecular mass near 50 kDa were found in the inclusion-body fraction of cell extracts
from expression strain BE273, and moderate amounts of a 50-kDa protein were also
found in the soluble fraction of cell extracts from this strain (Figure 4-1, lanes 3 and 5).
This is near the predicted molecular mass of His8-PduP (50 kDa). In contrast, the soluble
and inclusion-body fractions from control strain BE237 (expression vector without insert)
contained relatively little protein of 50 kDa molecular mass (Figure 4-1, lanes 2 and 4)
indicating that the observed 50-kDa protein produced by BE273 was His8-PduP.
The cell extracts analyzed by SDS-PAGE (Figure 4-1) were tested for
propionaldehyde dehydrogenase activity. Substantial activity (2.4 µmol min -1 mg -1
protein) was found in inclusion bodies isolated from strain BE273, and in the soluble
fraction of this strain (0.19 µmol min -1 mg -1 protein). However, little activity was found
in either the soluble or inclusion-body fractions of control strain BE237 (0.02 µmol min -1
mg -1 protein). For the inclusion-body fraction from the expression strain,
propionaldehyde dehydrogenase activity was linear with protein concentration (data not
shown). Furthermore, when the substrates, propionaldehyde, NAD + or HS-CoA were
omitted from the assay mixture, no activity was detected. Thus, these data provided
additional evidence that PduP is a CoA-acylating propionaldehyde dehydrogenase.
Purification of Recombinant His8-PduP Protein
It was observed that following storage of purified inclusion bodies at 4 °C for 24
hours a significant amount of active His8-PduP protein was eluted into the soluble

107
fraction. Following centrifugation of this preparation at 31,000 x g for 30 min.,
approximately 0.5 mg of soluble protein remained in the supernatant. Enzyme assays
showed that this supernatant contained 9.22 µmol min -1 mg -1 propionaldehyde
dehydrogenase activity. This sample was further purified by nickel-affinity
chromatography. The purity of the resulting preparation was analyzed by SDS-PAGE
followed by staining with Coomassie. Results indicated that the His8-PduP obtained was
highly purified (Figure 4-1, lane 6) and enzyme assays showed that the preparation was
enriched in propionaldehyde dehydrogenase activity, with a specific activity of 15.2 µmol
min -1 mg -1 protien. The reaction requirements for purified PduP were the same as those
described above for the partially purified enzyme. These results demonstrated that the
His8-PduP protein had propionaldehyde dehydrogenase activity.
Propionyl-CoA is a Product of the PduP Reaction
The fact that CoA-SH was required for the propionaldehyde dehydrogenase activity
of PduP suggested that propionyl-CoA was a reaction product. Several additional
experiments were conducted to test this possibility. Propionaldehyde dehydrogenase
reactions containing purified PduP protein were allowed to proceed to completion. Then,
reverse-phase HPLC was used to identify CoA-SH and CoA-derivatives present in
reaction mixtures. A single CoA compound with a retention time of 10.2 min was
detected. Further analyses showed that this compound co-eluted with authentic
propionyl-CoA following co-injection and resolution via C18 reverse-phase HPLC. This
indicated that propionyl-CoA was a product of the PduP aldehyde dehydrogenase
reaction. Next, the reaction product identified as propionyl-CoA by reverse-phase HPLC
was analyzed by MALDI-TOF MS. Three major peaks at m/z 824.8, 848.1, and 878.1

108
were observed in the mass spectrum. These peaks correspond to the [M+H] + ,
[M+H+Na] + , and [M+3NH4] + of propionyl-CoA, further supporting the assignment of
propionyl-CoA as a PduP reaction product. In addition, the propionaldehyde
dehydrogenase reaction catalyzed by the PduP enzyme was followed
spectrophotometrically at 232 nm, the wavelength at which thioester bonds
characteristically absorb (Dawson et al. 1969). Within experimental error, the specific
activity of purified PduP was the same when reactions were followed at 340 nm (NAD +
reduction) or 232 nm (thioester bond formation). Thus, based on the evidence described
above, we conclude that propionyl-CoA is a product of the PduP propionaldehyde
dehydrogenase reaction.
Preparation and Specificity of the Anti-PduP Antiserum
To obtain the antigen needed for preparation of antisera, His8-PduP was purified
from inclusion bodies isolated from expression strain BE273 by Ni 2+ -affinity
chromatography under denaturing conditions. SDS-PAGE indicated that the PduP
protein obtained was highly purified (data not shown). Denaturing conditions were used
to obtain antigen because a nondenaturing protocol was unavailable at the time.
Following purification, His8-PduP was used to prepare polyclonal anti-PduP antiserum in
rabbit. The antiserum obtained was subjected to a preadsorption procedure to remove
cross-reacting proteins (Warren et al. 2000). To determine the specificity of the adsorbed
anti-PduP antiserum, Western blot analysis was done on boiled cell lysates. The
anti-PduP antibody preparation recognized one major protein band in extracts from
wild-type S. enterica at approximately 50 kDa (Figure 4-2, lane 2); however, this band
was not detected in strain BE191, which contained a nonpolar pduP deletion, but was
otherwise isogenic to the wild-type strain (Figure 4-2, lane 2). This indicated that the

109
antiserum obtained was specific for the PduP protein. Furthermore, the anti-PduP
antibody preparation detected a 50-kDa band in boiled cell extracts from PduP expression
strain BE270 (pLAC22-PduP) but not in extracts from strain BE269 which carried the
pLAC22 expression plasmid without the insert but is otherwise isogenic to strain BE270
(Figure 4-2, lanes 4 and 5). This provided additional evidence that the antibody
preparation was specific for the PduP protein. The minor band observed in Figure 4-2,
lane 5, was most likely a degradation product of PduP that resulted from overproduction
since this band was not detected in extracts from the wild-type strain (Figure 4-2, lane 2).
Localization of PduP Immunoelectron Microscopy
Wild-type S. enterica and BE191 (∆pduP) were grown on
succinate/1,2-propanediol minimal medium which induces formation of the polyhedral
bodies involved in 1,2-propanediol degradation (Bobik et al. 1999, Havemann et al.
2002). Immunogold labeling of wild-type S. enterica and BE191 (∆pduP) was then
carried out using anti-PduP antiserum described above. In the micrograph (Figure 4-3),
the antibody-conjugated gold particles (solid black circles) indicate the location of the
PduP protein. Note that the gold particles localized to the polyhedral bodies (which
appear as uniformly stained regions within the cytoplasm) in the wild-type strain, but no
labeling was observed in the ∆pduP mutant, although polyhedra were present. These
results indicated the antibody preparation reacted specifically with the PduP protein
under the labeling conditions used and that the PduP protein localizes to the polyhedral
bodies. Additional electron microscopy studies of standard thin sections showed that the
∆pduP mutant formed normal-appearing polyhedra (standard thin sections result in
higher contrast than does fixation for immunolabeling and the fine structure of the

110
polyhedra is more easily discerned). This result indicated that PduP is nonessential for
polyhedral body formation.
Propionaldehyde Dehydrogenase Activity is associated with Purified pdu Bodies
To further investigate the subcellular localization of the PduP enzyme, the
polyhedral bodies involved in 1,2-propanediol degradation were purified and
propionaldehyde dehydrogenase activity was followed during the course of their
purification (Table 4-3). The polyhedra were purified by a combination of detergent
treatment and differential and density-gradient centrifugation as described in Havemann
and Bobik (2003). Electron microscopy and SDS-PAGE indicated that the polyhedra
obtained were highly purified (not shown). During the purification the specific activity
of the PduP propionaldehyde dehydrogenase increased 13.5-fold from 0.3 - 4.0 µmol
min -1 mg -1 protein (Table 4-3). This suggested that the polyhedral bodies comprised
about 7% of the total cell protein, which is consistent with previous electron microscopy
(Havemann et al. 2002). Furthermore, Western blots verified that the PduP enzyme was
associated with the purified polyhedra (not shown). Thus, enzyme assays and Western
blots with purified polyhedra supported the immunolabeling studies described above and
indicated that the PduP propionaldehyde dehydrogenase is associated with the polyhedral
bodies involved in 1,2-propanediol degradation.
Discussion
Prior biochemical studies indicated that a CoA-acylating propionaldehyde
dehydrogenase was required for AdoCbl-dependent 1,2-propanediol degradation (Toraya
et al. 1979, Obradors et al. 1988). Subsequent DNA sequence analyses of the pdu operon
identified a presumptive enzyme (PduP) that was 32% identical to the E. coli
aldehyde/alcohol dehydrogenase, AdhE (Bobik et al. 1999). In this report, genetic and

111
biochemical evidence was presented that established PduP as a CoA-acylating
propionaldehyde dehydrogenase involved in AdoCbl-dependent 1,2-propanediol
degradation by S. enterica.
Previously, Jorge Escalante's group proposed that two propionaldehyde
dehydrogenases might be involved in 1,2-propanediol degradation by S. enterica: one
that produces propionyl-CoA and a second that oxidizes propionaldehyde directly to
propionate (Palacios et al. 2003). The PduP enzyme studied here produced
propionyl-CoA as a reaction product and was inactive in the absence of HS-CoA showing
that it is incapable of oxidizing propionaldehyde directly to propionate under the assay
conditions used. It was also shown that an S. enterica strain with a nonpolar pduP null
mutation grew at one-third the rate of the wild-type strain. The residual growth of a
∆pduP mutant could have resulted from the activity of an alternative propionaldehyde
dehydrogenase, such as the one proposed to convert propionaldehyde directly to
propionate (Palacios et al. 2003), since the propionate produced by this enzyme could be
used as a carbon and energy source via the methyl citrate pathway (Horswill and
Escalante-Semerena 1997).
Previously, we established that S. enterica formed polyhedral bodies during
AdoCbl-dependent growth on 1,2-propanediol (Bobik et al. 1999). Immunolabeling
studies demonstrated that these bodies consisted of a protein shell (partly composed of
the PduA protein) AdoCbl-dependent diol dehydratase and additional unidentified
proteins (Bobik et al. 1999, Havemann et al. 2002). In this report, we presented several
lines of evidence that indicated PduP is polyhedral-body-associated. The finding that the
pdu polyhedra include both diol dehydratase (aldehyde-producing) and propionaldehyde

112
dehydrogenase (aldehyde-consuming) is consistent with the prior proposal that these
structures function to minimize aldehyde toxicity (Chen et al. 1994, Stojiljkovic et al.
1995, Havemann et al. 2002). Presumably, the shell of the polyhedral body could trap
the propionaldehyde produced by the PduCDE diol dehydratase until it is consumed by
the PduP propionaldehyde dehydrogenase. Sequestering propionaldehyde until it is
converted to propionyl-CoA might protect sensitive cytoplasmic components.
Alternatively, the inclusion of diol dehydratase and propionaldehyde dehydrogenase in
the pdu polyhedra might simply serve to juxtapose these enzymes in order to facilitate
propionaldehyde channeling, although it is unclear why such a complex structure would
be needed for this sole purpose. Prior studies suggested that the pdu polyhedra minimize
aldehyde toxicity by limiting aldehyde production through control of AdoCbl availability
(Havemann et al. 2002). Such a mechanism would be improved by either aldehyde
channeling or sequestration, and a combined mechanism that fine tunes propionaldehyde
production and consumption could explain the selective advantage provided by these
complex polyhedral bodies involved in 1,2-propanediol degradation.

113
Table 4-1. Bacterial strains
Species Strain Genotype
E. coli
BL21
(DE3) RIL
BE237
BE273
(E. coli B) F - ompT hsdS (rB - mB - ) dcm + Tet r gal λ (DE3) endA
Hte (argU ileY leuW Cam r )
BL21 (DE3) RIL/pET-41a (T7 expression vector without
insert, Kan r )
BL21 (DE3) RIL/pNL69 (T7expression vector with insert
encoding His8-PduP)
S17.1λpir recA(RP4-2-Tc::Mu) λpir
BE47 thr-480::Tn10dCam
S. enterica serovar Typhimurium LT2
TR6579
metA22 metE551 trpD2 ilv-452 hsdLt6 hsdSA29 HsdB -
strA120 GalE - Leu - Pro -
BE191 ∆pduP659
BE268 LT2/ pNL9 (pLAC22-pduP, Ap r )
BE269 ∆pduP659/ pLAC22 (vector without insert, Ap r )
BE270 ∆pduP659/pNL9 (pLAC22-pduP, Ap r )

114
Table 4-2. Propionaldehyde dehydrogenase activity in extracts from wild-type
S. enterica and strain BE191 (∆pduP).
Extract S/P a
Wild-type
BE191
(∆pduP)
Total
protein
(mg)
Propionaldehyde dehydrogenase activity
Specific activity
(µmol min -1 mg -1
protein)
Total activity
(µmol min -1 )
Total activity
(%)
S 45 0.004 0.18 1
P 25 1.15 29.23 99
S 51 ND b ND 0
P 30 0.011 0.33 100
a Assays were performed on the supernatant (S) and pellet (P) fractions following
centrifugation at 31,000 x g.
b ND indicates, none detected.

115
Table 4-3. Propionaldehyde dehydrogenase activity during polyhedral body purification
Sample
Crude extract
Protein
(mg)
Total activity
(µmol min -1 )
Specific activity a
(µmol min -1 mg -1
protein)
Yield
(%)
Fold -
purification
515 152 0.30 100 1.0
Detergent/salts
treatment
338 205 0.61 135 2.1
12,000 x g
super
338 169 0.50 111 1.7
48,000 x g
pellet
5.0 9.6 1.92 6.3 6.5
12,000 x g
super
Sucrose
4.3 8.6 2.00 5.7 6.8
density
gradient
0.3 1.2 4.00 0.8 13.6
a
Activities were calculated from the initial rate

97
66
45
31
22
14
7
116
kDa 1 2 3 4 5 6
PduP
Figure 4-1. SDS-PAGE analysis of His8-PduP. Lane 1: Molecular mass markers, lane 2:
12 µg soluble extract from the control strain (BE237), lane 3: 12 µg soluble
extract from the His8-PduP expression strain (BE273), lane 4: 12 µg
inclusion-body extract from the control strain, lane 5: 5 µg inclusion-body
extract from the His8-PduP expression strain, lane 6: 2 µg His8-PduP purified
by nickel-affinity chromatography. The control and the expression strains
were isogenic except that the control strain lacked the pduP coding sequence.

kDa
209
80
49
35
29
117
1 2 3 4
Figure 4-2. Western blot using the anti-PduP antiserum. Lane 1: Molecular mass
standards, lane 2: wild-type (Salmonella enterica), lane 3: ∆pduP (strain
BE191), lane 4: ∆pduP/pLAC22 (vector without insert), lane 5:
∆pduP/pLAC22-pduP. Each lane was loaded with 20 µl boiled whole cells
(OD600 = 1.0).
5

118
B
Figure 4-3. Immunogold localization of the PduP enzyme. A, Wild-type S. enterica; B,
strain BE191 (∆pduP). The 12 nm gold particles (small black circles)
indicate the location of the PduP protein. The arrows point to the polyhedral
bodies. Bars 100 nm

CHAPTER 5
CONCLUSIONS
In humans, cobalamin cofactors are required coenzymes for two enzymatic
reactions that are vital to human health. These coenzymes are used in propionate
metabolism and methionine biosynthesis, and in humans, deficiencies in cobalamin
metabolism result in methylmalonic aciduria and homocystinuria. Nine complementation
groups associated with such deficiencies have been identified. Historically, progress on
understanding these diseases has been slow due to difficulties in purifying the enzymes
involved as well as the lack of facile genetic methods. A bacterial model system was
used to circumvent some of these problems and allowed progress on the identification
and characterization of the genes and enzymes involved in these rare, but devastating
disorders.
Identification of the Bovine and Human Adenosyltransferase
In Chapter 2, S. enterica was used as a model system for the identification of the
bovine and human ATR cDNAs by complementation analysis. In this screening
technique, the bovine cDNA was isolated by complementing an ATR deficient
S. enterica strain. Subsequently, sequence similarity searches using the bovine ATR
cDNA identified a homologous human gene. Both human and bovine expression
libraries were screened; however, complementing clones were only found from the
bovine library. This could have been due to the relative abundance of ATR mRNA in
these systems. Ruminant livers have the highest reported concentration of B12-dependent
119

120
enzymes measured in any tissue. Similar methods could be applied for the identification
of additional cDNAs involved in cobalamin metabolism.
To determine whether the bovine and human cDNAs that we identified encode
ATRs, the cDNAs were cloned and over expressed in E. coli. Cell extracts from the
expression strains were then used for biochemical studies. Both the soluble and insoluble
fractions of the crude cell extracts of bovine and human expression strains were found to
have ATR activity. This suggests that the cDNAs we identified encode ATR enzymes.
Additional studies were used to show that the human cDNA could function as an ATR
in vivo. An S. enterica ATR mutant transformed with a plasmid-encoded source of the
human ATR was used for growth studies on 1,2-propanediol supplemented with HOCbl.
From these experiments, we showed that the human ATR restored wild-type phenotype
to an ATR-deficient S. enterica strain proving that the human ATR can function
physiologically in a heterologous host.
Previous studies have shown that patients with cblB methylmalonic aciduria lack
ATR activity (Fenton and Rosenberg 1981). To determine if mutations in the ATR
underlie cblB methylmalonic aciduria, expression of this enzyme was monitored in
normal and cblB mutant fibroblast cells. Western blot analysis using antibodies specific
for the ATR showed that expression of this protein was altered in fibroblast cells from
patients with cblB when compared to control cells. Additional studies conducted
concurrently in another lab showed that the ATR gene (MMAB) corresponded to the cblB
complementation group of methylmalonic aciduria (Dobson et al. 2002).
Biochemical Characterization of the Human Adenosyltransferase
In Chapter 3, we investigated the biochemical properties of the human ATR.
Dobson et al. previously analyzed the MMAB gene and identified two amino acid

121
substitutions that are common polymorphic variants in control cell lines (Dobson et al.
2002). Both of these variants were independently expressed, purified, and used for
biochemical analysis. In the studies conducted, the variants behaved similarly suggesting
that both are normal forms of the enzyme. The kinetic properties including Vmax, and Km
for ATP and cob(I)alamin were determined. The Vmax falls within the range of previously
reported ATRs including CobA, PduO, and T. acidophilum ATR (Suh and Escalante-
Semerena 1995, Johnson et al. 2001, Saridakis et al. 2004). The Km for cob(I)alamin
(1 µM) is physiologically significant when compared to the intracellular concentration of
B12 in human tissue (Schneider and Stroinski 1987). The Km for ATP is low when
compared to the physiological levels of this substrate but is considered relevant since
ATP is required for many different reactions in human tissue.
In this study, the nucleotide specificity of the human ATR variants was also
analyzed. It was found that ATR variants are highly specific for ATP (the physiological
substrate for adenosylation). This result is not surprising since catalysis with other
nucleotide substrates (GTP, CTP, and UTP) resulting in GCbl, CCbl, and UCbl are
generally inhibitory to AdoCbl-dependent enzymes and would be counter-productive
(Toraya and Mori 1999). Establishing the biochemical properties of control ATRs
provides a basis of comparison for the characterization of disease causing variants of this
enzyme.
Methionine Synthase Reductase is a Cob(II)alamin Reductase for the Human
Adenosyltransferase
In addition to analyzing the biochemical properties of the human ATR, we also
investigated whether human MSR can act as a cob(II)alamin reductase for the synthesis
of AdoCbl by the ATR (Chapter 3). Previous studies have shown that MSR functions as

122
a cob(II)alamin reductase for reductive activation of MS (Olteanu and Banerjee 2001).
Here, an MSR-ATR linked assay was developed and this provided biochemical evidence
that MSR can also function as a cob(II)alamin reductase for AdoCbl synthesis. The rate
of AdoCbl formation in these assays (0.96 nmol min -1 ) is physiologically significant
since humans require only small amounts of cobalamin (Schneider 1987). This rate is
also significant when compared to previous measurements of AdoCbl formed in human
fibroblast cells (0.021 pmol min -1 mg -1 ) (Fenton and Rosenberg 1981).
In mammalian systems, MSR is a cytosolic protein while ATR is localized to the
mitochondria. In these studies we demonstrated that MSR is as a cob(II)alamin reductase
for the ATR in vitro. The proposed requirement of MSR for AdoCbl synthesis
necessitates the presence of MSR in both the cytoplasm and the mitochondria. Recent
studies have shown that alternative transcript splicing of MSR produces protein with and
without a mitochondrial targeting sequence (Leclerc et al. 1999). These findings
strengthen the hypothesis that MSR has dual functionality in the cytoplasm and the
mitochondria.
Investigations also suggest an interaction between MSR and ATR. In MSR-ATR
linked assays containing iodoacetate (an alkylating agent that forms CMCbl in the
presence of cob(I)alamin), very little CMCbl was formed suggesting that MSR does not
release cob(I)alamin free in solution. In addition, HPLC analyses showed that AdoCbl
formation was favored even when a large excess of iodoacetate was added to the assays.
These studies indicate that an interaction between MSR and ATR triggers cobalamin
reduction.

123
Studies also indicated that humans have redundant systems for mitochondrial
cob(II)alamin reduction. Prior complementation analysis demonstrated that cblE (MSR)
disorders usually result in homocystinuria without methylmalonic aciduria with only one
reported case having both homocystinuria and mild methylmalonic aciduria (Wilson et al.
1999). The fact that most cblE patients lack methylmalonic aciduria suggests that there
are redundant mitochondrial cob(II)alamin reductase(s) involved in AdoCbl synthesis.
There is precedence for redundant cobalamin reductases in bacteria. The role of MSR
could be to provide extra capacity under certain conditions, such as a high protein diet,
rich in amino acids that are metabolized via the propionyl-CoA pathway.
Future Experimentation
The development of the screening method employed in this study can be used in
the future as a general technique for the isolation of other cDNAs involved in cobalamin
metabolism. Salmonella is an excellent model system for these studies since
AdoCbl-dependent diol dehydratase for 1,2-propanediol degradation has been well
studied. Once other enzymes involved in cobalamin metabolism are identified in
S. enterica, strains deficient in these activities can be used to screen mammalian
expression libraries by complementation. Hence, this technique may allow for the
identification of other genes crucial for cobalamin metabolism in eukaryotic systems.
Now that a purification method has been developed for the ATR, efforts can be
focused on crystallizing this protein. Determining the three dimensional structure of the
ATR will allow us to understand how mutations in this enzyme lead to dysfunction by
comparing amino acid residues in normal and mutant ATRs and mapping the altered
amino acid residues onto the structure. Potentially, this process would help lend insight
in identifying the residues of the ATR that are crucial for enzymatic activity.

124
Future studies should also be focused on determining the mode of interaction
between MSR and ATR. Approaches to verifying the interaction between these enzymes
include gel filtration, immuno-precipitation, and native PAGE analyses. All of these
techniques would verify a physical interaction between MSR and ATR and give a better
understanding of how this system functions in vivo.
Most importantly, knowledge of the ATR gene will allow for DNA based methods
of methylmalonic aciduria diagnosis following amniocentesis or chorionic villi sampling
and could aid in determining the mode of treatment. Additionally, identification of the
ATR gene makes possible the development of gene therapy as a treatment for cblB
methylmalonic aciduria.

LIST OF REFERENCES
Aberg, A., S. Hahne, M. Karlsson, A. Larsson, M. Ormo, A. Ahgren and B. M.
Sjoberg. 1989. Evidence for two different classes of redox-active cysteines in
ribonucleotide reductase of Escherichia coli. J. Biol. Chem. 264:12249-52.
Allen, S. G. H., R. W. Kellermeyer, R. L. Stjernholm and H. G. Wood. 1964.
Purification and properties of enzymes involved in the propionic acid fermentation.
J. Bacteriol. 87:171-187.
Altschul, S. F., W. Gish, W. Miller, E. W. Myers and D. J. Lipman. 1990. Basic local
alignment search tool. J. Mol. Biol. 215:403-410.
Altschul, S. F., T. L. Madden, A. A. Schäffer, J. Zhang, Z. Zhang, W. Miller and D.
J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein
database search programs. Nucleic Acids Res. 25:3389-3402.
Ampola, M. G., M. J. Mahoney, E. Nakamura and K. Tanaka. 1975. Prenatal therapy
of a patient with vitamin B12 responsive methylmalonic acidemia. N. Engl. J. Med.
293:313-317.
Baker, S. J. and V. I. Mathan. 1981. Evidence regarding the minimal daily requirement
of dietary vitamin B12. Am. J. Clin. Nutr. 34:2423-33.
Banerjee, R. 1997. The Yin-Yang of cobalamin biochemistry. Chem. Biol. 4:175-86.
Banerjee, R. (eds.). 1999. Chemistry and Biochemistry of B12. John Wiley & Sons, Inc,
New York.
Banerjee, R. and S. Chowdhury. 1999. Methylmalonyl-CoA mutase, p. 707-730. In R.
Banerjee (ed.), Chemistry and Biochemistry of B12. John Wiley & Sons, Inc., New
York.
Banerjee, R. and S. W. Ragsdale. 2003. The many faces of vitamin B12: catalysis by
cobalamin-dependent enzymes. Annu. Rev. Biochem. 72:209-47.
Banerjee, R. V., S. R. Harder, S. W. Ragsdale and R. G. Matthews. 1990. Mechanism
of reductive activation of cobalamin-dependent methionine synthase: an electron
paramagnetic resonance spectroelectrochemical study. Biochemistry 29:1129-35.
125

126
Banerjee, R. V., N. L. Johnston, J. K. Sobeski, P. Datta and R. G. Matthews. 1989.
Cloning and sequence analysis of the Escherichia coli metH gene encoding
cobalamin-dependent methionine synthase and isolation of a tryptic fragment
containing the cobalamin-binding domain. J. Biol. Chem. 264:13888-95.
Banerjee, R. V. and R. G. Matthews. 1990. Cobalamin-dependent methionine synthase.
Faseb. J. 4:1450-9.
Barker, H. A., H. Weisbach and R. D. Smyth. 1958. A Coenzyme containing
pseudovitamin B12. Proc. Natl. Acad. Sci. U S A 44:1039-1097.
Battersby, A. R. 1994. How nature builds the pigments of life: the conquest of vitamin
B12. Science 264:1551-7.
Bauer, C. B., M. V. Fonseca, H. M. Holden, J. B. Thoden, T. B. Thompson, J. C.
Escalante-Semerena and I. Rayment. 2001. Three-dimensional structure of
ATP:corrinoid adenosyltransferase from Salmonella typhimurium in its free state,
complexed with MgATP, or complexed with hydroxycobalamin and MgATP.
Biochemistry 40:361-74.
Berkowitz, D., J. Hushon, H. Whitfield, Jr., J. Roth and B. Ames. 1968. Procedure for
identifying nonsense mutations. J. Bacteriol. 96:215-220.
Bobik, T. A., M. Ailion and J. R. Roth. 1992. A single regulatory gene integrates
control of vitamin B12 synthesis and propanediol degradation. J. Bacteriol.
174:2253-66.
Bobik, T. A., G. D. Havemann, R. J. Busch, D. S. Williams and H. C. Aldrich. 1999.
The propanediol utilization (pdu) operon of Salmonella enterica serovar
Typhimurium LT2 includes genes necessary for the formation of polyhedral
organelles involved in coenzyme B 12-dependent 1,2-propanediol degradation. J.
Bacteriol. 181:5967-5975.
Bobik, T. A. and M. E. Rasche. 2003. HPLC assay for methylmalonyl-CoA epimerase.
Anal. Bioanal. Chem. 375:344-9.
Bobik, T. A. and R. S. Wolfe. 1989. Activation of formylmethanofuran synthesis in cell
extracts of Methanobacterium thermoautotrophicum. J. Bacteriol. 171:1423-1427.
Bobik, T. A., Y. Xu, R. M. Jeter, K. E. Otto and J. R. Roth. 1997. Propanediol
utilization genes (pdu) of Salmonella typhimurium: three genes for the propanediol
dehydratase. J. Bacteriol. 179:6633-9.
Booker, S., S. Licht, J. Broderick and J. Stubbe. 1994. Coenzyme B12-dependent
ribonucleotide reductase: evidence for the participation of five cysteine residues in
ribonucleotide reduction. Biochemistry 33:12676-85.

127
Bradford, 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.
Brady, R. O., E. G. Castanera and H. A. Barker. 1962. The enzymatic synthesis of
cobamide coenzymes. J. Biol. Chem. 237:2325-32.
Cauthen, S. E., M. A. Foster and D. D. Woods. 1966. Methionine synthesis by extracts
of Salmonella typhimurium. Biochem. J. 98:630-635.
Chen, P., D. I. Andersson and J. R. Roth. 1994. The control region of the pdu/cob
regulon in Salmonella typhimurium. J. Bacteriol. 176:5474-82.
Ciani, F., M. A. Donati, G. Tulli, G. M. Poggi, E. Pasquini, D. S. Rosenblatt and E.
Zammarchi. 2000. Lethal late onset cblB methylmalonic aciduria. Crit. Care Med.
28:2119-21.
Claros, M. G. and P. Vincens. 1996. Computational method to predict mitochondrially
imported proteins and their targeting sequences. Eur. J. Biochem. 241:770-786.
Conner, C. P., D. M. Heithoff, S. M. Julio, R. L. Sinsheimer and M. J. Mahan. 1998.
Differential patterns of acquired virulence genes distinguish Salmonella strains.
Proc. Natl. Acad. Sci. U S A 95:4641-5.
Cooper, B. A. and D. S. Rosenblatt. 1987. Inherited defects of vitamin B12 metabolism.
Annu. Rev. Nutr. 7:291-320.
Crouzet, J., S. Levy-Schil, B. Cameron, L. Cauchois, S. Rigault, M. C. Rouyez, F.
Blanche, L. Debussche and D. Thibaut. 1991. Nucleotide sequence and genetic
analysis of a 13.1-kilobase-pair Pseudomonas denitrificans DNA fragment
containing five cob genes and identification of structural genes encoding
Cob(I)alamin adenosyltransferase, cobyric acid synthase, and bifunctional
cobinamide kinase-cobinamide phosphate guanylyltransferase. J. Bacteriol.
173:6074-87.
Davis, R. W., D. Botstein and J. R. Roth. 1980. Advanced bacterial genetics : a manual
for genetic engineering. Cold Spring Harbor Laboratory, Cold Spring Harbor, New
York.
Dawson, R. M. C., D. C. Elliott, W. H. Elliott and K. M. Jones (eds.). 1969. Data for
biochemical research, 2nd ed. Oxford University Press, Oxford.
Debussche, L., M. Couder, D. Thibaut, B. Cameron, J. Crouzet and F. Blanche.
1991. Purification and partial characterization of cob(I)alamin adenosyltransferase
from Pseudomonas denitrificans. J. Bacteriol. 173:6300-6302.

128
Del Corral, A. and R. Carmel. 1990. Transfer of cobalamin from the cobalamin-binding
protein of egg yolk to R binder of human saliva and gastric juice.
Gastroenterologist 98:1460-6.
Dixon, M. M., S. Huang, R. G. Matthews and M. Ludwig. 1996. The structure of the
C-terminal domain of methionine synthase: presenting S-adenosylmethionine for
reductive methylation of B12. Structure 4:1263-75.
Dobson, C. M., T. Wai, D. Leclerc, H. Kadir, M. Narang, J. P. Lerner-Ellis, T. J.
Hudson, D. S. Rosenblatt and R. A. Gravel. 2002. Identification of the gene
responsible for the cblB complementation group of vitamin B12-dependent
methylmalonic aciduria. Hum. Mol. Genet. 11:3361-9.
Dobson, C. M., T. Wai, D. Leclerc, A. Wilson, X. Wu, C. Dore, T. Hudson, D. S.
Rosenblatt and R. A. Gravel. 2002. Identification of the gene responsible for the
cblA complementation group of vitamin B12-responsive methylmalonic acidemia
based on analysis of prokaryotic gene arrangements. Proc. Natl. Acad. Sci. U S A
99:15554-9.
Dolphin, D. 1982. B12. John Wiley & Sons, Inc., New York.
Drennan, C. L., S. Huang, J. T. Drummond, R. G. Matthews and M. L. Ludwig.
1994. How a protein binds B12: a 3.0 Å x-ray structure of B12-binding domains of
methionine synthase. Science 266:1669-1674.
Drummond, J. T. and R. G. Matthews. 1993. Cobalamin-dependent and cobalaminindependent
methionine synthases in Escherichia coli: two solutions to the same
chemical problem. Adv. Exp. Med. Biol. 338:687-692.
Drummond, J. T. and R. G. Matthews. 1994. Nitrous oxide inactivation of cobalamindependent
methionine synthase from Escherichia coli: characterization of the
damage to the enzyme and prosthetic group. Biochemistry 33:3742-50.
Fenton, W. A., R. A. Gravel and D. Rosenblatt. 2000. Disorders of propionate and
methylmalonate metabolism. In C. R. Scriver, A. L. Beaudet, W. S. Sly, et al. (ed.),
The Metabolic and molecular basis of inherited disease. McGraw-Hill, New York.
Fenton, W. A. and L. E. Rosenberg. 1978a. Genetic and biochemical analysis of human
cobalamin mutants in cell culture. Annu. Rev. Genet. 12:223-48.
Fenton, W. A. and L. E. Rosenberg. 1978b. Mitochondrial metabolism of
hydroxocobalamin: synthesis of adenosylcobalamin by intact rat liver
mitochondria. Arch. Biochem. Biophys. 189:441-7.
Fenton, W. A. and L. E. Rosenberg. 1981. The defect in the cblB class of human
methylmalonic acidemia: deficiency of cob(I)alamin adenosyltransferase activity in
extracts of cultured fibroblasts. Biochem. Biophys. Res. Commun. 98:283-289.

129
Fenton, W. A. and L. E. Rosenberg. 2000. Inherited Disorders of Cobalamin Transport
and Metabolism. In C. R. Scriver, A. L. Beaudet, W. S. Sly, et al. (ed.), The
Metabolic and molecular basis of inherited disease. McGraw-Hill, New York.
Fonseca, M. V., N. R. Buan, A. R. Horswill, I. Rayment and J. C. Escalante-
Semerena. 2002. The ATP : Co(I)rrinoid adenosyltransferase (CobA) enzyme of
Salmonella enterica requires the 2 '-OH group of ATP for function and yields
inorganic triphosphate as its reaction byproduct. J. Biol. Chem. 277:33127-33131.
Fonseca, M. V. and J. C. Escalante-Semerena. 2000. Reduction of Cob(III)alamin to
Cob(II)alamin in Salmonella enterica serovar Typhimurium LT2. J. Bacteriol.
182:4304-9.
Fonseca, M. V. and J. C. Escalante-Semerena. 2001. An in vitro reducing system for
the enzymic conversion of cobalamin to adenosylcobalamin. J. Biol. Chem.
276:32101-8.
Frey, P. A. and G. H. Reed. 2000. Radical mechanisms in adenosylmethionine- and
adenosylcobalamin-dependent enzymatic reactions. Arch. Biochem. Biophys.
382:6-14.
Friedmann, H. C. 1975. Biosynthesis of corrinoids, p. 75-103. In B. M. Babior (ed.),
Cobalamin. John Wiley and Sons, New York.
Fujii, K., J. H. Galivan and F. M. Huennekens. 1977. Activation of methionine
synthase: further characterization of flavoprotein system. Arch. Biochem. Biophys.
178:662-70.
Fujii, K. and F. M. Huennekens. 1974. Activation of methionine synthetase by a
reduced triphosphopyridine nucleotide-dependent flavoprotein system. J. Biol.
Chem. 249:6745-53.
Goulding, C. W. and R. G. Matthews. 1997. Cobalamin-dependent methionine
synthase from Escherichia coli: involvement of zinc in homocysteine activation.
Biochemistry 36:15749-57.
Goulding, C. W., D. Postigo and R. G. Matthews. 1997. Cobalamin-dependent
methionine synthase is a modular protein with distinct regions for binding
homocysteine, methyltetrahydrofolate, cobalamin, and adenosylmethionine.
Biochemistry 36:8082-91.
Gravel, R. A., M. J. Mahoney, F. H. Ruddle and L. E. Rosenberg. 1975. Genetic
complementation in heterokaryons of human fibroblasts defective in cobalamin
metabolism. Proc. Natl. Acad. Sci. U S A 72:3181-5.
Guest, J. R., M. A. Foster and S. Friedman. 1962a. Alternative pathways for
methylation of homocysteine by Escherichia coli. Biochem. J. 84:93P-94P.

130
Guest, J. R., E. L. Smith, D. D. Woods and S. Friedman. 1962b. A methyl analogue of
cobamide coenzyme in relation to methionine synthesis by bacteria. Nature
195:340-342.
Hall, D. A., T. C. Jordan-Starck, R. O. Loo, M. L. Ludwig and R. G. Matthews.
2000. Interaction of flavodoxin with cobalamin-dependent methionine synthase.
Biochemistry 39:10711-9.
Halpern, J. 1985. Mechanisms of coenzyme B12-dependent rearrangements. Science
227:869-75.
Harder, J., R. Eliasson, E. Pontis, M. D. Ballinger and P. Reichard. 1992. Activation
of the anaerobic ribonucleotide reductase from Escherichia coli by Sadenosylmethionine.
J. Biol. Chem. 267:25548-52.
Harlow, E. and D. Lane. 1988. Antibodies: a laboratory manual. Cold Spring Harbor
Laboratory, Cold Spring Harbor, New York.
Havemann, G. D. and T. A. Bobik. 2003. Protein content of polyhedral organelles
involved in coenzyme B12-dependent degradation of 1,2-propanediol in Salmonella
enterica serovar Typhimurium LT2. J. Bacteriol. 185:5086-95.
Havemann, G. D., E. M. Sampson and T. A. Bobik. 2002. PduA is a shell protein of
polyhedral organelles involved in coenzyme B12-dependent degradation of 1,2propanediol
in Salmonella enterica serovar Typhimurium LT2. J. Bacteriol.
184:1253-61.
Heithoff, D. M., C. P. Conner, U. Hentschel, F. Govantes, P. C. Hanna and M. J.
Mahan. 1999. Coordinate intracellular expression of Salmonella genes induced
during infection. J. Bacteriol. 181:799-807.
Hodgkin, D. C., J. Pickworth, J. H. Robertson, K. N. Trueblood, R. J. Prosen and J.
G. White. 1955. Crystal structure of the hexacarboxylic acid derived from B12 and
the molecular structure of the vitamin. Nature 176:325-328.
Horswill, A. R. and J. C. Escalante-Semerena. 1997. Propionate catabolism in
Salmonella typhimurium LT2: two divergently transcribed units comprise the prp
locus at 8.5 centisomes, prpR encodes a member of the sigma-54 family of
activators, and the prpBCDE genes constitute an operon. J. Bacteriol. 179:928-40.
Huennekens, F. M., K. S. Vitols, K. Fujii and D. W. Jacobsen. 1982. Biosynthesis of
the cobalamin coenzymes, p. 145-164. In D. Dolphin (ed.), B12, vol. volume 1.
John Wiley & Sons, Inc., New York.
Jarrett, J. T., D. M. Hoover, M. L. Ludwig and R. G. Matthews. 1998. The
mechanism of adenosylmethionine-dependent activation of methionine synthase: a
rapid kinetic analysis of intermediates in reductive methylation of cob(II)alamin
enzyme. Biochemistry 37:12649-58.

131
Johnson, C. L. V. J., E. Pechonick, S. D. Park, G. D. Havemann, N. A. Leal and T.
A. Bobik. 2001. Functional genomic, biochemical, and genetic characterization of
the Salmonella pduO gene, an ATP:cob(I)alamin adenosyltransferase gene. J.
Bacteriol. 183:1577-1584.
Jordan, A., E. Pontis, M. Atta, M. Krook, I. Gibert, J. Barbe and P. Reichard. 1994.
A second class I ribonucleotide reductase in Enterobacteriaceae: characterization of
the Salmonella typhimurium enzyme. Proc. Natl. Acad. Sci. U S A 91:12892-6.
Kapadia, C. R. 1995. Vitamin B12 in health and disease: part I--inherited disorders of
function, absorption, and transport. Gastroenterologist 3:329-44.
Kass, L. 1978. William B. Castle and intrinsic factor. Ann. Intern. Med. 89:983-91.
Kessler, D., I. Leibrecht and J. Knappe. 1991. Pyruvate formate-lyase-deactivase and
acetyl-CoA reductase activities of Escherichia coli reside on a polymeric protein
particle encoded by adhE. FEBS Lett. 281:59-63.
Kofoid, E., C. Rappleye, I. Stojiljkovic and J. R. Roth. 1999. The 17-gene
ethanolamine (eut) operon of Salmonella typhimurium encodes five homologues of
carboxysome shell proteins. J. Bacteriol. 181:5317-5329.
Kolhouse, J. F. and R. H. Allen. 1977. Recognition of two intracellular cobalamin
binding proteins and their identification as methylmalonyl-CoA mutase and
methionine synthetase. Proc. Natl. Acad. Sci. U S A 74:921-5.
Korotkova, N., L. Chistoserdova, V. Kuksa and M. E. Lidstrom. 2002. Glyoxylate
regeneration pathway in the methylotroph Methylobacterium extorquens AM1. J.
Bacteriol. 184:1750-8.
Korotkova, N. and M. E. Lidstrom. 2004. MeaB is a component of the methylmalonyl-
CoA mutase complex required for protection of the enzyme from inactivation. J.
Biol. Chem. 279:13652-8.
Leal, N. A., S. D. Park, P. E. Kima and T. A. Bobik. 2003. Identification of the human
and bovine ATP:Cob(I)alamin adenosyltransferase cDNAs based on
complementation of a bacterial mutant. J. Biol. Chem. 278:9227-34.
Leclerc, D., E. Campeau, P. Goyette, C. E. Adjalla, B. Christensen, M. Ross, P.
Eydoux, D. S. Rosenblatt, R. Rozen and R. A. Gravel. 1996. Human methionine
synthase: cDNA cloning and identification of mutations in patients of the cblG
complementation group of folate/cobalamin disorders. Hum. Mol. Genet. 5:1867-
74.
Leclerc, D., M. Odievre, Q. Wu, A. Wilson, J. J. Huizenga, R. Rozen, S. W. Scherer
and R. A. Gravel. 1999. Molecular cloning, expression, and physical mapping of
the human methionine synthase reductase gene. Gene 240:75-88.

132
Leclerc, D., A. Wilson, R. Dumas, C. Gafuik, D. Song, D. Watkins, H. H. Heng, J.
M. Rommens, S. W. Scherer, D. S. Rosenblatt and R. A. Gravel. 1998. Cloning
and mapping of a cDNA for methionine synthase reductase, a flavoprotein
defective in patients with homocystinuria. Proc. Natl. Acad. Sci. U S A 95:3059-
64.
Ledley, F. D., M. R. Lumetta, H. Y. Zoghbi, P. VanTuinen, S. A. Ledbetter and D.
H. Ledbetter. 1988. Mapping of human methylmalonyl CoA mutase (MUT) locus
on chromosome 6. Am. J. Hum. Genet. 42:839-46.
Lenhert, P. G. and D. C. Hodgkin. 1961. Structure of 5,6dimethylbenzimidazolylcobamide
coenzyme. Nature 192:937-938.
Ljungdahl, L. and H. Wood. 1982. Acetate biosynthesis. In D. Dolphin (ed.), B12. John
Wiley & Sons, New York.
Mahoney, M. J., A. C. Hart, V. D. Steen and L. E. Rosenberg. 1975.
Methylmalonicacidemia: biochemical heterogeneity in defects of 5'deoxyadenosylcobalamin
synthesis. Proc. Natl. Acad. Sci. USA 72:2799-2803.
Mahoney, M. J., L. E. Rosenberg, S. H. Mudd and B. W. Uhlendorf. 1971. Defective
metabolism of vitamin B12 in fibroblasts from children with methylmalonic
aciduria. Biochem. Biophys. Res. Commun. 44:375-81.
Mancia, F., N. H. Keep, A. Nakagawa, P. F. Leadlay, S. McSweeney, B. Rasmussen,
P. Bosecke, O. Diat and P. R. Evans. 1996. How coenzyme B12 radicals are
generated: the crystal structure of methylmalonyl-coenzyme A mutase at 2 Å
resolution. Structure 4:339-50.
Maniatis, T., E. F. Fritsch and J. Sambrook. 1982. Molecular Cloning; A Laboratory
Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.
Marsh, E. N. and C. L. Drennan. 2001. Adenosylcobalamin-dependent isomerases:
new insights into structure and mechanism. Curr. Opin. Chem. Biol. 5:499-505.
Marsh, E. N., N. McKie, N. K. Davis and P. F. Leadlay. 1989. Cloning and structural
characterization of the genes coding for adenosylcobalamin-dependent
methylmalonyl-CoA mutase from Propionibacterium shermanii. Biochem. J.
260:345-52.
Matsui, S. M., M. J. Mahoney and L. E. Rosenberg. 1983. The natural history of the
inherited methylmalonic acidemias. N. Engl. J. Med. 293:857-861.
Matthews, R. G. 1999. Cobalamin-dependent Methionine Synthase, p. 681-706. In R.
Banerjee (ed.), Chemistry and biochemistry of B12. John Wiley & Sons, Inc., New
York.

133
Mellman, I., H. F. Willard and L. E. Rosenberg. 1978. Cobalamin binding and
cobalamin-dependent enzyme activity in normal and mutant human fibroblasts. J.
Clin. Invest. 62:952-60.
Mellman, I., H. F. Willard, P. Youngdahl-Turner and L. E. Rosenberg. 1979.
Cobalamin coenzyme synthesis in normal and mutant human fibroblasts. Evidence
for a processing enzyme activity deficient in cblC cells. J. Biol. Chem. 254:11847-
11853.
Mellman, I. S., P. Youngdahl-Turner, H. F. Willard and L. E. Rosenberg. 1977.
Intracellular binding of radioactive hydroxocobalamin to cobalamin-dependent
apoenzymes in rat liver. Proc. Natl. Acad. Sci. U S A 74:916-20.
Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory,
Cold Spring Harbor, New York.
Miller, V. L. and J. J. Mekalanos. 1988. A novel suicide vector and its use in
construction of insertion mutations: osmoregulation of outer membrane proteins
and virulence determinants in Vibrio cholerae requires toxR. J. Bacteriol.
170:2575-83.
Minot, G. R. and W. P. Murphy. 1926. Treatment of pernicious anemia by a special
diet. J. Am. Med. Assoc. 87:470-476.
Nordlund, P. and H. Eklund. 1993. Structure and function of the Escherichia coli
ribonucleotide reductase protein R2. J. Mol. Biol. 232:123-64.
Obradors, N., J. Badía, L. Baldomà and J. Aguilar. 1988. Anaerobic metabolism of
the L-rhamnose fermentation product 1,2-propanediol in Salmonella typhimurium.
J. Bacteriol. 170:2159-2162.
Olteanu, H. and R. Banerjee. 2001. Human methionine synthase reductase, a soluble P-
450 reductase-like dual flavoprotein, is sufficient for NADPH-dependent
methionine synthase activation. J. Biol. Chem. 276:35558-63.
Padmakumar, R. and R. Banerjee. 1995. Evidence from electron paramagnetic
resonance spectroscopy of the participation of radical intermediates in the reaction
catalyzed by methylmalonyl-coenzyme A mutase. J. Biol. Chem. 270:9295-300.
Padmakumar, R. and R. Banerjee. 1997. Evidence that cobalt-carbon bond homolysis
is coupled to hydrogen atom abstraction from substrate in methylmalonyl-CoA
mutase. Biochemistry 36:3713-8.
Palacios, S., V. J. Starai and J. C. Escalante-Semerena. 2003. Propionyl coenzyme A
is a common intermediate in the 1,2-propanediol and propionate catabolic pathways
needed for expression of the prpBCDE operon during growth of Salmonella
enterica on 1,2-propanediol. J. Bacteriol. 185:2802-10.

134
Paschen, S. A. and W. Neupert. 2001. Protein import into mitochondria. IUBMB Life
52:101-12.
Pezacka, E. 1993. Identification of and characterization of two enzymes involved in the
intracellular metabolism of cobalamin. Cyanocobalamin β-ligand transferase and
micosomal cob(III)alamin reductase. Biochem. et. Biophys. Acta. 1157:167-177.
Pezacka, E., R. Green and D. W. Jacobsen. 1990. Glutathionylcobalamin as an
intermediate in the formation of cobalamin coenzymes. Biochem. Biophys. Res.
Commun. 169:443-450.
Price, G. D., S. Maeda, T. Omata and M. R. Badger. 2002. Modes of active inorganic
carbon uptake in the cyanobacterium, Synechococcus sp PCC7942. Functional
Plant Biology 29:131-149.
Price-Carter, M., J. Tingey, T. A. Bobik and J. R. Roth. 2001. The alternative electron
acceptor tetrathionate supports B12-dependent anaerobic growth of Salmonella
enterica serovar typhimurium on ethanolamine or 1,2-propanediol. J. Bacteriol.
183:2463-75.
Qureshi, A. A., D. S. Rosenblatt and B. A. Cooper. 1994. Inherited disorders of
cobalamin metabolism. Crit. Rev. Oncol. Hematol. 17:133-51.
Refsum, H., P. M. Ueland, O. Nygard and S. E. Vollset. 1998. Homocysteine and
cardiovascular disease. Annu. Rev. Med. 49:31-62.
Reichard, P. 1993. The anaerobic ribonucleotide reductase from Escherichia coli. J.
Biol. Chem. 268:8383-6.
Rickes, E. L., N. G. Brink, F. R. Koniuszy, T. R. Wood and K. Folkers. 1948.
Crystalline vitamin B12. Science 107:396-397.
Rondon, M. R., R. Kazmierczak and J. C. Escalante-Semerena. 1995. Glutathione is
required for maximal transcription of the cobalamin biosynthetic and 1,2propanediol
utilization (cob/pdu) regulon and for the catabolism of ethanolamine,
1,2-propanediol, and propionate in Salmonella typhimurium LT2. J. Bacteriol.
177:5434-9.
Rosenblatt, D. S., A. L. Aspler, M. I. Shevell, B. A. Pletcher, W. A. Fenton and M. R.
Seashore. 1997. Clinical heterogeneity and prognosis in combined methylmalonic
aciduria and homocystinuria (cblC). J. Inherit. Metab. Dis. 20:528-38.
Rosenblatt, D. S. and A. B. Cooper. 1990. Inherited disorders of vitamin B12 utilization.
Bioessays 12:331-334.
Rosenblatt, D. S. and W. A. Fenton. 1999. Inborn errors of cobalamin metabolism, p.
367-384. In R. Banerjee (ed.), Chemistry and Biochemistry of B12. John Wiley &
Sons, Inc., New York.

135
Rossi, M., J. P. Glusker, L. Randaccio, M. F. Summers, P. J. Toscano and L. G.
Marzilli. 1985. The Structure of a B12 coenzyme: Methylcobalamin Studies by Xray
and NMR Methods. J. Am. Chem. Soc. 107:1729-1738.
Roth, J. R., T. A. Bobik and J. G. Lawrence. 1996. Cobalamin (coenzyme B12):
synthesis and biological significance. Annual Rev. Microbiol. 50:137-181.
Sambrook, J., E. F. Fritsch and T. Maniatis. 1989. Molecular cloning : a laboratory
manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.
Saridakis, V., A. Yakunin, X. Xu, P. Anandakumar, M. Pennycooke, J. Gu, F.
Cheung, J. M. Lew, R. Sanishvili, A. Joachimiak, C. H. Arrowsmith, D.
Christendat and A. M. Edwards. 2004. The structural basis for methylmalonic
aciduria. The crystal structure of archaeal ATP:cobalamin adenosyltransferase. J.
Biol. Chem. 279:23646-53.
Schmieger, H. 1971. A method for detection of phage mutants with altered transducing
ability. Mol. Gen. Genet. 110:378-381.
Schneider, Z. 1987. The occurence and distribution of corrinoids, p. 157-223. In Z.
Schneider and A. Stroinski (ed.), Comprehensive B12. Walter de Gruyter, Berlin.
Schneider, Z. 1987. Purification and Estimation of Vitamin B12, p. 111-155. In Z.
Schneider and A. Stroinski (ed.), Comprehensive B12. Walter de Gruyter, Berlin.
Schneider, Z. and A. Stroinski (eds.). 1987. Comprehensive B12. Walter de Gruyter &
Co., Berlin.
Schrauzer, G. N. and E. Deutsch. 1969. Reactions of cobalt(I) supernucleophiles. The
alkylation of vitamin B12s cobaloximes(I), and related compounds. J. Am. Chem.
Soc. 91:3341-50.
Schrauzer, G. N., E. Deutsch and R. J. Windgassen. 1968. The nucleophilicity of
vitamin B12. J. Am. Chem. Soc. 90:2441-2.
Scott, A. I. 2003. Discovering nature's diverse pathways to vitamin B12: a 35-year
odyssey. J. Org. Chem. 68:2529-39.
Seetharam, B. 1999. Receptor-mediated endocytosis of cobalamin (vitamin B12). Annu.
Rev. Nutr. 19:173-95.
Segel, I. H. 1976. Biochemical calculations : how to solve mathematical problems in
general biochemistry, 2 nd ed. Wiley, New York
Shibata, N., J. Masuda, T. Tobimatsu, T. Toraya, K. Suto, Y. Morimoto and N.
Yasuoka. 1999. A new mode of B12 binding and the direct participation of a
potassium ion in enzyme catalysis: X-ray structure of diol dehydratase. Structure
Fold Des. 7:997-1008.

136
Shively, J. M. and R. S. English. 1991. The carboxysome, a prokaryotic organelle: a
mini-review. Can. J. Bot. 69:957-962.
Shively, J. M., R. S. English, S. H. Baker and G. C. Cannon. 2001. Carbon cycling:
the prokaryotic contribution. Curr. Opin. Microbiol. 4:301-306.
Shively, J. M., G. van Keulen and W. G. Meijer. 1998. Something from almost
nothing: carbon dioxide fixation in chemoautotrophs. Annu. Rev. Microbiol.
52:191-230.
Small, I., N. Peeters, F. Legeai and C. Lurin. 2004. Predotar: A tool for rapidly
screening proteomes for N-terminal targeting sequences. Proteomics 4:1581-90.
Smith, E. L. and L. F. J. Parker. 1948. Purification of anti-pernicious anaemia factor.
Biochem. J. 43:R8-R9.
Stabler, S. P. 1999. B12 and Nutrition, p. 343-365. In R. Banerjee (ed.), Chemistry and
biochemistry of B12. John Wiley & Sons, Inc., New York.
Stojiljkovic, I., A. J. Baumler and F. Heffron. 1995. Ethanolamine Utilization in
Salmonella Typhimurium - Nucleotide-Sequence, Protein Expression, and
Mutational Analysis of the Ccha Cchb Eute Eutj Eutg Euth Gene-Cluster. J.
Bacteriol. 177:1357-1366.
Suh, S. J. and J. C. Escalante-Semerena. 1993. Cloning, sequencing and
overexpression of cobA, which encodes ATP:corrinoid adenosyltransferase in
Salmonella typhimurium. Gene 129:93-97.
Suh, S.-J. and J. C. Escalante-Semerena. 1995. Purification and initial characterization
of the ATP:corrinoid adenosyltransferase encoded by the cobA gene of Salmonella
typhimurium. J. Bacteriol. 177:921-925.
Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin and D. G. Higgins. 1997.
The ClustalX windows interface: flexible strategies for multiple sequence
alignment aided by quality analysis tools. Nucleic Acids Res. 24:4876-4882.
Toraya, T., S. Honda and S. Fukui. 1979. Fermentation of 1,2-propanediol and 1,2ethanediol
by some genera of Enterobacteriaceae, involving coenzyme B12dependent
diol dehydratase. J. Bacteriol. 139:39-47.
Toraya, T. and K. Mori. 1999. A reactivating factor for coenzyme B12-dependent diol
dehydratase. J. Biol. Chem. 274:3372-7.
Vitols, E., G. A. Walker and R. M. Huennekens. 1966. Enzymatic conversion of
vitamin B 12s to a cobamide coenzyme, α-(5,6dimethylbenzimidazolyl)deoxyadenosylcobamide
(adenosyl-B12). J. Biol. Chem.
241:1455-1461.

137
Vlasie, M., S. Chowdhury and R. Banerjee. 2002. Importance of the histidine ligand to
coenzyme B12 in the reaction catalyzed by methylmalonyl-CoA mutase. J. Biol.
Chem. 277:18523-7.
Vogel, H. J. and D. M. Bonner. 1956. Acetylornithase of Escherichia coli: partial
purification and some properties. J. Biol. Chem. 218:97-106.
Walker, G. A., S. Murphy and F. M. Huennekens. 1969. Enzymatic conversion of
vitamin B12a to adenosyl B12: evidence for the existence of two separate reducing
systems. Arch. Biochem. Biophys. 134:95-102.
Walter, D., M. Ailion and J. Roth. 1997. Genetic characterization of the pdu operon:
use of 1,2-propanediol in Salmonella typhimurium. J. Bacteriol. 179:1013-22.
Warren, J. W., J. R. Walker, J. R. Roth and E. Altman. 2000. Construction and
characterization of a highly regulable expression vector, pLAC11, and its
multipurpose derivatives, pLAC22 and pLAC33. Plasmid 44:138-51.
Watanabe, F., Y. Nakano, S. Maruno, N. Tachikake, Y. Tamura and S. Kitaoka.
1989. NADH- and NADPH-linked aquacobalamin reductases occur in both
mitochondrial and microsomal membranes of rat liver. Biochem. Biophys. Res.
Commun. 165:675-9.
Watanabe, F., Y. Oki, Y. Nakano and S. Kitaoka. 1987. Purification and
characterization of aquacobalamin reductase (NADPH) from Euglena gracilis. J.
Biol. Chem. 262:11514-8.
Watanabe, F., H. Saido, R. Yamaji, K. Miyatake, Y. Isegawa, A. Ito, T. Yubisui, D.
S. Rosenblatt and Y. Nakano. 1996. Mitochondrial NADH- or NADPH-linked
aquacobalamin reductase activity is low in human skin fibroblasts with defects in
synthesis of cobalamin coenzymes. J. Nutr. 126:2947-2951.
Watkins, D., N. Matiaszuk and D. S. Rosenblatt. 2000. Complementation studies in
the cblA class of inborn error of cobalamin metabolism: evidence for interallelic
complementation and for a new complementation class (cblH). J. Med. Genet.
37:510-3.
Watkins, D. and D. S. Rosenblatt. 1988. Genetic heterogeneity among patients with
methylcobalamin deficiency. Definition of two complementation groups, cblE and
cblG. J. Clin. Invest. 81:1690-4.
Watkins, D. and D. S. Rosenblatt. 1989. Functional methionine synthase deficiency
(cblE and cblG): clinical and biochemical heterogeneity. Am. J. Med. Genet.
34:427-34.
Watkins, D. and D. S. Rosenblatt. 2001. Cobalamin and inborn errors of cobalamin
absorption and metabolism. The Endocrinologist 11:98-104.

138
Watkins, D., M. Ru, H. Y. Hwang, C. D. Kim, A. Murray, N. S. Philip, W. Kim, H.
Legakis, T. Wai, J. F. Hilton, B. Ge, C. Dore, A. Hosack, A. Wilson, R. A.
Gravel, B. Shane, T. J. Hudson and D. S. Rosenblatt. 2002.
Hyperhomocysteinemia due to methionine synthase deficiency, cblG: structure of
the MTR gene, genotype diversity, and recognition of a common mutation, P1173L.
Am. J. Hum. Genet. 71:143-53.
Weissbach, H., A. Peterkofsky, B. G. Redfield and H. Dickerman. 1963. Studies on
the Terminal Reaction in the Biosynthesis of Methionine. J. Biol. Chem. 238:3318-
24.
Weissbach, H., B. Redfield and A. Peterkofsky. 1961. Conversion of vitamin B12 to
coenzyme B12 in cell-free extracts of Clostridium tetanomorphum. J. Biol. Chem.
236:PC40-2.
Weissbach, H., B. G. Redfield and A. Peterkofsky. 1962. Biosynthesis of the B12
coenzyme: requirements for release of cyanide and spectral changes. J. Biol. Chem.
237:3217-22.
Wilson, A., D. Leclerc, D. S. Rosenblatt and R. A. Gravel. 1999. Molecular basis for
methionine synthase reductase deficiency in patients belonging to the cblE
complementation group of disorders in folate/cobalamin metabolism. Hum. Mol.
Genet. 8:2009-16.
Zass, R., D. Leupold, M. A. Fernandez and U. Wendel. 1995. Evaluation of prenatal
treatment in newborns with cobalamin-responsive methylmalonic acidaemia. J. Inh.
Met. Dis. 18:100-101.

BIOGRAPHICAL SKETCH
Nicole Aurora Leal was born to Raul and Adrianne Leal, on August 2, 1976, in
Santo Domingo, Dominican Republic. Nicole and her family moved to New York City
during the summer of 1979; and 2 years later, to South Florida, where she learned
English while attending Pinecrest Elementary School. She attended Plantation Key
Middle School, and Coral Shores High School, in the Florida Keys. It was during middle
school that Nicole first starting taking an active interest in music and science. During her
final years of middle school, the tragic loss of her brother Ryan to cancer awakened a
desire to study the life sciences. In high school, she continued to pursue both music and
science. During this time, Nicole was a member of the honors choir, competed in
regional competitions, and was selected to represent her school at choral camp at the
University of Miami. She was a key member of the Coral Shores High School Odyssey
of the Mind team, where she competed at both the regional and state levels winning the
Ranatra Fusca award for creativity in consecutive years. Throughout High School,
Nicole volunteered extensively with Youthwish, a nonprofit organization helping
children displaced by Hurricane Andrew; and Habitat for Humanity. She earned a degree
in microbiology and cell science (with a minor in chemistry) at the University of Florida,
in Gainesville, in the Spring of 1999. During this time, she conducted undergraduate
research in the laboratory of Dr. Thomas Bobik. In the Summer of 1999, she continued
working in Dr. Bobik’s laboratory, studying B12-dependent 1,2-propanediol degradation
in Salmonella. She was accepted into Graduate School in the Fall of 1999, and continued
139

140
her work on B12-dependent propanediol degradation, with the successful identification
and partial characterization one of the metabolic enzymes involved in this pathway. In
the summer of 2001, she switched her focus of study to B12 metabolism in humans, with
specific interest in identifying the enzymes involved in AdoCbl formation, which is the
basis of this dissertation. She plans to continue her academic study and intellectual
development as a postdoctoral fellow.