.m». V ..n «firflfl‘

 

2.110! ,1.

.1 . 1;

‘ “ fi.....§a.§.¥.

 

THESKs

IllllfillllllillllUlHill!lUIllilllllHINIIU’JIJIUHIUHII

2600 31293 02048 9138

LIBRARY
Michigan State
University

 

 

 

 

PLACE iN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requeSted.

 

DATE DUE

DATE DUE

DATE DUE

 

JUN 2 8 2004

 

 

 

 

 

 

 

 

 

 

 

 

 

11/00 c/CIRCIDatoDUOpBS-p. 14

 

HOMOLOGY MODELING OF THE ACCESSORY SUBUNIT OF
DROSOPHILA MITOCHONDRIAL DNA POLYMERASE AND
SUBUNIT INTERACTION STUDIES
By

Li Fan

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Biochemistry

2000

ABSTRACT
HOMOLOGY MODELING OF THE ACCESSORY SUBUNIT OF
DROSOPHILA MITOCHONDRIA DNA POLYMERASE AND
SUBUNIT INTERACTION STUDIES
By

Li Fan

Mitochondrial DNA (mtDNA) replication is essential for the
maintenance of the mitochondrial genome and the functions of
mitochondria. The key enzyme in mtDNA replication is mitochondrial
DNA polymerase or pol y. The native enzyme isolated from embryos of
the fruit fly Drosophila is a heterodimer of a catalytic subunit (125 kDa)
and an accessory subunit (35 kDa). The catalytic subunit contains DNA
polymerase and 3’-5’ exonuclease activities. Little is known about the
function or properties of the accessory subunit, although available data
suggest that the accessory subunit is important for the catalytic and
structural integrity of pol y.

A computational modeling method was used to predict the structure
of the C-terminus (residues 254-361) of the accessory subunit based on
sequence homology with the anticodon binding domain of class IIa
aminoacyl-tRNA synthetases. The property Of binding to the stem-loop of
cognate tRNA molecules in the latter implies that the accessory subunit may

have the potential of binding to the unusual RNA primers at mtDNA

replication initiation sites and serving a role as a primer recognition factor.
This structural model also shows similarity to the structure of E. coli
thioredoxin, the processivity factor of T7 DNA polymerase, suggesting that
the accessory subunit may be the processivity factor of pol y as well.
Physical interactions between the two subunits of Drosophila pol V
have been investigated by both in vivo reconstitution and in vitro protein
overlay assays. Results with various deletion mutants suggest that the
accessory subunit consists of three domains, termed the N-, M-, and C-
domains. Both the M- and C-domains appear to comprise the main contacts
involved in subunit interaction, possibly through multiple sites with the
exonuclease (exo) domain and part of the spacer between the exo and
polymerase (pol) domains in the catalytic subunit. Furthermore, the N-
domain may modulate the subunit assembly through weak interaction with
the pol domain. Sequence comparison suggests that the M-domain of the
accessory subunit shares significant similarity with the RNase H domain of
HIV-1 reverse transcriptase (RT). Based on these results and previous
Observations that pol Y is capable of synthesizing DNA using an RNA
template as is a property of RT, we propose that the overall conformation
and arrangement of functional domains in the heterodimeric pol 7 complex

resemble those of HIV-RT.

(Z0 my family

iv

ACKNOWLEDGMENTS

I would like to thank all the people who helped me in these past few
years. Special thanks to Dr. Laurie Kaguni, my adviser and mentor, who
has bestowed on me her characteristics of perfectionism. None of this work
could have been accomplished without her support and encouragement. I
would also like to thank other members in my thesis committee: Drs.
Michael Garavito, Leslie Kuhn, Alex Raikhel, and Steven Triezenberg who
provided many thought-provoking discussions and were always a source of
useful suggestions as well as encouragement. Special thanks to Dr. Kuhn
for her guidance on the computational modeling project. I would also like
to thank Dr. J on Kaguni for many things including his provocative
comments and suggestions during our regular joint group meetings, for use
of his reagents and instruments. I thank Dr. John Wang and Dr. Suzanne
Thiem for their help with cell culture and baculovirus—related work, and
Dr. K. Padmanabhan for general computer assistance.

It was a real pleasure to have the chance to work together with
fellow graduate student Paul Sanschagrin from Dr. Kuhn’s lab on the
computational modeling. Paul is a true team player.

I would also like to thank all of the members in both Kaguni labs for
their help and companionship, and in particular, Carol Farr, Kevin Carr,

and former Ph. D. student Yuxun Wang.

I acknowledge the Department of Biochemistry, the Graduate
Program Director Dr. Pamela Fraker, and Graduate Programs secretary
Julie Oesterle for their support and help.

Finally, I would like to thank my family for their love, support and

understanding.

vi

 

TABLE OF CONTENTS

LIST OF TABLES ........................................................................... x
LIST OF FIGURES ......................................................................... xi
LIST OF ABBREVIATIONS ........................................................ xiii
CHAPTER I
INTRODUCTION ............................................................................ l
Mitochondrial DNA ................................................................ 4
Mitochondrial DNA replication ........................................... 10
I. General rules for DNA replication ............................ 10
11. Overview of mtDNA replication ............................... 12
III. Specialized mechanisms for initiation Of
mtDNA replication .................................................. 13
Drosophila as an animal model for
mtDNA replication studies ................................................... 18
Mitochondrial DNA polymerase and
other DNA polymerases ....................................................... 20
I. Mitochondrial DNA polymerase ............................... 24
II. E coli pol I and pol III ............................................. 27
III. Bacteriophage T7 DNA polymerase .......................... 31
IV. HSV DNA polymerase ............................................. 34
V. HIV reverse transcriptase ......................................... 36
VI. Nuclear DNA polymerases ....................................... 37
Overview ............................................................................... 41
CHAPTER II

HOMOLOGY MODELING OF THE CARBOXYL-TERMINAL
DOMAIN OF THE ACCESSORY SUBUNIT OF

DROSOPHILA MITOCHONDRIAL DNA POLYMERASE .......... 43
INTRODUCTION .................................................................. 44
MATERIALS AND METHODS ............................................ 48

vii

Sequence Analysis of the Accessory
Subunit of Drosophila pol y ............................................... 48
Structural Modeling of the Accessory Subunit ..................... 49

RESULTS AND DISCUSSION ............................................. 52

Identification of Sequence Homologs of

the Accessog Subunit of Pol y .......................................... 52
Modeling the Accessory Subunit of P01 1

Based on Structural Homology with

tRN A Synthetases ............................................................. 54
Biochemical Implications Of the Structural Model ................ 67
The accessory subunit of pol y as a
specialized primer recognition factor ........................ 67
The accessory subunit of pol y as a
processivity clamp ................................................... 70

CHAPTER III
SUBUNIT INTERACTIONS IN DROSOPHILA

MITOCHONDRIAL DNA POLYMERASE ................................... 76
INTRODUCTION .................................................................. 77
MATERIALS AND METHODS ............................................ 80

Materials ....................................................................... 80
Enzymes and Proteins .............................................. 80
Nucleotides and Nucleic Acids .................................. 80
Bacterial Strains ...................................................... 81
Insect Cells and Tissue Culture Medium .................... 81
Chemicals ............................................................... 81

Methods ........................................................................ 82
Construction of plasmids for expression
of pol y—fl mutants in E. coli: .................................... 82
Bacterial overexpression and
preparation of cell lysates: ....................................... 83
Purification of His-Tag fusion proteins
with Ni-NTA affinity chromatography: ..................... 84
Protein overlay assay: .............................................. 84
Construction of Recombinant Baculoviruses: .............. 85

Phosphocellulose chromatography for
monitoring the assembly of pol y—Ot and

pol y—fi mutants in SF9 cells: .................................... 90
Gel filtration chromatography: ................................. 92

viii

RESULTS .............................................................................. 93
The accessory subunit interacts through

multiple sites with the catalytic subunit ............................... 93
The accessory subunit of Drosophila pol 7 possibly
consists of three domains with different functions ................ 99

The interactions between residues 1-490 Of

the catalytic subunit and the C- and M-domains of

the accessory subunit are the main force of

subunit interactions in Drosophila pol y ............................ 105
The N-domain of the accessory subunit may modulate

subunit interactions through weak interaction with

the pol domain of the catalytic subunit .............................. 111
DISCUSSION ....................................................................... 117
CHAPTER IV
FUTURE RESEARCH .................................................................. 125
ACKNOWLEDGMENT .......................................................... 133
BIBLIOGRAPHY .......................................................................... 134

ix

Table 2.1

Table 2.2

Table 3.1

Table 3.2

Table 3.3

LIST OF TABLES

BLAST SEARCH RESULTS ............................................. 53
SUMMARY OF FOLD RECOGNITION RESULTS ............ 57

PCR PRIMERS FOR CONSTRUCTION OF
POL ’y—a MUTATIONS BY QUICKCHAN GE
MUTAGENESIS .............................................................. 88

PCR PRIMERS FOR CONSTRUCTION OF
POL Y—B MUTATIONS BY QUICKCHANGE
MUTAGENESIS .............................................................. 89

FOLD RECOGNITION RESULTS OF THE M-
DOMAIN OF POL y—B .................................................. 103

LIST OF FIGURES

Figure 1.1 Human mtDNA map .................................................. 7
Figure 1.2 Replication model for mammalian mtDNA ........... 14
Figure 1.3 Drosophila mtDNA map ......................................... 19

Figure 1.4 Topology and overall fold of
T7 DNA polymerase ...................................................................... 33

Figure 2.1 Structure-based sequence alignment
between the accessory subunit of Drosophila pol 'y and
Thermus thermophilus glycyl-tRNA synthetase .......................... 55

Figure 2.2 Sequence alignment between the accessory
subunit of Drosophila pol 7 (fly) and human pol 7 (man) .......... 61

Figure 2.3 Sequence comparison of the accessory
subunit (pol y—B ) of Drosophila pol y with class Ila
aminoacyl-tRNA synthetases ......................................................... 62

Figure 2.4 Homology modeling of the C-terminal
domain of the accessory subunit of Drosophila pol y ................ 64

Figure 2.5 Comparison of the structures of the
accessory subunit of pol y, the N-terminal domain
of 5' subunit of E. coli DNA polymerase III, and
E. coli thioredoxin. ....................................................................... 71

Figure 3.1 Schematic diagram of QuickChange
mutagenesis .................................................................................... 86

Figure 3.2 Interaction between pol ‘Y—OL and
pol y—B(l-318) in Sf9 cells ........................................................... 94

Figure 3.3 Interactions of pol y—ot with
pol y—B mutants in Sf9 cells .......................................................... 96

Figure 3.4 Summary of interactions between
pol y—oc and pol y—B mutants in Sf9 cells ..................................... 98

xi

Figure 3.5 Proposed structural domains of
Drosophila pol y—B ....................................................................... 101

Figure 3.6 Sequence homology between the M-domain

of the accessory subunit (pol 7—13) of Drosophila

mitochondrial DNA polymerase and the RNase H

domain (RT-RH) of HIV reverse transcriptase .......................... 104

Figure 3.7 Interactions of various pol y—or mutants
with pol 7—11 in Sf9 cells .............................................................. 106

Figure 3.8 Domain Interactions between Drosophila
pol y—or and pol y—B coexpressed in Sf9 cells ............................. 109

Figure 3.9 In vitro interactions of pol 'y—oc(9-1145)
with pol y—B mutants by protein overlay assay .......................... 112

Figure 3.10 In vitro interactions of pol y—or mutants
with pol 7-13 and mutants ............................................................. 115

Figure 3.11 Proposed complex for Drosophila pol y
at initiation site of mtDNA replication ...................................... 122

Figure 4.1 The accessory subunit of Drosophila pol y
binds to RNA primer in the presence of the catalytic
subunit. ......................................................................................... 129

Figure 4.2 The accessory subunit binds to ssDNA
independent of the catalytic subunit ........................................... 132

xii

aaRS

A+T

ATP

bp(s)
C-terminus
C-domain
D-loop

D. melanogaster
DNA
dNTP
DTT

E. coli
Gly-RS
HIV

HSV

kbp

kDa
M-domain

mRNA

LIST OF ABBREVIATIONS

aminoacyl-tRN A synthetase
deoxyadenylate and thymidylate
adenosine triphosphate

base pair(s)

carboxyl-terminus

residues 254-362 of pol y—B
displacement loop

Drosophila melanogaster
deoxyribonucleic acid
deoxynucleoside triphosphate
dithiothreitol

Escherichia coli

glycyl-tRNA synthetase

human immunodeficiency virus
herpes simplex virus

kilobase pairs

kiloDalton

residues 128-240 of pol 'y—B

messenger RNA

xiii

mtDNA
mtSSB
N-terminus
N-domain
Ni-NTA
nt

PCNA
PCR

p01 7

pol 'y—Ot
p01 7-13
Pro—RS

RF-C
RT
SSB
ssDNA
SV40
tRN A
UV

X. laevis

mitochondrial DNA

mitochondrial single-stranded-DNA binding protein
amino terminus

residues 1—120 Of pol y—B

Nickel chelated nitrilo-tri-acetic acid
nucleotide

proliferating cell nuclear antigen
polymerase chain reaction
mitochondrial DNA polymerase
catalytic subunit of pol y

accessory subunit of pol 'y
prolyl-tRNA synthetase

ribonucleic acid

replication factor C

reverse transcriptase

single-stranded DNA-binding protein
single-stranded DNA

simian virus 40

transfer RNA

ultraviolet light

Xenopus laevis

xiv

CHAPTER I

INTRODUCTION

Mitochondria are the central organelles of cellular energy
metabolism. They are intimately involved with most of the major metabolic
pathways used by a cell to build, break down, and recycle its molecular
building blocks such as purines and pyrimidines for RNA and DNA
synthesis. One principal mitochondrial function is to produce ATP
(adenosine triphosphate) through the process of oxidative phosphorylation
(OXPHOS) (Wallace, 1992). Most human cells contain hundreds of
mitochondria. These cytoplasmic organelles are believed to have evolved
about 1.5 billion years ago from eubacteria-like endosymbionts in a
primitive eukaryotic cell as the host (Gray, 1989). Modern mitochondria
still retain some features that reflect their endosymbiotic origin. These
include a double-membrane structure and a circular mitochondrial DNA
genome with mitochondrial-specific replication, transcription, translation,
and protein assembly systems (for reviews see (Attardi, 1985; Gray, 1989;
Clayton, 1991)). Mitochondria are the only cellular organelles to have their
own DNA other than the nuclei in animal cells. Mitochondrial DNA
(mtDNA) encodes proteins critical for oxidative phosphorylation (Wallace,
1992). Therefore, the proper maintenance and faithful replication of
mtDNA molecules are very important for mitochondria to achieve their
biological functions.

Mitochondrial defects were first noticed as rare childhood disorders

(for review see (Luft, 1994)). It is now clear that mitochondrial diseases

are manifested as an extraordinarily broad spectrum of clinical problems,
commonly involving the most energy-dependent organs such as brain,
heart, skeletal muscle, kidney, endocrine glands and bone marrow
(Wallace, 1994). Many common diseases such as diabetes and ischemic
heart disease have a defective mitochondrial basis (Wallace, 1999). Aging-
related diseases such as Parkinson’s disease and Alzheimer’s disease may
result in part from mitochondrial dysfunction. Mitochondrial medicine has
achieved great progress in the past years (Luft, 1994). A breakthrough in
mitochondrial pathophysiology occurred in 1988 when mutations in
mtDNA were reported to be associated with a form of young-adult
blindness, Leber’s hereditary Optic neuropathy (LHON) (Wallace et al.,
1988), and progressive muscle disorders (Holt et al., 1988). Since then,
more than 50 pathogenic mtDNA base substitution mutations and hundreds
of mtDN A rearrangement mutations (deletions and insertions) have been
identified in a variety of degenerative diseases (MITOMAP:
http://www.gen.emory.edu/mitomap.html ). Some mitochondrial diseases
involving mtDNA mutations may be inherited through the maternal side of
the family as almost all mitochondria come from the mother (Wallace,
1995). In addition, mitochondrial toxicity caused by long term treatment
with antiviral nucleoside analogue (ANA) drugs was found to resemble
those of genetic mitochondrial diseases. These ANAs include zidovudine

(AZT), zalcitabine (ddC), didanosine (ddI), stavudine (D4T), 2’-deoxy-3’-

thiacytidine (3TC), and 1—(2’-deoxy-2’fluoro—l-B-D-arabinofuranosyl 5-
iodo)uracil (FIAU) for treatment of human immunodeficiency virus type-1
(HIV-1) and hepatitis B virus (for review see (Lewis and Dalakas, 1995)).
The mitochondrial toxicity was at least partially due to their inhibition of
mitochondrial DNA polymerase, a key enzyme for mtDNA replication
(Chen and Cheng, 1989; Chen et al., 1991). These results gave evidence for
direct connections between mtDNA replication defects and mitochondrial
dysfunction. Therefore, studies on mtDN A replication and related enzymes

and proteins have major medical relevance.

Mitochondrial DNA

Mitochondrial DNA molecules represent less than 1% Of the total
cellular DNA in a typical animal cell, but the copy number of mtDNA
molecules is in great excess over the nuclear chromosomal DNA molecules
(Clayton, 1982). Each mitochondrion has several mtDN A molecules.
Mitochondrial DNA is located in the matrix which is encircled by a double-
membrane structure. The inner membrane is folded into cristae toward the
matrix. The outer membrane separates the organelle from the cytosol.
There is an intermembrane space between the outer and inner membranes.
There are about 3000 proteins in each mitochondrion, virtually all of them
are encoded by nuclear genes (Lill et al., 1996). These proteins are

synthesized on cytosolic ribosomes and have to be imported into the

mitochondrial subcompartments (Schatz, 1996). Many of these proteins
have a mitochondrial target sequence at their N-terminus and the target
sequences are usually removed after the proteins are delivered to the
correct locations (Lill et al., 1996). Proteins encoded by mtDNA are
expressed inside the mitochondrion which has its own machinery for gene
expression and DNA replication (Clayton, 1982; Clayton, 1992).

Many mitochondrial DNA molecules from animals, plants, fungi and
protozoa have been well characterized over the years (Attardi and Schatz,
1988; Gray, 1989; Leblanc et al., 1997). Sequences for more than 70
complete mitochondrial DNA genomes are now available from the
GenBank database. Mitochondrial DNA has the same basic role in all
eukaryotes: it encodes rRNA and tRN A components of a mitochondrial
protein synthesizing machinery (Attardi and Schatz, 1988; Gray, 1989).
However, proteins for gene expression and mtDNA replication are encoded
by nuclear genes. The mtDNA genome also contains genes encoding
proteins involved in the electron transport chain for oxidative
phosphorylation. Although conservative in basic genetic function, mtDNA
differs radically in structure among and even within the four traditional
eukaryotic kingdoms (Animalia, Plantae, Fungi and Protista). Some
mitochondrial genomes also contain additional genes encoding proteins
associated with mitochondrial ribosomes and/or proteins involved in

mitochondrial RNA processing or intron transposition (Gray, 1989).

Mitochondrial genome size varies over more than a 150-fold range, from
14.3 kbp in the nematode worm, Ascaris suum (Wolstenholme et al.,
1987), to an estimated high of 2400 kbp in the cucurbit, Cucumis melo
(muskmelon) (Gray, 1989). Most mtDNAs are circular but linear mtDNAs
have been found in some species (Nosek et al., 1998). Nevertheless, with a
few notable exceptions, the overall genetic content and gene order in
mtDNA are remarkably conserved among animal species (Wolstenholme,
1992).

Animal mitochondrial genomes range in size from 16 to 19 kbp and
can be isolated as closed circular double-stranded DNA molecules (Clayton,
1991). Because the mtDNA is supercoiled, its replication and transcription
can be inhibited by ethidium bromide. The two strands of vertebrate
mtDNA can be separated in denaturing cesium chloride gradients, due to a
strand bias in guanine and thymidine (G+T) content, and are thus defined
as the heavy strand (H-strand) and the light strand (L-strand), respectively.
A major form observed for vertebrate mtDNA is the displacement-loop
(D-loop) molecule, which contains a short piece of nascent H strand at the
H-strand origin (OH) of DNA replication (Clayton, 1982).

The human mitochondrial genome is about 16.5 kbp which encodes
13 essential polypeptides of oxidative phosphorylation plus two rRNAs and
twenty-two tRNAs necessary for gene expression within the mitochondrion

(Wallace, 1992) (Figure 1.1). The proteins encoded are seven subunits of

Figure 1.1 Human mtDNA map. The mtDNA comprises 16,569
bps with numbering start at OH and proceeding counterclockwise around
the circular map. The function of each gene is identified by shading
according to the designations within the circle. The tRNA genes are
indicated by the letter of their cognate amino acid. Transcription directions
at the promoters (PH and PL) as well as the directions at the two DNA
replication initiation sites (OH and OL) are indicated by arrows.
Mitochondrial disease-related mutations including one 5 kb deletion and
several point mutations are presented. Point mutations are named by the
associated diseases followed by DNA sequence number and the mutated
amino acid residue. Diseases abbreviations are: LHON, Leber’s hereditary
Optic Neuropath; NARP, Neurogenic muscle weakness, Ataxia, and
Retinitis Pigmentosa; MERRF, Myoclonic Epilepsy Ragged-Red Fiber
Disease; MELAS, Mitochondrial Encephalomyopathy, Lactic Acidosis, and
Strokelike epsodes; LDYT, LHON and dystonia; DEAF, deafness.

 

 

 

 

The Hum Mitchondr‘tal DNA

 

cytochrome '. .
Mum

W 3°"?me mammama

   

 

 

 

NADH dehydrogenase (NDl, ND2, ND3, ND4, ND4L, NDS, and ND6),
the apocytochrome b component (Cyt b) of ubiquinol cytochrome-c
reductase, three subunits of cytochrome—c oxidase (COI, C011, and COIII),
and two subunits of ATP synthase (ATPase 6 and 8). Gene Cyt b and C01
are common to all the mitochondrial genomes whose coding capacity has
been defined completely (Okimoto and Wolstenholme, 1990). The H-strand
functions as the template for the small (128) and large (16S) rRNAs, 12 of
the polypeptides, and 14 of the tRNAs. The L-strand is the template for one
protein and 8 tRNAs (for details see Figure 1.1). The D-loop region
contains two transcription promoters, one for the H-strand (PH) and the
other for the L-strand (PL) in the opposite direction. MtDN A is transcribed
symmetrically from these two promoters (for reviews see (Clayton, 1984;
Clayton, 1992)). The two transcripts encompass the entire genome
(Montoya et al., 1981). As the tRNAs are transcribed, they fold into their
three-dimensional structures and are cleaved out of the transcript by a
processing enzyme comparable to RNase P. Because most protein coding
genes and the two rRN A genes are flanked by tRNA genes, cleavage of the
tRN As also releases the rRNAs and the mRN A molecules from the
transcript (Ojala et al., 1981). Most mRNAs lack 5’ and 3’ nontranslated
sequences and polyA tails are added to the mRN As posttranscriptionally
(Clayton, 1984). There is a weak bi-directional transcriptional terminator

(nps 3237-3249) in the tRNA-leu(UUR) gene downstream of the 16S rRNA

gene. Hence, an excess of rRNA transcripts from H—strand transcription is
produced over mRNAs and most L-strand transcripts terminate before the
rRNA genes are transcribed (Clayton, 1992). Mitochondrial protein
synthesis, like bacterial protein synthesis, is initiated with
formylmethionine, and is sensitive to the bacterial ribosome inhibitor
chloramphenicol. However, mitochondria use a different genetic code from
the universal genetic code: the Opal stop codon (UGA) and UGG are used
for tryptophan while the arginine codons AGA and AGG are used for stop
codons. This special mtDN A genetic code limits the expression of mtDNA
to within the mitochondrion because most mitochondrial mRNAs contain
multiple UGA codons and can not be translated in the cytoplasmic

compartment.

Mitochondrial DNA replication
1. General rules for DNA replication

DNA replication has been well studied in many systems including
both prokaryotes and eukaryotes (Kornberg and Baker, 1992). Various
DNA replication mechanisms have some basic common features.

DNA replication is always a semiconservative process. This means
that two progeny molecules are produced from one parental DNA molecule
and each progeny molecule has one strand derived from the parental DNA

duplex and one novel strand with complementary nucleotide sequence.

10

DNA replication usually starts at specific sites within the genome,
called the origins of DNA replication. Initiation of DNA replication is
usually the major control point for DNA replication just as the initiation
from the promoter in transcription. In bacteria, a single origin is used for
initiation of DNA replication. Reinitiation at the same origin can occur
before the first round of replication is complete for rapid growth. In
eukaryotic chromosomes, replication begins at many origins located at
different positions along the chromosomal DNA. The initiation of DNA
replication at each origin is controlled for once per cell cycle. Origins of
replication serve to import specificity and increase the efficiency of
initiation of DNA replication by providing loci for assembly Of
multiprotein complexes to Open locally the DNA duplex for access to the
template strands.

Replication fork movement is uni- or bi-directional. The initial
opening of the DNA duplex at the origin allows the establishment of one or
two replication forks (the growing ends of the nascent DNA strand
including the replicative enzyme complex ) loaded with replicative
proteins. If one replication fork is established, replication proceeds from
the origin in one direction (unidirectional DNA replication). If two
replication forks are produced, replication proceeds in both directions with

the two forks moving away from each other.

11

A primer is essential to start DNA synthesis by DNA polymerases.
All of the known DNA polymerases are incapable of starting chains de
novo. In most DNA replication systems, DNA replication is primed by
short RNA chains. The initiator RNA is later excised and replaced with
DNA.

DNA strands are elongated in the 5 ’-3’ direction by DNA
polymerases through the addition of nucleotide monomers.

Many enzymes and proteins other than DNA polymerases are
required for the completion of DNA synthesis. In general, these include
primases, helicases, topoisomerases, ligases, and single-stranded DNA-

binding proteins (SSB).

11. Overview Of mtDNA replication

The total number of mtDNA molecules in a particular cell may vary
depending on the type of the cell and the developmental stage of the cell,
suggesting that mtDNA replication is somehow regulated. However, it is
believed that mtDN A replication is not directly coupled to cell cycle
regulation, in contrast with the cell-cycle controlled nuclear DNA
replication process (Shadel and Clayton, 1997). MtDN A molecules can
replicate in any and all phases of the cell cycle in mouse L cells

(Bogenhagen and Clayton, 1977).

12

Our current understanding of mtDNA replication is based largely on
early electron microscopic study of mtDNA replication intermediates from
mammalian mitochondria (for reviews see (Clayton, 1982; Clayton, 1991;
Clayton, 1992)). There are two initiation sites of mtDNA replication, one
for each DNA strand. H-strand replication starts from the origin (OH) in
the D-loop region. The replication fork extends to one direction using the
L-strand as the template. The parental H-strand is displaced when the
replication fork moves. The displaced H-strand is coated by mitochondrial
single-stranded DNA-binding protein (mtSSB) for protection. The
initiation of the L-strand replication does not start until the H-strand
replication fork passes by the initiation site of L-strand (OL), which is two-
thirds of the genome away from OH. The L—strand replication fork moves
in the opposite direction back toward the D-loop using the displaced H-
strand as the template. The completion of mtDNA replication results in two
semiconserved supercoiled closed circular mtDNA molecules (Figure 1.2).
In sum, mtDNA replication is asynchronous in firing at the two initiation

sites, asymmetric at the replication fork, and unidirectional.

III. Specialized mechanisms for initiation of mtDNA replication.
Two different mechanisms of initiation are believed to occur at the
two origins in mtDNA replication (Clayton, 1982; Shadel and Clayton,

1997). The initiation of the H-strand replication starts in the D-loop region

13

 

 

 

Figure 1.2 Replication model for mammalian mtDNA. Thick
solid lines: parental heavy (H) strands. Thin solid lines: parental light (L)
strands. Thick dashed lines: progeny H strands. Thin dashed lines: progeny
L strands. The order of replication is clockwise starting at D mtDNA, the
D-loop containing molecule. C mtDNA: supercoiled mtDNA molecule. OH
and OL : initiation sites for H- and L-strand synthesis, respectively. The
double arrows reflect the equilibrium between D mtDNA and C mtDNA.
Carat: interruption of at least one phosphodiester bond in the H strand of
the or progeny molecule. [3 Gpc: gapped circular progeny molecule. E
mtDNA: complete progeny molecule

14

(Chang and Clayton, 1985; Chang et al., 1985). A specialized primer is
synthesized from the L-promoter by mitochondrial RNA polymerase.
Transcription factors mtTFA and/or mtTFB are believed to be important
for the transcription from the L-promoter (Shadel and Clayton, 1993; J ang
and Jaehning, 1991). It is not clear if the transcriptional event is specific
for DNA replication initiation or simply a normal L—strand transcription.
In either case, the transcript is processed for mtDNA replication at specific
sites in the D-loop region to produce a correct 3’-OH end for mtDNA
polymerase to extend. The transcripts form a specialized structure
involving the two separated DNA strands in the D-loop (Lee and Clayton,
1997). Folded structural features and three conserved sequence blocks
(CBS 1, II, and III) are essential for the processing of the transcripts into
primers (Lee, 1996). The mitochondrial RNA processing nuclease (MRP),
an enzyme similar to RNase P, is needed for this process (Chang and
Clayton, 1987a; 1987b).

The involvement of processed RNA transcripts in initiation of DNA
replication has been observed in other systems (Kornberg and Baker,
1992). Initiation of bacteriophage T7 DNA replication (Fuller et al., 1983)
and E. coli plasmid Col E1 replication are two other examples of systems
which utilize processed transcripts synthesized by RNA polymerase as
primers for DNA synthesis. It is quite interesting that the secondary

structure of the transcripts plays a role in the regulation of Col E1 DNA

15

replication (Kornberg and Baker, 1992). Col E1 is a 6.6-kbp, high-COpy-
number E. coli plasmid (about 20 copies per cell) containing resistance to
colicin E1. Col E1 replicates unidirectionally from a unique origin and
relies entirely on host-encoded proteins. The host RNA polymerase
synthesizes the primer for DNA replication. The transcription of the
primer (termed RNA 11) is initiated at a promoter which is 555 bp
upstream from the origin. RNA 11 spontaneously forms an RNA—DNA
hybrid with the template in the origin region. RNA-DNA hybrid formation
opens the duplex DNA, allowing the replication enzymes to be loaded. The
formation Of a specific RNA secondary structure is required for RNA-
DNA hybrid formation, and the template sequence at the origin affects the
efficiency with which this hybrid is utilized for initiation. RNase H, a
specific enzyme for cleavage of RNA in RNA-DNA hybrids, generates the
3’-OH ends in the origin that serve as primers for DNA synthesis by E. coli
DNA polymerase III (pol III). The proper RNA secondary structure is
crucial for efficient cleavage by RNase H and for primer utilization. In
addition, modulation of the RNA 11 secondary structure by a trans-acting
factor, RNA 1, regulates replication. RNA I is a 108 nt RNA molecule
transcribed in the RNA 11 sequence region from the strand opposite that
used to produce the RNA II primer. The antisense RNA inhibits primer
formation by RNA 11 by complexing with RNA 11 and blocking its folding

into the conformation required for RNA-DNA hybrid formation.

16

However, if the RNA-DNA hybrid is formed, the inhibition of RNA I is
not effective any more.

The initiation of the L-strand replication of mtDNA synthesis is quite
different from initiation at OH (Clayton, 1982). The OL sequence is
approximately 30 nucleotides in size and is nested in a cluster of five tRNA
genes. The origin itself has the potential to form a stem-loop structure once
the region is single-stranded during H-strand replication (Hixson et al.,
1986). This specialized secondary structure plays a role in the initiation of
the L-strand DNA replication. Early studies revealed that a putative human
mtDNA primase could recognize OL and initiate priming and DNA
synthesis (Wong and Clayton, 1985a; 1985b). Daughter L-strand synthesis
is primed by RNA synthesis complementary to the T—rich loop of OL. It has
been proposed that the presumptive primase produces a short RNA primer
which is then extended by mitochondrial DNA polymerase to complete the
L-strand synthesis. The transition of RNA synthesis to DNA synthesis
occurs near the base of the stem in OL. However, in chicken mtDNA
molecules, the lack of this OL region in the cluster of tRNA genes implies
that the tRNA genes themselves may play a role in the initiation of L-strand

synthesis (Desjardins and Morais, 1990).

17

Drosophila as an animal model for mtDNA replication studies
Drosophila mtDNA is very similar to the human counterpart in gene
organization and structure (Figure 1.3). There are no introns in Drosophila
mtDNA, and few intergenic nucleotides. The genes found in Drosophila
mtDNA are the same as those in human mtDNA, but there are differences
in gene arrangement and in the relative proportions of the two strands of
the molecule that serve as templates for transcription (for details see Figure
1.3). MtDNAs isolated from flies and Drosophila tissue culture cells are
supercoiled circular molecules (Rubenstein et al., 1977) but contain no
detectable D-loop forms (less than 1%) (Goddard and Wolstenholme, 1978;
Goddard and Wolstenholme, 1980) which, in contrast, is the major form of
mtDNA in vertebrates (>75%) (Robberson and Clayton, 1972). A
noncoding “A+T-rich” region, so called because it is rich (90-96%) in
adenine (A) and thymine (T), substitutes for the function Of the D-loop
region in the vertebrate mtDNA (Goddard and Wolstenholme, 1978;
Goddard and Wolstenholme, 1980). This A+T-rich region shows species-
specific size variation from 1.0 kbp in D. virilis (Goddard and
Wolstenholme, 1978) to 4.8 kbp in D. simulans (Goddard and
Wolstenholme, 1980). This region also shows great variation in nucleotide
sequence compared to the rest of mitochondrial genome. The mtDNA from

D. melanogaster is about 19.6 kbp with a 4.6 kbp A+T-rich region (Lewis

18

DROSOPHILA

MITOCHONDRIAL DNA

 

Figure 1.3 Drosophila mtDNA map. Drosophila melanogaster
mtDNA contains 19,517 bps including a control region (A+T). The genes
are named as those in human mtDNA with arrows in the coding regions
indicating the direction of transcription. The direction of mtDNA leading
strand replication is indicated by the arrow at OR in the control region.

19

et al., 1994). Early studies including electron microscopic analysis of
mtDNA replication intermediates suggested that the mechanism of
Drosophila mtDNA replication is very similar to that of human mtDN A
replication (Goddard and Wolstenholme, 1978; Goddard and
Wolstenholme, 1980). The basic features of the unidirectional, asymmetric
and asynchronous mechanism of mammalian mtDNA replication are well
conserved in Drosophila mtDNA replication. Drosophila has been a very
useful model for studies of developmental biology and other biological
studies for a long time. The well-developed genetic tools in Drosophila will
be very useful to assess in the context of the organism as a whole the results
from in vitro biochemical studies on mtDNA replication, and to study
cellular phenotypic consequences of deficiencies in mtDNA replication and

genome maintenance.

Mitochondrial DNA polymerase and other DNA polymerases

Many enzymes are believed to be involved in mtDNA replication but
only few have been identified. The key enzyme is mitochondrial DNA
polymerase, generally called pol 7. Other known enzymes involved in
mtDNA synthesis include mtRN A polymerase and primase for priming at
either OH or OL, and mtSSB which helps to protect the displaced single
strand during mtDNA synthesis. Other enzymes such as helicases,

topoisomerases, ligases are also expected to contribute to mtDNA

20

replication based on studies of nuclear DNA replication and prokaryotic

DNA replication (Kornberg and Baker, 1992).

Mitochondrial DNA polymerase is a member of a large superfamily
of DNA polymerases. DNA polymerases play an essential role in nucleic
acid metabolism, including DNA replication, repair and recombination. A
great number of DNA polymerases have been identified from various
species including eukaryotes, prokaryotes and viruses (Kornberg and
Baker, 1992). These DNA polymerases show both sequence diversity and
variety in subunit components. For example, the human DNA polymerase
[3 (pol B) is a single polypeptide of 40 kDa, while E. coli pol III
holoenzyme has at least 20 subunits consisting of 10 distinct polypeptides
with sizes varying from 9 kDa to 130 kDa. However, all DNA polymerases
share a basic common mechanism that is to add deoxynucleotides onto the
growing end of a DNA primer strand. Some general rules apply to all
DNA polymerase action (Kornberg and Baker, 1992):

0 DNA polymerase can not start DNA synthesis de novo: a primer,
usually a short RNA with a 3’-OH terminus, annealed to the template is
essential;

0 The added nucleotide is a monomer, activated as a 5’-triphosphate;

0 The phosphodiester bond is formed by the nucleophilic attack of the 3’-
OH group of the primer terminus on the a-phosphorus of the activated

dNTP;

21

0 Correct base pairing is required at the primer terminus as well as in
matching of the incoming nucleotide to the template;
0 Chain elongation proceeds in the 5’ to 3’ direction;

0 Polarity of the growing chain is opposite to that of the template.

The DNA polymerase often serves as the core of a holoenzyme
complex such as the or subunit in E. coli DNA pol 111. All other subunits
are added to this core subunit to facilitate its function. The complexity of a
particular DNA polymerase depends largely on the role Of that DNA
polymerase. In general, DNA polymerases for genomic DNA replication
(e.g. E. coli pol III) are much more complex than those for DNA repair
(e. g. human pol B). Two activities are added to replicative DNA
polymerases: one is the proofreading 3’-5’ exonuclease activity such as the
8 subunit in E. coli pol III to enhance the fidelity of genome duplication,
and the other is a processivity factor such as the B subunit of E. coli pol III
which enables the DNA polymerase to add many deoxynucleotides to the
growing DNA strand without dissociation to ensure the completion of DNA
replication within permitted time. In most cases, the 3’-5’ exonuclease
activity is contained in the same polypeptide with the polymerase activity
but in different structural domains.

The availability of DNA polymerase sequence data in the past years

allows the compilation and alignment of all the DNA polymerase protein

22

sequences. According to their similarities to the three E. coli DNA
polymerases (pol I, II, and III), Ito and Braithwaite classified the DNA-
dependent DNA polymerases into four families: A, B, C, and X (Ito and
Braithwaite, 1991; Braithwaite and Ito, 1993). Despite the sequence
diversity Of DNA polymerases, there are a small number Of crucial active-
site residues conserved for all DNA polymerases or within an individual
polymerase family (Braithwaite and Ito, 1993). Crystal structures for
several DNA polymerases including that of the Klenow fragment (01113,
1985; Beese et al., 1993a; 1993b) of E. coli pol 1, T7 DNA polymerase
(Doublie et al., 1998), Thermus aquaticus DNA polymerase (Taq DNA
polymerase) (Kim et al., 1995), Bacillus stearothermophilus DNA
polymerase (Kiefer et al., 1997), and rat DNA pol B (Sawaya et al., 1994;
Pelletier et al., 1994) also suggest that they have a similar overall structure,
exemplified by the structure of the polymerase domain of the Klenow
fragment of E. coli pol I. The pol domain of the Klenow fragment can be
divided into three subdomains, named “palm,” “fingers,” and “thumb” for
their anatomical analogy to a right hand. The structure has a cleft in the
middle to hold the template and the primer strand. The palm subdomain
contains a B-sheet that forms the base of the cleft and usually contains the
catalytic residues; the fingers subdomain forms one wall of the cleft, while

the thumb subdomain forms the other side of the cleft. This structural

23

similarity reflects very well the conserved polymerase activity of all DNA
polymerases (Steitz, 1998).

However, different DNA polymerases are distinct enzymes: different
proteins from different organisms are different; even within a particular
cell, different DNA polymerases are distinguished from each other
biochemically and structurally. The following sections will describe some
details about some well studied DNA polymerases which will provide
information on how different DNA polymerases fulfill their unique
functions in a particular system. Particular attention will be devoted to
mitochondrial DNA polymerase and to DNA polymerases that are similar

to mitochondrial DNA polymerase.

I. Mitochondrial DNA polymerase

Mitochondrial DNA polymerase, pol y, usually accounts for less than
1% of total cellular DNA polymerase activity in somatic cells (Weissbach,
1979). P01 7 is distinguished from other cellular DNA polymerases by its
sensitivity to N-ethylmaleimide and dideoxynucleoside triphosphates and its
optimal activity under high salt conditions. P01 7 has been purified from
several sources including chick embryos (Yamaguchi et al., 1980),
Drosophila melanogaster embryos (Wemette and Kaguni, 1986), porcine
liver (Mosbaugh, 1988), Xenopus laevis oocytes (Insdorf and Bogenhagen,

1989a; 1989b), and HeLa cells (Gray and Wong, 1992). The purified

24

enzyme from all these sources has a proofreading 3’-5’ exonuclease activity
(Kunkel and Soni, 1988; Kaguni and Olson, 1989; Kunkel and Mosbaugh,
1989; Insdorf and Bogenhagen, 1989b; Gray and Wong, 1992). Pol y
purified from D. melanogaster embryos is a heterodimer comprising
subunits of 125 kDa and 35 kDa. Human pol y isolated from HeLa cells
consists of two subunits of 140 kDa and 54 kDa (Gray and Wong, 1992).
P01 7 isolated from other sources indicated a native enzyme of 160-180
kDa (Yamaguchi et al., 1980; Insdorf and Bogenhagen, 1989a; Mosbaugh,
1988). These results together suggest that animal pol y is a heterodimer
complex of two subunits of 125-140 kDa and 35-54 kDa. The large subunit
of pol 7 contains both polymerase and 3'-5' exonuclease activities (Lewis et
al., 1996; Longley et al., 1998).

The gene, MIPl, for the catalytic subunit of mitochondrial DNA
polymerase was initially identified from S. cerevisae (Foury, 1989). It
encodes a protein Of 143.5 kDa. The cDNAs of the catalytic subunit from
fission yeast (Ropp and Copeland, 1995), fly (Lewis et al., 1996), and
mammals (Ropp, 1996; Lecrenier et al., 1997) have been isolated.
Consistent with the biochemical data, these catalytic subunits have
conserved motifs for 3’-5’ exonuclease and polymerase activities identified
in E. coli pol I. The cDNA of the small subunit of D. melanogaster pol y
and its mammalian homologs were isolated more recently (Wang et al.,

1997). The cDNA for the small subunit pol y of X. laevis is now also

25

available (Carrodeguas et al., 1999). However, no homologues have been
identified in the complete yeast or nematode genomes.

The polymerization and exonucleolytic mechanism of pol y from D.
melanogaster embryos has been investigated in detail (Wemette et al.,
1988; Kaguni et al., 1988; Olson and Kaguni, 1992; Williams et al., 1993;
Williams and Kaguni, 1995). Recent development of reconstitution Of
human pol y with recombinant proteins will definitely help biochemical
characterizations of human pol y (Lim, et al., 1999). The enzyme catalyzes
DNA synthesis on a variety of DNA substrates. Interestingly, pol y can also
catalyze DNA synthesis using poly(rA)'oligo(dT) as a substrate (Wemette
and Kaguni, 1986; Longley et al., 1998), a characteristic of reverse
transcriptase. Drosophila pol 7 replicates on single-stranded DNA of
natural DNA sequence with the highest substrates specificity in a
quasiprocessive manner, and with a high degree Of nucleotide insertion
fidelity. Pol y misincorporates 1 nucleotide per 100-500 thousand
nucleotides incorporated (Wemette et al., 1988), similar to the error rates
of the nuclear enzymes pol e and pol 8 in the presence of proliferating cell
nuclear antigen (PCNA) (see below). The intrinsic 3’-5’ exonuclease
activity of Drosophila pol 'y is mispair-specific, showing about 15-fold
higher efficiency of hydrolysis of 3’-terminal mispair than 3’-terminal base
pairs under nonpolymerization conditions (Olson and Kaguni, 1992).

Under DNA polymerizing conditions, pol 7 does not extend a 3’-mispair

26

even in the presence of a large excess of the next correct nucleotide (Olson
and Kaguni, 1992). Maximal activity of pol y is obtained when assayed
under moderate salt while the highest processivity is observed under low
salt conditions (Williams et al., 1993). In the presence of mtSSB, pol y
exhibits maximal DNA polymerase activity and processivity. The
enhancement Of pol 'y processivity by mtSSB is optimal at high salt
conditions (Williams and Kaguni, 1995). Mitochondrial SSB stimulates 30-
fold the rate of initiation Of DNA strands by enhancing the primer
recognition and binding by pol y in addition to its role for clearing
secondary structures of the template (Farr et al., 1999). The rate of DNA
synthesis mediated by D. melanogaster pol y is about 2 nucleotides per
second (Wemette et al., 1988), comparable to the rate of 4.5 nucleotides

per second observed in vivo (Clayton, 1982).

II. E coli pol I and pol III

As the first DNA polymerase discovered, E. coli DNA polymerase I
is the most abundant DNA polymerase in bacterial cells and serves roles in
DNA replication, recombination, and repair (Kornberg and Baker, 1992).
It is the most well studied DNA polymerase. The achievements of
biochemical characterization and structural determination of E. coli pol I
have provided tremendously helpful information for studies of other DNA

polymerases. The main function of pol I is to fill short gaps and it shows

27

moderate processivity. Pol I is a single-subunit enzyme of 103 kDa. It can
be separated into two fragments by partial proteolysis. The large fragment
(67 kDa), termed the Klenow fragment or enzyme, contains 3’-5’
exonuclease and DNA polymerase activities, while the small fragment (36
kDa) contains a 5’-3’ exonuclease activity. The 5’-3’ exonuclease functions
both in replication and repair for the removal of the RNA primers that
initiate nascent DNA synthesis and the removal of damaged bases,
respectively. The overall structure of Klenow fragment determined by
crystallography (Beese et al., 1993a; 1993b) showed very clearly that the
67-kDa molecule consists of two domains: a larger carboxyl (C)-terminal
domain with a very prominent cleft containing the polymerase active site,
and a smaller globular amino (N)-terminal domain for the 3’-5’
exonuclease activity. The 3’-5’ exonuclease site is located sufficiently close
to the polymerase site to allow the 3’ terminus of the growing chain to
shuttle back and forth between them. Thus, polymerization and editing can
both proceed without release of DNA from the enzyme (Joyce and Steitz,
1994).

Pol I is the prototype of DNA polymerase family A (Ito and
Braithwaite, 1991). Sequence comparison has indicated that the important
amino acid residues for both the 3’-5’ exonuclease and polymerase active
centers are conserved from pol I to pol 'y enzymes (Lecrenier et al., 1997 ).

The established structural details of the Klenow fragment of E. coli pol I

28

are a valuable asset for structure-function study on pol 'y enzymes.
However, the spacer region separating the exo and pol domains is much
larger in pol y enzymes than that in the Klenow enzyme (Lecrenier et al.,
1997). This lengthy spacer of pol 7 may provide the enzyme with special
properties which are lacking in the Klenow enzyme.

The replicative pol III is the most complex holoenzyme among all
the DNA polymerases. Fortunately, tremendous progress has been achieved
in the past years to resolve the subunit composition and to understand in
great detail the interactions between subunits as well as the functions of
most subunits. Therefore, pol 111 provides an excellent model for studies of
other DNA polymerases. Pol III is a complex of about two copies each of
10 distinct polypeptides organized as an asymmetric dimer of about 900
kDa (McHenry, 1988; Kornberg and Baker, 1992; Kelman and O'Donnell,
1995). The core of p01 III consists of three subunits (or, 8, and 0). The 130-
kDa 0t subunit is the polymerase (Maki et al., 1985), and the 27.5-kDa 8
subunit is a 3’-5’ proofreading exonuclease (Scheuermann and Echols,
1984). The or subunit encoded by gene polC in E. coli is the prototype of
DNA polymerase family C (Ito and Braithwaite, 1991). The 0t and 8
subunits form a tight 1:1 complex, resulting in 2—fold increase in
polymerase but 50- to 100-fold increase in 3’-5’ exonuclease (Maki and
Kornberg, 1987). The function of the 86-kDa 0 subunit is not defined yet.

The 0 subunit interacts with the 8 but not or subunit (Studwell-Vaughan and

29

O'Donnell, 1993). Pol III holoenzyme contains two cores which act
coordinately at a replication fork to replicate both strands of the duplex
bacterial chromosome (Maki et al., 1988). Each core has a dimer of the B
subunit as a processivity factor (Burgers et al., 1981). The structure of the
B subunit reveals how it functions as the processivity factor. In the crystal
structure, two B subunits appear as a ring-shaped head-tO-tail dimer with a
central cavity of sufficient diameter to accommodate duplex DNA (Kong et
al., 1992). There is room in the cavity for 1-2 layers of water molecules
between the DNA and the B subunit dimer which may insulate B from local
interaction with DNA so that the B dimer can move along the DNA as a
sliding clamp. The B dimer does not assemble onto DNA by itself. A clamp
loader, the 7 complex, is needed to load B clamps onto DNA at the cost of
ATP hydrolysis (Onrust et al., 1991). The 7 complex is composed of five
distinct subunits in the stoichiometry 1651537111111 (Maki and Kornberg,
1988). The crystal structure of the 5’ subunit has also been determined
(Guenther et al., 1997). The 5’ subunit consists of three domains which are
arranged in a C-shaped architecture. The 5’ (36.9 kDa) amino acid
sequence (Dong et al., 1993) shows homology to y (47 kDa) (Flower and
McHenry, 1990). The 7 subunit is the catalytic component of the 7 complex
which has weak DNA-dependent ATPase activity and is stimulated by B
clamps (Onrust et al., 1991). Nucleotides are believed to bind to y at an

interdomain interface within the inner surface of the predicted C shape

30

structure of the 'y subunit (Guenther et al., 1997). A dimer of 1? subunit (71
kDa) is believed to tether the 7 complex to two (180 cores (McHenry,
1982). The 7 complex and the B clamps together mediate the grip of pol III
holoenzyme to primed DNA in the presence of ATP. The effects of
holoenzyme assembly on the biochemical properties of the enzyme are
remarkable (for review see (Kelman and O'Donnell, 1995)). The rate of
DNA synthesis by the enzyme increases from 8 nucleotides per second for
the or subunit to 20 nucleotides per second for the three-subunit core and to
750 nucleotides per second for pol III holoenzyme. The processivity of the
enzyme increases from 11 nucleotides per association for the core to

several thousand nucleotides by pol III holoenzyme.

III. Bacteriophage T7 DNA polymerase

T7 DNA polymerase appears to be very similar to mitochondrial
DNA polymerase. It is a two-subunit hybrid: the catalytic subunit is the 80-
kDa polypeptide of T7 gene 5, and the accessory subunit is E. coli
thioredoxin from the host cell. Sequence analysis has indicated that T7 g5p,
like the catalytic subunit of mitochondrial DNA polymerase, belongs to
DNA polymerase family A (Ito and Braithwaite, 1991). T7 g5p contains
both 3’-5’ exonuclease and polymerase but shows only modest polymerase
activity and low processivity (1—15 nucleotides/association) (Tabor et al.,

1987). After bacteriophage T7 infects a bacterial cell, T7 g5p annexes

31

thioredoxin from the host cell to form a 1:1 complex (Kd~5 nM) (Modrich
and Richardson, 1975), which immediately converts T7 g5p into a
polymerase of high processivity (thousands of nucleotides/association)
(Huber et al., 1987). E. coli thioredoxin is a single polypeptide of 108
amino acid residues (11.7 kDa) and functions mainly as a protein disulfide
oxido—reductase or a substrate for reductive enzymes (Holmgren, 1985).
However, the redox capacity of thioredoxin is not required for stimulation
of DNA polymerase activity but the oxidized thioredoxin (~Cys32-S-S-
Cys35-) can not bind to T7 g5p well (Adler, 1979). The X—ray structure
(Holmgren et al., 1975) of thioredoxin was determined and it is not a ring
structure like B dimer. Therefore, there is no clamp and clamp loader
involved, which is quite different from the mechanisms observed in E. coli
pol III holoenzyme. The crystal structure of the T7 DNA polymerase-DNA
complex (Figure 1.4) has been determined (Doublie et al., 1998). The
structure of T7 g5p is very similar to that of the Klenow fragment of E.
coli pol I, with an amino-terminal 3’-5’ proofreading exonuclease domain
and a carboxyl-terminal polymerase domain. Sequence alignment with the
Klenow enzyme has shown that there is a 71-residue insertion in T7 g5p in
the region between the 3’-5’ exonuclease and the polymerase domain. This
insertion is located in the tip of the thumb of the pol domain and forms an
extended loop that binds to thioredoxin (Doublie et al., 1998).

Interestingly, when the same 71-residue fragment was inserted into the

32

Thumb

Fingers

 

Figure 1.4 Topology and overall fold of T7 DNA polymerase.
T7 DNA polymerase complex (1t7p) bound with a primer (yellow)
and template (red) DNA is represented. The polymerase domain is
represented in magenta with fingers, thumb, and palm subdomains
labeled separately, and exonuclease domain in blue. Thioredoxin is
represented in golden yellow. Image is presented in color.

33

corresponding region in the Klenow enzyme by protein engineering, the
mutated Klenow enzyme also interacted with thioredoxin and showed
enhanced processivity (Bedford, 1997). Biochemical data indicate that
thioredoxin increases by 80-fold the affinity of T7 DNA polymerase
binding to the primer-template junction (Huber et al., 1987) but how this
happens is not clear. In the crystal structure, thioredoxin and the extended
binding loop in T7 g5p appear to be flexibly tethered to the thumb of T7
g5p, and they could presumably swing across the DNA-binding groove to
encircle the primer-template exiting the polymerase. In this way, they
could secure the primer-template in the polymerase, either by forming a
lid over the active site that functions like a slide clamp within the
polymerase, or by making electrostatic interactions with the DNA (Doublie

et al., 1998).

IV. HSV DNA polymerase

With the exception of the papovaviruses and parvoviruses, which
utilize cellular DNA polymerases for their replication, all known DNA
viruses that infect animal cells encode their own DNA polymerases. In
general, viral DNA polymerases share considerable sequence homology
with eukaryotic DNA polymerase 0t, 5, and e, and they belong to DNA
polymerase family B (Ito and Braithwaite, 1991). Herpes simplex virus

(HSV) DNA polymerase is the most well studied viral DNA polymerase.

34

HSV DNA polymerase is distinguished from most cellular polymerases by
its stimulation by relatively high concentrations of salt, its relatively low
Km values for dNTPs, and its sensitivity to a variety of inhibitors
(Weissbach et al., 1973). HSV DNA polymerase is a heterodimer which
consists of the products of the viral UL30 and UL42 genes (Hernandez and
Lehman, 1990). The polypeptide encoded by UL30 is the polymerase
catalytic subunit which also contains an intrinsic 3’-5’ exonuclease activity
and an intrinsic RNase H activity (Marcy et al., 1990; Crute and Lehman,
1989). Limited proteolysis showed that the amino-terminal half of the pol
molecule is sufficient for its 3’-5’ exonuclease and RNase H activities
(Weisshart et al., 1994). An approximately 12-kDa carboxyl-terminal
structural domain has DNA-binding activity required for polymerase
activity (Digard et al., 1993; Weisshart et al., 1994). The catalytic subunit
alone is much less active than the holoenzyme for polymerization on
certain primer templates (Gottlieb et al., 1990). Mutational analysis
indicates that the catalytic subunit can suffer deletions as large as 227
amino acid residues at the amino terminus and 58 amino acid residues at
the carboxyl terminus without loss of polymerase activity in vitro (Dorsky
and Crumpacker, 1988; Haffey et al., 1990). However, deletion of the 58
amino acid residues at the carboxyl terminus impairs association with
UL42 (Digard et al., 1993). The smaller subunit encoded by UL42 is a

processivity factor (Gottlieb et al., 1990). The UL42 subunit has intrinsic

35

affinity for double-stranded DNA (Gallo et al., 1988; Chow and Coen,
1995). The current model suggests that UL42 simultaneously binds pol and
DNA and thereby increases affinity for primer template. The interaction of
HSV pol and UL42 is not only required for processive DNA synthesis in

vitro, it is also required for viral DNA replication in vivo (Digard et al.,

1993).

V. HIV reverse transcriptase

RNA viruses contain a specialized DNA polymerase which can
mediate DNA synthesis on either RNA or DNA templates. A well studied
example is the human immunodeficiency virus type 1 reverse transcriptase
(HIV-l RT). HIV-1 RT consists of two subunits of 66 kDa (p66) and 51
kDa (p51) in 1:1 ratio (Joyce and Steitz, 1994). P66 has a polymerase
domain at the N—terminus and an RNase H domain at the C—terminus. The
two domains. are connected by a flexible nonessential region. The pol
domain has three subdomains similar to the fingers, palm, and thumb
subdomains in the Klenow enzyme of E. coli pol 1. P51 is derived from
p66 by loss of the RN ase H domain. Even though their amino acid
sequences are identical, the polymerase domains are arranged differently in
the two subunits, with p66 forming a large active-site cleft and p51
forming an inactive closed structure. The polymerase active site of p66

contains a catalytic triad of aspartic acid residues conserved in many DNA

36

polymerases (Joyce and Steitz, 1994). HIV-RT has no 3’—5’ exonuclease
activity for proofreading and shows a very high error rate of
polymerization: a misincorporation rate in vitro Of 1 per 2000 to 4000
nucleotides polymerized (Preston et al., 1988).

HIV-1 RT has an essential role in the viral replication and has been a
major target for drug design. Antiviral nucleoside analogs such AZT and
ddI are presumed to bind directly to the polymerase active site as
competitive inhibitors (Kohlstaedt et al., 1992). However, long term
clinical usage of these drugs have shown sever mitochondrial defects due to
the inhibitory effects on pol y. This may suggest that HIV-RT shares more

similarities with pol 7 than previously expected.

VI. Nuclear DNA polymerases

In eukaryotic cells, DNA polymerases are required to maintain the
integrity Of the genome during processes such as DNA replication, DNA
repair, and DNA recombination. It has been known for a long time that
there are four distinct DNA polymerases, named DNA polymerase or, B, 5,
8 (Wang, 1991), in the nuclei of eukaryotes. The number of known nuclear
DNA polymerases has increased to eight in recent years (for reviews see
(Burgers, 1998; Hubscher et al., 2000)). Genetic studies of yeast have
demonstrated that pols 0t, 5, and e are all essential for cell viability and

nuclear DNA replication (Campbell, 1993). In contrast, strains harboring a

37

deletion or mutation of POL4 (the pol B gene) are viable (Leem et al.,
1994). The newly identified members (termed pols C, (zeta), n (eta), 0
(theta), and t (iota)) are involved in various pathways for DNA repair
(Baynton and Fuchs, 2000; Hubscher et al., 2000).

DNA pol or is the only enzyme with an associated DNA primase
(Wang, 1991). It is required for both the initiation of DNA replication and
priming for the discontinuous synthesis of Okazaki fragments on the
lagging strand (Waga et al., 1994). Pol on is a moderately processive
enzyme in MgH-catalyzed reactions (Copeland and Wang, 1991) and has no
3’-5’ exonuclease activity for proofreading. The human pol 0t consists of
four subunits: a catalytic subunit of 180 kDa, a 70-kDa subunit, and two
subunits of 49 kDa and 58 kDa that contain the DNA primase activity.
Similar subunit components are found in all eukaryotes examined (Wang,
1991). Although no enzymatic activity has been associated with p70
subunit, biochemical studies have shown that it may play a role in the
assembly of the primosome (Collins et al., 1993). Both the human p180 and
p70 subunits are phosphorylated predominantly during late G2 and M
phases of the cell cycle (Nasheuer et al., 1991), suggesting that the
phosphorylation of pol 0t might play a role either in coordinating DNA
replication and mitosis or in resetting the replication apparatus for the next

S phase.

38

DNA polymerase B is a single polypeptide of 39 kDa (Wang, 1991).
It is the only eukaryotic cellular DNA polymerase with determined three-
dimensional structure (Sawaya et al., 1994; Pelletier et al., 1994). Pol B is
a distributive enzyme with long gapped DNA as a substrate but can
synthesize DNA processively in short gapped substrates (Prasad et al.,
1994), in agreement with its role in base excision repair (Matsumoto et al.,
1998). Pol B consists of two domains connected by a protease-sensitive
hinge region. The amino-terminal 8-kDa domain functions in binding
single-stranded DNA (Kumar et al., 1990), recognition of a 5'-phosphate in
gapped DNA structures, and as a 5'-deoxyribose phosphate (dRP) lyase
(Prasad et al., 1998a; 1998b). The carboxyl-terminal 31-kDa domain has
approximately 5% of the polymerase activity of the intact pol B (Kumar et
al., 1990). This domain has fingers, palm, and thumb subdomains similar
to those of the Klenow fragment of E. coli pol I (Pelletier et al., 1994).
The 8-kDa domain is composed of two pairs of antiparallel helices and is
attached to the 31-kDa domain at the tip of the finger subdomain by a
flexible hinge. The 8-kDa domain moves closer to the palm subdomain
when pol B binds to DNA primer template in a closed conformation
(Pelletier et al., 1994). DNA pol B has no intrinsic 3’-5’ exonuclease
activity. Structural data suggest that pol B may enhance fidelity by an

induced fit mechanism in which correct base pairing between template and

39

incoming dNTP induces alignment of catalytic groups for catalysis (via
thumb closure), but incorrect base pairing will not (Sawaya et al., 1997).

DNA polymerase 5 plays important roles in DNA replication,
nucleotide excision repair, base excision repair and VDJ recombination
(Hindges and Hubscher, 1997). Pol 5 is required for both in vivo DNA
replication and SV40 cell-free DNA replication (Waga et al., 1994). There
is considerable subunit diversity between enzymes from different sources.
P01 5 from mammalian cells and D. melanogaster consists of two subunits
(Wang, 1991) while pol 5 from Schizosaccharomyces pombe has at least
four subunits (Zuo et al., 1997) and pol 5 from S. cerevisae is a hexamer
formed by two heterotn'mers (Burgers and Gerik, 1998). All these
enzymes have a low processivity in the absence of proliferating cell nuclear
antigen. PCNA functions as a processivity factor. It exists as a trimer
forming a ring structure similar to the B clamp of E. coli pol III even
though there is no significant sequence similarity between them (Krishna et
al., 1994).

DNA polymerase 8 is distinguished from pol 5 because of its high
processivity in the absence of PCNA (Wang, 1991). The purified human
pol 8 consists of two subunits (Kesti et al., 1993). Mutational analyses
suggest that pol e is essential for completion of S phase (Araki et al., 1992)

but the role of pol e in DNA replication is not clear.

40

Overview

In sum, mitochondrial DNA polymerase appears more similar
overall to bacteriophage T7 DNA polymerase and viral DNA polymerases
than cellular DNA polymerases from both prokaryotes and eukaryotes.
This could be the result of natural selection. The endosymbiotic origin of
mitochondria in evolution parallels the infection of cells by viruses.
Therefore, mitochondria may utilize enzymes for their biogenesis just as
effectively as those viruses. Mitochondrial DNA polymerase is a two-
subunit enzyme, and so are T7 DNA polymerase, HSV DNA polymerase,
and HIV-RT, as compared to the multiple-subunit complexes like E. coli
pol III, pol or and pol 5. The catalytic subunit of pol y and T7 DNA
polymerase belong to the same DNA polymerase family A. Biochemically,
mitochondrial DNA polymerase is stimulated by high salt concentration as
is HSV DNA polymerase. Mitochondrial DNA polymerase can also catalyze
DNA synthesis on an RNA template, a characteristic of reverse
transcriptase. However, mitochondrial DNA polymerase is the least well
studied in terms of structure-function relationships. Even the subunit
composition is not completely resolved in many species. Little was known
about the function of the accessory subunit when this project began. Data
about protein-protein interactions between the two subunits in

mitochondrial DNA polymerase are completely absent.

4]

My thesis research aims to explore the function(s) of the accessory
subunit in Drosophila mitochondrial DNA polymerase and how it interacts
with the catalytic subunit. Different strategies have been adopted to fulfill
the aims. In chapter 11, a computational model of the C-terminal domain of
the accessory subunit is described. The structural model was developed
based on the sequence homology between the C-terminal domain of the
accessory subunit and the anticodon binding domain of prolyl-tRN A
synthetase. According to the characteristics of this structural model, we
have predicted a dual role of the accessory subunit as a processivity factor
and a primer recognition factor during mtDNA replication. In chapter 111,
physical interactions between the two subunits of Drosophila mitochondrial
DNA polymerase were investigated with recombinant proteins of various
mutations both by in vivo reconstitution in Sf9 cells and by in vitro protein
overlay assays. The results indicate that the assembly Of the heterodimer of
pol 7 occurs through multiple contact sites between the two subunits and
results in a complex with overall similarity to the structure of HIV-RT
holoenzyme. Chapter IV is the brief description of some preliminary
results on the biochemical properties of the accessory subunit, and outlines

future plans to extend the results described in this thesis.

42

CHAPTER II

HOMOLOGY MODELING OF
THE CARBOXYL-TERMINAL DOMAIN OF
THE ACCESSORY SUBUNIT OF
DROSOPHILA MITOCHONDRIAL DNA POLYMERASE

The work described in this chapter was carried out in collaboration with

Paul Sanschagrin and Dr. Leslie A. Kuhn

Department of Biochemistry, Michigan State Univeristy

43

INTRODUCTION

Mitochondrial DNA polymerase (pol y) is the key enzyme involved
in the replication of the mtDNA genome, and its catalytic features appear
to be conserved from yeast to man. Previous studies from this laboratory
have shown that Drosophila pol y is a heterodimer of catalytic and
accessory subunits of 125 and 35 kDa, respectively (Olson et al., 1995;
Wemette and Kaguni, 1986). The large catalytic subunit contains both 5'-
3' DNA polymerase and 3'-5' exonuclease activities (Lewis et al., 1996).
Typical of replicative DNA polymerases, the two-subunit Drosophila pol y
is both highly processive and highly accurate in nucleotide polymerization
(Wemette et al., 1988). With the high processivity determined in in vitro
biochemical studies, Drosophila pol 7 could replicate completely mtDNA
upon binding of the initiating primer on each DNA strand. Thus, the
recruitment of pol 'y to the initiating primer and maintenance of enzyme
processivity may be critical for efficient and faithful mitochondrial DNA
replication.

The accessory subunit of the heterodimeric pol 'y from Drosophila
embryos has been shown to be required to maintain the structural integrity
and/or catalytic efficiency of the holoenzyme (Lewis et al., 1996; Wang et
al., 1997). Based on amino acid sequence analysis, the catalytic subunit

shares a high degree of sequence similarity with other DNA polymerase

44

catalytic cores, both from the p01 7 family (Lecrenier et al., 1997; Lewis et
al., 1996) and among other members of the family A class of DNA
polymerases that includes the two-subunit bacteriophage T7 DNA
polymerase (Braithwaite and Ito, 1993). In contrast, the cDNA for the
accessory subunit encodes a novel protein (Wang et al., 1997). However,
human, mouse and rat homologs of the Drosophila cDNA have been
identified from the EST (Expressed Sequence Tag) databases, providing
evidence for a common heterodimeric structure among animal pol 73
(Wang et al., 1997).

In parallel with biochemical studies of pol y and its isolated subunits,
we have pursued molecular modeling of the accessory subunit to elucidate
its possible functions. Although protein function is best determined
experimentally, it can also be predicted by sequence alignment of a query
protein with proteins of known functions (Dujon, 1996; Koonin and
Mushegian, 1996; Oliver, 1996). This is possible because similar protein
sequences tend to have similar functions, although exceptions occur
(Orengo et al., 1994). One way to enhance sequence-based predictions of
function is to determine or predict the three-dimensional (3D) structure of
a protein. The 3D structure of a protein generally provides more
information about its function than its sequence, because function is more
directly dependent on structure than on sequence, and structure is more

conserved in evolution than sequence (Chothia and Lesk, 1986). In this

45

case, the structure, rather than the amino acid sequence, is compared
against those of the protein structure database (Protein Data Bank). If there
are one or more significant structural homologs, the novel protein is
predicted to have molecular properties similar to the homologs
(Zarembinski et al., 1998). Recent advances in molecular and structural
biology and in computer simulation have made it possible to predict 3D
structure of a protein with satisfactory accuracy without experimental
determination. In recent years, fold recognition or homology modeling of
the 3D structure of a protein sequence has become particularly popular
(Abagyan et al., 1994; Bowie et al., 1991; Bryant and Lawrence, 1993;
Fischer and Eisenberg, 1996; Flockner et al., 1995; Godzik and Skolnick,
1992; Jones et al., 1992; Jones et al., 1999; Russell et al., 1996). In this
regard, blind testing has shown that fold recognition methods can be very
effective (Flockner et al., 1997; Shortle, 1997; Shortle, 1999).

Here we present a 3D structural model for the carboxyl-terminal
domain of the accessory subunit of Drosophila pol 7 based on the fold
recognition method developed by Fischer and Eisenberg (Fischer and
Eisenberg, 1996). This structural model shares a striking structural
similarity with the anticodon binding domain of several class IIa aminoacyl
tRNA synthetases. That this domain has been shown to be involved in
tRNA binding in the latter suggests that the accessory subunit of pol 7 plays

a role in recognition of the unusual RNA primers generated at the

46

mitochondrial DNA replication origin (Lee and Clayton, 1998), to guide
the catalytic subunit to the primer terminus. In addition, the accessory
subunit model shows structural homology with thioredoxin, the accessory
subunit of T7 DNA polymerase, despite their lack of amino acid sequence
homology. Thus, we predict a dual role for the accessory subunit as a
processivity clamp and as a primer recognition factor/clamp loader in

mtDNA replication.

47

MATERIALS AND METHODS

Sequence Analysis of the Accessory Subunit of Drosophila pol y

The SwissProt protein sequence database was searched using the
BLAST algorithm (Altschul et al., 1990) through the NCBI web server
(http://www.ncbi.nlm.nih.gov/) for homologs to the sequence of the
accessory subunit of Drosophila pol y (Wang et al., 1997). Sequence
alignments were obtained with the Gap and Bestflt programs of the
University of Wisconsin Genetics Computer Group software package
(GCG; (Devereux et al., 1984)) using the Needleman and Wunsch

algorithm (with gap weight 5 and length weight 4). Residues 254-361 of

the accessory subunit Of Drosophila pol 'y (pol y—B) were aligned using Gap

to residues 288-386 of the sequence from the crystal structure of the
Thermus thermophilus prolyl tRNA synthetase (Pro-RS; generously
provided by S. Cusack Of EMBL-Grenoble; (Cusack et al., 1998)),

corresponding to an independent folding domain in E. coli pro-RS

identified by the BLAST search as being similar to the pol y-B sequence.

Minor modifications of this alignment were made so that insertions and

deletions fell into regions between secondary structures.

48

Structural Modeling of the Accessog Subunit

Initial prediction of the three-dimensional (3D) fold of the accessory
subunit of Drosophila pol V was obtained from the Protein Fold
Recognition Server (http://www.doe-
mbi.ucla.edu/people/frsvr/preds.html), based on the algorithm of Fischer
and Eisenberg (Fischer and Eisenberg, 1996). This approach uses both the
secondary structure assigned to the sequence by PhDsec ((Rost and Sander,
1994), http://www.embl-heidelberg.de/predictprotein/predictprotein.html)
and the Gonnet amino acid substitution matrix to assess the similarity
between the probe sequence (e. g., the accessory subunit) and sequences
from the Protein Data Bank (PDB; (Bernstein et al., 1977; Sussman et al.,
1998)) of three-dimensional protein structures. The use of both primary
and predicted secondary structures in the search enhances its sensitivity in
identifying the protein with the most similar 3D fold to that Of the probe.
When tested on a set of 68 proteins whose 3D structures were known but
ignored, the fold recognition server can recognize successfully the 3D fold
of a new sequence in 70% of the cases (Fischer and Eisenberg, 1996).

Three-dimensional modeling was carried out using Molecular
Simulations’ Insight” and Biopolymer software on a Silicon Graphics

Indig02 graphics workstation. Using the sequence alignment between the

accessory subunit of D. melanogaster pol y (pol y-B) and the 2.4 Angstrom

49

crystal structure of residues 288-386 of T. thermophilus pro-RS, the

backbone of pro-RS was used as a structural template, and side chains

differing in pol y—B were substituted individually. For two of the three

short loops between regular secondary structures that differ between pol y-

B and pro-RS, the Searchloop function in Biopolymer was used to identify
a loop of known structure with appropriate length, geometry, and packing
to use as a structural template; for the third loop, between B strands 4 and
5, the corresponding loop in the structure of T. thermophilus histidyl
tRNA synthetase (His-RS; PDB entry ladj; (Aberg et al., 1997)) was used
as a template, followed by side-chain substitution where necessary. When
side-chain substitutions resulted in steric collisions, they were resolved by
minimal torsional angle adjustments in Insight”, followed by energy
minimization on the loop regions only, by 100 steps of steepest descents
minimization with the CVFF forcefield using Discover software

(Molecular Simulations; San Diego, CA). The stereochemistry of the final
structural model of the C-terminal domain of pol y—B was validated by
Procheck (Laskowski et al., 1993; Morris et al., 1992), and the favorability
Of amino acid environments within this structure was assessed using the 3D
Profile method (Luthy et al., 1992) via the Verify3D server

(httpz/lwww.doe-mbi.ucla.edu/Services/Verify3D.html). T hree-

dimensional structural comparisons between the accessory subunit model

50

and the available crystallographic structures of thioredoxin and the 5'
subunit of pol III employed the Dali server ((Holm and Sander, 1993),

http://www.embl—ebi.ac.uk/dali).

51

RESULTS AND DISCUSSION

Identification of Sequence Homologs of the Accessory Subunit of Pol J!
The accessory subunit Of Drosophila pol y represents a novel protein
in the protein and DNA databases (Wang et al., 1997). However, BLAST
search in the SwissProt database suggested moderate sequence homology
between a region of the accessory subunit of Drosophila pol y (pol y—B) and
aminoacyl-tRN A synthetases (aaRS) from both prokaryotes and eukaryotes
(Table 2.1). The top ten matches, with up to 35% identity over ~40
residues (data not shown), were all members of the class II aminoacyl-
tRNA synthetase family (Cusack, 1995). Although the sequence similarities
are not so great, these results suggest that the accessory subunit of pol y
may have evolved from the same ancient as those aminoacyl-tRNA
synthetases, implying that they may still have some function and structural
similarities. The most similar was a probable glycyl-tRN A synthetase from
Methanococcus jannaschii (SwissProt accession code Q57681), with a score
of 84 and likelihood of 0.0018 over two aligned segments, reflecting 31%
identity to residues 211-254 of pol y-B and 25% identity to residues 290-
355 (data not shown). The BLAST sequence alignments (data not shown)
between the accessory subunit of Drosophila pol y and the rest aaRSs in
Table 2.1 were mostly limited to the carboxy (C)-terrninus (residues 290-

360) of pol y—B.

52

Table 2.1 BLAST SEARCH RESULTS

(Database: non-redundant SwissProt sequences)

Smallest
Sum

High Probability
Sequences producing High-scoring Segment Pairs: Score P (N) N
sp|Q5768l|SYG_METJA PROBABLE GLYCYL—TRNA SYNTHETASE ........... 84 0 0018 2
sp|P47493|SYG_MYCGE GLYCYLTRNA SYNTHETASE .................... 72 0.15 1
splP36431|SYP_CHLTR PROBABLE PROLYL-TRNA SYNTHETASE ........... 72 0.15 1
sp|P18255|SYT1_BACSU THREONYL-TRNA SYNTHETASE 1 ................ 72 0.15 1
sp|P70076|SYH_FUGRU HISTIDYL-TRNA SYNTHETASE .................. 71 0.20 1
sp|P16659|SYP_ECOLI PROLYL-TRNA SYNTHETASE .................... 70 0.26 1
splp49590|SYHH_HUMAN HISTIDYL—TRNA SYNTHETASE HOMOLOG .......... 67 0.55 1
sp|P75000|SYP_ZYMMO PUTATIVE PROLYL—TRNA SYNTHETASE ........... 66 0.67 1
sp|P10179|SYG_SHPO PUTATIVE GLYCYL-TRNA SYNTHETASE ........... 64 0.88 1
sp|P12081|SYH_HUMAN HISTIDYL-TRNA SYNTHETASE .................. 64 0.88 1
splpo7178|SYHrMBSAU HISTIDYL-TRNA SYNTHETASE .................. 63 0.95 1
spIQ61035ISYH_MOUSE HISTIDYL-TRNA SYNTHETASE .................. 63 0.95 1

The first column indicates the database (SwissProt (sp)), accession code,
and the symbol of protein connected by "_" with the source (species). The
second column indicates the full name (or part of it due to space) of
individual proteins. High score is the parameter used by BLAST to rank
the results. N is the number of segments aligned by sequence comparison.
P(N) is the smallest probability of sequence alignment of N segments.

53

Modeling the Accessory Subunit of pol y Based on Structural Homology

with tRN A Synthetases

To obtain insights into the structure and function of the accessory
subunit of pol y, three-dimensional fold recognition (Fischer and
Eisenberg, 1996) was used to identify the best structural match for pol y—B,
based on the best alignment between its sequence and secondary structures
with those of all the distinct structural folds in the Protein Data Bank. For
71% of test sequences, this fold recognition algorithm is able to identify the
correct structural fold (Fischer and Eisenberg, 1996). However, the major
benefit of the fold recognition server in the case of pol y—B was the
identification Of a structural homolog of the pol y—B protein that was not
yet included in sequence databases and therefore not identified by BLAST
search. Using the sequence of the accessory subunit, the server indicated
3D structural similarity for pol y-B residues 24-359 with residues 73-504
of the glycyl-tRNA synthetase from T. thermophilus (PDB entry lati;
(Logan et al., 1995)) (Figure 2.1). The Z-score was 11.5, more than twice
the threshold value (4.8+/-1.0) for significant similarity (Table 2.2).

Aminoacyl-tRNA synthetases are a group of 20 enzymes that catalyze
the attachment of each amino acid to its cognate tRN A (Amez and Moras,
1997; Cusack, 1995). The 20 aaRS enzymes can be divided into two
different families of 10 members each, based on their amino acid sequence

and structural similarity, and crystallographic structures have been

54

Figure 2.1 Structure-based sequence alignment between the
accessory subunit of Drosophila pol y and Thermus
thermophilus glycyl-tRNA synthetase. The sequence alignment
between the accessory subunit (pol y—B) and glycyl-tRNA synthetase (Gly-
RS) was provided by fold recognition service (http://www.doe-
mbi.ucla.edu/people/frsvr). Secondary structural elements are indicated by
hhhh for alpha helices and bbb for beta strands. The structural elements for
pol 'y—B were based on PHD predication while the structural elements for
Gly-RS were assigned by X-ray structural determination (PDB entry lati).
Conserved motif II and III in class II aminoacyl-tRNA synthetases are
boxed in doted line and dashed line, respectively. Conserved residues in
aminoacyl-tRN A synthetases are in bold letters. The region corresponding
to anticodon binding domain is shadowed for both sequences.

55

2 85am

 

 

 

one mmamaw

emu n.» Hem

 

Benzeneenen .---mam:.-mm.

Annnnnnn awn nan Benenz an annngnnannn nannnnnnn
mwm OgmmmH>>m>h1¥09mmflm>¢dm09mmmzmd>m<BH04¢m0QxfimmOQQ.mDBm04HOmqumwmmhmwdHO>B€M mmN mMI thaw
_ __ _ . __ _ _ _ _ _ _ _ __ _
«mm.mH>mmH>Emo>Vmwem¢omqmm....oxoo>Hm>...amaom>¢>mo¢oowmemHoH<owo ............... _n_n.> How
Mmmmmmmaw «a pagan oneeneneneg anann
nannnncnnn .......................a.......nuwu.mmfimm.:.z:.mmammnm

 

mmN dwmfimmmmmoowm>aZMMme. .Emozzmdmmm>3>m23>mom0mm>mwmHm20mmum>mmHmZMQBHWZMhdMUH. mHN mmlxfiw

__ L_ __ _ _._ _

osa ..Zuzqom<qnum>Hm>mmm2mmqmzzzHemmmozz>mmmq¢mmmm>qmaeeqmqemoxH<mqmmmamemoemommW no nu» How
an: annnh ......................m ........ mmmmm:.mmmmm .................... mm

 

aspen Baneeeeneneeenggg annennnae manna nannn annnnnnengegnnenng
#HN O<H0mwammm9<nq>zmmz>mHOO<Bmmmdw<qumUOmmUMJQOhZEZthmm9320>2mfl<mfi<mm®m>ma> Mb WMI>HO
__ __ _ _ _ L _ __ _ _ _
mm QABAUfidfiMQmQM>AXOEZZMhOO.mommmmm0H2>mZHmmMB<OQmddqmmqmfigmooqqxdwmmommaqmdx vm QIF Hom
Baneesannennn an 5555 ..................... Baggage

 

56

 

Table 2.2 SUMMARY OF FOLD RECOGNITION RESULTS

CONFIDENCE
METHOD LEVEL FIRST HIT SCORE
BLAST SEQ e-lO to e-5 sw: SYG_METJA 0.0084
FASTA SEQ ? SW:SYG_METJA
MOMENT No. OF TMHS: 0
GON+PREDSS 4.8 +/- 1.0 latia (ttGly-RS) 11.80
GON+PREDSS+MULT 5.0 +/- 1.0 latia 9.19
GON+PREDSS+LOCAL 6.0 +/- 1.0 latia 14.75
GON 4.8 +/— 1.0 latia 9.68

METHOD, method used for analysis. MOMENT, a method of
transmembrane helices (TMHS) prediction. GON, a Gonnet algorithm of
sequence comparison. GON+PREDSS, analysis based on sequence and
predicted secondary structure by PHD program. MULT, multiple sequence
alignment. GON+PREDSS+LOCAL, a Smith-Waterman alignment
algorithm. CONFIDENCE LEVEL, a threshold value for reliable results
by indicated methods. FIRST HIT, the best candidate presented by PDB
code or SwissProt (SW) Symbol.

57

determined for members of both classes ((Cusack, 1995; Cusack et al.,
1998), and refs therein). Besides Gly-RS, class II aaRSs include Ser-, Thr-,
PrO-, His-, Gly-, Asp-, Asn-, Lys-, Phe-, and Ala-RS. All are
homodimers, with the catalytic domain built around a six-stranded
antiparallel B sheet. Most class II aaRSs have three conserved, class-
defined motifs: motifs 2 and 3 are responsible for ATP binding and
catalysis, whereas motif 1 is involved in dimerization essential for the
enzymatic activity. Based on amino acid sequence comparison, Thr-, Pro-,
His-, and most Gly-RS (including Gly-RS from T. thermophilus) have a
homologous C-terminal domain (shadowed in Figure 2.1) attached to the
catalytic core, identifying them as members of subclass IIa. This domain
interacts with the anticodon loop and stem of the tRNA substrate, as
elucidated by crystallography (Cusack et al., 1998).

The sequence-structure alignments from fold recognition (Figure
2.1) reveal that residues 254—360 of Drosophila pol y—B map onto the
anticodon binding domain of class II aaRSs. The match of residues 24-253
of pol y—B to regions of the catalytic core of the synthetase proved more
difficult to interpret, as critical motifs 2 and 3 (boxed in Figure 2.1) in the
synthetase were not matched, and generally the alignment was substantially
fragmented. Two B strands in the catalytic core of Gly-RS are absent from
pol y-B, which suggests that it would be difficult for the two to fold the

same in this region. In addition, the motif 1 structure in Gly-RS is missing

58

in pol y—B. These results indicate that pol 'y-B lacks the central core .
structure for catalysis of tRNA aminoacylation.

While these results suggest that the accessory subunit of pol y is not
enzymatically homologous to the aaRS family, it is likely to share
structural homology with the anticodon binding domain of the synthetases,
based on significant sequence identity in this region. In order to select the
best template for homology modeling, we compared the sequence of the C-
terrninal region of Drosophila pol 'y—B with the sequences of different
anticodon binding domain of class Ila aminoacyl-tRN A synthetases with
known structures (data not shown). Of the known structures of aaRS's, T.
thermophilus (tt) Pro-RS has the highest sequence identity to pol y-B, with
28% identity (42.7% similarity) between the 81 C-terminal residues of
Drosophila pol y—B and residues 302-381, which form the anticodon
binding domain in ttPro-RS (Cusack et al., 1998). The 3D structure of this
C-terminal region Of pol 'y-B is therefore expected to have a closely
superimposible 3D structure with that of the anticodon-binding domain.
This expectation is based on exceeding the degree of sequence identity
required for structural similarity found from thousands of pairwise
sequence and structural alignments between known protein structures in the
PDB (Sander and Schneider, 1991). Once an appropriate structural

template has been identified based on sequence identity, a structural

59

homology model may be built by using this template as the main chain, and
substituting side chains according to the sequence alignment.

We utilized the anticodon binding domain of T. thermophilus Pro-RS
(coordinates provided by S. Cusack, EMBL—Grenoble) as the structural
template to build by homology modeling the three-dimensional structure of
the C-terminal domain of the accessory subunit of pol 7. Amino acid
sequence comparison indicated that the C-terminal domain of the accessory
subunit of Drosophila pol 'y is also the most conserved region as compared
with its human and murine homologs, supporting its functional importance
(Wang, 1998). For example, the human and Drosophila amino acid
sequences of pol 'y—B have less than 20% identity over the whole length of
the polypeptide but have more than 31% identity over the last 120 amino
acid residues (Figure 2.2). Modeling was based on the Bestfit sequence
alignment between pol 'y-B and Pro-RS (Figure 2.3), adjusted so that
deletions or insertions fell in loop regions rather than in regular secondary
structures. Residues 288-301 of Pro-RS were aligned manually to pol 'y-B
based on the fold recognition alignment between Gly-RS and pol 'y-B. The
aligned region shows 23-29% pairwise sequence identity between the aaRS
sequences, which is the same as the identity percentage between pol y—B
and ttPro-RS (Figure 2.3) Based on this alignment of ttPro-RS to pol y—B,
InsightII was used to substitute pol y-B side chains on the ttPro-RS main-

chain structure, followed by side-chain and loop optimization (as

60

man
f 1y
man
fly
man
fly
man
f 1y
man
fly
man
f 1y
man
fly
man

fly

Figure 2.2

50

44

99

94

148

144

198

194

248

244

298

288

345

338

MVDLGGGVHGAVFPVDALHHKPSPLLPGDSAFRLVS.AETLREILQDKEL

....... MSRIQRCFKSLASAGFFRTVEDNKLELLSHGREYAKLLQQHWT

SKEQLVAFLENVLKTSGKLR.ENLLHGALEHYVNCLDLVNKRLPYGLAQI

RLRPLAAHLGATREPINPVNIQRFSFPQSQQFRNNFQKLVKDHPRKAKCP

GVCFHPVFDTKQIRNGVKSI.GEKTEASLVWFTPPRTSNQWLDFWLRHRL

TLLKHQSTCSGPTSHSLFAIKGPTLHLTTDFLVEPHRALEHFYNMQRESK

QWWRKFAMSPSNFSSSDCQDEEGRKGTNFTTIFPWGKELIETLWNLGDHE

IWWMRLSSNPSRYRIVPCDLAEDLNPNDYQAIDIRTSYGDAGEVAVEQLS

LLHMYPGNVSKLHGRDGRKNVVPCVLSVNGDLDRGMLAYLYDSFQLTENS

LVRIVDDKDFRLPDARTGETVQPTVIRSVIELETTTCALLLDGCDHGRDS

FTRKKNLHRKVLKLHPCLAPIKVALDVGRGPTLELRQVCQGLFNELLENG

..QSLLLHRyLAPYQQQIACVESDSEL.,..SADLSDLCQHLKHVLNHAG

..ISVWPGYLETMQSS.LEQLYSKYDEMSILFTVLVTETTLENGLIHLRS

W

RDTTMKEMMHISKLKDFLIKYISSAKNV 372

RDTRLAETIHISDVPDYLLNIFKN.... 361

49

98

93

147

143

197

193

247

243

297

287

344

337

Sequence alignment between the accessory subunit
of Drosophila pol 7 (fly) and human pol 7 (man). The alignment
was obtained using the Gap program in the UW-GCG package (Devereux
et al., 1984). The amino acid sequence identity and similarity over the

entire length of the polypeptide are 19.7% and 26.5%, respectively.

Identity and similarity for the most conserved region (underlined for the

fly sequence ) are 32% and 42%, respectively. Identical residues are

connected by bars between the two sequences while similar residues are
connected by dots. Every 10 residues are marked by dots at the top of the

human sequence.

61

.555 Son By a 98 80:5: 8m 5 me @8365 new @8383:

6.5 8.5825 1550qu gasooom

$53 .3 8 86838 555.658 Buaosbm 8 56584

ESE E as as, 684% :65 863 552. a e wens .52 .55 m9: @425
885:»... 425-1582 use 232% 8 53C. a: be... 5.: $59 6.8555

<72“;wa mo moocoscom .953 .

.R B 4:29»on owmxomm 000-33 55 Mo 5505 “Seem

was $52 .a 8 8565485 65655 475-an F: 33852: .352: BE
~65:me 5m 55 8:358 n1» 3m 33449.er 68333:? 4733-30555 4: mas—o 55
> _ea 5.33395 me A ml» 35 355.5 E8884 2: we 5.6.2588 oases—Gum Md 953..“

mmm E Eco 838.50 B

.148 L. _Irml:

 

5.4 404 54. mm 50.45445 .554 1111111111 4.1T

mom «3 mmm

wwm 4naxwm 4mm 44a emmrowqmm4. move .......... m-

Hmm *ZMWHZA NOW gm“

 

. Nd . _

5554mm 111111 5.545. 1
55554452 ...... OB W55
555555 ..... moommnqg -

qzmewq mm4amm4zxeemHOmommm 1 w

mm 28 :5 :4 5 338.80 .4... :4 5 338.50 I

        

    

 

Sum 535.4.
H3. 536.5
mam 5345

51:3.

.52 .......... 45.52 __ . 3m 5:085

. to _
Bmm mm d>4mm 111111 .
.5445 44%. weHmmmzunux>

5 85553505 4. Go» mmm 9:095

m m m mmmflmee
m m m 536.5
m w m 5845

  

62

described in Methods). The structural model was finally evaluated using
two techniques, Procheck, which evaluates stereochemistry (favorable bond
lengths, angles, and atomic packing), and Verify3D, which assesses the
favorability of residue environments within the folded structure. Procheck
found all residues to have favored or highly favored main-chain dihedral
(<I>,‘I’) angles, and the model to have significantly better overall
stereochemistry (G-factor of 0.07) than typical crystal structures solved at
the resolution of the Pro-RS structure (2.4 Angstroems). Evaluation of
residue environments by Venfy3D gave the accessory subunit model
structure a score matching those of higher resolution (2.0 Angstroem)
crystal structures.

The structural model of the C-terminal domain of the accessory
subunit of Drosophila pol 7 reveals a rare a/B fold, comprising a five-
stranded mixed [3 sheet surrounded by four on helices (Figure 2.4A), found
only in the anticodon binding domains of class IIa aaRSs (Cusack et al.,
1998). The main-chain of the resulting model of the C-terminal region of
pol y—B (residues 255-361) is the same as that of the anticodon binding
domain of ttPro-RS (0.21 Angstroem RMSD), except for three loop
insertions: loop 1 between strand B1 strand and helix a1, loop 2 between
[32 and a2, and loop 3 between [34 and [35 (Figure 2.3 and Figure 2.43).
The anticodon binding pocket of the class II tRN A synthetases, which faces

the viewer in the lower half of Figure 2.4B, is formed predominantly by

63

Figure 2.4 Homology modeling of the C-terminal domain of
the accessory subunit of Drosophila pol y. A) Molecular model of
the C-terminal domain of the accessory subunit of Drosophila pol 'y. The
main-chain ribbon is shown along with protein side-chain atoms colored
according to chemistry: Negatively-charged residues and derivatives (Asp,
Asn, Glu, Gln), red; positively-charged residues (His, Lys, Arg), blue;
small hydrophobic residues (Leu, Ile, Val, Met), green; large hydrophobic
residues (Phe, Tyr, Trp), purple; cysteine, yellow; and other residues
(Pro, Ala, Gly, Ser, Thr), grey. The termini of the CL helices and B strands
are numbered according to the Drosophila pol 'y-B sequence in Figure 2.3,
with the N- and C- termini labeled as (N) and (C). B) Comparison of the
structures of the modeled C-terminal domain of the accessory subunit of
Drosophila pol 'y and the anticodon binding domain of T. thermophilus
prolyl-tRN A synthetase, used as a template for the pol 'y-B model. The
accessory subunit is colored according to sequence conservation between
the Drosophila and human sequences shown in Figure 2.2; residues
conserved between the two sequences are in yellow with surface-accessible
side chains shown, conservatively substituted residues are in green, non-
conserved residues are in blue, and insertions in the fly sequence relative to
the human sequence are in red. (There are no deletions in the fly sequence
relative to human). The Pro-RS main-chain ribbon is colored by sequence
conservation among the three related aminoacyl-tRN A synthetases shown in
Figure 2.3, with conserved residues in dark pink, similar residues in light
pink, and non-conserved residues in grey.

64

361m)

B)

 

Figure 2.4 (images are presented in color)

65

the central [3 sheet and by a-helices 2 and 3 (Cusack et al., 1998); this [3
sheet appears at center, with a helix 2 appearing as the long helix at right
and helix 3 as the short helix at lower left. Whereas residue conservation
is often associated with side chains that bind substrates or ligands, and thus
the anticodon binding pocket might be expected to be conserved amongst
aaRSs and between aaRSs and the accessory subunit, this is explicitly not
the case. Different tRNA synthetases must recognize different anticodons,
and the accessory subunit of pol 7 must recognize at least two distinct
primers; thus their ligand binding residues are expected to vary to confer
this specificity. Of the residues identical in the aligned region between pol
y-B and ttPro-RS (Figure 2.3), half are surface-exposed and half are
somewhat or completely buried in the protein, and do not colocalize to a
specific region of the structure; for the additional residues that are
chemically similar but not identical, ~2/3 are exposed and the remainder
are buried (Figure 2.4). Conserved residues at the surface of this domain
in ttPro-RS are not associated with an interface with the region N-terminal
region to this domain, as the two domains are connected only by a loose
tether in class Ila synthetases (Cusack et al., 1998). Conserved surface-
exposed residues of pol y—B (yellow side chains in Figure 2.4B) may be
involved in interactions with the N-terminal domain of the accessory
subunit, or with the catalytic subunit or nucleic acid substrates. An RN A-

binding structural fold in the C-terminus of the accessory subunit

66

corresponding to that in class Ila aaRSs was unanticipated, but is consistent
with a function of the accessory subunit in interacting with RNA molecules
containing tRNA-like structures.

The modeled region of the accessory subunit also includes two
previously noted sequence motifs: a putative zinc finger and a putative
leucine zipper (Wang et al., 1997). The segment containing the leucine
zipper sequence in the model structure (residues 319-347) consists of one
a-helical segment and two short [3 strands (Figure 2.4A), and thus it can
not function as a leucine zipper. The segment for the putative Zn finger
motif in pol y—B (residues 256-286) consists of a [3 strand, a loop, and an on
helix (Figure 2.4A), which is similar to the structure observed for some
known Zn fingers (Pavletich and Pabo, 1991). However, the side chains of
Cys and His residues in the pol 'y-B structural model are not positioned such

that they could bind zinc.

Biochemical Implications of the Structural Model

The accessory subunit of pol ’y as a specialized primer recognition factor
Mitochondrial DNA replication is thought to involve different
priming mechanisms at the two origins. The leading DNA strand origin is
located within the major non-coding region of the mtDNA genome in all
animals examined to date (Clayton, 1982). In man and mouse, transcripts

initiated from the L-strand promoter are processed for use as primers for

67

H-strand synthesis; RN Ase MRP cleaves the transcripts at one of several
discrete sites at OH, approximately 100 nucleotides from the promoter
(Chang and Clayton, 1987; Lee and Clayton, 1997). The 3'-termini of
these processed transcripts are then presumably elongated by pol 7, but data
regarding the mechanism of pol 7 binding at the template-primer junction
are lacking. Our molecular model suggests that the accessory subunit may
play a role in the recognition of these unusual RNA-DNA hybrids, to
recruit pol y to the template-primer junction. In particular, we propose
that by virtue of its conserved C-terminal domain, homologous to the
anticodon binding domain in class IIa aaRSs, the accessory subunit
recognizes secondary structural features in the processed RNA, thereby
enhancing the template-primer binding affinity of pol ”y.

A similar mode of RNA-DNA hybrid recognition by pol y may
occur at the lagging DNA strand origin. Priming of L-strand synthesis in
mammals appears to be catalyzed by a primase (Wong and Clayton, 1985a)
that recognizes a specific stem-loop structure formed in the displaced
parental H-strand (Wong and Clayton, 1985b). Because the lagging DNA
strand initiation site is nested within a cluster of five tRNA genes, the
involvement of tRN A molecules cannot be excluded. Bird mtDNA
genomes lack the non-coding sequence within the tRNA cluster that
contains the stem-loop structure, suggesting a replication function for the

tRN A genes (or a corresponding tRNA) per se (Desjardins and Morais,

68

1990). In Drosophila, the lagging DNA strand initiation site has been
mapped within the non-coding A+T region, and in several species a stem-
loop structure‘has been identified that may serve a replication function
(Clary and Wolstenholme, 1987). Thus, although only in mammals has the
stem-loop structure been documented to function in lagging DNA strand
initiation, the situation may be similar in birds and in Drosophila. If so,
the function of the accessory subunit of pol y in specialized primer
recognition is likely very similar in both leading and lagging strand
initiation in mtDNA replication, and generally conserved in animals.

The postulated role of the accessory subunit of pol y as a primer
recognition factor parallels the function of the non-homologous 7 complex
of E. coli DNA polymerase III and eukaryotic replication factor C (RFC).
In contrast to the unidirectional and asymmetric mode of mtDN A
replication, chromosomal replication in both prokaryotes and eucaryotes
proceeds bidirectionally with symmetric synthesis of leading and lagging
DNA strands. This involves multiple cycles of recognition of short
oligonucleotide primers by the 7 complex and RFC, respectively. The 7
complex and RFC are multi—subunit proteins that share both sequence and
structural similarity (Guenther et al., 1997; O'Donnell et al., 1993). Our
structural model of the C-terminus of the accessory subunit of pol 7 shows
some similarity with the crystal structure determined for the 5' subunit of

the 7 complex, which forms a C-shaped structure comprising three domains

69

(Guenther et al., 1997). While the N-terminal domain of 5' is not highly
similar to the C-terminal domain of pol y—B based on Dali comparison
(statistical Z-score of 1.0, below the significance threshold of 2.0), they are
qualitatively similar, sharing an (x/B fold with a central [3 sheet (Figure 2.5,
left). This is consistent with our hypothesis of the accessory subunit as a
primer recognition factor, and may also indicate a general structure-

function theme in primer recognition factors in various replicative systems.

The accessory subunit of pol yas a processivity clamp

The structural and functional features of Drosophila pol y are similar
to those of the bacteriophage T7 DNA polymerase. The T7 DNA
polymerase and pol y catalytic cores are members of the family A group of
DNA polymerases (Braithwaite and Ito, 1993). T7 DNA polymerase has a
low intrinsic processivity, dissociating from the DNA after incorporation
of only a few nucleotides (Huber et al., 1987). Upon infection, T7 DNA
polymerase recruits a host-encoded protein, E. coli thioredoxin (12 kDa),
to form a heterodimer that is capable of polymerizing thousands of
nucleotides without dissociation (Huber et al., 1987). Likewise, we have
found that the heterodimeric pol 'y is a highly processive enzyme (Wemette
et al., 1988). In the E. coli and eucaryotic replicative systems, pol III and
pol 5 use ring-shaped accessory subunit complexes, [3 (Kong et al., 1992)

and PCNA (Krishna et al., 1994), respectively, to convert them into highly

7O

Figure 2.5 Comparison of the structures of the accessory
subunit of pol y, the N-terminal domain of 5' subunit of E. coli
DNA polymerase III , and E. coli thioredoxin. The N-terminal
domain of DNA polymerase III 5' (PDB entry 1a5t (Guenther et al.,
1997)) is shown in blue, pol y—B in magenta, and thioredoxin (PDB entry
1t7p, chain B (Doublie et al., 1998)) in green, with residues interacting
with the catalytic subunit of T7 polymerase shown in yellow. Optimally
similar orientations were produced using Dali (Holm and Sander, 1994;
Holm and Sander, 1993).

71

€28 E 68:89:. 0.3 woman—C m.~ ensur—

.U
j\. /

)

55090.5 :8 .m , .. .n-___ @989:on <20

2L. mmmdoEzoq <20

 

72

processive enzymes. These "processivity clamps" are directed to the
template-primer by the 7 complex and RFC, which actually serve dual roles
in primer recognition and clamp loading.

The recently determined crystal structure of a T7 DNA polymerase -
thioredoxin - DNA complex revealed that thioredoxin is not a ring-shaped
protein (Doublie et al., 1998). Instead, thioredoxin binds to the tip of the
thumb in the T7 DNA polymerase structure, and from this position
presumably clamps the template-primer. Corresponding biochemical
studies indicate that this association results in an ~80-fold increase in the
affinity of T7 DNA polymerase for the primer terminus (Huber et al.,
1987). Thus, the accessory subunit of T7 DNA polymerase apparently
serves a dual function in primer binding and processivity enhancement. In
fact, the mechanism of initiation at the T7 primary DNA replication origin
shares similarity with that at the leading DNA strand origin in
mitochondrial replication. In T7 replication, T7 RNA polymerase initiates
transcription at a site upstream of the origin and proceeds through it; T7
DNA polymerase then displaces the RNA polymerase and extends the 3'-
end of the transcript (Kornberg and Baker, 1992). We find that although
the accessory subunit of pol 7 and thioredoxin have low sequence
similarity, our structural model of the C-terminus of the accessory subunit
is highly similar to the structure of thioredoxin (Figure 2.6, right), which

consists of a central core of five [3 strands surrounded by four helices.

73

While the thioredoxin (PBD entry 1t7p) and the C—terminal domain of pol
'y-B do not have identical fold topologies, Dali structural alignment
indicates significant structural similarity between them (Z-score of 2.7
over 71 residues).

Herpes simplex virus type 1 DNA polymerase is also a two-subunit
enzyme (Gottlieb et al., 1990; Hernandez and Lehman, 1990). The catalytic
subunit is a member of the family B DNA polymerase class (Braithwaite
and Ito, 1993). Its accessory subunit, the processivity factor UL42, binds
double-stranded DNA (Gallo et al., 1988). Comparative sequence analysis
reveals that residues 115-174 of UL42 bear 35% identity with the
anticodon binding domain (residues 329-383) of Thermus thermophilus
ProRS (data not shown). Furthermore, UL42 residues 80-289 aligned with
pol 'y-B residues 105-300, and of the 45 of these residues contained in the
structural model, 31% are identical. Residues 129-163 and 202-337 in
UL42 are required for stimulation of DNA polymerase activity (Monahan
et al., 1993), and the former and several specific residues within the latter
are also required for DNA binding by UL42 (Chow and Coen, 1995).
Taken together, these findings support our hypothesis of the role of the
accessory subunit in pol 7 function.

In sum, the C-terminal domain of the accessory subunit of
Drosophila pol 7 reveals a rare Ot/B fold topology, comprising a five-

stranded mixed B sheet surrounded by four on helices, that is found only in

74

the anticodon binding domains of class IIa aaRSs (Cusack et al., 1998).
This structure may endow pol 'y-B with the capacity to bind RNA molecules
with tRNA-like structures, as in aminoacyl tRNA synthetases. This and the
structural similarity with the 5' subunit of E. coli DNA polymerase III and
particularly with thioredoxin suggest that the accessory subunit may play a
dual role in loading pol 7 onto the primer terminus and in enhancing its
processivity. Considering the fact that all of the proteins required for its
replication are the products of nuclear genes, it is not surprising that the
mitochondrion has evolved to require only a single polypeptide to perform
the functions provided by both the multi-subunit clamp loaders and
processivity clamps in procaryotic and nuclear replicative systems.
Furthermore, this conservative mechanism for a replicative cellular DNA
polymerase can likely be generalized to animal pol ’ys, because our
sequence-based structural comparisons indicate that the proposed C-

terminal structure is conserved in the accessory subunits from fly to man.

75

CHAPTER III

SUBUNIT INTERACTIONS IN
DROSOPHILA MITOCHONDRIAL DNA POLYMERASE

76

INTRODUCTION

Replicative DNA polymerases are generally multiprotein complexes
with a polymerase catalytic subunit associated with one or more accessory
subunits that bring additional functions for effective DNA synthesis
(Kornberg and Baker, 1992). It is well known that protein-protein
interactions play an important role in the overall function of a DNA
polymerase complex. For example, the association of E. coli thioredoxin
with T7 DNA polymerase not only increases tremendously the processivity
of the holoenzyme but also enhances significantly the overall activity
(Tabor et al., 1987; Huber et al., 1987). In HIV-RT, the presence of p51 is
essential for the enzymatic activity although the active sites reside in p66
(Kohlstaedt et al., 1992; Jacobo—Molina et al., 1993; Joyce and Steitz,
1994).

The heterodimer of Drosophila mitochondrial DNA polymerase
provides an excellent model for the studies of the function and structure of
a DNA polymerase complex. The large catalytic subunit of pol 7 contains
both DNA polymerase and 3’-5’ exonuclease activities (Kaguni and Olson,
1989; Lewis et al., 1996). The molecular model presented in Chapter 11
suggests that the accessory subunit may have dual roles as a primer
recognition factor and a processivity factor. The Drosophila pol 'y

heterodimer has been reconstituted with recombinant proteins expressed in

77

baculovirus-infected Sf9 cells, and exhibits biochemical properties
indistinguishable by in vitro biochemical assays from those of the native
enzyme isolated from fly embryos (Wang and Kaguni, 1999). However, the
recombinant catalytic subunit alone is nearly devoid of DNA polymerase
activity on gapped DNA substrates, demonstrating that the accessory
subunit may be important for the overall activity of pol ’y.

In this chapter, I have investigated the subunit interactions in the
heterodimer of Drosophila pol 'y. I first monitored subunit assembly in
vivo with recombinant subunits containing various deletion mutations in
Sf9 cells. The results show that the accessory subunit interacts with the
catalytic subunit through multiple contact sites in different regions. Subunit
interactions were investigated further by in vitro protein overlay assays
using purified recombinant proteins expressed in E. coli, and proteins
prepared by coupled in vitro transcription and translation. These results
together suggest that the accessory subunit of Drosophila pol 7 consists of
three domains: an N-terminal domain (aa1-120), an intermediate (M)
domain (aa128-240), and a C-terminal domain (aa254-361). Sequence
analysis indicates that the M-domain shares significant sequence similarity
with the RNase H domain of HIV-RT. One or more motifs in theM-
domain and the C-domain interact mainly with the exo domain (aa1-430)
and part of the spacer region (aa431-750) in the catalytic subunit. Deletion

of the first 100 amino acid residues in the N-domain appears to enhance the

78

interaction between the remainder of the accessory subunit and the catalytic
subunit, suggesting that the N-domain may modulate overall subunit

interactions in Drosophila pol y.

79

MATERIALS AND METHODS

Materials

Enzymes and Proteins - Restriction enzymes were purchased from
GIBCO-BRL and New England BioLabs. Deep Vent PCR DNA polymerase
was purchased from New England BioLabs. Pfu DNA polymerase was
purchased from Stratagene. Polyclonal antibodies were raised against
recombinant pol y—a or B expressed in bacteria and were purified by the
method of Olson et a1. (Olson et at., 1995). The in vitro transcription and
translation (TNT ) kit was purchased from Promega.

Nucleotides and Nucleic Acids - Unlabeled deoxyribonucleoside
triphosphates were purchased from P-L Biochemicals. [y-3ZP] ATP was
purchased from ICN. Plasmid pET-l 1a, pET-16b, and pET-23d DNAs
were purchased from N ovagen. The cDNAs of the two subunits (pol 'Y-Ol
and B) of Drosophila pol 7 were cloned previously into pET-l 1a and pET-
16b vectors (Wang, 1998). Plasmid pETGA-l 1a and pETGA-l6b are pol
y—a expression vectors while pETGB-l 1a and pETGB-16b are pol 'y—B
expression vectors. The pETGA-11a(NL) plasmid was used for expression
of a non—leader (the deletion of amino acid residues 2-9) pol y-a subunit.
Baculovirus transfer vector pVL1392/1393, and linear modified
baculovirus AcMNPV DNA (BaculoGold) were purchased from

PharMingen. Wild type cDNAs pol y—oz and pol 'y—B were cloned

80

separately into transfer vectors and named as pVLGA and pVLGB,
respectively (Wang, 1998). Synthetic oligodeoxynucleotides as indicated in
the text were synthesized in an Applied Biosystems Model 477
oligonucleotide synthesizer.

Bacterial Strains - E. coli XL-l Blue (recAI, endAI, gyrA96, thi,
hst17, supE44, relAI, lac, (F ’poAB, lachZ M15, Tn] 0 (tet’)) ) was used
for PCR cloning. E. coli BL21 DE3 (ompT, rB'MB' ) was used for the
expression of recombinant proteins cloned in pET-l 1a, pET-l6b, and pET-
23d vectors.

Insect Cells and Tissue Culture Medium - Sf-9 (Spodoptera
frugiperda) cells were passed from the stored stocks (Wang, 1998). TC-
100 insect cell culture medium and fetal bovine serum were purchased
from GIBCO-BL.

Chemicals - Isopropylthio-B-D-galactoside (IPTG), nitro blue
tetrazolium (NTB), 5-bromo-4-chlorol-3-indolyl phosphate (BCIP), and
tryptose broth were purchased from Sigma. Sodium metabisulfite and
leupeptin were purchased from the J. T. Baker Chemical Co. and the
Peptide Institute, Minoh-Shi, Japan, respectively. SeaKem ME low melting
agarose was purchased from FMC Bioproducts. Antibiotics amphotericine,
penicillin-G, streptomycin, and the protease inhibitor phenylmethylsulfonyl

fluoride (PMSF) were purchased from Sigma. Ni-NTA resin was

81

purchased from Qiagen. [358]—L—methionine was purchased from ICN

Pharmaceuticals.

Methods

Construction of plasmids for expression of pol x-B mutants in E. coli:

Four pol 'y-B mutants were constructed from pol y—B cDNA in

pETGB-l 1a by polymerase chain reactions (PCR) using different primers

as follows:

pol y—B(1-318):

p01 y—B(l-262)

pol 'y-B(41-355):

pol y—B(lOl-355):

5'-cgctggacatATGAGTCGCATACAACG-3'
NdeI
3'-TATACGACCCGTATGGAatcctaggacc-5'
BamI-II

5'-cgctggacatATgiAGTCGCATACAACG-3'
NdeI
3'-CACC’I'IAACGGACACACatcctaggatc-S'
BamH I

5'-agaccATGgAGCATTGGACACGACTACG-3'
NcoI
3'---GCTACATGGCCTGATAAATgagctcgtg---5'
XhoI

5'-agaccATgigctACTTGCTCTGGTCCCAC-3'
NcoI
3'---GCTACATGGCCTGATAAATgagctcgtg---5'
XhoI

82

Capitalized sequences are the cDNA sequence of pol y—B while lower case
letters are sequences added as linkers for cloning. A typical PCR was
carried out with 50—100 ng pETGB-l 1a plasmid DNA in a 50 ul reaction
mixture under conditions recommended by New England BioLabs for the
Deep Vent enzyme. PCR products of pol y—B(l-318) and pol y—B(1-262)
were gel purified, digested with NdeI and BamHI and inserted into the
vector pET-l 1a or pET-16b to produce N-terminal His-tagged fusion pol
'y—B mutants. Pol y—B(41-355) and pol y—B(101-355) were subcloned into
pET-23d at its NcoI and XhoI sites to produce C-terminal His-tagged
fusion proteins. A fifth mutant pol 'y—B(101-295) was prepared by
digestion of pol y—B(101-355) cDNA with NcoI and EcoRI and then
inserted into pET-23 at its NcoI and HincII sites by standard recombinant

DNA methods.

Bacterial overexpression and preparation of cell lysates:

Recombinant plasmid containing BL21(DE3) cells were grown at 37
0C with aeration in 1 liter of Luria broth containing 100 jig/ml ampicillin.
When the optical density at 600 nm reached 0.8-1.0, IPTG was added to 1
mM, and the culture was incubated for additional four hours. Cells were
harvested by centrifugation at 5000 x g for 15 minutes. Cell pellets were

either stored at -80 °C or were processed directly for Ni-NTA affinity

purification.

83

Purification of His-Tag fusion proteins with Ni-NTA affinity

chromatography:

The purification followed the procedure for denaturing purification
of insoluble proteins recommended by Qiagen. Briefly, cell pellets were
resuspended in Buffer B (8 M urea, 0.1 M Na-phosphate, 0.01 M Tris-HCl,
pH 8.0) at 5 ml per gram of pellet. Cells were lysed by stirring 1 hour at
room temperature. The lysate was collected by centrifugation at 10,000 x g
for 15 min at 4 0C. The lysate was loaded onto a 2 ml Ni-NTA column at a
flow rate of 10-15 ml/hr. The column was then washed with 10 column
volumes of Buffer B, followed by 5 column volumes of Buffer C (8 M
urea, 0.1 M Na-phosphate, 0.01 M Tris-HCl, pH 6.3). The His-tag fusion

proteins were eluted with 300 mM imidazole in Buffer C.

Protein overlay assays:

Recombinant pol 'y—B proteins were expressed in E. coli and purified
by Ni-NTA affinity chromatography as described in the above. Samples
containing 1-2 ug of pol y—B proteins were subjected to 12% SDS-PAGE
and transferred to nitrocellulose membranes PROTRAN (Schleicher &
Schuell, Keene, NH) for 4 to 6 hours at 200 mA in a TE Series Transphor
Electrophoresis Unit (Hoefer Scientific Instruments, San Francisco, CA).
Prestained protein standards (New England BioLabs) were used as

molecular weight markers. The nitrocellulose membranes were trimmed to

84

minimal sizes and blocked with 0.5% BSA (Sigma) in TBST containing 2.5
mM MgCl2 and 2 mM DTT for 45 to 60 minutes at room temperature
followed by a TBST wash for 5 minutes. The blots were then incubated at
4 °C overnight with 35S-Met labeled protein probes as indicated in the text
in TBST containing 1 mM PMSF, 10 mM sodium metabisulfite, 2 mg/ml
leupeptin, and 2 mM DTT. The blots were washed five times with TBST
for 10-15 minutes each at room temperature, and were then dried and
scanned by Phosphor Imager(Molecular Dynamics, CA). Quantitation of
individual bands was performed with Imagequant computer software

(Molecular Dynamics, CA).

Construction of Recombinant Baculoviruses:

Baculovirus transfer vectors containing various pol 'y—(x and pol y—B
mutants were prepared mainly by QuickChange mutagenesis (see Figure
3.1). There are at least three important advantages in QuickChange
mutagenesis. First, it is a one-step cloning procedure; second, it requires
only one PCR cycle per clone to avoid err accumulation; finally, it uses a
highly accurate PCR polymerase, Pfu DNA polymerase, to ensure a very

low intrinsic PCR error rate.

85

Step 1
Plasmid Preparation

Step 2
Temperature Cycling

Step 3
Digestion

Step 4
Transformation

 

 

 

Gene in plasmid with
target site (@i for mutation

Denature the plasmid and anneal the

oligonucleotide primers (,4 containing
the desired mutation (x)

Using the nonstrand-displacing
action of PfuTurbo DNA polymerase.
extend and incorporate the
mutagenic primers resulting

in nicked circular strands

Digest the methylated, nonmutated
parental DNA template with Dpnl

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

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

(from online httpzllwww.stratagene.com/manuals/200518.pdf)

Figure 3.1 Schematic diagram of QuickChange mutagenesis

A typical PCR was carried out in a 50 ul volume with 2 units of Pfu
polymerase in 1x Pfu polymerase buffer (Stratagene, CA) in the presence
of 200 11M of the four dNTPs and 50 ng of DNA template (pVLGA or
pVLGB). A specific primer pair was used for each mutant as indicated in
Table 3.1 and 3.2. The DNA template was first denatured at 95 °C for 45
seconds, followed by 20-25 three-step cycles: 95 0C for 45 seconds, 50 °C
for 1 minute, and 68 °C for 2 minutes per kbp DNA template. The reaction
mixture was then digested with 10 units of Dpnl for 2 hours at 37 °C to
eliminate the methylated parental DNA template. A 2 ul aliquot was used
for transformation of E. coli competent cells (XL-1 blue) by
electroporation using an E. coli Pulser (BioRad, CA). The cloned plasmid
DNA was confirmed by restriction enzyme mapping and DNA sequencing.
Mutant pol 'y—B(1-200) was cloned by standard recombinant DNA methods
into pVL1392 at restriction sites of PstI and BamHI using the PCR product

of the following primer pair:

5’-—gatctgcagATGAGTCGCATACAACGA——3’ PstI site is underlined
3’—~CACCTGCTGTTTCTGAtcgtaggacc-—5’ BamHI site is underlined

Lower case letters are added sequences for purpose of cloning.

87

Table 3.1 PCR PRIMERS FOR CONSTRUCTION OF POL 'y—OL
MUTATIONS BY QUICKCHANGE MUTAGENESIS

pol ydxA(483-533):

5’--GGAGATTAAGGATTCTGGA“CTAGAGGATGACGAAGAGCCG--3’
3’--CCTCTAATTCCTAAGACCTAGATCTCCTACTGCTTCTCGGC--5’

pol Y4LA(483-674):

5’--GGAGATTAAGGATTCTGGA“AATGGACCTTCCTTTCGAGTG-—3’
3’—-CCTCTAATTCCTAAGACCTATTACCTGGAAGGAAAGCTCAC--5’

pol yeaA(536—581):
5’--GGGAAGATGAAATTCTAGAGATTTCGCAGCGATTCAGAAGG--3’
3'--CCCTTCTACTTTAAGATCTC“AAAGCGTCGCTAAGTCTTCC--S’

pol ¥4LA(413—470):

5’--CCAACTGGGAGCGGTACATAAAAGCCTCTTCCTACAGTGG--3'
3’--GGTTGACCCTCGCCATGTATATTCGGAGAAGGATGTCACC--5'

pol y4LA(27—126):
5'——GCAGTAGCGTGAAGATCTTT“ATTGCCAAGGAACAAGTGCAGC——3’
3’—-CGTCATCGCACTTCTAGAAA“TAACGGTTCCTTGTTCACGTCG—-5’

pol yeaA<27-492):

5’--GCAGTAGCGTGAAGATCTTT“AAGTTTCAGCACCTCTATGAT--3’
3’-—CGTCATCGCACTTCTAGAAAATTCAAAGTCGTGGAGATACTA--5’

pol Y4LA(128—404):

5'--TTGAAGAGCACTTCCACAACATTAAATTCCAACTGGGAGCGG--3’
3’-—AACTTCTCGTGAAGGTGTTGTAAATTAAGGTTGACCCTCGCC--5’

pol YLaA(27—742):

5'--GCAGTAGCGTGAAGATCTTTAAATGAATTCACTGGCGAAAAGTGCC--3’
3’--CGTCATCGCACTTCTAGAAAATTACTTAAGTGACCGCTTTTCACGG--5'

A

indicates deletions.

88

Table 3.2 PCR PRIMERS FOR CONSTRUCTION OF POL y—B
MUTATIONS BY QUICKCHANGE MUTAGENESIS

pol y—BA(213-246):
5’—-CCGATGCGCGAACGGGTGAA“CTCTTACACCGCGTTCTG——3’
3’——GGCTACGCGCTTGCCCACTTAGAGAATGTGGCGCAAGAC-—S’

pol y—BA(37-242):
5'--CATGGAAGGGAATACGCCAAAATCGCAATCCCTGCTCTTACAC--3'
3'--GTACCTTCCCTTATGCGGTTTAAGCGTTAGGGACGAGAATGTG--5’

pol ‘y—BA(61-104):
5'--GCCACGAGAGAACCCATAAATACCCACTAGCCATTCCCTATTTG--3'
3'--CGGTGCTCTCTTGGGTATTTAAGGGTGATCGGTAAGGGATAAAC--5'

pol 'y—BA(129-160):
5’--CGGACTTCCTAGTGGAACCAATGCGATTTGGCTGAGGATCTC--3’
3’--GCCTGAAGGATCACCTTGGTAACGCTAAACCGACTCCTAGAG--5'

pol y—BA(264-274):

5’--TGTGGAATTGCCTGTGTGGAA“CTTTGTCAGCATCTGAAACAT-—3’
3’--ACACCTTAACGGACACACCTT“GAAACAGTCGTAGACTTTGTA--5’

pol y—BA(327-346):
5'--ACACTTGTTATAAACGAGCAAACACATAAGCGATGTACCGGAC--3'
3'--TGTGAACAATATTTGCTCGTTAGTGTATTCGCTACATGGCCTG--5'

pol y—B(1-349):
5’--CCATACACATAAGCtGATGTACCGGACT—-3’
3'-—GGTATGTGTATTCGaCTACATGGCCTGA——5’

LSZP:
5’-- CCTTGGCTGCGCACCC(T)GGGCGCCACGAG--3’
3’-- GGAACCGACGCGTGGG(A)CCCGCGGTGCTC--5’
L38VL45V:
5’-- CCAAACTGG(T)TGCAACAGCATTGGACACGAG(C)TACGTCCC--3'
3’-- GGTTTGACC(A)ACGTTGTCGTAACCTGTGCTC(G)ATGCAGGG--5’

A indicates deletions, bold letters replace the letters in parenthesis, and
lower case letters are insertion.

89

Transfer vectors containing various mutants were prepared and
purified by standard CsCl sedimentation methods. Aliquots (5 ug) of pure
plasmid DNA were used with BaculoGoldTM DNA to prepare recombinant
viruses following the procedure in the commercial manuals (Pharmingen,
CA). Individual viruses were amplified from a single plaque to make

recombinant viral stocks for expression and other applications.

Phosphocellulose chromatography for monitoring the assembly of pol y—oc

and pol y—B mutants in SF9 cells:
Cell extracts of ~ 6 x 107 Sf9 cells infected with a pair of pol y—oc

and pol y—B mutant containing recombinant viruses as indicated in the text
were prepared as described by Wang (Wang, 1998) with some
modifications. Briefly, 60 ml of Sf9 cells in two 150 cm2 flasks were co-
infected with recombinant viruses at a multiplicity of infection of five each
at 27 °C for 45-50 hours. The cells were harvested by centrifugation of 5
minutes at 2 krpm in a Beckman TH-4 rotor at room temperature. The
pellets were washed with 30 ml of cold phosphate buffer saline (pH 7.5)
and recentrifuged. The pellets were resuspended on ice in 5 ml HB.28
buffer (15 mM Hepes pH 8.0, 2 mM CaClz, 5 mM KCl, 0.5 mM EDTA,
280 mM sucrose, 1 mM PMSF, 0.5 mM DTT, 10 mM Na bisulfite, 2 ug/ml
leupeptin) and lysed by 20 strokes in a Dounce homogenizer. The

homogenate was centrifuged at 3000 rpm for 7 minutes at 3 °C in an SS34

9O

rotor. This step was repeated twice. The supematants were combined and
centrifuged at 10 krpm for 15 minutes at 3 °C in an SS34 rotor. The
supernatant was centrifuged further at 100,000 x g for 30 minutes at 3 0C
to obtain a soluble cell extract. After the salt concentration of the cell
extract was adjusted to 60-80 mM (measured by conductivity) by addition
of 1M potassium phosphate buffer (pH7.6) ( about 0.6-1.0 ml), the extract
was applied to a 2.5 ml phosphocellulose column packed in a 3 m1 syringe
in the cold room.

The phosphocellulose column was pre-equilibrated with 60 mM
potassium phosphate buffer (60 mM potassium phosphate, pH7.6, 10%
glycerol, 2 mM EDTA, 1 mM PMSF, 0.5 mM DTT, 10 mM Na bisulfite, 2
jig/ml leupeptin). The soluble extract was loaded at 1 column volume per
hour, followed by washes with 5-10 column volumes of 60 mM potassium
phosphate buffer. Proteins were eluted with 5 column volumes of 300 mM
potassium phosphate buffer and collected in 0.4 m1 fractions. Aliquots of
various fractions were analyzed by SDS-PAGE and immunoblot analysis
with pol y subunit—specific polyclonal antibodies. The salt concentrations of
the various fractions were monitored by conductivity measurement with

CDM80 Conductivity Meter (Radiometer, OH).

91

Qel filtration chromatography:
Aliquots (0.2-0.4 ml) of the pool of peak fractions from

phosphocellulose chromatography above were applied to a gel filtration
column which was prepared by packing Sephacryl-SOO Superfine resins
(Pharmacia, Sweden) into a glass column (~1 cm x 80 cm). The column
was equilibrated with running buffer (50 mM Na phosphate, pH 7.5, 10%
glycerol, 1 mM PMSF, 0.5 mM DTT, 10 mM Na bisulfite, 2 jig/ml
leupeptin). The column was calibrated with a mixture of Dextran blue
2000, yeast alcohol dehydrogenase, phosphorylase b, bovine serum
albumin, carbonic anhydratase, and cytochrome C (100-200 ug each).
Eluted fractions were collected in volume of 0.2-0.3 ml. Aliquots of 0.2 ml
were analyzed by SDS-PAGE and immunoblot analysis after precipitation

with TCA in the presence of 2 pg of cytochrome C as a carrier.

92

RESULTS

The accessory subunit interacts through multiple sites with the
catalytic subunit.

In order to identify the regions important for physical interaction
between the two subunits of Drosophila mitochondrial DNA polymerase,
we monitored complex assembly between individual mutants of the
accessory subunit and the wild type catalytic subunit coexpressed in Sf9
cells by infection with recombinant viruses. An example is presented in
Figure 3.2 to show the interaction between mutant pol ‘y—B(l-318) and pol
y—or. When an extract from Sf9 cells expressing the two recombinant
proteins was applied to a phosphocellulose column, mutant pol y—B(l-3l8)
was coeluted with the catalytic subunit at ~200 mM KPO4 (fraction 10 in
Figure 3.2A). Gel filtration chromatography of this peak fraction
confirmed the formation of a complex (fractions 19-23 in Figure 3.2B)
larger than the free catalytic subunit (fractions 25-29 in Figure 3.2B). A
relatively large portion of free catalytic subunit indicates a dissociation of
the two proteins during chromatography, suggesting a weak interaction
between mutant pol y—B(l-3l8) and pol 'y—a. As demonstrated in Figure
3.2C, the interaction between pol 'y—B(l-3l8) and pol 'Y-Ol is much weaker

than the interaction of pol y—B with pol 'y—or. In the presence of a 2 fold

93

A) In p. uvoBlotoomparrpxcmlom ahmov Move
T 3 6 8101214161821

poly—(1Q - Pour—“'-

pol yB(l-318)I; I" .-

B) In u ovoBlororbZZOO ystpmov xnmu ounymamw
1719 21 23 25 27 29 31 33 35 37 39 41 L

POW-o. - -1.”- ’ ’ fl

 

 

 

 

 

 

P01 74311318)

 

 

 

X) In u ovoBlotoormoat-noxemiocn ahmov Mama
w m8 mmvxe «MOM-B

T36 8101214161820

 

poly—a...— ub-n—v-wm‘ 7*

 

 

a~-

 

 

pol H3
pol 74311-318) g -

 

Figure 3.2. Interaction between pol y—oc and pol 'y—B(l-318)
in Sf9 cells. Cell extracts from about 60 millions of Sf9 cells infected
with two recombinant viruses expressing pol y—a and pol y—B(1-318) (A
and B) or three recombinant viruses expressing pol 'y—or, pol ”y-B, and pol
y—B(1-318) (C) were analyzed by phosphocellulose chromatography as
described in Methods. The peak sample (fractions 9-11) in (A) was
analyzed by S200 gel filtration chromatography (B). Aliquots from elution
fractions were analyzed by 10% SDS-PAGE and immunoblot with subunit-
specific polyclonal antibodies. Numbers on the top of each lane are the
number of individual elution fractions. T is for total proteins from 10,000

Sf9 cells.

94

excess pol y—B, pol y—B(1-318) no longer coeluted with pol y—oc upon
phosphocellulose chromatography whereas wild type pol y—B was coeluted
with pol 'y—oc at ~200 mM KPO4, suggesting the interaction between pol 7—
on and mutant pol 'y—B(1-318) is reduced to less than 30%. These results
also indicate that the elution profile upon phosphocellulose chromatography
can be used to monitor directly the interactions between an individual pol
y—B mutant and pol 'y—oc when they are coexpressed in Sf9 cells. In fact, in
the absence of the catalytic subunit, neither the wild type nor any of the
mutants of the accessory subunit bound to the phosphocellulose column
(data not shown). In this way, if the expression level of various pol 'y—B

‘ mutants is kept at the same level, their relative amounts in the peak
fractions of pol y—or coeluted from phosphocellulose columns will represent
the relative strength of a particular mutant interacting with pol 'y—(x, when
the level of pol y—or in the elution fractions is also the same.

Various mutants of pol y—B were expressed pairwise with the
catalytic subunit in Sf9 cells. In most cases, the expression level of these
recombinant protein pairs was within 2-fold variation except for the
mutant pol y—BA(37-242) (lane 2 in Figure 3.3A). The interactions of these
pol 'y—B mutants with pol ‘y—oc were monitored by phosphocellulose
chromatography. The immunoblotting results (Figure 3.3B) of the peak

fractions eluted from phosphocellulose columns show various levels of the

95

A) lmmunoblot of recombinant proteins expressed in Sf9 cells

12345678910

l:——-—---—:l

 

 

 

5..

“v
“r

 

 

 

B) lmmunoblot of phosphocellulose peak fractions

 

 

 

 

 

   

   
  

    

C)
A {3 6?
z =13 55
E > N 7‘ ‘-
‘n ‘3 - N a
100 _1 5 8
> A N <
'6 co <0 21’ en.
.5 80 ‘3 a a
C A I
2% 60 :2 $2 5'3
2 v A or) 91’
fig 40 :38 i a
2:5 «“3 “1
=- 20 2,“:
2 nfl
2 0

Figure 3.3 Interactions of pol y—or with pol y—B mutants in
819 cells. Soluble cytoplasmic extracts from Sf9 cells coinfected with
recombinant baculoviruses expressing pol y—or and various pol y—B mutants
as indicated were chromatographed on phosphocellulose as described under
“Methods”. The expression level of individual recombinant proteins was
monitored by 12% SDS-PAGE and immunoblotting of total cell lysates
(A). The peak fractions eluted at ~200 mM KPO4 were pooled and
subjected to 10% SDS-PAGE and immunoblot analysis (B). The relative
strength of individual pol y—B mutant interacting with pol ‘Y—a is plotted
according to the relative intensities of individual pol y—B bands after
normalization of the pol 'y—or bands (C). Immunoblotting assays were
conducted separately with subunit-specific polyclonal antibodies.

96

pol y—B mutants in the presence of similar levels of pol y—oc, indicating
different affinities for interaction of pol 'y—B mutants with pol y—(x. Mutant
pol y—BA(37-242) (lane 2 in Figure 3.3B) was not observed in the elution
fractions because of its reproducibly very low level of expression, which
may suggest that subunit interaction is also important for the stability of
pol 'y—B per se. Mutant pol y—B(l-200) (lane 3 in Figure 3.3) shows a very
weak interaction (<10%) with pol y—OL. Other mutants exhibit interactions
from 20% to 80% as compared to wild type pol y—B, while pol y—B(1-349)
(lane 10 in Figure 3.3B &C) shows the strongest interaction with pol y—or
(96%). The result that mutant pol 'y—B(1-3l8) retains only 20-30% of wild
type interaction with pol y—or (lanes 4 and 7 in Figure 3.3B &C) is in
agreement with the estimation of <30% by the competitive interaction assay
shown in Figure 3.2C. Results for these mutants and others tested are
summarized in Figure 3.4. Mutant BA(2-9), having the putative
mitochondrial target sequence interrupted, shows no interaction with the
catalytic subunit, indicating that protein sorting is important for in vivo
subunit interaction in mitochondrial DNA polymerase. The extremely weak
interaction between mutant pol y—B(1-200) and the catalytic subunit
indicates that the deleted region (residues 201-361) is important for
protein-protein interaction whereas the region of residues 1-200 provides

limited subunit-subunit interaction (<10%). Furthermore, mutant pol y—

97

pol Hi mutants interactions with

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

palm
[ 1 wild type strong
I . 1 L52P strong
1 1 11(2-9) "OM
1——1 J a A(61-104) strong
L 1 1 1 A(129-160) weak
1 1 1 1 A(213-246) weak
L H 1 A(264-274) mu
1 j n A(327-346) weak
L j 1.349 strong
L L 1-318 weak
L .. J 1-318/L38VL45V weak
m . 1 1-318/L52P weak
L 1 1-200 voryweak
[—1 1 1 1(37-241) none
Figure 3.4 Summary of interactions between pol y—or and pol

y—B mutants in Sf9 cells. The interactions were monitored by
phosphocellulose chromatography as described in Methods. Individual pol
y—B mutants are presented in schematic boxes. Dots indicate residue
substitution mutations. Lines are for deletion mutations.

98

BA(129-l60) also shows reduced subunit interaction. In contrast, both
mutant pol y—BA(61-104) and mutant pol y—B(1-349) retain strong
interaction with pol y—oc. These results suggest that the interaction motifs
are localized mainly within residues 129-349 of the accessory subunit.
Within this region of 221 amino acid residues, deletions such as A(213-
246), A(264-274), A(327-346), and A(319-361) (B(1-318) show various
reductions in subunit interaction, indicating multiple sites are involved in
subunit-subunit interaction in the Drosophila pol y heterodimer.

In order to test if the putative leucine zipper (Lx6Lx6Lx6l) at the
position of residues 38-59 is involved in subunit-subunit interaction, the
interaction of pol y—oc with several mutants (pol 'y—B/L52P, pol y—B(1-
318)/L52P, pol y—B(1-318)/L38VL45V) with leucine residues changed into
proline or valine was monitored. These amino acid residue-substitution
mutations show no significant effects on protein-protein interaction (Figure
3.4 and Figure 3.3), indicating that the leucine zipper motif plays no role

in subunit-subunit interaction in the Drosophila pol 7 complex.

The accessory subunit of Drosophila pol 7 possibly consists of
three domains with different functions.

The above subunit interaction results have shown that the accessory
subunit interacts with the catalytic subunit through several contact sites

located largely in the region of residues 129-349. Based on these results

99

and the molecular modeling presented in Chapter II, I divide the accessory
subunit into three domains shown in Figure 3.5. The N-terminal region
(aa1-l20, N-domain) in its entirety may not be important for subunit
interaction because a deletion of 45 amino acid residues in the middle of
this domain in mutant BA(61-104) did not affect the subunit interaction
(Figure 3.4). The C-terminal region (aa254-361, C-domain) shows
sequence homology (~30% identity) with the anticodon binding domain of
class Ila aminoacyl-tRNA synthetases and was predicted to fold similarly to
provide the accessory subunit with the property of binding to specialized
RNA primers (Fan et al., 1999). The intermediate region (aa128-240)
forms the M-domain. Both the C- and M-domains are involved in subunit
interactions. The secondary structures predicted by the PHD program
within different domains fit quite well with the three-domainal assignment
based on the results of the biochemical subunit interaction studies (Figure
3.5). According to the PHD prediction, the N-domain comprises three
helices (H1, H2, and H3) and three beta strands (S1, S2, and S3). The first
helix H1 may function as the mitochondrial targeting signal whereas H2
and H3 possibly form a helix-loop-helix (l-ILH) motif which is a common
DNA binding structure. The M-domain consists of two helices and four
beta strands. The M-domain is connected with the C-domain by a short

helix H6.

100

 

 

 

 

 

 

 

 

 

 

Figure 3.5 Proposed structural domains of Drosophila pol 'y—B.
Three structural domains are proposed based on results from Figure 3.4
and previous computer-modeling (Fan, et at, 1999). The amino terminal
region of 120 amino acid residues forms the N-domain. Residues 128-240
form the M-domain in the middle of the polypeptide. The carboxyl
terminal region of residues 254—361 is the C-domain as predicted by
computer modeling. The secondary structural elements are based on results
from PHD predication. Bars represent B-strands and helices are for alpha
helices. Lines represent undefined structural elements or loops.

101

In order to test if the assigned domains can form independent
structures, the same fold recognition analysis used for development of the
C-terminal structural model was applied independently to the N-domain
and the M-domain. The folding of the N-terminal domain resembles the
structure of MarA (Z score of 5.6 over the threshold value of 5.0 +/- 1.0),
a member of the AraC transcriptional activator family with a helix-loop-
helix (HLH) DNA binding motif (Rhee et al., 1998). The significance of
the HLH motif in the accessory subunit is not clear at present. The fold
recognition analysis did not provide good matches for the M-domain. The
top ten candidates have Z-scores close to the threshold value but not higher
(Table 3.3). However, the RNase H domain of HIV-RT shows significant
sequence homology (30% identity over 102 amino acid residues) with the
M-domain by the structure-based sequence alignment of the fold
recognition analysis (not shown). Sequence analysis conducted with the
GCG program (Figure 3.6) reveals that the M-domain of the accessory
subunit shares 27% amino acid sequence identity and 39% similarity with
the whole RNase domain of HIV-RT. In addition, three acidic amino acid
residues (shadowed in Figure 3.6) which are important for RNase H
activity are also conserved or replaced by similar residues in the M-
domain. This level of sequence homology between the M-domain and the
RNase H domain of HIV-RT is comparable to the sequence homology

conserved among known RNase H homologs. For example, a

102

ucwficmflam mnu CH mmnnflmmn Hmoflucwflfl mo manucmunwm NBHBZmQH
UHOM wsu Ou vwcmflam who? umzu wocwdem >HwSU EOHM mwdwflmwn mo Hwnfidc one mewzmq

o.H 1\+ o.m mo wHOUmIN m we @Hogmwuau wucmwflmsoo mflu .Uonqu mflnu Sufi: .mGHOM

Ham mo As3onm uocv mmHOUm 3mu mo Cofludnflnumflfi may mcflm: Umudmfioo ma wuoumrN one mmOUmIN

nonunucwen sumac cflmuonm zHamu

ma or mm. NMJQEOU Am.mm.¢.m.0.mv mmfiBDE mEdSMHmomU
DH mw mm. mmdmmmm24MBd>EmOm BmSthBIQVZOHmBmE
NH mh Hm. HH mmddu Zdzbm AHOHO ammo .fimQV amaldqm
ma Nb mm. Emmeiz mmB m0 mmDEUDmEm A<ZOHmZmSHO mmmme
om NOH mm. mmmm>mm >Hm m0 ZHdEOQ m mmdmdUDZOmHm
mm mm mm. mm49¢mmm0mmom>m mBDD >Hm
mm vm mh. ZHmflomm XHmB<2 HH1>QBm mmE m0 mmDEUDmBm
mm mm Ho. Ami< mZHémUV mm<QHm0904440149mm
hm Hm ma. Am1< mZHfimUV mmdQHmOBU¢J<OI<Bmm
vm om mm. Dds %m .0325 ho ZOHB<ZHEMMEMD mmDBUDmBm
552%: :55: mmeomd 22:56me

mQHDUUSHUW QM 03...“ MO meMCHmuHOOU wflu HOW mGUOU xgflm MUMQ CHwUOHnm Odom

d ZOUH OH
4 BEmH m
d mADH m
0mmH b
d mmma m
< BDQH m
m>bH v
< AOmH m
d AUmH N
d DEDH H

Z—<=U GAOR M2;

Aautism.m_o:.BE-ooc.333\\”9E 85:8 3 Began:

an» Add no 222092 was no medammm zoEzooomm 30m 3 2%...

103

structure-based sequence alignment has shown previously that E. coli
RNase H shares 28% identity with a yeast RN ase H enzyme, 27% with
RNase H of Moloney murine leukemia virus, 19% with RNase H

of Rous sarcoma virus, and 24% with the RNase H domain of HIV-RT
(Yang, et al., 1990). The sequence homology suggests that the M-domain
may have a structure similar to that of the RNase H domain of HIV-RT.
Thus it is plausible that the accessory subunit is a multifunctional protein

with different functions for different domains.

* w t * i:
RT—RH 4 3 2 EPIVGAETF GAANRETKLGKAGYVTNKGRQKWP . . LTNTTN . . . . Q 47 5
| l | I | - | = - | = = | |

1301 7‘8 127 EPHRALEHFY . . . QRESKIWWMRLSSNPSRYRIVPCDLAEDLNPNDYQ l7 3
'k * 'k * *

RT—RH 47 6 K LQAIY ..... LALQDSGLEVNIVT QYALGIIQAQPDKSESELVN. 519

=- -|-= |||| =| H II

po 1 7‘13 1 7 4 A IRTSYGDAGEVAVEQLSL . VRIVD DFRL ...... PDARTGETVQP 2 1 6
* * * *

RT— RH 5 2 O QI I EQLIKKEKVYLAWVPAHKGIGGNEQVDKLVSAGIRKIL 5 6 0

[301 Y‘B 2 l 7 TVIRSVIELETTTCALLLDGCDHGRDSQ . SLLLHRVLA 2 5 3

Figure 3.6 Sequence homology between the M-domain of the accessory

subunit (pol y—B) of Drosophila mitochondrial DNA polymerase and
the RNase H domain (RT-RH) of HIV reverse transcriptase. The

alignment was based on the Gap alignment of GCG program and adjusted
manually. I) indicates identical amino acid residues. :) indicates highly
similar amino acid residues. .) is for similar amino acid residues. The two
sequences share 26.6% identity and 38.4% similarity over the alignmed
region. Every ten residues are marked on the top by (*). Three acidic
residues important for RT-RH activity are shadowed. The sequences

are numbered as in pol y—B and HIV reverse transcriptase, respectively.

104

The interactions between residues 1-490 of the catalytic subunit
and the C- and M-domains of the accessory subunit are the main
force of subunit interactions in Drosophila pol ”y

Reconstitution of altered holoenzymes with various mutants of the
catalytic subunit and the wild type accessory subunit in Sf9 cells was also
monitored by phosphocellulose chromatography as described in the above
section. The results are summarized in Figure 3.7B. Several examples of
these interactions are shown in Figure 3.7A and in lanes indicated by "W"
in Figure 3.8. The relative strength of individual pol Y-(l mutants
interacting with the accessory subunit has not been tested directly because
all of pol Y—(I mutants were eluted at overlapping salt concentrations in the
presence or absence of the accessory subunit during phosphocellulose
chromatography (data not shown). All of the mutants including large
deletions at different regions spanning the entire pol 'y—oc polypeptide show
interaction with the accessory subunit except mutant 0tA(2-9) which has the
mitochondrial target sequence deleted in Figure 3.7. This re-enforces the
fact that protein sorting plays an important role in in vivo reconstitution of
Drosophila mitochondrial DNA polymerase. These results argue that the
catalytic subunit interacts with the accessory subunit through multiple
contacts.

Based on sequence comparison with the Klenow fragment of E. coli

DNA pol I, the catalytic subunit of Drosophila pol y can be divided

105

A) lmmunoblot data of elution fractions from phosphocellulose chromatography

 

 

poly—B-> - "" "I“ “-

 

 

 

B) Summary of interactions between pol y—B and pol y—a mutants in Sf9 cells

pol 7-0L mutants interactions with

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

pol r-B
[ fl WT(1-1145) yes
J 1 A(2-9) none
11 1 I A(27-126) yes
Cfi 1 J A(127-404) yes
L fl I 1 A(413-470) yes
[ 1 r l A(483-533)(A12) yes
[ 1 I l A(536-581)(AH) yes
L 1 I l A(483-674)(AS) yes
1 fl (1-490) yes
1"] L J A(27—492) yes

 

Figure 3.7 Interactions of various pol y—a mutants with pol y—B in Sf9
cells. Wild type pol y—B and pol y—a mutants were expressed in pairs in Sf9
cells. The interactions were monitored by phosphocellulose chromatography
as described in the text. Representative peak elution fractions are shown in
(A). The results of various pol y—a mutants interacting with pol y—B are
summarized in (B). Individual pol y—a mutants are represented by schematic
boxes, lines represent deletion mutations.

106

approximately into three domains: the exonuclease (exo) domain (aa1-430),
the spacer domain (aa431-750), and the polymerase (pol) domain (aa751-
1145). Several mutants with deletions in the spacer domain were
investigated to test if the spacer domain is important for subunit
interactions, because the corresponding part in T7 DNA polymerase
provides the motif for interaction with thioredoxin. The putative leucine
zipper at position 483-53 in the spacer and the residues around this putative
leucine zipper are not essential for subunit interactions because mutants pol
y—OLA(483-533), A(423-470), A(536-581) and A(483-674) show interaction
with pol y—B as shown in Figure 3.7. The results are consistent with the
above result that the leucine zipper of pol y—B is not involved in subunit
interactions. Furthermore, residues 1-419 at the N—terminus of pol 'y—Ot are
sufficient to allow complex formation with pol y—B. Interestingly, mutant
pol 'y—ocA(27-492) also forms a complex with pol 'y—B. The interactions of
pol y—B with pol y—or(l-490) and pol y—orA(27-492) were investigated
further (see below).

From the results described in the previous section, we have located
the interaction regions in the accessory subunit to the M- and C-domains.
In order to understand how the domains of pol 'y—a interact with the
domains of pol 'y—B, we have monitored by phosphocellulose
chromatography the reconstitution of three mutated catalytic subunit

proteins with the wild type and two mutated accessory subunit proteins

107

 

expressed in pairs in Sf9 cells. Mutant pol y—BA(129-160) with a deletion r
of 32 amino acid residues in the M-domain and mutant pol y—B(1-318) with
a deletion of 43 residues in the C-domain were chosen for the tests because
these deletions likely disrupt the conformation of the individual domain
where the deletion occurs. Therefore, we can exclude the interactions by
the M-domain and monitor only the interactions by the C-domain of mutant
pol y—BA(129-160) with the pol y—oc proteins. Similarly, we monitor only
the interactions mediated by the M-domain of mutant pol 'y—B(1-318). As a
control, wild type pol 'y—a was tested pairwise with pol y—B (W), pol y—
BA(129-160) (AM), and pol y—B(1-318) (AC) (Figure 3.8B). As indicated
by the relative intensities of the bands for the three pol 'y—B proteins, pol
y—BA(129-160) and pol y—B(1-318) show weaker interaction with pol y—or
than pol y—B does, about 60% and 23%, respectively (Figure 3.8C).
Similarly, the interaction between BA(129-161) or B(1-318) and pol 'y—
0t(1-490) is much weaker, only about 20% of that between pol y—a(1-490)
and pol y—B (Figure 3.8B & C). These results indicate that the exo domain
and part of the spacer (aa431-490) of the catalytic subunit interact with
both the M- and C-domains of the accessory subunit because subunit
interaction was reduced by 80% through disruption of either domain in the
accessory subunit. Similar results were obtained for pol y—OLA(483-674),
which deletes most of the spacer of the catalytic subunit. The results

suggest that the region of aa675-1145, which contains mainly the pol

108

A)

W —N—._—_1C POW-l3 B-(l -361)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

AM F' 7 1 Q poly— BA(129 160)
1c fi . poly-1m- 318)
1
cm %— pawn-1145)
1—-————1 r————j aA(483-674)
r—-———'1 01(1-490)
l'l fl 4J aA(27-492)
B) pol y—or 01(1—490) aA(483-674) aA(27-492)
WAMAC WAM AC WAC AM WAM AC
“M"? an:
‘*%“nn
" fi . :5 It , -
l ”A I. 17 ._. “3" .1123 .
C)
«5 100
a = 80
6
Es 60
g E 40
13.5 20
a
E o

 

pol y-a (1(1—490) aA(483-674) aA(27-492)

Figure 3.8 Domain interactions between Drosophila pol y—oc
and pol y—B coexpressed in Sf9 cells. Recombinant proteins pol y—B
(W), pol y—BA(129—l60) (AM), pol 'y—B(1-318) (AC) were examined for
interaction with pol ‘Y—(X, pol y—a(l-490), pol y—orA(483-674), and pol y—
aA(27-492) by pairwise coinfections in Sf9 cells. Reconstitution of
heterodimeric complex was monitored by phosphocellulose
chromatography. Aliquots of elution peak fractions were analyzed by
immunoblotting. The bands shown by immunoblot were scanned. The
relative intensities of the bands of pol y—B proteins are converted into
relative strength of protein-protein interaction after normalization to the
pol y—oc bands within each group assuming 100% interaction for wild type
pol y—B. The interactions are not compared between different groups.

109

domain of the catalytic subunit, may not provide additional contacts with
the accessory subunit. These data argue that the interactions between the C-
and M-domains of pol 'y—B and the region of aa1-490 of pol y—a are the
dominant force for subunit interactions. In contrast, the same mutations in
the M- and C- domains did not show significant differences (about 10%
reduction) for the interactions of these pol y—B mutants with pol y—OLA(27-
492). This suggests that the region of aa493-1146 (part of the spacer and
the whole pol domain) of the catalytic subunit might interact with residues
in the N-domain of the accessory subunit in the absence of the exo domain.
Alternatively, because the pattern of pol y—ocA(27-492) interaction with
these three pol y—B proteins is different from that of the wild type pol y—or
interaction with those proteins , the deletion of residues 27-492 might
change the conformation of the remainder of the catalytic subunit, and
result in different interactions than those which occur in the native
Drosophila pol y complex. In either case, the results suggest that the
interactions between the region of aa1-490 of the catalytic subunit and the
C- and M-domains of the accessory subunit are the dominant force for

subunit interactions in Drosophila pol "y.

110

The N-domain of the accessory subunit may modulate subunit
interactions through weak interaction with the pol domain of the
catalytic subunit.

The above results are based on in vivo reconstitution of altered
holoenzymes with recombinant proteins in Sf9 cells. I have also tested in
vitro physical interactions with purified recombinant proteins by protein
overlay assays, in order to confirm the in vivo results and to investigate
further the involvement of the N-domain of the accessory subunit in
subunit interactions. As shown in Figure 3.9, different pol 'y—B mutants
were expressed in E. coli as His-tagged fusion proteins and were purified
by Ni-NTA affinity chromatography. These purified recombinant proteins
were then resolved by 12% SDS-PAGE and transferred to nitrocellulose
membranes. After the proteins were renatured as described under
“Methods”, the membranes were probed with 35S-labeled pol ‘Y-(X.(9—l 146)
protein prepared by in vitro transcription/translation with rabbit
reticulocyte lysate. After phosphor image scanning, bands detected reveal
the interactions between the corresponding pol y—B protein and the probe.
The relative intensities of different bands resulting from interaction
between the 35S—labeled pol y—oc probe and the pol y—B proteins on the
membrane indicate the relative strength of the interactions. As shown in

Figure 3.9B, the nonleader pol 'y—OL(9-1146) interacts with his-tagged wild

lll

A) Proteins on the membranes

 

 

 

 

 

 

 

 

 

 

l - fl 0] — 1—361
2 —| - go] ;—EE41—35)5) 1: a
3 —1:1- pol y—B(101—355) 2- 5
4 —1=—o poly—[3(101—295) 3— -6
51 fi— pol 'y— 51—318; 4.-
6 01::— pol y— 1—262 -

0 HIS-tag Irnmunoblot

B) Protein overlay assay

relative strength of
interaction

 

4 5 6

.a
N
00

proteins on membrane

Figure 3.9 In vitro interactions of pol y—a(9-l 145) with pol y—B

mutants by protein overlay assay. One microgram of each purified His-
tagged pol y—B proteins was resolved by 12% SDS-PAGE and transferred
to nitrocellulose membrane. The proteins were then renatured and blotted

with 35S-labeled pol y—a(9-1 145) prepared by in vitro transcription/trans-
lation as described in “Methods”. The membrane was scanned by Phos-
phor Imager. A) schematic diagram of His-tagged pol y—B proteins and
immunoblot results. Lines represent deletions. B) results of protein over-
lay assay. The numbers under each lane represent the relative intensities
of individual bands in the same order. Data analysis was performed with
ImagQuant software. The relative strength of protein-protein interactions
is plotted based on band intensity.

112

type pol y—B on the membrane, indicating that the mitochondrial targeting
sequence is not directly involved in subunit-subunit interaction, which
might be misinterpreted by in vivo reconstitution-based results shown in
Figure 3.4 and Figure 3.7. In agreement with in vivo subunit interaction
results described in the above, the interaction of pol y—OL(9-1145) with pol
y—B (1-318) is only about 14% of that of pol 'y—a(9-1145) with wild type
pol y—B, confirming the conclusion that the C-terminal domain is important
for subunit-subunit interaction. Furthermore, the interaction between pol
y—B(101-295) and pol y—a(9-1145) (18% in Figure 3.9B) is nearly the
same as that of pol y—B(1-318) with pol y—oc(9—1145), indicating that the
N-domain (at least residues 1-100) is not involved directly in subunit
interactions. The results also indicate that the M-domain (here including
residues 101-295) is sufficient for limited subunit interaction (18%). This
is also in agreement with the earlier conclusion that the M-domain is
important for subunit interactions. Mutant pol y—B(1-262) also shows
reduced interaction (40%) with pol 'y—or(9-1145); however, it is not clear
why deletion of 100 residues at the C-terminus has less effect than deletion
of 43 residues in mutant pol y—B(1-318).

In contrast, deletions of the N-terminal domain appear to enhance
protein-protein interaction because mutant pol 'y—B(4l—355) and pol y—
B(101-355) show 3 to 4 fold stronger interactions with pol 'y—0t(9—l 145)

than does the wild type pol 'y—B (Figure 3.9B). This implies that the

113

presence of the N-domain prevents the two subunits from approaching each
other closely, possibly by space exclusion with parts other than the
interaction regions in pol y—or. If this is true, the negative effects (or the
enhancement by deletions) of the N-terminal domain will be less significant
for smaller mutated pol y—oc probes which still retain the interaction
regions. The results shown in Figure 3.10 meet this expectation for probes
pol y—a (460-1145), pol 'y—OL (1-360), and pol y—or (460-743). The
enhancement of A(1-100) in pol y—B gradually decreases from 4-fold to 3-
fold to 2-fold to 1-fold as the masses of these probes decrease. The
enhancement of A(1-40) decreases from 3-fold to 2-fold to 1-fold in a
similar way. These results imply that the N-domain of pol y—B may play a
role to modulate subunit-subunit interaction in Drosophila pol y, either to
keep the two subunits in the required orientations for productive protein-
protein interaction or to maintain the complex in an active conformation.
In addition, these three probes (pol y—or (460-1146), pol 'y—oc (1-360) and
pol y—oc (460-743)) also show reduced interactions with pol y—B(1—318)
and pol 'y—BA(1-262), once again supporting the conclusion that the C-
domain of pol 'y—B is important for subunit interactions.

Probe pol y—or (742-1145) shows an entirely different pattern of
interactions with these pol y—B proteins. The deletions at the C-domain of
pol 'y—B show little effect on protein-protein interactions between pol 'y—0t

(742-1145) and the mutants pol 'y—B(l-262) or pol 'y—B(1-318) on the

114

A)

B) Protein overlay assays

10'9““ “@500“? O"!

35S-labeled probes

pol y—a(9-360)

 

 

 

a-pol
a-s/pol
a-s

2....—
3....

500 -

- pol y—a(742-1 145)

 

 

fl pol y—a(460-1 145)

 

 

 

 

 

 

 

pol 'y—a(460-743)

 

 

 

 

 

 

 

 

 

a-s/pol or-exo a-S or-pol
7 '1 , . : E . ,, ,1
100 100 100 100
217 21 101 43 127 10 160 90
255 22 I58 67 130 22 86 105
16 57\ 8 41

 

400
300
200

100

 

a-s/pol

 

 

a-exo

 

(1-8

Figure 3.10 In vitro interactions of pol y—a mutants with pol y—B
and mutants. Four pol y—a mutants were used as probes to interact

with His tagged pol y—B proteins (labeled as 1-6) on nitrocellulose
membrane as described in F igure-3.9. A) schematic diagram of four pol y—
a proteins. Lines represent deletions. B) results of protein overlay assays.
As a control, the results of pol y—or(9-1145)(or) is also presented in the plot.

115

membrane, indicating that there is little physical interaction between the C-
domain of pol y—B and the pol domain of pol 'y—OL, in agreement with the in
vivo results obtained for pol y—orA(27-492) and pol 'y—B(1-318) in Sf9
cells. Pol y—oc(742-1145) shows weak interaction with residues 41-100 at
the N -terminus of pol y—B, because its interaction with pol y—B(41-355)
drops from 160% to 86% for interaction with pol 'y—B(101-355). Pol y—
01(742-1 145) may also interact weakly with pol y—B(101-295) (41% in
Figure 3.10B). However, the different pattern of pol y—OL(742-1145)
interactions with the pol y—B mutants as compared to pol y—or(9-1145)
might again indicate that the deletion of residues 1-742 may change the
conformation of the rest of the polypeptide, and result in interactions

different from those observed in the native Drosophila pol y.

116

DISCUSSION

The results presented here demonstrate the complexity of subunit
interactions in Drosophila mitochondrial DNA polymerase. The physical
interaction between the two subunits involves multiple contact sites, which
may explain the biochemical finding that subunit association in the native
pol y is so tight that separation can not be achieved without denaturation of
the enzyme (Olson et al., 1995). These physical interactions are likely
required to keep the catalytic subunit in an active conformation, because
the recombinant catalytic subunit purified either from E. coli or from
baculovirus-infected Sf9 cells exhibits very low catalytic activity, whereas
the recombinant holoenzyme complex isolated from baculovirus-infected
Sf9 cells is indistinguishable from the native pol y (Wang, 1998).

The current studies on subunit interactions have provided some
insight about the structure of the accessory subunit, which is emerging as
one of the hot topics in this area of research. The results suggest that the
accessory subunit can be divided into three domains: an N-domain of
residues 1-120, an M-domain of residues 128-240, and a C-domain of
residues 254-361. Both the M- and C-domains are important for
interactions with the exo domain and part of the spacer of the catalytic
subunit. The N-terminal domain is not involved in direct interactions with

these regions in pol y—a but its presence may play a role in coordinating

117

the major interactions of the C- and M-domains of pol 'y—B with pol y—OL.
The N-domain may interact with the pol domain of pol y—oc in a way
different from the interactions of the C- and M-domains with the exo
domain and spacer in pol y—oc. Thelatter likely contributes the dominant
force in subunit interactions in the pol 7 complex, whereas the former may
hold the two subunits in the required orientation or conformation for
formation of an active enzyme complex. These hypotheses are in agreement
with preliminary data from this laboratory that a complex formed by pol
y—OL and pol y—BA(6l-104) in Sf9 cells has virtually no DNA polymerase
activity. Other mutants described in this study with deletions in the M- and
C-domains of pol y—B show weak interaction with pol 'y—OL, and can not be
purified as functional complexes with pol y—oc for further biochemical
characterization.

It was not a surprise to us that the C-terminal domain of the
accessory subunit contributes to the subunit association. Previously, we
used computational modeling to predict that the C-domain likely folds
similarly to the anticodon domain of class Ila amino acyl-tRN A synthetases
(Fan et al., 1999). In class IIa aminoacyl-tRN A synthetases, the anticodon
binding domain is also involved in the dimerization which is essential in
most cases for the activity of class IIa aminoacyl-tRNA synthetases (for
reviews see (Amez and Morase, 1997; Cusack, 1995)). Several deletions in

the C-domain as well as in the M-domain reduce the subunit interaction,

118

suggesting that each domain has multiple regions interacting with pol y—a,
or that the conformation of the individual domain is very important for
subunit interactions. Unfortunately, the nature of most mutations in the
accessory subunit makes it impossible to distinguish the direct involvement
of the deleted regions in subunit-subunit interaction from the effects of the
structural changes of the domains caused by these deletions. According to
the predicted secondary structures in Figure 3.5, most mutations in the
accessory subunit include deletions of one or two beta strands and alpha
helices which most likely interrupt the domainal structure. An exception is
deletion A(264-274) which includes a part of a loop plus less than two turns
of an alpha helix (H7 in Figure 3.5) in the C-domain. So mutant BA(264-
274) may still have a C-terminal domain folded similarly but nonetheless
shows a weak interaction with the catalytic subunit, suggesting that residues
aa264-274 are involved directly in subunit association. Interestingly, a
structurally corresponding region (aa31-36) in thioredoxin, presumably the
counterpart of pol 'y—B in T7 DNA polymerase complex, is actually
involved in interaction with T7 DNA polymerase as shown in the
determined X-ray structure.

It is noteworthy that the sequence homology between the M-domain
and RN ase H domain of HIV-RT is an unique property of the accessory
subunit of Drosophila mitochondrial DNA polymerase; this similarity is

not conserved in other accessory subunit homologs of animal mitochondrial

119

 

DNA polymerases. As noted earlier, the most conserved region (or
domain) is the C-domain among all the identified accessory subunit
homologs (Wang, 1998; Fan et al., 1999). This may suggest that the
mechanisms of mtDNA replication in different systems are similar but not
identical. This prediction is in agreement with the different structures of
mtDNA molecules from different sources (for review see (Gray, 1989)).
MtDNA molecules with a D-loop form have been isolated from mammals
but not from fruit flies, even though Drosophila mtDNA molecules contain
an ‘A+T’ region which is believed to function similarly to the D-loop
region of mammalian mtDNA (Clayton, 1982; Goddard and Wolstenholme,
1980). The position and structure of the initiation site for the lagging
strand mtDNA synthesis is also different in flies than in mammals.

The results presented in this chapter allows parallels to be drawn
between Drosophila mitochondrial DNA polymerase and HIV reverse
transcriptase. HIV-RT is a heterodimer of p66 and p51, and both proteins
are encoded by the same pol gene. The p66 subunit contains both a
polymerase and an RNase H domain (Kohlstaedt et al., 1992; J acobo-
Molina et al., 1993). The RNase H domain itself does not show activity
even though it has a stable independent structure (Davies et al., 1991).
However, p66 or the reconstitution of p51 and RNase H domain
polypeptide provides RNase H activity (Hostomsky et al., 1991; Prasad et

al., 1991). By comparison, the M-domain of pol ‘y—B itself may not be

120

 

sufficient for RN ase H activity. Other domains in the accessory subunit of
pol 'y or even some parts from the catalytic subunit may be essential for the
RNase H activity if it exists in the heterodimeric mitochondrial DNA
polymerase. In the heterodimer of HIV-RT, p66 and p51 have quite
different structures with p66 providing both the pol and RN ase H activities.
However, the presence of p51 is essential for the polymerase activity. HIV—
RT starts minus strand DNA synthesis with human tRNA(Lys-3) whose 3’
end partially unfolds and forms 18 base pairs of duplex with the viral RNA
primer binding site. In order to allow plus strand DNA synthesis, the RNA
template is then removed by its RNase H activity that leaves an
oligoribonucleotide to serve as a primer for plus strand DNA synthesis.
These features in both minus and plus strand DNA synthesis share some
similarities with the initiation of mtDNA replication. In fact, mitochondrial
DNA polymerases from fly and man have been shown to have reverse
transcriptase activity (Wemette and Kaguni, 1986; Longley et al., 1998).
Based on these similarities, I propose a model of the heterodimeric pol 7
complex (Figure 3.11) which resembles the structure of the HIV-RT
complex (Kohlstaedt, et al., 1992). The catalytic subunit of pol y is divided
into the exo, spacer and pol domains in the model. Because pol y has
reverse transcriptase activity, the pol domain of pol y—oc may also fold like

the pol domain of p66 in HIV-RT. The exo domain together with the

121

 

 

(from Science 256: 1787 (1992))

 

Figure 3.11 Proposed complex for Drosophila pol y at
initiation site of mtDNA replication. In the working model, the
catalytic subunit of pol 'y is divided into three domains: exonuclease (exo),
spacer (S), and polymerase (pol). The accessory subunit comprises N—, M—,
and C-domains as described in the text. The domain arrangement is based
on the results of subunit interactions, on structural modeling, and on
similarities between pol y and HIV-RT. The substrates RNA (thin line) and
DNA(thick line) are positioned to fit the hypothetical functions of
individual domains. The proposed model resembles the initiation complex
of HIV-RT at the initiation site of minus strand replication (top) proposed
by Kohlstaedt, ct al.(l992).

122

Q

J

'1
,.-I

spacer region provide the corresponding regions of p51 in HIV-RT. The
M-domain of pol y—B corresponds to the RNase H domain of p66 in HIV-
RT. Thus, in this model, the pol domain and the M-domain are contributed
by different subunits in Drosophila poly but have a similar orientation as
the pol domain and the RN ase H domain in p66 of HIV-RT. The C-domain
of pol y—B is connected to the bottom of the M-domain to keep both
domains in contact with the exo domain and the spacer of pol y—or. It is less
clear how the N-domain of pol y—B is oriented with respect to pol y~0t but
the biochemical studies suggest that there is weak interaction between the
N-domain and the pol domain, which may maintain the six domains in the
two subunits in the orientation required for an active pol y enzyme. It
seems plausible that Drosophila pol 'y may initiate mtDNA replication by a
mechanism similar to that of HIV minus strand initiation (Kohlstaedt, et al.,
1992). As proposed previously by the structural model (Fan, et al., 1999),
the C-terrninal domain of pol y—B may bind to a stem-loop structure of
RNA transcribed by mitochondrial RNA polymerase, thereby placing the
holoenzyme of Drosophila pol y at the initiation site in a correct
conformation. Once this happens, the M-domain could function as an

RN ase H to process the RNA into a primer with an extendible 3’-OH end.
Pol y—oc then extends the primer to start DNA synthesis. In this proposed

model, the M-domain would interact with the novel duplex DNA during

123

strand elongation, which may increase the interaction between the

holoenzyme and DNA substrate and thereby enhance its processivity.

124

CHAPTER IV

FUTURE RESEARCH

125

 

Biochemical analysis and computer-aided modeling have been
applied in the research described in this thesis to investigate the structure-
function relationships of Drosophila mitochondrial DNA polymerase. The
results presented here reveal for the first time subunit interactions in a
mitochondrial DNA polymerase complex from any source. The results
suggest that the accessory subunit of Drosophila pol 7 comprises three
domains: termed N-, M-, and C-domains. The two subunits of Drosophila
pol y interact through multiple sites in different regions. The interactions
of the M- and C-domains of the accessory subunit with the exo domain and
spacer of the catalytic subunit are the main forces for the formation of the
heterodimeric complex of Drosophila pol 'y. The N-domain of the
accessory subunit may coordinate the overall subunit-subunit interaction to
keep the complex in an active conformation. These results provide valuable
information on the overall domain interactions in Drosophila pol 7. Future
research should focus on extending these results to identify interaction
motifs or specific residues involved in subunit-subunit interaction.

Because of the complexity of the subunit interactions in Drosophila
pol y as discussed in chapter III, the deletion mutations described in this
thesis may not distinguish the effects of the mutations on subunit interaction
from structural defects caused by these mutations. Different strategies may
be used for future research. I would propose to set up a phage display

library for initial screening of possible peptides which are the favored

126

 

substrates of the individual subunit. In the past decade, phage-display has
developed into one of the most popular and powerful methods for protein-
protein interaction studies. A strategy similar to that used for mapping the
protein-protein interactions between c-myb and its coactivator CBP
(Kiewitz and Wolfes, 1997) would suit perfectly our purposes here. The
approach would be to clone random cDNA segments from each subunit of
Drosophila pol 7 into a modified phagemid to construct a mini-library for
expression of many fragments in both subunits of pol y. This mini-library
will then be propagated as phages and screened with either individual
subunit or the domains of each subunit as the bait. A pool of peptides
binding to the bait would be obtained. By analyzing the sequences of these
peptides, we can then identify the regions in individual subunits which have
sequences matching those of the peptides, and pursue selective site-directed
mutagenesis to study subunit-subunit interaction by the methods described

_ in this thesis. In this way, we will limit the number of mutations to a
reasonable level and find most of the important motifs and residues for
subunit interactions.

Sequence and structural analysis presented in this thesis predict that
the accessory subunit may have properties of binding to DNA or RNA
molecules, and in particular, RNA with stem-loop like secondary
structures. These predictions provide useful information for further

characterization of the accessory subunit. I suggest to use two systems to

127

explore the biochemical properties of the accessory subunit of Drosophila
pol “y.

In one approach, native pol y or recombinant pol 7 complex would
be the main source for biochemical characterization. Although the two
subunits are associated in the complex, UV-crosslinking experiments such
as those shown in Figure 4.1 will be very useful to test if the accessory
subunit itself can bind to RNA or DNA molecules. As indicated in Figure
4.1, the pol 7 complex was incubated with an RNA molecule (p608), or
p60S annealed to an oligodeoxyribonucleotide DK114 as a mimic of
unprocessed RNA-DNA hybrid at the initiation site of mtDNA replication.
Under the standard conditions for DNA polymerase activity assay, the
accessory subunit but not the catalytic subunit can be crosslinked to p60S
by UV irradiation, indicating strong interactions between the accessory
subunit and the p60S molecule. The results confirm our prediction that the
accessory subunit can bind to RNA molecules. Annealing oligonucleotide
DK114 to p60S did not affect the binding. In addition, in the presence of
dATP, a new crosslinked complex with slower mobility was formed as
indicated by the filled arrow in Figure 4.1, suggesting that an RNA
molecule extended by the pol 7 was crosslinked to the accessory subunit.
This preliminary result supports our prediction about the role of the
accessory subunit in the initiation of mtDNA replication. Two different

strategies can be used for future research. One is to study the substrate

128

A) Sequence and structure of primer-template substrate:

Sequence of RNA primer (p60S) (142 nt)

5'—gggcgaauug gguaccggaa uugccuuugc caaacuuuua
gaaguaaagu augcuuaugg aauuauggau uuuuuguugu
uuuuguuguu uguuuagauc gaauuccugc agcccggggg

auccacu aguucuagagcggcc-B' 142

Sequence of template (DK114) (27 nt):
3 ' -TCAAGATQTQGQCETTATAATATATA- 5 '

Predicted secondary structure of ntl-127 in p60S:

10

-G AAUU CCG

GGCG GGGUA

CCGU UCCGU G
UC -——U UAA

30 20

40
AAAC G AA

UUUUA AAGUA
AAGGU UUCGU G

---- A AU
so 50
7o 80 90
UUA UUGUUU --—U UUU
UGGAUUUUUUG UUGU GUUUG
ACCUAGGGGGC GACG UAAGC A
-—- ----CC UCCU UAG
120 110 100
Figure 4.1

129

B) UV-crosslinking of pol y—B to 32P-labeled p60S

- - + + dATP
+ - + - pol y
+ + + + sub

 

Figure 4.1 The accessory subunit of Drosophila poly binds to RNA
primer in the presence of the catalytic subunit. Purified recombinant
poly heterodimer from baculovirus-infected Sf9 cells was incubated
under conditions of standard DNA polymerase activity assay with a
primer (p60S)-template (DK114) subustrate (sub). RNA p60S was labeled
with 32F at 5’-end and annealed at 3 ’-end with DNA DK114 where
underlined nucleotides form base-pairs (A). The secondary structure of un-
annealed sequence of p60S was predicted by Foldrna program of GCG.
The reaction mixture was incubated for 15 minutes at 30 OC, and then
irradiated under UV light (305 nm) for 15 minutes on ice. The mixture
was finally resolved by 10% SDS-PAGE (B). The gel was dried and
subjected to Phosphor Imager analysis. +/- indicate the presence or absence
of reaction components. Protein sizes are based on prestain markers from
BioRad. Arrows point to crosslinked RNA-pol y—B products.

130

 

specificity in this protein-RNA interaction by varying either the sequence
or the secondary structure of the RNA substrate. Another one is to map the
binding regions in the accessory subunit. One way to do this is to increase
the scale of the UV-crosslinking reaction so that the amount of the
crosslinked RNA-protein complex is suitable for protease digestion. The
digestion products can then be analyzed by two-dimensional gel
electrophoresis and mass spectrum analysis. The interaction regions of the
accessory subunit will be tracked by the attached 32P-labeled RNA.

A second approach to study biochemical activity of the accessory
subunit, independent of the catalytic subunit, would prepare 35S-labeled
accessory subunit by in vitro transcription /translation using rabbit
reticulocyte lysate. During the course of this project, I found that a
significant portion of the accessory subunit prepared in this way is soluble,
whereas other methods of expressing the accessory subunit alone failed. As
shown in Figure 4.2, after the gradient sedimentation, the accessory subunit
appears as a clear peak at a position corresponding to its size of 40 kDa
although there is some unfolded or aggregated material pelleted at the
bottom of the gradient. The accessory subunit prepared in this way shows
weak binding to a ssDNA column (Figure 4.2). In the future, various
mutated forms of the accessory subunit can be prepared and examined for
their capacity to bind to ssDNA in order to map the interaction regions or

residues.

131

 

A) Glycerol Gradient Sedimentation of pol y—B

Glycerol concentration

 

30% a 12%

P 5 8 11 1417 202326293235 3841 L

 

 

 

 

 

B) Binding of pol y—B to an ssDNA-cellulose column

 

 

 

 

 

 

     

600mM
250mM KCl
mM
50m M

 

 

 

 

L 1 2 3 4 5 678910111213 141516171819202122232425

Figure 4.2 The accessory subunit binds to ssDNA
independent of the catalytic subunit. The accessory subunit protein
was prepared by in vitro transcription/translation with rabbit reticulocyte
lysate (Promega, CA) in the presence of 35S-labeled methionine. A sample
of 200 1.11 reaction mixture was subjected to glycerol gradient sedimentation
(A) or ssDNA-cellulose chromatography as described by Wang (Wang,
1998). The fractions were analyzed by 12% SDS-PAGE and
autoradiography.

132

ACKNOWLEDGMENT

I would like to thank Dr. Donna Koslowsky of the Department of

Microbiology for providing the p60S RNA and DK114 oligonucleotides.

133

 

BIBLIOGRAPHY

134

 

7-

Abagyan, R., Frishman, D., and Argos, P. (1994). Recognition of distantly
related proteins through energy calculations. Proteins 19, 132-40.

Aberg, A., Yaremchuk, A., Tukalo, M., Rasmussen, B., and Cusack, S.
(1997). Crystal structure analysis of the activation of histidine by Thermus
thermophilus histidyl-tRNA synthetase. Biochemistry 36, 3084-94.

Adler, S. a. M., P. (1979). T7-induced DNA polymerase. J. Biol. Chem.
245, 11605—11614.

Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J.
(1990). Basic local alignment search tool. J. Mol. Biol. 215, 403-10.

Araki, H., Ropp, P. A., Johnson, A. L., Johnston, L. H., Morrison, A., and
Sugino, A. (1992). DNA polymerase II, the probable homolog of
mammalian DNA polymerase epsilon, replicates chromosomal DNA in the
yeast Saccharomyces cerevisiae. EMBO J. I 1, 733-40.

Amez, J. G., and Moras, D. (1997). Structural and functional
considerations of the aminoacylation reaction. Trends Biochem. Sci. 22,
211-6.

Attardi, G. (1985). Animal mitochondrial DNA: an extreme example of
genetic economy. Int. Rev. Cytol., 93-145.

Attardi, G., and Schatz, G. (1988). Biogenesis of mitochondria. Annu. Rev.
Cell. Biol. 4, 289-333.

Baynton, I., and Fuchs, R. P. (2000). Lesions in DNA: hurdles for
polymerases. Trends Biochem. Sci. 25, 74-9.

Bedford, E., Tabor, S., and Richardson, C. C. (1997). The thioredoxin
binding domain of bacteriophage T7 DNA polymerase confers processivity
on Escherichia coli DNA polymerase 1. Proc. Natl. Acad. Sci. USA 94,
479-484.

Beese, L. S., Derbyshire, V., and Steitz, T. A. (1993a). Structure of DNA

polymerase I Klenow fragment bound to duplex DNA. Science 260, 352-
355.

Beese, L. S., Friedman, J. M., and Steitz, T. A. (1993b). Crystal structures
of the Klenow fragment of DNA polymerase I complexed with

135

deoxynucleoside triphosphate and pyrophosphate. Biochemistry 32, 14095-
101.

Bernstein, F. C., Koetzle, T. F., Williams, G. J ., Meyer, E. F., Jr., Brice,
M. D., Rodgers, J. R., Kennard, O., Shimanouchi, T., and Tasumi, M.
(1977). The Protein Data Bank. A computer-based archival file for
macromolecular structures. Eur. J. Biochem. 80, 319-24.

Bogenhagen, D., and Clayton, D. A. (1977). Mouse L cell mitochondrial
DNA molecules are selected randomly for replication throughout the cell
cycle. Cell 11, 719-27.

Bowie, J. U., Luthy, R., and Eisenberg, D. (1991). A method to identify
protein sequences that fold into a known three- dimensional structure.
Science 253, 164-70.

Braithwaite, D. K., and Ito, J. (1993). Compilation, alignment, and
phylogenetic relationships of DNA polymerases. Nucleic Acids Res. 21,
787-802.

Bryant, S. H., and Lawrence, C. E. (1993). An empirical energy function
for threading protein sequence through the folding motif. Proteins 16, 92-
1 12.

Burgers, P. M. (1998). Eukaryotic DNA polymerases in DNA replication
and DNA repair. Chromosoma 107, 218-27.

Burgers, P. M., and Gerik, K. J. (1998). Structure and processivity of two

forms of Saccharomyces cerevisiae DNA polymerase delta. J. Biol. Chem.
273, 19756-62.

Burgers, P. M., Kornberg, A., and Sakakibara, Y. (1981). The dnaN gene
codes for the beta subunit of DNA polymerase III holoenzyme of
escherichia coli. Proc. Natl. Acad. Sci. USA 78, 5391-5.

Campbell, J. L. (1993). Yeast DNA replication. J. Biol. Chem. 268, 25261-
25264.

Carrodeguas, J. A., and Bogenhagen, D. F. (2000). Protein sequences
conserved in prokaryotic aminoacyl-tRNA synthetases are important for
the activity of the processivity factor of human mitochondrial DNA
polymerase. Nucleic Acids Res. 28, 1237-1244.

136

 

Carrodeguas, J. A., Kobayashi, R., Lim, S. E., Copeland, W. C., and
Bogenhagen, D. F. (1999). The accessory subunit of Xenopus laevis
mitochondrial DNA polymerase gamma increases processivity of the

catalytic subunit of human DNA polymerase gamma and is related to class
H aminoacyl-tRNA synthetases. Mol. Cell. Biol. 19, 4039-46.

Chang, D. D., and Clayton, D. A. (1987a). A mammalian mitochondrial

RNA processing activity contains nucleus- encoded RNA. Science 235,
1 178-84.

Chang, D. D., and Clayton, D. A. (1987b). A novel endoribonuclease
cleaves at a priming site of mouse mitochondrial DNA replication. EMBO
J. 6, 409-17.

Chang, D. D., and Clayton, D. A. (1985). Priming of human mitochondrial
DNA replication occurs at the light-strand promoter. Proc. Natl. Acad. Sci.
USA 82, 351-355.

Chang, D. D., Hauswirth, W. W., and Clayton, D. A. (1985). Replication
priming and transcription initiate from precisely the same site in mouse
mitochondrial DNA. EMBO J. 4, 1559-67.

Chen, C. H., and Cheng, Y. C. (1989). Delayed cytotoxicity and selective
loss of mitochondrial DNA in cells treated with the anti-human

immunodeficiency virus compound 2',3'- dideoxycytidine. J. Biol. Chem.
264, 11934-7.

Chen, C. H., Vazquez-Padua, M., and Cheng, Y. C. (1991). Effect of anti-
human immunodeficiency virus nucleoside analogs on mitochondrial DNA
and its implication for delayed toxicity. Mol. Pharmacol. 39, 625-8.

Chothia, C., and Lesk, A. M. (1986). The relation between the divergence
of sequence and structure in proteins. EMBO J. 5, 823-6.

Chow, C. S., and Coen, D. M. (1995). Mutations that specifically impair
the DNA binding activity of the herpes simplex virus protein UL42. J.
Virol. 69, 6965-71.

Clary, D. O., and Wolstenholme, D. R. (1987). Drosophila mitochondrial
DNA: conserved sequences in the A + T-rich region and supporting

evidence for a secondary structure model of the small ribosomal RNA. J.
Mol. Evol. 25, 116-125.

137

 

Clayton, D. A. (1991). Replication and transcription of vertebrate
mitochondrial DNA. Annu. Rev. Cell. Biol. 7, 453-78.

Clayton, D. A. (1982). Replication of animal mitochondrial DNA. Cell 28,
693-705.

Clayton, D. A. (1992). Transcription and replication of animal
mitochondrial DNAs. Int. Rev. Cytol., 217-32.

Clayton, D. A. (1984). Transcription of the mammalian mitochondrial
genome. Annu. Rev. Biochem., 573-94.

Collins, K. L., Russo, A. A., Tseng, B. Y., and Kelly, T. J. (1993). The
role of the 70 kDa subunit of human DNA polymerase alpha in DNA
replication. EMBO J. 12, 4555-4566.

Copeland, W. C., and Wang, T. S. (1991). Catalytic subunit of human
DNA polymerase alpha overproduced from baculovirus-infected insect

cells. Structural and enzymological characterization. J. Biol. Chem. 266,
22739-22748.

Crute, J. J ., and Lehman, I. R. (1989). Herpes simplex-1 DNA polymerase.
Identification of an intrinsic 5'-3' exonuclease with ribonuclease H activity.
J. Biol. Chem. 264, 19266-70.

Cusack, S. (1997). Aminoacyl-tRNA synthetases. Curr. Opin. Struct. Biol.
7, 881-9.

Cusack, S. (1995). Eleven down and nine to go. Nat. Struct. Biol. 2, 824-
31.

Cusack, S., Yaremchuk, A., Krikliviy, I., and Tukalo, M. (1998).

tRNA(Pro) anticodon recognition by Thermus thermophilus prolyl-tRNA
synthetase. Structure 6, 101-8.

Desjardins, P., and Morais, R. (1990). Sequence and gene organization of
the chicken mitochondrial genome. A novel gene order in higher
vertebrates. J. Mol. Biol. 212, 599-634.

Devereux, J ., Haeberli, P., and Smithies, O. (1984). A comprehensive set
of sequence analysis programs for the VAX. Nucleic Acids Res. 12, 387-
95.

138

 

Digard, P., Bebrin, W. R., Weisshart, K., and Coen, D. M. (1993). The
extreme C terminus of herpes simplex virus DNA polymerase is crucial for

functional interaction with processivity factor UL42 and for viral
replication. J. Virol. 67, 398-406.

Dong, Z., Onrust, R., Skangalis, M., and O'Donnell, M. (1993). DNA
polymerase III accessory proteins. 1. holA and holB encoding delta and
delta‘. J. Biol. Chem. 268, 11758-65.

Dorsky, D. I., and Crumpacker, C. S. (1988). Expression of herpes
simplex virus type 1 DNA polymerase gene by in vitro translation and
effects of gene deletions on activity. J. Virol. 62, 3224-32.

Doublie, S., Tabor, 8., Long, A. M., Richardson, C. C., and Ellenberger,
T. (1998). Crystal structure of a bacteriophage T7 DNA replication
complex at 2.2 A resolution. Nature 391, 251-8.

Dujon, B. (1996). The yeast genome project: what did we learn? Trends
Genet. 12, 263-70.

Fan, L., Sanschagrin, P. C., Kaguni, L. S., and Kuhn, L. A. (1999). The
accessory subunit of mtDNA polymerase shares structural homology with
aminoacyl-tRNA synthetases: Implications for a dual role as a primer

recognition factor and processivity clamp. Proc. Natl. Acad. Sci. USA 96,
9527-9532.

Farr, C. L., Wang, Y., and Kaguni, L. S. (1999). Functional interactions of
mitochondrial DNA polymerase and single— stranded DNA-binding protein.
Template—primer DNA binding and initiation and elongation of DNA
strand synthesis. J. Biol. Chem. 274, 14779-85.

Fischer, D., and Eisenberg, D. (1996). Protein fold recognition using
sequence-derived predictions. Protein Sci. 5, 947-55.

Flockner, H., Braxenthaler, M., Lackner, P., Jaritz, M., Ortner, M., and
Sippl, M. J. (1995). Progress in fold recognition. Proteins 23, 376-86.

Flockner, H., Domingues, F. S., and Sippl, M. J. (1997). Protein folds

from pair interactions: a blind test in fold recognition. Proteins Suppl. 1,
129-33.

139

 

Flower, A. M., and McHenry, C. S. (1990). The gamma subunit of DNA
polymerase III holoenzyme of Escherichia coli is produced by ribosomal
frameshifting. Proc. Natl. Acad. Sci. USA 87, 3713-7.

Foury, F. (1989). Cloning and sequencing of the nuclear gene MIPl

encoding the catalytic subunit of the yeast mitochondrial DNA polymerase.

J. Biol. Chem. 264, 20552-20560.

Fuller, C. W., Beauchamp, B. B., Engler, M. J ., Lechner, R. L., Matson,
S. W., Tabor, S., White, J. H., and Richardson, C. C. (1983). Mechanisms
for the initiation of bacteriophage T7 DNA replication. Cold Spring Harb.
Symp. Quant. Biol. 47 Pt 2, 669-79.

Gallo, M. L., Jackwood, D. H., Murphy, M., Marsden, H. S., and Parris,
D. S. (1988). Purification of the herpes simplex virus type 1 65-kilodalton
DNA-binding protein: properties of the protein and evidence of its

association with the virus-encoded DNA polymerase. J. Virol. 62, 2874-83.

Goddard, J. M., and Wolstenholme, D. R. (1978). Origin and direction of
replication in mitochondrial DNA molecules from Drosophila
melanogaster. Proc. Natl. Acad. Sci. USA 75, 3886-90.

Goddard, J. M., and Wolstenholme, D. R. (1980). Origin and direction of
replication in mitochondrial DNA molecules from the genus Drosophila.
Nucleic Acids Res. 8, 741-57.

Godzik, A., and Skolnick, J. (1992). Sequence-structure matching in
globular proteins: application to supersecondary and tertiary structure
determination. Proc. Natl. Acad. Sci. USA 89, 12098-102.

Gottlieb, J., Marcy, A. I., Coen, D. M., and Challberg, M. D. (1990). The
herpes simplex virus type 1 UL42 gene product: a subunit of DNA
polymerase that functions to increase processivity. J. Virol. 64, 5976-87.

Gray, H., and Wong, T. W. (1992). Purification and identification of

subunit structure of the human mitochondrial DNA polymerase. J. Biol.
Chem. 267, 5835-5841.

Gray, M. W. (1989). Origin and evolution of mitochondrial DNA. Annu.
Rev. Cell. Biol. 5, 25-50.

140

Guenther, B., Onrust, R., Sali, A., O'Donnell, M., and Kuriyan, J. (1997).
Crystal structure of the delta' subunit of the clamp-loader complex of E.
coli DNA polymerase 111. Cell 91, 335-45.

Haffey, M. L., Novotny, J ., Bruccoleri, R. E., Carroll, R. D., Stevens, J.
T., and Matthews, J. T. (1990). Structure-function studies of the herpes
simplex virus type 1 DNA polymerase. J. Virol. 64, 5008-18.

Hernandez, T. R., and Lehman, I. R. (1990). Functional interaction

between the herpes simplex-1 DNA polymerase and UL42 protein. J. Biol.
Chem. 265, 11227-32.

Hindges, R., and Hubscher, U. (1997). DNA polymerase delta, an essential
enzyme for DNA transactions. Biol. Chem. 378, 345-62.

Hixson, J. E., Wong, T. W., and Clayton, D. A. (1986). Both the conserved
stem-loop and divergent 5'-flanking sequences are required for initiation at

the human mitochondrial origin of light- strand DNA replication. J. Biol.
Chem. 261, 2384-90.

Holm, L., and Sander, C. (1994). The FSSP database of structurally aligned
protein fold families. Nucleic Acids Res. 22, 3600-9.

Holm, L., and Sander, C. (1993). Protein structure comparison by
alignment of distance matrices. J. Mol. Biol. 233, 123-38.

Holmgren, A. (1985). Thioredoxin. Annu. Rev. Biochem. 54, 237-71.

Holmgren, A., Soderberg, B. O., Eklund, H., and Branden, C. I. (1975).
Three-dimensional structure of Escherichia coli thioredoxin-S2 to 2.8 A
resolution. Proc. Natl. Acad. Sci. USA 72, 2305-9.

Holt, 1. J ., Harding, A. E., and Morgan-Hughes, J. A. (1988). Deletions of
muscle mitochondrial DNA in patients with mitochondrial myopathies.
Nature 331, 717-9.

Huber, H. E., Tabor, S., and Richardson, C. C. (1987). Escherichia coli
thioredoxin stabilizes complexes of bacteriophage T7 DNA polymerase and
primed templates. J. Biol. Chem. 262, 16224- 16232.

Hubscher, I., Nasheuer, I., and Syvaoja, J. E. (2000). Eukaryotic DNA
polymerases, a growing family. Trends Biochem. Sci. 25, 143-147.

141

 

Insdorf, N. F., and Bogenhagen, D. F. (1989a). DNA polymerase gamma
from Xenopus laevis. I. The identification of a high molecular weight

catalytic subunit by a novel DNA polymerase photolabeling procedure. J.
Biol. Chem. 264, 21491-21497.

Insdorf, N. F., and Bogenhagen, D. F. (1989b). DNA polymerase gamma
from Xenopus laevis. II. A 3'-5' exonuclease is tightly associated with the
DNA polymerase activity. J. Biol. Chem. 264, 21498-21503.

Ito, J., and Braithwaite, D. K. (1991). Compilation and alignment of DNA
polymerase sequences. Nucleic Acids Res. 19, 4045-4057.

Ito, J ., and Braithwaite, D. K. (1990). Yeast mitochondrial DNA
polymerase is related to the family A DNA polymerases. Nucleic Acids
Res. 18, 6716.

Jacobo-Molina, A., Ding, J ., Nanni, R. G., Clark, A. D., Jr., Lu, X.,
Tantillo, C., Williams, R. L., Kamer, G., Ferris, A. L., Clark, P., Hizi, A.,
Hughes, S. H., and amold, E. (1993). Crystal structure of human
immunodeficiency virus type 1 reverse transcriptase complexed with
double-stranded DNA at 3.0 A resolution shows bent DNA. Proc. Natl.
Acad. Sci. USA 90, 6320-6324.

Jang, S. H., and Jaehning, J. A. (1991). The yeast mitochondrial RNA
polymerase specificity factor, MTFl, is similar to bacterial sigma factors.
J. Biol. Chem. 266, 22671-7.

Johnson, A. A., Tsai, Y., Graves, S. W., and Johnson, K. A. (2000).
Human mitochondrial DNA polymerase holoenzyme: reconstitution and
characterization. Biochemistry 39, 1702-1708.

Jones, D. T., Taylor, W. R., and Thornton, J. M. (1992). A new approach
to protein fold recognition. Nature 358, 86-9.

Jones, D. T., Tress, M., Bryson, K., and Hadley, C. (1999). Successful
recognition of protein folds using threading methods biased by sequence
similarity and predicted secondary structure. Proteins 37, 104-111.

Joyce, C. M., and Steitz, T. A. (1994). Function and structure relationships
in DNA polymerases. Annu. Rev. Biochem. 63, 777-822.

142

Kaguni, L. S., and Olson, M. W. (1989). Mismatch-specific 3'-5'
exonuclease associated with the mitochondrial DNA polymerase from
Drosophila embryos. Proc. Natl. Acad. Sci. USA 86, 6469-6473.

Kaguni, L. S., Wemette, C. M., Conway, M. C., and Yang-Cashman, P.
(1988). Structural and catalytic features of the mitochondrial DNA
polymerase from Drosophila melanogaster embryos. In Cancer Cells:
Eukaryotic DNA Replication, T. J. Kelly and B. Stillman, eds. (Cold
Spring Harbor: Cold Spring Harbor Laboratory), pp. 425-432.

Kelman, Z., and O'Donnell, M. (1995). DNA polymerase III holoenzyme:
structure and function of a chromosomal replicating machine. Annu. Rev.
Biochem. 64, 171-200.

Kesti, T., Frantti, H., and Syvaoja, J. E. (1993). Molecular cloning of the
cDNA for the catalytic subunit of human DNA polymerase epsilon. J. Biol.
Chem. 268, 10238-45.

Kiefer, J. R., Mao, C., Hansen, C. J ., Basehore, S. L., Hogrefe, H. H.,
Braman, J. C., and Beese, L. S. (1997). Crystal structure of a thermostable

Bacillus DNA polymerase I large fragment at 2.1 A resolution. Structure
5, 95-108.

Kiewitz, A., and Wolfes, H. (1997). Mapping of protein-protein
interactions between c-myb and its coactivator CBP by a new phage display
technique. FEBS Lett. 415, 258-62.

Kim, Y., Eom, S. H., Wang, J., Lee, D. 8., Sub, S. W., and Steitz, T. A.
(1995). Crystal structure of Thermus aquaticus DNA polymerase. Nature
376, 612-6.

Kohlstaedt, L. A., Wang, J., Friedman, J. M., Rice, P. A., and Steitz, T. A.
(1992). Crystal structure at 3.5 A resolution of HIV-1 reverse transcriptase
complexed with an inhibitor. Science 256, 1783-1790.

Kong, X. P., Onrust, R., O'Donnell, M., and Kuriyan, J. (1992). Three-
dimensional structure of the beta subunit of E. coli DNA polymerase III
holoenzyme: a sliding DNA clamp. Cell 69, 425-37.

Koonin, E. V., and Mushegian, A. R. (1996). Complete genome sequences
of cellular life forms: glimpses of theoretical evolutionary genomics. Curr.
Opin. Genet. Dev. 6, 757-62.

143

 

 

Kornberg, A., and Baker, T. A. (1992). DNA Replication, Second Edition
(New York: W. H. Freeman and Company).

Krishna, T. S., Fenyo, D., Kong, X. P., Gary, S., Chait, B. T., Burgers,
P., and Kuriyan, J. (1994). Crystallization of proliferating cell nuclear
antigen (PCNA) from Saccharomyces cerevisiae. J. Mol. Biol. 241, 265-8.

Kumar, A., Abbotts, J., Karawya, E. M., and Wilson, S. H. (1990).
Identification and properties of the catalytic domain of mammalian DNA
polymerase beta. Biochemistry 29, 7156-7159.

Kumar, A., Widen, S. G., Williams, K. R., Kedar, P., Karpel, R. L., and
Wilson, S. H. (1990). Studies of the domain structure of mammalian DNA

polymerase beta. Identification of a discrete template binding domain. J.
Biol. Chem. 265, 2124-2131.

Kunkel, T. A., and Mosbaugh, D. W. (1989). Exonucleolytic proofreading
by a mammalian DNA polymerase. Biochemistry 28, 988-995.

Kunkel, T. A., and Soni, A. (1988). Exonucleolytic proofreading enhances
the fidelity of DNA synthesis by chick embryo DNA polymerase-gamma. J.
Biol. Chem. 263, 4450-4459.

Laskowski, R. A., MacArthur, M. W., Moss, D. S., and Thornton, J. M.
(1993). PROCHECK: a program to check the stereochemical quality of
protein structures. J. Appl. Cryst. 26, 283-91.

Leblanc, C., Richard, O., Kloareg, B., Viehmann, S., Zetsche, K., and
Boyen, C. (1997). Origin and evolution of mitochondria: what have we
learnt from red algae? Curr. Genet. 31, 193-207.

Lecrenier, N., Van Der Bruggen, P., and Foury, F. (1997). Mitochondrial

DNA polymerases from yeast to man: a new family of polymerases. Gene
185, 147-52.

Lee, D. Y., and Clayton, D. A. (1996). Properties of a primer RNA-DNA
hybrid at the mouse mitochondrial DNA leading-strand origin of
replication. J. Biol. Chem. 271, 24262-24269.

Lee, D. Y., and Clayton, D. A. (1998). Initiation of mitochondrial DNA
replication by transcription and R-loop processing. J. Biol. Chem. 273,
30614-21.

144

 

Lee, D. Y., and Clayton, D. A. (1997). RNase mitochondrial RNA
processing correctly cleaves a novel R loop at the mitochondrial DNA
leading-strand origin of replication. Genes Dev. 11, 582-92.

Leem, S. H., Ropp, P. A., and Sugino, A. (1994). The yeast Saccharomyces

cerevisiae DNA polymerase IV: possible involvement in double strand
break DNA repair. Nucleic Acids Res. 22, 3011-7.

Lewis, D. L., Farr, C. L., Farquhar, A. L., and Kaguni, L. S. (1994).
Sequence, organization, and evolution of the A+T region of Drosophila
melanogaster mitochondrial DNA. Mol. Biol. Evol. 11, 523-538.

Lewis, D. L., Farr, C. L., Wang, Y., Lagina, A. T. Jr., and Kaguni, L. S.
(1996). Catalytic subunit of mitochondrial DNA polymerase from

Drosophila embryos. Cloning, bacterial overexpression, and biochemical
characterization. J. Biol. Chem. 271 , 23389-94.

Lewis, W., and Dalakas, M. C. (1995). Mitochondrial toxicity of antiviral
drugs. Nat. Med. 1, 417-22.

Lill, R., Nargang, F. E., and Neupert, W. (1996). Biogenesis of
mitochondrial proteins. Curr. Opin. Cell. Biol. 8, 505-12.

Lim, S. E., Longley, M. J., and Copeland, W. C. (1999). The
Mitochondrial p55 Accessory Subunit of Human DNA Polymerase gamma
Enhances DNA Binding, Promotes Processive DNA Synthesis, and Confers
N- Ethylmaleimide Resistance. J. Biol. Chem. 274, 38197-38203.

Logan, D. T., Mazauric, M. H., Kern, D., and Moras, D. (1995). Crystal

structure of glycyl-tRNA synthetase from Thermus thermophilus. EMBO
J. 14, 4156-67.

Longley, M. J ., Ropp, P. A., Lim, S. E., and Copeland, W. C. (1998).
Characterization of the native and recombinant catalytic subunit of human
DNA polymerase gamma: identification of residues critical for exonuclease
activity and dideoxynucleotide sensitivity. Biochemistry 3 7, 10529-39.

Luft, R. (1994). The development of mitochondrial medicine. Proc. Natl.
Acad. Sci. USA 91 , 8731-8738.

Luthy, R., Bowie, J. U., and Eisenberg, D. (1992). Assessment of protein
models with three-dimensional profiles. Nature 356, 83-5.

145

Maki, H., Horiuchi, T., and Kornberg, A. (1985). The polymerase subunit
of DNA polymerase III of Escherichia coli. 1. Amplification of the dnaE

gene product and polymerase activity of the alpha subunit. J. Biol. Chem.
260, 12982-6.

Maki, H., and Kornberg, A. (1987). Proofreading by DNA polymerase III
of Escherichia coli depends on cooperative interaction of the polymerase
and exonuclease subunits. Proc. Natl. Acad. Sci. USA 84, 4389-92.

Maki, H., Maki, S., and Kornberg, A. (1988). DNA Polymerase III
holoenzyme of Escherichia coli. IV. The holoenzyme is an asymmetric
dimer with twin active sites. J. Biol. Chem. 263, 6570-8.

Maki, S., and Kornberg, A. (1988). DNA polymerase III holoenzyme of
Escherichia coli. 11. A novel complex including the gamma subunit
essential for processive synthesis. J. Biol. Chem. 263, 6555-60.

Marcy, A. I., Olivo, P. D., Challberg, M. D., and Coen, D. M. (1990).
Enzymatic activities of overexpressed herpes simplex virus DNA

polymerase purified from recombinant baculovirus-infected insect cells.
Nucleic Acids Res. 18, 1207-15.

Matsumoto, Y., Kim, K., Katz, D. S., and Feng, J. A. (1998). Catalytic
center of DNA polymerase beta for excision of deoxyribose phosphate
groups. Biochemistry 3 7, 6456-64.

McHenry, C. S. (1988). DNA polymerase III holoenzyme of Escherichia
coli. Ann. Rev. Biochem. 57, 519-550.

McHenry, C. S. (1982). Purification and characterization of DNA
polymerase III'. Identification of tau as a subunit of the DNA polymerase
III holoenzyme. J. Biol. Chem. 257, 2657-63.

Modrich, P., and Richardson, C. C. (1975). Bacteriophage T7
Deoxyribonucleic acid replication in vitro. A protein of Escherichia coli

required for bacteriophage T7 DNA polymerase activity. J. Biol. Chem.
250, 5508-14.

Monahan, S. J ., Barlam, T. F., Crumpacker, C. S., and Parris, D. S.
(1993). Two regions of the herpes simplex virus type 1 UL42 protein are

required for its functional interaction with the viral DNA polymerase. J.
Virol. 67, 5922-31.

146

Montoya, J ., Ojala, D., and Attardi, G. (1981). Distinctive features of the
5'-terminal sequences of the human mitochondrial mRN As. Nature 290,
465-70.

Morris, A. L., MacArthur, M. W., Hutchinson, E. G., and Thornton, J. M.
(1992). Stereochemical quality of protein structure coordinates. Proteins
12, 345-64.

Mosbaugh, D. W. (1988). Porcine pol garnmma. Nucleic Acids Res. 16,
5645-5659.

Nasheuer, H. B, Moore, A., Wahl, A. F., and Wang, T. S. (1991). Cell
cycle-dependent phosphorylation of human DNA polymerase alpha. J. Biol.
Chem. 266, 7893-903.

Nosek, J., Tomaska, L., Fukuhara, H., Suyama, Y., and Kovac, L. (1998).
Linear mitochondrial genomes: 30 years down the line. Trends Genet. 14,
184-8.

O'Donnell, M., Onrust, R., Dean, F. B., Chen, M., and Hurwitz, J. (1993).
Homology in accessory proteins of replicative polymerases--E. coli to
humans. Nucleic Acids Res. 21, 1-3.

Ojala, D., Montoya, J ., and Attardi, G. (1981). tRNA punctuation model of
RNA processing in human mitochondria. Nature 290, 470-474.

Okimoto, R., and Wolstenholme, D. R. (1990). A set of tRNAs that lack
either the T psi C arm or the dihydrouridine arm: towards a minimal
tRNA adaptor. EMBO J. 9, 3405-11.

Oliver, S. G. (1996). From DNA sequence to biological. Nature 379, 597-
600.

Ollis, D. L. B., P., Hamlin, R. Xuong, N. G. and Steitz, T. A. (1985).
Structure of large fragment of Escherichia coli DNA polymerase I
complexed with dTMP. Nature 313, 762-766.

Olson, M. W., and Kaguni, L. S. (1992). 3'-5' exonuclease in Drosophila
mitochondrial DNA polymerase. Substrate specificity and functional

coordination of nucleotide polymerization and mispair hydrolysis. J. Biol.
Chem. 267, 23136-23142.

147

Olson, M. W., Wang, Y., Elder, R. H., and Kaguni, L. S. (1995). Subunit
structure of mitochondrial DNA polymerase from Drosophila embryos.
Physical and immunological studies. J. Biol. Chem. 270, 28932-7.

Onrust, R., Stukenberg, P. T., and O'Donnell, M. (1991). Analysis of the
ATPase subassembly which initiates processive DNA synthesis by DNA
polymerase III holoenzyme. J. Biol. Chem. 266, 21681-6.

Orengo, C. A., Jones, D. T., and Thornton, J. M. (1994). Protein
superfamilies and domain superfolds. Nature 372, 631-4.

Pavletich, N. P., and Pabo, C. O. (1991). Zinc finger-DNA recognition:
crystal structure of a Zif268-DNA complex at 2.1 A. Science 252, 809-17.

Pelletier, H., Sawaya, M. R., Kumar, A., Wilson, S. H., and Kraut, J.
(1994). Structures of ternary complexes of rat DNA polymerase beta, a
DNA template-primer, and ddCTP. Science 264, 1891-1903.

Prasad, R., Beard, W. A., Chyan, J. Y., Maciejewski, M. W., Mullen, G.
P., and Wilson, S. H. (1998a). Functional analysis of the amino-terminal 8-
kDa domain of DNA polymerase beta as revealed by site-directed

mutagenesis. DNA binding and 5'-deoxyribose phosphate lyase activities. J.
Biol. Chem. 273, 11121-6.

Prasad, R., Beard, W. A., Strauss, P. R., and Wilson, S. H. (1998b).
Human DNA polymerase beta deoxyribose phosphate lyase. Substrate
specificity and catalytic mechanism. J. Biol. Chem. 273, 15263-70.

Prasad, R., Beard, W. A., and Wilson, S. H. (1994). Studies of gapped
DNA substrate binding by mammalian DNA polymerase beta. Dependence
on 5'-phosphate group. J. Biol. Chem. 269, 18096-101.

Preston, B. D., Poiesz, B. J ., and Loeb, L. A. (1988). Fidelity of HIV-1
reverse transcriptase. Science 242, 1168-71.

Rhee, S., Martin, R. G., Rosner, J. L., and Davies, D. R. (1998). A novel
DNA-binding motif in MarA: the first structure for an AraC family
transcriptional activator. Proc. Natl. Acad. Sci. USA 95, 10413-8.

Robberson, D. L., and Clayton, D. A. (1972). Replication of mitochondrial
DNA in mouse L cells and their thymidine kinase - derivatives:

displacement replication on a covalently-closed circular template. Proc.
Natl. Acad. Sci. USA 69, 3810-4.

148

Ropp, P. A., and Copeland, W. C. (1996). Cloning and characterization of

the human mitochondrial DNA polymerase, DNA polymerase y. Genomics
36, 449-458.

R0pp, P. A., and Copeland, W. C. (1995). Characterization of a new DNA
polymerase from Schizosaccharomyces pombe: a probable homologue of
the Saccharomyces cerevisiae DNA polymerase gamma. Gene 165, 103-7.

Rost, B., and Sander, C. (1994). Combining evolutionary information and
neural networks to predict protein secondary structure. Proteins 19, 55-72.

Rubenstein, J. L., Brutlag, D., and Clayton, D. A. (1977). The
mitochondrial DNA of Drosophila melanogaster exists in two distinct and
stable superhelical forms. Cell 12, 471-82.

Russell, R. B., Copley, R. R., and Barton, G. J. (1996). Protein fold

recognition by mapping predicted secondary structures. J. Mol. Biol. 25 9,
349-65.

Sander, C., and Schneider, R. (1991). Database of homology-derived
protein structures and the structural meaning of sequence alignment.
Proteins 9, 56-68.

Sawaya, M. R., Pelletier, H., Kumar, A., Wilson, S. H., and Kraut, J.
(1994). Crystal structure of rat DNA polymerase beta: evidence for a
common polymerase mechanism. Science 264, 1930-1935.

Sawaya, M. R., Prasad, R., Wilson, 8. H., Kraut, J ., and Pelletier, H.
(1997). Crystal structures of human DNA polymerase beta complexed with

gapped and nicked DNA: evidence for an induced fit mechanism.
Biochemistry 36, 11205-15.

Schatz, G. (1996). The protein import system of mitochondria. J. Biol.
Chem. 271 , 31763-6.

Scheuermann, R. H., and Echols, H. (1984). A separate editing exonuclease
for DNA replication: the epsilon subunit of Escherichia coli DNA
polymerase III holoenzyme. Proc. Natl. Acad. Sci. USA 81, 7747-51.

Shadel, G. S., and Clayton, D. A. (1997). Mitochondrial DNA maintenance
in vertebrates. Annu. Rev. Biochem. 66, 409-35.

149

 

Shadel, G. S., and Clayton, D. A. (1993). Mitochondrial transcription
initiation. Variation and conservation. J. Biol. Chem. 268, 16083-6.

Shortle, D. (1997). Structure prediction: folding proteins by pattern
recognition. Curr. Biol. 7, R151-4.

Shortle, D. (1999). Structure prediction: The state of the art. Curr. Biol. 9,
R205-9.

Steitz, T. A. (1998). A mechanism for all polymerases Nature 391 , 231-2.

Studwell-Vaughan, P. S., and O'Donnell, M. (1993). DNA polymerase III
accessory proteins. V. Theta encoded by holE. J. Biol. Chem. 268, 11785-
91.

Sussman, J. L., Lin, D., Jiang, J., Manning, N. O., Prilusky, J., Ritter, O.,
and Abola, E. E. (1998). Protein Data Bank (PDB): Database of Three-

Dimensional Structural Information of Biological Macromolecules. Acta
Cryst. D54, 1078-1084.

Tabor, S., Huber, H. E., and Richardson, C. C. (1987). Escherichia coli
thioredoxin confers processivity on the DNA polymerase activity of the
gene 5 protein of bacteriophage T7. J. Biol. Chem. 262, 16212-16223.

Waga, S., Bauer, G., and Stillman, B. (1994). Reconstitution of complete

SV40 DNA replication with purified replication factors. J. Biol. Chem.
269, 10923-34.

Wallace, D. C. (1992). Diseases of the mitochondrial DNA. Annu. Rev.
Biochem. 61, 1175-1212.

Wallace, D. C. (1994). Mitochondrial DNA mutations in diseases of energy
metabolism. J. Bioenergy and Biomembranes 26, 241-250.

Wallace, D. C. (1995). Mitochondrial DNA variation in human evolution,
degenerative disease, and aging. Am. J. Hum. Genet. 5 7, 201-223.

Wallace, D. C. (1999). Mitochondrial diseases in man and mouse. Science
283, 1482-8.

Wallace, D. G, Singh, G., Lott, M. T., Hodge, J. A., Schurr, T. G., Lezza,
A. M., Elsas, L. J. d., and Nikoskelainen, E. K. (1988). Mitochondrial

150

DNA mutation associated with Leber's hereditary optic neuropathy.
Science 242, 1427-30.

Wang, T. S. (1991). Eukaryotic DNA polymerases. Annu. Rev. Biochem.
513-552.

Wang, Y. (1998). Molecular cloning, baculovirus expression and
reconstitution of Drosophila mitochondrial DNA polymerase. Ph. D.
Dissertation. Michigan State University.

Wang, Y., Farr, C. L., and Kaguni, L. S. (1997). Accessory subunit of
mitochondrial DNA polymerase from Drosophila embryos. Cloning,

molecular analysis, and association in the native enzyme. J. Biol. Chem.
272, 13640-6.

Wang, Y., and Kaguni, L. S. (1999). Baculovirus expression reconstitutes
Drosophila mitochondrial DNA polymerase. J. Biol. Chem. 274, 28972-7.

Weissbach, A. (1979). The functional roles of mammalian DNA
polymerase. Arch. Biochem. Biophys. 198, 386-396.

Weissbach, A., Hong, S. C., Aucker, J ., and Muller, R. (1973).
Characterization of herpes simplex virus-induced deoxyribonucleic acid
polymerase. J. Biol. Chem. 248, 6270-7.

Weisshart, K., Kuo, A. A., Hwang, C. B., Kumura, K., and Coen, D. M.
(1994). Structural and functional organization of herpes simplex virus

DNA polymerase investigated by limited proteolysis. J. Biol. Chem. 269,
22788-96.

Wemette, C. M., Conway, M. C., and Kaguni, L. S. (1988). Mitochondrial
DNA polymerase from Drosophila melanogaster embryos: kinetics,

processivity, and fidelity of DNA polymerization. Biochemistry 27, 6046-
6054.

Wemette, C. M., and Kaguni, L. S. (1986). A mitochondrial DNA
polymerase from embryos of Drosophila melanogaster. Purification,

subunit structure, and partial characterization. J. Biol. Chem. 261, 14764-
14770.

Williams, A. J., and Kaguni, L. S. (1995). Stimulation of Drosophila

mitochondrial DNA polymerase by single-stranded DNA-binding protein.
J. Biol. Chem. 270, 860-865.

151

Williams, A. J., Wemette, C. M., and Kaguni, L. S. (1993). Processivity of
mitochondrial DNA polymerase from Drosophila embryos. Effects of
reaction conditions and enzyme purity. J. Biol. Chem. 268, 24855-24862.

Wolstenholme, D. R. (1992). Animal mitochondrial DNA: structure and
evolution. In Mitochondrial Genomes, D. R. Wolstenholme and K. W.
Jeon, eds. (New York: Academic Press), pp. 173-216.

Wolstenholme, D. R., Macfarlane, J. L., Okimoto, R., Clary, D. O., and
Wahleithner, J. A. (1987). Bizarre tRNAs inferred from DNA sequences

of mitochondrial genomes of nematode worms. Proc. Natl. Acad. Sci. USA
84, 1324-8.

Wong, T. W., and Clayton, D. A. (1985a). Isolation and characterization of
a DNA primase from human mitochondria. J. Biol. Chem. 260, 11530-5.

Wong, T. W., and Clayton, D. A. (1985b). In vitro replication of human

mitochondrial DNA: accurate initiation at the origin of light-strand
synthesis. Cell 42, 951-8.

Yamaguchi, M., Matsukage, A., and Takahashi, T. (1980). Chick embryo
DNA polymerase gamma. Purification and structural analysis of nearly
homogeneous enzyme. J. Biol. Chem. 255, 7002-9.

Yang, W., Hendrickson, W. A., Crouch, R. J., and Satow, Y. (1990).
Structure of ribonuclease H phased at 2 A resolution by MAD analysis of
the selenomethionyl protein. Science 249, 1398-405.

Zarembinski, T. I., Hung, L. W., Mueller-Dieckmann, H. J ., Kim, K. K.,
Yokota, H., Kim, R., and Kim, S. H. (1998). Structure-based assignment of
the biochemical function of a hypothetical protein: a test case of structural
genomics. Proc. Natl. Acad. Sci. USA 95, 15189-93.

Zuo, S., Gibbs, E., Kelman, Z., Wang, T. S., O'Donnell, M., MacNeill, S.
A., and Hurwitz, J. (1997). DNA polymerase delta isolated from

Schizosaccharomyces pombe contains five subunits. Proc. Natl. Acad. Sci.
USA 94, 11244-9.

152

"‘11111111111111