. biwflmwmfi

— ‘

 

 

 

 

 

, NE «fin...
. t _ x,

\ “w;
fim .
E .. ,

Fa.
.0

g

.1. 5.3..
C: ..

v4
. anfi.»

.K

 

 

:x.l}!:.
I . is..\115.
‘ «C.2.~.:I.)~

n! It! it.

x
. 13.35."

“ul- 5.

1;"!
I ististt
a r).l(...3 iexl.
.I;I...\.ltl...lfa

“its - LIBRARY
,\ j 9, Michigan. State
' ‘ University

-

This is to certify that the
dissertation entitled

THE ROLE OF HUMAN REV7, THE ACCESSORY SUBUNIT
OF HUMAN DNA POLYMERASE C, IN CELL SURVIVAL AND
DNA DAMAGE INDUCED MUTAGENESIS

presented by

JESSICA A. NEAL

has been accepted towards fulfillment
of the requirements for the

Biochemistry and
Ph.D. degree in Molecular Biology

 

 

flaw/fl ,7/‘i4/n/W

Major Profésso‘r‘sSignaturE“

émggf 02%!

Date

MSU is an Affirmative Action/Equal Opportunity Employer

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K;IPrqlAcc&PresICIRClDateDm.indd

THE ROLE OF HUMAN REV7, THE ACCESSORY SUBUNIT OF HUMAN DNA
POLYMERASE 2;, IN CELL SURVIVAL AND DNA DAMAGE INDUCED
MUTAGENESIS
By

Jessica A. Neal

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Biochemistry and Molecular Biology

2008

ABSTRACT
THE ROLE OF HUMAN REV7, THE ACCESSORY SUBUNIT OF HUMAN DNA
POLYMERASE :, lN CELL SURVIVAL AND DNA DAMAGE INDUCED
MUTAGENESIS
By
Jessica A. Neal
The presence of DNA damage can pose a considerable threat to cell survival
because the high fidelity polymerases that are charged with the task of
replicating DNA are often unable to synthesize past sites of damage. A
considerable amount of the DNA damage incurred is removed by repair
mechanisms. Nevertheless, repair processes are often ovenivhelmed by the
abundance of DNA damage and consequently, some lesions persist during DNA
replication. To overcome the presence of DNA lesions during replication, cells
also employ damage tolerance mechanisms that promote cell survival by
facilitating replication of damaged DNA. One important strategy for tolerating
DNA damage is translesion synthesis (TLS). Translesion synthesis refers to the
process of direct replicational bypass of DNA damage, which is conducted by a
group of specialized DNA polymerases capable of inserting nucleotides opposite
DNA lesions. Although TLS polymerases promote cell survival by avoiding
replication arrest, many of these enzymes demonstrate reduced fidelity and as a
result, also contribute to mutagenesis. The first of such TLS polymerases to be
discovered was DNA polymerase zeta (Pol 2;). The function of Pol t; has been

well characterized in yeast and has been implicated in both spontaneous and

damage-induced mutagenesis in that organism. However, the role of human Pol

i; in such mutagenesis is much less well understood. To examine the role of
hRev7, the accessory subunit of hPol C in the mutagenesis of human cells, we
generated cell strains with significantly reduced levels of hRev7 and compared
them to their parental strain and a vector control strain for cell survival, induction
of mutations, and ability to progress through the cell cycle following exposure to
UV radiation. Cells with reduced hRev7 progressed through S-phase more slowly
and were more sensitive to the cytotoxic effects of UV radiation than the controls.
In addition, cell strains with reduced hRev7 demonstrated a significantly reduced
frequency of UV-induced mutations. These results strongly support the
hypothesis that hRev7 is required for TLS past UV-induced DNA lesions. When
these same cell strains were assayed for their response to the biological effects
of benzo[a]pyrene diol epoxide (BPDE), cell strains with reduced hRev7
demonstrated a BPDE-induced delay in progression through S-phase and were
also more sensitive to its cytotoxic effect. Surprisingly however, the frequency of
mutations induced by BPDE in cell strains with reduced hRev7 did not differ from
those of the control stains, indicating that hRev7 is not required for translesion
synthesis past BPDE-induced DNA lesions. Taken together, these results
demonstrate that the protective role hRev7 plays for cells exposed to BPDE is
independent of the role of this protein in TLS. Combined with the results of the
UV study, these data suggest that hRev7 has at least two distinct cellular
functions: one as the accessory subunit of hPol C, and an additional, as of yet
uncharacterized role, in protecting cells from the cytotoxic effects of DNA

damage.

ACKNOWLEDGEMENTS

I would like to thank Dr. Veronica Maher and Dr. Justin McCormick for accepting
me into the Carcinogenesis Laboratory. l have learned a lot about science and
about what it takes to be a good mentor from them. In particular I am struck by
their ability to understand that life happens, even to graduate students. Although
scientific research is a huge priority to them, it never comes before the people
who conduct it. They are truly two of the most compassionate people I have ever

met.

I thank the members of my committee Dr. Zach Burton, Dr. Rob Larkin and Dr.
Kathy Meek. A friendly and helpful committee is a tremendous asset for a
graduate student, and I am grateful for the support and guidance they have

offered throughout this study.

I would also like to thank the members of the Carcinogenesis Laboratory, both
past and present, for all of their help and advice about science and about life. I
would especially like to thank Katie Fletcher who has spent the last three years
working with me. Katie is a talented scientist and a gifted student, who I am sure
will make a wonderful physician. I would also like to offer a special note of thanks
to Kristin McNally who started the hRev7 project and very generously passed it to

me.

iv

Finally, I would like to thank my family. My parents have provided constant help
and encouragement throughout all of my life. They taught me the value of
education and have always been there to support me in everything I have done.
My family has always cheered me on through wins and cheered me up after

losses. You can not put a price on the unconditional love of a family! Thank you!

My husband, Jake, may be the strongest person I have ever met...he is definitely
the most patient. The constant love and support that he has given me throughout
the last five years has been invaluable. I can truly say that I would not be here

without him. I love you Jake, and thank you so very much!

TABLE OF CONTENTS

LIST OF TABLES ................................................................................................. ix
LIST OF FIGURES ................................................................................................ x
INTRODUCTION ................................................................................................... 1
CHAPTER I: Literature Review ........................................................................... 6
I. DNA Repair ................................................................................................... 10
A. Base Excision Repair .............................................................................. 1O
1. Base Excision Repair Pathways ......................................................... 11
a. Glycosylases ................................................................................... 14
b. Apurinic/Apyrimidinic (AP) Endonucleases ..................................... 16
c. Post-Incision Events in Base Excision Repair ................................ 17
B. Nucleotide Excision Repair ..................................................................... 18
1. Global Genomic Excision Repair in Prokaryotes ................................ 19
2. Transcription-Coupled Nucleotide Excision Repair in Prokaryotes ..... 23
3. Global Genomic Nucleotide Excision Repair in Eukaryotes ............... 23
4. Transcription-Coupled Nucleotide Excision Repair in Eukaryotes ...... 27
5. Nucleotide Excision Repair Deficiency Diseases ................................ 29
C. Mismatch Repair ..................................................................................... 33
1. Mismatch Repair in Prokaryotes ......................................................... 33
2. Mismatch Repair in Eukaryotes .......................................................... 37
3. Mismatch Repair Deficiency Leads to Cancer .................................... 39
||. DNA Damage Tolerance .............................................................................. 41
A. DNA Damage Tolerance Mechanisms in Prokaryotes ............................ 42
1. The SOS Response to DNA Damage ................................................. 42
2. Translesion Synthesis in Prokaryotes ................................................. 44
a. DNA Polymerase V (UmuD’ZC) ...................................................... 45
b. DNA Polymerase lV (DinB) ............................................................ 48
3. Damage Tolerance in Prokaryotes ..................................................... 51
B. Damage Tolerance Mechanisms in Eukaryotes ..................................... 54
1. Translesion Synthesis in Eukaryotes .................................................. 56
a. DNA Polymerase c ......................................................................... 58
b. Rev1 ............................................................................................... 71
0. DNA Polymerase Tl ......................................................................... 77
d. DNA Polymerase L .......................................................................... 84
e. DNA Polymerase x ......................................................................... 88
2. Damage Avoidance in Eukaryotes ...................................................... 93
REFERENCES ................................................................................................ 98

vi

CHAPTER II: hRev7, putative subunit of hPolI;, plays a critical role in
survival, induction of mutations, and progression through S-phase of

UV(254nm)-irradiated human fibroblasts ........................................................... 132
ABSTRACT .................................................................................................... 133
INTRODUCTION ........................................................................................... 135
MATERIALS AND METHODS ....................................................................... 139

Cell culture ................................................................................................ 139
Cell strains ................................................................................................. 139
Derivation of cell strains with reduced hRev7 ........................................... 139
Preparation of nuclear protein extract and Western blot analysis ............. 140
Assay of UV cytotoxicity ............................................................................ 141
Assay for the frequency of UV-induced mutations .................................... 141
Hypoxanthine phosphoribosyltransferase (HPRT) mutation spectrum
analysis ..................................................................................................... 142
Cell synchronization and flow cytometry analysis ..................................... 143
Statistical Methods .................................................................................... 144
RESULTS ...................................................................................................... 145
Efficient reduction of hRev7 protein using siRNA ..................................... 145
Effect of reduction of hRev7 protein on the sensitivity of the cell strains
to the cytotoxic effect of UV(254nm) radiation ............................................... 148
Effect of reduced hRev7 on the frequency of UV-induced mutations ....... 148
Effect of decreased hRev7 protein on kinds of UV-induced mutations ..... 151
Evidence that reduction of expression of hRev7 protein in human
fibroblasts results in a UV-induced delay in traversing S—phase ............... 154
DISCUSSION ................................................................................................ 158
ACKNOWLEDGEMENTS .............................................................................. 162
REFERENCES .............................................................................................. 163

CHAPTER III: hRev7, the accessory subunit of hPol C, protects human
fibroblasts from the cytotoxic effects of benzo[a]pyrene diol epoxide, but is
not essential for translesion synthesis past DNA damage induced by this

agent ................................................................................................................ 166
ABSTRACT .................................................................................................... 167
INTRODUCTION ........................................................................................... 169
MATERIALS AND METHODS ....................................................................... 173

Cell Culture ................................................................................................ 173
Cell Strains ................................................................................................ 173
Western blot analysis ................................................................................ 174
Exposure of cells to chemical mutagens ................................................... 176
Exposure of cells to ultraviolet radiation .................................................... 177

vii

Exposure of cells to ionizing radiation ....................................................... 177

Cell survival assay ..................................................................................... 178
Mutagenesis assay .................................................................................... 179
Analysis of the cell cycle progression of BPDE or UV treated cells by
flow cytometry ........................................................................................... 180
CHAT selection ......................................................................................... 181
RESULTS ...................................................................................................... 182
Effect of reduced hRev7 protein on the survival of human fibroblasts
exposed to BPDE and on the frequency of BPDE-induced mutations ...... 182
Effect of reduced hRev7 on cell cycle progression following BPDE
treatment ................................................................................................... 185
Effect of reduced hRev7 on the survival of cells exposed to DNA
damaging agents ....................................................................................... 191
Effect of hRev3 antisense expression on BPDE-induced cytotoxicity
and mutagenicity ....................................................................................... 194
Effect of hRev3 antisense expression on UV-induced cytotoxicity and
mutagenicity .............................................................................................. 197
Effect of hRev3 antisense expression on cell cycle progression following
UV or BPDE treatment .............................................................................. 200
DISCUSSION ................................................................................................ 204
ACKNOWLEDGEMENTS .............................................................................. 208
REFERENCES .............................................................................................. 209

viii

LIST OF TABLES
CHAPTER II

Table 1: Effect of reduced expression of hRev7 on the types of base
substitution mutations induced by UV(254nm) radiation ......................... 153

ix

LIST OF FIGURES

Images in this dissertation are presented in color.
CHAPTER I:
Figure 1: The base excision repair pathway ....................................................... 13
Figure 2: The two pathways of prokaryotic nucleotide excision repair (NER) ..... 22

Figure 3: The global genomic nucleotide excision repair pathway in

eukaryotes .......................................................................................... 26
Figure 4: The transcription-coupled nucleotide excision repair pathway in

eukaryotes .......................................................................................... 31
Figure 5: The prokaryotic mismatch repair pathway ........................................... 36
CHAPTER II:

Figure 1: Western blot analysis of hRev7 protein levels in the cell strains ....... 147

Figure 2: Effect of reduced expression of hRev7 on the survival of cell strains
exposed to UV(254nm) radiation and on the frequency of UV-induced

mutations .......................................................................................... 150
Figure 3: Flow cytometry analysis of UV-irradiated cells .................................. 156
CHAPTER III:
Figure 1: Western blot analysis of the level of hRev7 protein ........................... 184

Figure 2: Reducing the level of hRev7 in human fibroblasts renders them more
sensitive to the cytotoxic effect of BPDE, but does not affect their
frequency of BPDE-induced mutations ............................................. 187

Figure 3: Reducing the level of hRev7 in human fibroblasts results in a BPDE-
induced delay in progression through the cell cycle ......................... 190

Figure 4: Reducing the level of hRev7 in human fibroblast cells renders them
more sensitive to the cytotoxic effects of a variety of different types
of DNA damaging agents .................................................................. 193

Figure 5: Expression of high levels of hRev3 antisense does not affect the cell
survival or induced mutation frequency of human fibroblasts treated
with BPDE ......................................................................................... 196

Figure 6: Expression of high levels of hRev3 antisense does not affect the
survival of human fibroblasts exposed to UV radiation, but does
cause a significant reduction in UV-induced mutations .................... 199

Figure 7: Expression of high levels of hRev3 antisense has no affect on the

progression of human fibroblasts through the cell cycle after BPDE
or UV treatment ................................................................................ 202

xi

INTRODUCTION

Cellular DNA is continually damaged both by endogenous and environmental
agents, causing a variety of DNA lesions. Many of these lesions are repaired by
mechanisms that remove them directly, such as base excision repair or
nucleotide excision repair. However, repair may be SIOW or the DNA damage
may be extensive, resulting in DNA lesions that persist during replication. If the
high fidelity, replicative polymerase complex encounters DNA damage that

blocks elongation, potentially fatal stalling or arrest of DNA replication may occur.

To mitigate the threat of cell death due to replication arrest, cells have also
evolved strategies for tolerating DNA damage. One strategy, called damage
avoidance, refers to a mechanism in which a homologous piece of DNA, for
example the newly replicated daughter strand of the sister duplex, is temporarily
used as a template to replicate around a DNA lesion. Because damage
avoidance does not require utilization of the lesion-containing portion of the DNA
as a template, but instead, makes use of an undamaged, homologous DNA
segment, it is typically considered to be an error-free mechanism of damage

tolerance (for review see [1]).

Alternatively, cells may tolerate the presence of DNA damage by employing a
process of direct replicational lesion bypass, called translesion synthesis (T LS).
Translesion synthesis requires the use of one or more specialized DNA

polymerases to replicate past the DNA lesion until the high fidelity replicative

polymerases are able to resume function. These specialized TLS polymerases
possess active sites that are typically less restrictive then those of the classical
replicative polymerases, which allows them to incorporate nucleotides opposite
sites of DNA damage [2-5]. However, one consequence of the more open active
sites of these polymerases is that they are also prone to erroneous nucleotide
incorporation, and are therefore commonly associated with an increased

frequency of mutations (reviewed in [6]).

One such specialized TLS polymerase is DNA polymerase zeta (Pol C).
Polymerase t is composed of two subunits, a catalytic subunit, called Rev3, as
well as an accessory subunit, called Rev7 [7]. In 2002, our laboratory
demonstrated that hRev3, the catalytic subunit of hPol C, is involved in
mutagenesis induced by UV radiation as well as by the chemical carcinogen
benzo[a]pyrene diol epoxide (BPDE) [8]. However, much less is known about the
function of hRev7, the accessory subunit of hPol C, in such mutagenesis.
Therefore, the focus of my research, which is presented in this dissertation, was

to determine the role of hRev7 in human mutagenesis.

Chapter I consists of a review of the literature concerning two of the major
biological responses to DNA damage, DNA repair and DNA damage tolerance,
with a special focus on the process of translesion synthesis. Chapter II consists
of a manuscript published in the April 2008 issue of DNA Repair [9], concerning

the role of hRev7 in cell survival, induction of mutations, and cell cycle

progression of UV-irradiated human fibroblasts. In that manuscript I contributed
data to Table 1 and I also contributed notably to the data interpretation and to the
writing of the manuscript. Chapter III consists of a manuscript being prepared for
submission to DNA Repair, detailing the role of hRev7 in cell survival, induction
of mutations and cell cycle progression of BPDE-treated human fibroblasts. I
carried out all of the experiments described, wrote the paper, and the

interpretation of the data was essentially mine.

REFERENCES:

1. Andersen, P. L., Xu, F., and Xiao, W. (2008). Eukaryotic DNA damage
tolerance and translesion synthesis through covalent modifications of
PCNA. Cell Res. 18. PP. 162-173.

2. Ling, H., Boudsocq, F., Woodgate, R., and Yang, W. (2001). Crystal
structure of a Y-family DNA polymerase in action: a mechanism for error-
prone and lesion-bypass replication. Cell 107, pp. 91-102.

3. Silvian, L. F., Toth, E. A., Pham, P., Goodman, M. F., and Ellenberger, T.
(2001). Crystal structure of a DinB family error-prone DNA polymerase
from Sulfolobus solfatan'cus. Nat. Struct. Biol. 8, pp. 984-989.

4. Trincao, J., Johnson, R. E., Escalante, C. R., Prakash, S., Prakash, L.,
and Aggarwal, A. K. (2001). Structure of the catalytic core of S. cerevisiae
DNA polymerase eta: implications for translesion DNA synthesis. Mol. Cell
8, pp. 417-426.

5. Zhou, B. L., Pata, J. D., and Steitz, T. A. (2001). Crystal structure of a
DinB lesion bypass DNA polymerase catalytic fragment reveals a classic
polymerase catalytic domain. Mol. Cell 8, pp. 427-437.

6. McCulloch, S. D., and Kunkel, T. A. (2008). The fidelity of DNA synthesis
by eukaryotic replicative and translesion synthesis polymerases. Cell Res.
18, pp. 148-161.

7. Nelson, J. R., Lawrence, C. W., and Hinkle, D. C. (1996). Thymine—
thymine dimer bypass by yeast DNA polymerase zeta. Science 272, pp.
1646-1649.

8. Li, Z., Zhang, H., McManus, T. P., McCormick, J. J., Lawrence, C. W., and
Maher, V. M. (2002). hREV3 is essential for error-prone translesion
synthesis past UV or benzo[a]pyrene diol epoxide-induced DNA lesions in
human fibroblasts. Mutat. Res. 510, pp. 71-80.

9. McNally, K., Neal, J. A., McManus, T. P., McCormick, J. J., and Maher, V.
M. (2008). hRev7, putative subunit of hPolzeta, plays a critical role in
survival, induction of mutations, and progression through S-phase, of

UV(254nm)—irradiated human fibroblasts. DNA Repair (Amst) 7, pp. 597-
604.

CHAPTER I

Literature Review

Cellular Responses to DNA Damage

Cellular DNA is subjected to a variety of different types of agents that result in
unfavorable chemical and structural modifications. Even under normal
physiological conditions, DNA is subject to insult by water and reactive oxygen,
resulting in the production of a variety of “spontaneous” DNA lesions (for review
see reference [1]). For example, DNA hydrolysis can result in cleavage of the N-
glycosidic bond, resulting in release of the DNA base and production of an
apurinic or apyrimidinic site. Estimates indicate that in mammalian cells as many
as 10,000 purines and 500 pyrimidines are lost due to hydrolysis each day [2-4].
Abasic sites can be cytotoxic if they block DNA replication, and because they are
noninstructional lesions they have the potential to be highly mutagenic [5]. An
additional consequence of DNA hydrolysis is removal of the exocyclic amino
group of a DNA base i.e. deamination. Most frequently, cytosine is converted to
uracil; however, adenine and guanine can also be deaminated resulting in
conversion to hypoxanthine and xanthine, respectively. Loss of the amino group

changes the base pairing properties of DNA and as result can be miscoding,

leading to mutation. Reactive oxygen species, generated as the byproducts of
various metabolic processes, are another major source of spontaneous damage
to DNA. Thus far, over 80 distinct types of oxidative DNA damage have been
characterized, including oxidized bases, purine aldehyde adducts, DNA-protein
cross-links, and DNA strand breaks [6, 7]. Even the process of DNA replication
can result in a number of errors that, if left unrepaired, would cause a significant

increase in mutagenesis.

In addition to endogenously generated DNA damaging agents, cellular DNA is
also exposed to a variety of physical and chemical damaging agents that come
from external sources (reviewed in [1]). Ultraviolet (UV) radiation from exposure
to sunlight can cause adjacent pyrimidine residues to become fused, resulting in
the formation of the cytotoxic and mutagenic, cyclobutane pyrimidine dimer or the
pyrimidine-pyrimidone (6-4) photoproduct. Ionizing radiation from cosmic rays
and from radioactive materials in the earth causes a broad spectrum of DNA
lesions, including damaged bases and highly lethal DNA double-strand breaks.
Finally, a host of chemical species can react with DNA to form a variety of
potentially mutagenic lesions, including an assortment of alkylating agents, and
cross-linking agents, as well as chemicals that are metabolized into electrophilic

reactants, such as aromatic amines, and polycyclic aromatic hydrocarbons.

Given that DNA damage is so ubiquitous in nature and the fact that it poses such

a significant threat to genomic stability, it is not surprising that cells have evolved

multiple strategies to respond to DNA damage. Many DNA lesions are repaired
by mechanisms that either reverse or remove the damage directly. Repair of
damaged DNA by excision mechanisms can result in release of the lesion either
as a free base, referred to as base excision repair, or as a lesion-containing
oligonucleotide, referred to as nucleotide excision repair. A third distinct form of
excision repair, called mismatch repair, refers to the process of specifically

removing mispaired bases from DNA (reviewed in [8]).

In addition to DNA repair, damage tolerance mechanisms exist, that allow the cell
to mitigate the potentially lethal consequences of DNA damage, i.e. arrested
DNA replication, without physical removal of the damage from DNA. One such
tolerance strategy is called damage avoidance. ln damage avoidance, replication
arrest is prevented by using an undamaged, homologous portion of the DNA as a
template to replicate around the DNA lesion. Alternatively, cells may employ a
mechanistically distinct process of damage tolerance, called translesion
synthesis. Translesion synthesis involves the temporary use of specialized DNA
polymerases, capable of inserting nucleotides opposite DNA lesions, to replicate
past sites of DNA damage. Because the specialized polymerases that perform
translesion synthesis often exhibit relaxed fidelity, a property that confers the
ability to replicate past DNA lesions, translesion synthesis is sometimes

associated with an increase in mutagenesis (see [9] for review).

DNA repair and DNA damage tolerance represent two distinct, but equally
important biological responses to DNA damage, which function to promote cell
survival and maintain genomic stability in the face of constant DNA damage.
Deficiencies in these processes have been linked to a number of different
diseases, including several cancer predispositions. Consequently, DNA repair
and damage tolerance have been and continue to be the focus of a significant
amount of scientific investigation. The following is a review of the current

literature concerning these two biological responses to DNA damage.

I. DNA Repair

To prevent the harmful consequences of DNA damage, including induction of
mutations and premature cell death, a variety of mechanisms have evolved that
are responsible for repairing an enormous spectrum of DNA lesions. DNA repair
specifically refers to those cellular responses that result in the restoration of
damaged DNA to normal nucleotide sequence and DNA structure (see for
example [1, 8]). DNA repair can be divided into several broad classes based on
mechanism. Direct repair refers to processes in which the damage to the DNA is
reversed by an enzyme in a single step reaction. Excision DNA repair
mechanisms are defined as those in which DNA is repaired by removal, i.e.
excision, of the damaged portions, followed by resynthesis using the
complementary strand as a template. Finally, strand breaks can be repaired by
mechanisms involving recombination or by end-joining mechanisms. Although
each one of these DNA repair mechanisms are of fundamental importance to
maintaining genomic integrity, the primary focus of this review is on the repair
processes that operate on damaged or inappropriate nitrogenous bases in DNA.
Specifically, the mechanisms of base excision repair, nucleotide excision repair

and mismatch repair are covered in detail.

A. Base Excision Repair

10

Base excision repair (BER) is considered to be the most frequently utilized
pathway for removal of DNA lesions. The base excision repair pathway is
involved in the repair of many different types of DNA lesions, including those
generated by hydrolytic deamination of bases, oxidative damage, alkylating

agents and ultraviolet light [10].

1. Base Excision Repair Pathways

The mechanism of base excision repair in prokaryotes and eukaryotes is well
understood (for review see [1, 10, 11]). Base excision repair is initiated by
enzymes called DNA glycosylases, which recognize DNA lesions and cleave the
N-glycosidic bond linking the damaged bases to the DNA backbone (Figure 1).
This cleavage event results in excision of the damaged base and thus, this repair
process, where lesions are released as free bases, is referred to as “base

excision repair”.

DNA glycosylases fall into two general Classes depending on their mechanism of
catalysis. The first class of glycosylases, called monofunctional glycosylases,
eliminate the damaged nucleoside, leaving behind an abasic site. The second
class of glycosylases are called bifunctional glycosylases because, in addition to

hydrolysis of N-glycosidic bonds, these enzymes support an apurinic/apyrimidinic

11

Figure 1 - The base excision repair pathway. Base excision repair is initiated
by DNA glycosylases, which recognize DNA lesions and cleave the N-glycosidic
bond linking the damaged bases to the DNA backbone. Cleavage by
monofunctional glycosylases leaves an abasic (AP) Site. Bifunctional
glycosylases possess an additional AP lyase activity that results in cleavage of
the DNA backbone on the 3’ side of the newly generated abasic site. Immediately
following removal of the damaged base by a DNA glycosylase, an AP
endonuclease nicks the DNA backbone immediately 5’ of the AP site. When AP
endonuclease acts on the product of a monofunctional DNA glycosylase,
cleavage results in a free 3’-OH and 5’-deoxyribosephosphate (dRP) group. At
this point, BER continues through one of two distinct pathways termed short-
patch repair (right) and long-patch repair (left). In short patch repair, the
remaining one-nucleotide gap is filled by a repair DNA polymerase and the 5’-
dRP group is removed by an enzyme possessing deoxyribosephosphate lyase
(dRPase) activity. Finally, the DNA ends are sealed by a DNA ligase. In long-
patch repair, a more extensive patch of DNA is synthesized, displacing the strand
containing the 5’dRP group. This 5’ flap is cleaved by a flap endonuclease and
the resulting DNA ends are sealed by DNA ligase. When an AP endonuclease
cleaves the product of a bifunctional DNA glycosylase, the blocked 3’ end is
removed and repair polymerization and ligation can occur directly. This figure

was adapted from reference [10].

 

 

 

 

 

 

 

 

5! % 3!
5’

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3’
R if:
Monofunctional Bifunctional
DNA glycosylase DNA glycosylase
5’ 3’ 5’ 3’
3, L 59 3, l 5!
AP endonuclease AP endonuclease
59 dRP 39 59 H P 39
3, I 59 3, I 59
DNA DNA
synthesis synthesis
dR
dRP
9 9 9 H 9
5 H 3 5 0 3 DNA
39 5’ 3r 5’ SYDthBSIS
Flap
dRPase endonuclease
5! H F. 3) 5! H p 3’
I +—
3' 5' 3' k 5'
DNA ligase
5’ 3’
3’ 5’

13

(AP) lyase activity that results in cleavage of the DNA backbone on the 3’ side of
the newly generated abasic Site. The lyase activity of bifunctional glycosylases
generates a 3’ end that lacks a hydroxyl and must be removed before DNA

synthesis can occur.

After removal of the damaged base by a DNA glycosylase, processing of the
abasic site is initiated by an AP endonuclease. These enzymes generate a nick
in the DNA backbone immediately 5’ of the abasic site. When AP endonuclease
acts on the product of a monofunctional DNA glycosylase, cleavage of the DNA
backbone 5’ to the abasic site results in a free 3’-OH and 5’-
deoxyribosephosphate (dRP) group. At this point, BER continues through one of
two distinct pathways, termed short-patch repair and long-patch repair. In short
patch repair, the one-nucleotide gap that remains after excision is filled by a
repair DNA polymerase, and the 5’-dRP group is removed by an enzyme
possessing deoxyribosephosphate lyase (dRPase) activity. Finally, the DNA
ends are sealed by a DNA ligase. ln long-patch repair, a more extensive length
of DNA (2-20 nucleotides) is synthesized, displacing the strand containing the
5’dRP group. This 5’ flap is then cleaved by a flap endonuclease, and the
resulting DNA ends are sealed by DNA ligase. When an AP endonuclease
cleaves the product of a bifunctional DNA glycosylase, the blocked 3’ end is

removed and repair polymerization and ligation can occur directly.

a. DNA Glycosylases

14

The base excision repair pathway is utilized to repair many different types of
DNA lesions, including those generated by hydrolytic deamination of bases,
oxidative damage, alkylating agents and ultraviolet light. To distinguish between
such a large variety of DNA lesions, each species possesses many different DNA
glycosylases that recognize one, or at most a few, specific types of DNA lesions
based on their specific structural folds and motifs. DNA glycosylases have been
divided into five classes based on their structure and substrate specificities
(reviewed in [1, 10, 11]). The uracil-DNA glycosylase family consists of
monofunctional DNA glycosylases, including the uracil N-DNA glycosylases
(UNGs), the mismatch specific uracil-DNA glycosylases (MUGs), the thymine-
DNA glycosylases (TDGS), and the single-strand-selective mono-functional
uracil-DNA glycosylase (SMUGS). Uracil-DNA glycosylases recognize uracil or
thymine mismatches in DNA. The Fpg/Nei superfamily, defined by the E. coli
enzymes Fpg (also known as MutM) and endonuclease VIII (also known as Nei),
specializes in recognizing oxidative DNA damage. This superfamily was thought
to be unique to bacteria until 2002, when three homologous Nei-like proteins
(NEIL1-NEIL3) were identified in humans [12-14]. Members of the helix-hairpin-
helix (HhH) superfamily are characterized by a core helix-hairpin-helix motif
found in many DNA-dependent enzymes [15]. The HhH superfamily is the most
functionally diverse superfamily of DNA glycosylases. It includes both
monofunctional and bifunctional glycosylases, that recognize a variety of different
types of DNA lesions [10]. In fact, the HhH superfamily is so diverse that

glycosylases from this superfamily cover nearly the entire spectrum of lesions

15

repaired by BER [11]. Finally, there are two single member superfamilies that
have no known structural relatives: the alkyladenine glycosylase (AAG)
superfamily, which is defined by the DNA glycosylase AAG, a monofunctional
glycosylase that recognizes alkylation damage [16], and the endonuclease V
superfamily, which is defined by endonuclease V from the T4 virus, a small
bifunctional glycosylase that recognizes UV-induced thymine-thymine dimers

[17].

b. ApuriniclApyrimidinic (AP) Endonucleases

In contrast to the multitude of DNA glycosylases that exist, each organism
possesses only a few distinct AP endonucleases. AP endonucleases have thus
far been divided into four structural classes. Classes l and II are defined by the
bacterial AP endonucleases, Xth (also known as exonuclease Ill) and Nfo (also
known as endonuclease IV) (reviewed in [10]). Xth is the predominant AP
endonuclease in E. coli. In addition to AP endonuclease activity, Xth also
possesses 3’—>5’ exonuclease activity for which it was originally characterized
[18]. The major human AP endonuclease, APE1, and its recently discovered
paralog, APE2, are homologous to Xth [19, 20]. Bacteria contain a second AP
endonuclease, Nfo, although it is thought to account for only 10% of the total
bacterial AP endonuclease activity [21, 22]. The major S. cerevisiae AP
endonuclease, APN1, is homologous to Nfo [23]. Two groups have also reported
the existence of a third class of AP endonucleases, defined by the UV damage

endonuclease, UVDE, which is found in some fungi and bacteria [24, 25]. More

16

recently, a fourth class of AP endonucleases has been defined based on the

discovery of a third human AP endonuclease, called PALF [26].

c. Post-incision Events in Base Excision Repair

When an AP endonuclease acts on the product of a monofunctional DNA
glycosylase, cleavage of the DNA backbone 5’ to the abasic site results in a one
nucleotide gap and a dRP group that must be removed. If the base excision
repair process proceeds through the short-patch repair pathway, the one-
nucleotide gap is filled by a repair DNA polymerase, and the dRP group is
removed by an enzyme possessing dRPase activity. In bacteria, the major DNA
repair polymerase, Pol I is used for gap filling in short-patch BER [27]. Bacterial
Pol I also possesses a 5’—>3’ activity and can degrade the blocked dRP terminus.
However, several other enzymes from E. coli, including exonuclease l, RecJ, and
Fpg have also been proposed as E. coli dRPases [28-30]. In mammalian short-
patch BER, DNA polymerase [3 can fill in the one-nucleotide gap and can
subsequently remove the 5’ dRP group from a blocked terminus [31]. The
dRPase activity of Pol [3 is apparently required for repair of at least some types of
DNA lesions because the hypersensitivity of cells lacking Pol [3 to methyl
methane sulfonate can be complemented by overexpression of the dRPase
domain of Pol [3, but not by overexpression of the polymerase domain [32]. Once
the dRP group is removed, the newly generated DNA ends are sealed by DNA

ligase l in E. coli, or the DNA ligase 3a-XRCC1 complex in humans.

17

lf processing of the dRP residue is not efficient, for example, because the dRP
residue becomes reduced, long-patch BER can occur. In long-patch BER, a
more extensive patch of DNA is synthesized displacing the strand containing the
5’dRP group. This displaced strand must then be cleaved by a flap
endonuclease. In E. coli, if the 5’ dRP residue cannot be removed, Pol I
displaces the dRP-containing strand and then cleaves the strand using its
5’—>3’exonuclease activity [33]. In human cells, if Pol [3 is unable to remove the 5’
dRP group, a polymerase switch may occur and Pol 8 or Pol 8 together with
accessory factors RFC, PCNA and RPA add several nucleotides generating a
flap, which is then removed by the flap endonuclease, FEN1 [34]. In humans,
ligation of the DNA ends generated in long-patch BER is achieved by DNA ligase
I rather than DNA ligase 3a (as in short-patch BER), probably as a consequence

of its interaction with PCNA.

Finally, if an AP endonuclease acts on the product of a bifunctional DNA
glycosylase, it removes the blocked 3’ end, leaving a free 3’-OH which is suitable
for repair polymerization and ligation by Pol l and DNA ligase l in E. coli, and by

Pol [3 and Ligase 3a in humans.

B. Nucleotide Excision Repair

The idea that DNA can be repaired by the physical removal of a lesion was first

suggested in the 19603, when two independent groups reported that in bacteria,

18

UV-induced cyclobutane thymine-thymine dimers (T-T dimers) were eliminated
from DNA by excision [35, 36]. It is now understood that excision of damaged
sequences from DNA is one step in a complex pathway that has come to be
known as nucleotide excision repair (NER). The nucleotide excision repair
process works through a “cut and patch” mechanism whereby a stretch of DNA
containing damage is excised and then restored by DNA repair synthesis
(reviewed in [37]). Nucleotide excision repair is the primary repair process for
removal of a variety of bulky DNA adducts, such as those formed by exposure to
UV radiation i.e. T—T dimers and 6-4 photoproducts, as well as those formed by
exposure to chemicals such as benzo[a]pyrene or acetylaminofluorene. In
addition, single base lesions can be removed through the NER pathway, and
therefore, NER can function as a back-up system for lesions escaping base
excision repair. Because NER is required for repair of such a diverse set of
genotoxic DNA lesions, defects in NER can be detrimental to an organism’s
survival. Nucleotide excision repair consists of two mechanistically similar
pathways termed global genomic NER (GG-NER) and transcription-coupled NER
(TC-NER). The global genomic NER pathway removes DNA lesions both from
transcriptionally active and transcriptionally silent portions of the genome,
discriminating the frequency of lesion removal only based on chromatin status. In
contrast, transcription-coupled NER is highly specific for removal of DNA lesions

generated in genes that are being actively transcribed (reviewed in [38]).

1. Global Genomic Nucleotide Excision Repair in Prokaryotes

19

The nucleotide excision repair process consists of four basic steps: (1)
recognition of damaged DNA, (2) excision of the DNA damage by dual incisions,
(3) DNA synthesis to repair the resulting gap, and (4) ligation. Prokaryotic NER is
primarily controlled by the Uvr proteins, which were originally identified based on
their ability to complement strains of E. coli defective in the ability to repair DNA

lesions induced by UV radiation [39].

In prokaryotes, the damage recognition factor is generated in solution, when two
subunits of the UvrA protein dimerize in an ATP-dependent manner and then
bind to the DNA helicase, UvrB (Figure 2). Once this “damage recognition
complex” is formed, UvrA, which is highly specific for binding damaged DNA,
binds at or near the site of a DNA lesion. When the damage recognition complex
has identified a DNA lesion, UvrB, using its helicase activity, unwinds and bends
the DNA at the Site of damage. Following this conformational change, UvrA
dissociates resulting in formation of the UvrB-DNA pre-incision complex. The pre-
incision complex is then able to be recognized and bound by the endonuclease,
UvrC. The UvrC endonuclease generates two incisions in the DNA; a 3’ incision,
4-5 nucleotides from the site of the lesion, followed by a 5’ incision that occurs
exactly 8 nucleotides from the lesion. This results in excision of the DNA
oligonucleotide containing the lesion, but it remains bound to the post-incision
complex until the Uer helicase displaces both the UvrC protein and the lesion-

containing oligonucleotide. The UvrB protein remains bound to the single

20

Figure 2 - The two pathways of prokaryotic nucleotide excision repair
(NER). In prokaryotes, global genomic NER (left) is initiated when the damage
recognition complex, consisting of two subunits of the UvrA protein and one
subunit of the UvrB DNA helicase, recognizes and binds to DNA at the site of
damage. Transcription-coupled NER (right) is initiated when a lesion on the
transcribed strand of DNA blocks the progression of RNA polymerase. The
stalled RNA polymerase is identified and bound by the transcription-repair
coupling factor (TRCF) at the site of DNA damage. Upon binding the damaged

DNA, TRCF displaces RNA polymerase along with the unfinished transcript. The

UvrAzB complex then recognizes TRCF bound at the site of DNA damage.

Although the mechanisms for lesion recognition differ between the two classes of
NER, once the DNA lesion has been identified, the two pathways converge and
proceed through the same series of mechanisms. First, the area surrounding the
lesion is unwound and bent by the UvrB helicase. The UvrB-DNA complex is
then bound by the endonuclease UvrC. The UvrC endonuclease generates an
incision on either side of the DNA damage. After the DNA oligonucleotide
containing the lesion has been excised, it remains attached until the Uer
helicase displaces both the UvrC protein and the lesion-containing
oligonucleotide. The UvrB protein remains bound to the single stranded DNA to
prohibit degradation, until it is displaced by DNA polymerase I. Finally, the
resultant gap is filled by DNA polymerase I, and the newly synthesized DNA end

is sealed by DNA ligase. This figure was adapted from reference [38].

21

5!
3!

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

,.
,.

rR—rl
UVBTUWAT 5'

3’

 

 

 

 

[UvrAJ [um +>\

5’
3
5!

3’

 

 

 

 

 

 

 

 

 

 

 

 

0/01»

 

 

@333

 

 

 

 

 

 

 

 

UvrA UvrA

4,-

 

 

 

:; muffin... :3 mm
W T

: Illlligfillll ;;

 

 

 

 

 

 

 

gums * Vfi :

3’ Incision

 

 

 

 

 

 

.WWW

 

 

 

 

 

 

 

III

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

22

2. Transcription-Coupled Nucleotide Excision Repair in

Prokaryotes

In E. coli, DNA lesions that occur in actively transcribed portions of the genome
are preferentially repaired [41]. Such transcription-coupled repair occurs via an
NER pathway that overlaps the global genomic NER pathway described
previously. In transcription-coupled NER, however, recognition of the DNA lesion
is initiated by a stalled RNA polymerase complex (Figure 2). If RNA polymerase
becomes stalled by the occurrence of a DNA lesion, its very presence inhibits

global genome NER by interfering with the ability of the damage recognition
complex, UvrAzB, to bind to the site of DNA damage. In this case, the
transcription-repair coupling factor, TRCF binds to the site of DNA damage and
displaces RNA polymerase along with the unfinished transcript. The UvrAzB

complex recognizes TRCF bound at the site of DNA damage. Although the initial
recognition step differs in TC-NER, requiring a stalled RNA polymerase and
TRCF, excision of the DNA lesion, repair synthesis, and ligation proceed via the

same process as in GG-NER [38].

3. Global Genomic Nucleotide Excision Repair in Eukaryotes

Nucleotide excision repair in eukaryotes proceeds through a similar set of

reaction mechanisms as those used by prokaryotic NER, beginning with

23

recognition of DNA damage, followed by excision of the DNA lesion via dual
incisions, and finally gap-filling DNA synthesis. However, eukaryotic NER
requires the function of a larger set of proteins. Surprisingly, although the
mechanistic properties of the eukaryotic NER proteins are highly similar to the
prokaryotic Uvr proteins, there is no evolutionary relationship between the

proteins required for NER in prokaryotes and eukaryotes [42].

Most researchers agree that, in human NER, the formation of the pre-incision
complex requires at least four factors, replication protein A (RPA), XPA, the XPC—
RAD23B complex and the transcription factor TFIIH. However, it remains unclear
which of these proteins is the true damage recognition factor. All of these factors
Show a preference for binding damaged DNA, but the order of assembly of these
factors (if a specific order of assembly is necessary) remains unproven [43].
Adding to the already confusing assembly scheme in human NER is the fact that
the damaged DNA binding protein (DDB) has been shown to play a role in
recognition of cyclobutane pyrimidine-pyrimidine dimers (CPDs), the most

frequent lesion induced by UV radiation [44].

Currently, the most commonly accepted scheme for assembly of the pre-incision
complex (Figure 3) is as follows. First, the site of DNA damage is recognized by
the XPC-RADZ3B complex in collaboration with the DDB complex. After the DNA

lesion has been recognized, XPC-RADZBB and/or the DDB complex recruit the

24

Figure 3 - The global genomic nucleotide excision repair pathway in
eukaryotes. The site of a DNA lesion is recognized and bound by the XPC-
RAD23B complex in collaboration with the DDB complex. After the DNA lesion
has been identified, the transcription factor TFIIH is also recruited to the Site of
the DNA lesion. TFIIH unwinds DNA at the lesion site using the helicase activity
of its subunits XPB and XPD. The single-stranded DNA generated by the
combined action of XPC-RAD23B and TFIIH facilitates recruitment of the single-
stranded binding protein, RPA, along with XPA, to verify the presence of DNA
damage. The XPG nuclease makes the first incision 3’ of the lesion and then the
XPF-ERCC1 complex makes an incision 5’ of the lesion. The damaged DNA is
released as a 25-30 residue oligonucleotide. The resulting DNA gap is filled by
DNA polymerase 8/s along with the accessory factors, PCNA and RFC. In the
final step of NER, the newly synthesized DNA end is sealed by DNA ligase I.

This figure was adapted from reference [45].

 

:IIIIIIII‘I’IIIIIIII:

 

 

vTFl-H

:IIIIIIIIIIIII

transcription factor TFIIH to the DNA lesion. TFIIH unwinds DNA at the site of the
lesion using the helicase activity of subunits XPB and XPD, creating short
stretches of single-stranded DNA surrounding the lesion. The Single-stranded
DNA generated by the combined action of XPC-RAD23B and TFIIH facilitates
recruitment of the single-stranded binding protein, RPA, along with XPA, to verify
the presence of DNA damage. The opening of the pre—incision complex results in
25-30 residues of single-stranded DNA, which is the substrate for cleavage by
the nucleases XPG and XPF-ERCC1. The XPG nuclease makes the first incision
3’ of the lesion, and then the XPF-ERCC1 complex makes an incision 5’ of the
lesion. The damaged DNA is then released as a 25-30 residue oligonucleotide.
The resulting DNA gap is filled by DNA polymerase 8/s along with the accessory
factors, proliferating cell nuclear antigen (PCNA) and replication factor C (RFC).
In the final step of NER, the newly synthesized DNA end is sealed by DNA ligase

I (for review see [43, 45]).

4. Transcription-Coupled Nucleotide Excision Repair in
Eukaryotes

The mechanism of transcription-coupled NER in mammalian cells is much less
well understood then that of prokaryotes. However, many proteins thought to play
a role in TC-NER have been identified. The two most well known factors that
participate in human TC-NER are the CSA and CSB proteins. These two proteins
were identified based on their ability to complement impaired TC-NER in cells
derived from patients suffering from Cockayne syndrome (see below) [46]. The

CSB protein interacts with the transcription machinery dynamically, and this

27

interaction is stabilized in the presence of DNA damage [47]. CSB serves as the
transcription-repair coupling factor by recruiting the pre-incision NER factors and
the CSA protein to the site of DNA damage [48, 49]. The CSA protein has been
shown to be part of an E3-ubiquitin ligase complex also consisting of DDB1,
Culin 4A and Roc1/be1. In response to UV radiation, the CSA complex was
shown to associate with the COP9 signalasome (CSN), resulting in inactivation of
the ubiquitin ligase activity of the CSA complex. These data suggests that
although CSA is absolutely required for TC-NER, the ubiquitin ligase function of

the CSA complex is dispensable [50].

The XBA2 and the HMGN1 proteins also play a role in mammalian TC-NER.
Both factors have been shown to interact with stalled RNA polymerase II
(RNAPII) complexes in a UV and CS-dependent manner [48]. The XBA2 protein
has also been shown to interact with the XPA protein, suggesting that XBA2
functions as a scaffold factor in TC-NER [51]. HMGN1 is a nucleosome binding
protein that is thought to activate the pre-incision complex by facilitating
chromatin remodeling and activating histone hyperacetylation by the histone
acetyltransferase, p300 [48]. Changes in chromatin structure are known to occur
during NER and remodeling of the DNA/RNAPII interface may be required for
repair factors to access the DNA lesion and/or to facilitate replication restart [52,
53]. Finally, unlike prokaryotic TC-NER, in mammalian TC-NER, RNAPII likely
remains bound at the site of the DNA lesion [48, 54-57]. This would be expected

to be a barrier to access and repair the DNA lesion. This barrier might be

28

removed by RNAPII backtracking in the presence of the transcription factor,
TFIIS. It has been speculated that TFIIS is recruited to a stalled RNAPII complex
in a CS—dependent manner shifting the transcription bubble backwards allowing
access to the DNA lesion [37]. After the DNA lesion is repaired, TFIIS can also
stimulate cleavage of the extruded mRNA to reposition the 3’ end by RNAPII

allowing for transcription restart [58].

Based on these studies, a tentative model for TC-NER in humans has been
suggested (Figure 4). If the RNAPII encounters a DNA lesion that it is unable to
bypass during transcription, CSB forms a stable complex with stalled RNAPII.
CSB facilitates chromatin remodeling by recruiting the histone acetyltransferase,
p300 and also promotes recruitment of the pre-incision NER factors to the DNA
damage site. CSB is also required for recruitment of the CSA-DDB1 E3-ubiquitin
ligase/CSN complex. Once both CSB and the CSA complex are present,
HMGN1, XAB2 and the TFIIS proteins are recruited to the site of DNA damage to

facilitate further chromatin remodeling and transcription restart [37, 48].

5. Nucleotide Excision Repair Deficiency Diseases

At least three categories of disease result from impaired NER, i.e., xeroderma
pigmentosum (XP), Cockayne syndrome (CS), and trichothiodystrophy (TT D).
Xeroderma pigmentosum is caused by defects in the mammalian NER system

[59]. There are seven XP complementation groups, designated XP-A through

29

Figure 4 - The transcription-coupled nucleotide excision repair pathway in
eukaryotes. The CSB protein interacts with the transcription machinery
dynamically, and this interaction is stabilized in the presence of transcription-
blocking DNA damage. CSB serves as the transcription-repair coupling factor by
recruiting the pre-incision NER factors, RPA, XPA, XPG and the transcription
factor TFIIH, as well as the histone acetyl transferase p300, and the CSA-DDB-
CSN complex to the site of DNA damage. CSA facilitates further chromatin
remodeling by recruiting HMGN1, XABZ and TFIIS. These remodeling events
enable cleavage on each side of the DNA lesions by the XPF-ERCC1 complex.
Gap filling synthesis is then conducted by the Pol O/E complex, and the resultant
DNA ends are ligated by DNA Ligase l. After the DNA lesion is repaired, TFIIS
can stimulate cleavage of the extruded mRNA to reposition the 3’ end allowing

for transcription restart. This figure was adapted from references [37, 48].

 

 

  
 
 

POI ble
RFC, PCNA
Ligase I

3’ .
TFIIS
Transcriptionfl
Restart

 

 

31

XP-G, that are the result of mutations in seven different proteins that are involved
in NER (note that one additional complementation group, XP-V, results from
mutations in DNA polymerase Ti and will be discussed in the upcoming section
on translesion synthesis). Because NER is impaired in XP patients, they retain
DNA damage and as a result, demonstrate severe photosensitivity and are at

least a 1000 times more susceptible to UV-induced skin cancer.

Cockayne syndrome is a rare autosomal recessive disorder initially described in
the 19305 by pediatrician Edward Cockayne [60]. Cockayne syndrome results
from defects in either the CSA or much more frequently, the C88 protein [46].
Both the CSA and C88 proteins are required for TC-NER, and as a
consequence, one hallmark of CS is the inability of cells to resume RNA
synthesis following exposure to various DNA damaging agents [61]. Patients with
CS exhibit a variety of symptoms, including dwarfism, impaired development of
the nervous system, premature aging, and photosensitivity, but in contrast to XP

patients, CS patients are not more susceptible to cancer [62].

Trichothiodystrophy is caused by mutations in the XPD, XPB, or TTDA proteins,
all of which are subunits of the transcription factor TFIIH. Because TFIIH plays a
role in transcription as well as in NER, it is unclear whether the phenotype of TTD
patients results from defects in transcription or in DNA repair [45, 63]. Patients
with ‘I'I'D demonstrate developmental and neurological abnormalities similar to

patients with Cockayne syndrome but in addition, exhibit the hallmark brittle hair

32

and nails and scaly skin, which result from a defect in synthesis of sulfur-rich

proteins.

C. Mismatch Repair

DNA mispairs can occur as a result of several processes, most commonly by
misincorporation of nucleotides by the replicative DNA polymerases during DNA
synthesis (reviewed in [1]). If mismatches in DNA are not corrected, they can
lead to mutations. The mismatch repair pathway (MMR) is responsible for
correcting most DNA mispairs, thereby preventing the introduction of mutations.
Unlike other DNA repair processes, where the target for repair, a DNA lesion,
can be identified based on its abnormal structure, for mismatch repair the
mismatched base is a normal constituent of DNA. As a result, the key component
for avoiding mutations by mismatch repair is identifying which of the mispaired
bases was erroneously generated. Inactivation of the mismatch repair system
results in Spontaneous mutation rates that are 50-1000 times higher than normal,
indicating the importance of the MMR pathway for protecting against

mutagenesis (reviewed in [64]).

1. Mismatch Repair in Prokaryotes

The MMR pathway has been most extensively studied and is best characterized

in E. coli (for review see [1, 64, 65]). As described above, the greatest barrier to

33

correctly repairing a DNA mismatch is directing repair to the newly synthesized
daughter strand. In E. coli, the problem of strand discrimination is reconciled by
use of methyl-directed mismatch repair. Because newly synthesized DNA in E.
coli is subject to methylation at GATC sites, and because this methylation occurs
slightly behind DNA synthesis, the difference in methylation status between the

parent and daughter DNA strand permits strand discrimination [66].

The mismatch repair pathway in E. coli proceeds through four basic steps:
incision of the unmethylated strand near the site of the mispair; excision of the
DNA strand containing the mismatch; DNA synthesis to fill the resulting gap; and
ligation (Figure 5). The first step of the MMR pathway requires recognition of the
DNA mismatch. In E. coli, the “mismatch recognition factor” is a homodimer of
the MutS protein, which recognizes and binds to single base DNA mispairs as
well as small insertion/deletion loops generated by replication slippage [67, 68].

Once bound to the DNA mispair, MutS recruits a homodimer of MutL in an ATP-

dependent manner [69]. After the MutSz-Muth complex has been formed, the

MutH endonuclease is activated and nicks the unmethylated strand of
hemimethylated DNA at a GATC site. The nicked DNA strand serves as a signal
to direct the ensuing excision repair enzymes to remove the mismatch-containing
portion of the unmethylated strand. Cleavage of the unmethylated DNA strand by
MutH is bidirectional and as such, can occur either 5’ or 3’ of the DNA mispair
[70, 71]. DNA helicase ll recognizes the nicked strand and unwinds the DNA

toward the mismatch [72]. Depending on the position of the DNA nick relative to

34

Figure 5 - The prokaryotic mismatch repair pathway. A homodimer of the
MutS protein, recognizes and binds to single base DNA mispairs (indicated in
red) as well as small insertion/deletion loops generated by replication slippage.

Once bound to the DNA mispair, MutS recruits a homodimer of MutL. The

MutSz-Muth complex recruits the MutH endonuclease, which nicks the

unmethylated strand of hemimethylated DNA at a GATC Site. The nicked DNA
strand serves as a signal to direct the ensuing excision repair enzymes to
remove the mismatch-containing portion of the unmethylated strand. Cleavage of
the unmethylated DNA strand by MutH is bidirectional and can occur either 5’
(right) or 3’ (left) of the DNA mispair. DNA helicase ll recognizes the nicked
strand and unwinds the DNA toward the mismatch. Depending on the position of
the DNA nick relative to the mismatch, the displaced portion of the nicked strand
is degraded from the nick to Slightly past the DNA mispair either by the 3’—>5’
exonucleases Exol, Exonl, or Eon (right) or the 5’-—>3’ exonucleases Exonl or
RecJ (left). The resulting DNA gap is then stabilized by single-stranded DNA
binding protein, and filled by DNA polymerase III. The DNA ends are sealed by

DNA ligase. This figure was adapted from reference [64].

5!
3’

5!
3’

3!

5’
3’

CH3 CH3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5, G 3’
3, A 5.
MutS, \
MutL, ATP
MUISZ MUtSz

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3’ 5’ -
2 9 9 / I
(MugE MutL) 5 3 (Mutg Matt.)
1 MutH, ATP 1

 

 

 

 

 

 

 

MUISZ MUISZ
CH CH CH CH
I 3 m 3 3’ 5’ 3 I 3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

........ .. ]

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

39
5’

3’
5!

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3’ 5’
5' 3’
Exol, ExoVIl 6)
l or Eon l
IIIIGI 3’ 5’ GIIIII 3’
5’ 3’ 5’
DNA Pol III, 888,
DNA ligase
CH3 CH3 CH3 CH3
G 3’ 5’ G 3’
g 59 3! Q 5!

36

the mismatch, the displaced portion of the nicked strand is degraded from the
nick to slightly past the DNA mispair, either by the 3’—>5’ exonucleases Exol,
Exonl, or Eon or the 5’-—>3’ exonucleases Exonl or RecJ [73, 74]. The
resulting DNA gap is then stabilized by single-stranded DNA binding protein, and
filled by DNA polymerase III. Finally, the DNA ends are sealed by DNA ligase

completing the E. coli MMR pathway [75].

2. Mismatch Repair in Eukaryotes

The mismatch repair process is thought to be highly conserved from prokaryotes
to eukaryotes with similarities including substrate specificity, bidirectionality, and
nick-directed strand specificity (for review see [1, 64, 65]). Many of the eukaryotic
MMR proteins have been identified based on their homology to the E. coli MMR
proteins. In humans, five MutS homologs (MSH2-MSH6) and four MutL
homologs (MLH1, MLH3, PMS1 and PMS2) have been identified, although the
primary function of some of these proteins seems to be in processes other than
MMR [76-85]. In humans, unlike in E. coli, the MutS and MutL homologs function
as heterodimers. Human MSH2 can dimerize with either MSH6 to form MutSa
[86, 87] or with MSH3 to form MutSfi [88]. Similar to their homologs in E. coli,
both MutSa and MutSB are mismatch recognition factors. MutSa recognizes
single base mismatches as well as small insertion/deletion loops, whereas MutSB
seems to be involved only in recognition of larger insertion/deletion loops [87,

89]. Of the four MutL homologs, only MutLa, composed of a heterodimer of

37

MLH1 and PMSZ, is required for MMR in humans [90]. In addition to the MutS
and MutL homologs, several other factors have been demonstrated to be
required for human MMR, including the exonuclease, Exo1 [91, 92]; the single-
stranded binding protein, RPA [93, 94]; proliferating cellular nuclear antigen
(PCNA) [95-97]; replication factor C (RFC) [98, 99]; DNA Polymerase 8 [100];

and DNA ligase | [99].

Several studies have demonstrated that the initial steps of human MMR are
similar to those in E. coli. Specifically, MutSa or MutSB first binds to a DNA

mismatch and then recruits MutLa in an ATP-dependent manner [97, 101-104].
In E. coli, after the MutSz-Muth complex has been formed, the MutH

endonuclease is activated and nicks the unmethylated newly synthesized DNA
strand at a GATC site, thereby conferring strand discrimination. However, in
eukaryotic cells, strand discrimination must occur through a distinct mechanism
because eukaryotes do not posses the Dam methylase enzyme [1]. Although the
signals that direct strand discrimination in human cells are unknown, strand
specificity is thought to be nick-directed because in human cell-free extracts

nicked DNA is sufficient to direct strand specificity [105, 106].

In E. coli, the excision step of MMR requires the function of DNA helicase II and
multiple exonucleases. In contrast, in human MMR there appears to be no
requirement for a DNA helicase function [107-109], and only one exonuclease,

Exo1, has been convincingly implicated thus far [64]. Similar to bacteria, in

38

humans, excision occurs both in a 5’—>3’ and a 3’—>5’ direction based on the
placement of the strand discriminating nick, however in humans the two
processes occur through distinct mechanisms. When a nick is placed 5’ of the
mispair, MutSa stimulates the 5’—>3’ exonuclease activity of Exol, which begins
excision at the 5’ nick and continues until the mispair has been removed [89].
Termination of the exonuclease activity after removal of the mispair is dependent
on RPA [89]. This 5’-directed excision reaction mechanism requires the activity of
only three proteins, Mutsa, Exol and RPA, although MutLa does enhance the
mismatch dependence of the reaction [89]. In contrast, when the nick is placed 3’
of the mispair, the activity of at least two other factors, RFC and PCNA are
required [98]. In 3’-directed excision, RFC and PCNA suppress the 5’—>3’
exonuclease activity of Exol and activate a cryptic 3’—>5’ exonuclease activity,
resulting in removal of the mispair [98]. It has been suggested that the
directionality of excision is regulated by the orientation in which PCNA is loaded
[64]. In other words, a different face of the PCNA clamp would be oriented toward
a mispair located 5’ of the nick than would be oriented if the mispair was located

3’ of the nick.

3. Mismatch Repair Deficiency Leads to Cancer

The primary role of the MMR pathway is to remove incorrectly paired bases or
insertion/deletion loops generated during DNA synthesis, thereby offering

protection against mutation. Deficiencies in MMR have been associated with an

increased rate of cancer development. Most commonly, defects in MMR have
been associated with development of hereditary nonpolyposis colon cancer (for
review see [1, 110]). Hereditary nonpolyposis colon cancer (HNPCC) or Lynch
Syndrome [111] is an autosomal dominant disorder and is one of the most
common cancer predisposition diseases of humans. Patients with HPNCC have
an ~80% risk of developing colon cancer in their lifetime. Thus far, mutations in at
least seven genes associated with MMR, including MLH1, MSH2, MSH6, PMS1,
PMSZ, MLH3, and EXO1 have been linked to HNPCC. Mutations in mismatch
repair genes are thought to contribute to HNPCC by causing an accumulation of
deletion or insertion mutations in simple, repetitive microsatellite sequences, a

hallmark of HNPCC.

4o

ll. DNA Damage Tolerance

As described above, many of the lesions generated in DNA can be repaired by
excision of the damaged portion of the DNA. However, if DNA repair is slow or if
the DNA damage is extensive, lesions may persist into DNA replication. The
presence of lesions during DNA synthesis can cause replication to arrest, which
can ultimately lead to cell death. To avoid the threat of cell death, cells have also
evolved strategies for coping with the DNA lesions that remain during DNA
replication. Because these mechanisms do not result in the physical removal of
the DNA damage, but rather prevent the potentially lethal consequences of
arrested replication, they are collectively referred to as DNA damage tolerance
mechanisms. One such damage tolerance mechanism is translesion synthesis
(TLS). Translesion synthesis is the process in which specialized DNA
polymerases insert nucleotides opposite DNA lesions, thereby alleviating
replication arrest. Depending on which specialized polymerase is involved, the
type of DNA lesion being bypassed, and even the sequence context surrounding
the DNA lesion, TLS can be either error-free or error-prone. In many organisms,

error-prone TLS is a major source of mutations (reviewed in [112]).

In contrast to TLS, where the damaged portion of the DNA is used as a template
to direct DNA synthesis, cells can also employ a second method of damage
tolerance that makes use of a strand switch mechanism, whereby a homologous

copy of the damaged DNA strand is temporarily used as an error-free template to

41

replicate past the DNA lesion. In this mechanism, synthesis across from the DNA
lesion is not required, and thus this mechanism is often referred to as DNA
damage avoidance. Because an undamaged, homologous strand of DNA is used
as the template to direct DNA synthesis, damage avoidance is typically an error-

free mechanism of damage tolerance (for review see [9]).

A. DNA Damage Tolerance Mechanisms in Prokaryotes

Although the existence of DNA damage tolerance mechanisms was recognized
only relatively recently. Many of the important advances that led to their
discovery were based on studies of the mechanism of mutagenesis in E. coli
(reviewed in [1]). As it turns out, many of the mechanisms of DNA damage
tolerance first elucidated in bacteria are conserved in eukaryotes, including
humans. Because of the speed and ease with which bacterial systems can be
manipulated, prokaryotes are still considered to be among the most valuable
models available for studying DNA damage tolerance mechanisms. Moreover,
the knowledge gained from such studies has fundamentally changed our
understanding of the ability of an organism to sustain life in the face of consistent

exposure to DNA damaging agents.

1. The SOS Response to DNA Damage

42

In 1974, Miroslav Radman hypothesized that E. coli possess a DNA repair
system, which he called “SOS repair” (now more commonly known as the SOS
response), that is repressed under normal physiological conditions, but that can
be induced in response DNA damage [113]. The SOS response system was the
first DNA damage-induced regulatory network to be characterized. It is now
known that the E. coli SOS response system controls the expression of more
than 40 genes than encode proteins required for a variety of DNA damage

response mechanisms (for review see [1]).

Just as Radman suggested in 1974, the SOS response system in E. coli is
repressed under normal physiological conditions by the LexA protein (see [1] for
a more thorough review). In the absence of DNA damage or replication stress,
the LexA protein binds to the operator sequences (SOS boxes) of genes
controlled by the SOS response system and represses their transcription.
However, if E. coli are exposed to DNA damaging agents or replication is
interrupted, the SOS response system can be mobilized through activation of the
RecA protein. The RecA protein becomes activated when it polymerizes along
single-stranded regions of DNA that are generated by discontinuous DNA
synthesis. These RecA nucleoprotein filaments stimulate LexA autodigestion,
resulting in inactivation of the LexA repressor function and ultimately, LexA
degradation. As the level of the LexA repressor decreases, the SOS genes begin
to be derepressed. Genes whose operators are bound relatively weakly by LexA,

are the first to be fully induced. If DNA damage and replication stress persist,

43

additional RecA nucleoprotein filaments are generated and RecA dependent
cleavage of LexA continues. If the amount of LexA declines to a very low level,
then even the genes whose operators are bound tightly by LexA are expressed.
As the cell is able to complete DNA synthesis and eliminate single-stranded DNA
gaps, the amount of activated RecA decreases and as a consequence, the
amount of LexA repressor is elevated. As the amount of LexA accumulates, the

SOS genes are once again repressed.

One important advantage of the E. coli SOS response is that the SOS genes are
derepressed sequentially, based on how tightly the LexA repressor binds to their
operator sequence (reviewed in [1]). This mechanism of chronological
derepression of the SOS genes allows damage response mechanisms to be
activated preferentially. For example, SOS-induced responses such as
nucleotide excision repair, are induced first. Potentially mutagenic responses,
such as translesion synthesis, are induced later, only as a final effort to rescue
the cell from persistent DNA damage. Several other SOS-induced responses that
destabilize the E. coli genome, such as adaptive mutagenesis, can be induced to
offer long-term survival benefits by increasing the probability of acquiring a

preferential genetic change.

2. Translesion Synthesis in Prokaryotes

44

It has long been known that induction of the SOS response system can be
accompanied by a 100-fold increase in mutations [114]. However, for many years
the source of this so-called SOS mutagenesis remained unknown. It is now
recognized that the mechanistic basis for SOS mutagenesis is translesion
synthesis. As described above, translesion synthesis allows the cell to avoid
replication arrest by employing specialized polymerases to synthesized DNA past
lesions that would typically block the major replicative polymerases. However,
many of the TLS polymerases replicate DNA with a relaxed fidelity. As a result,
the price of increased cell survival sometimes comes at the cost of increased
mutagenesis. In E. coli, TLS is primarily conducted by two enzymes, which were
first referred to as DinB and UmuC, but are now known as DNA polymerases IV

and V.

a. DNA Polymerase V (UmuD’ZC)

In 1977, two independent laboratories screened for mutations in E. coli that could

abolish UV-induced SOS mutagenesis. These screens identified mutations that

could be mapped to three different loci in E. coli: the ch+ locus and the IexA+

locus (which were known regulators of the SOS response), as well as a

previously uncharacterized locus called umuC+ (UV-induced mutability) [115,
116]. It was later demonstrated that the umuC+ locus is actually composed of two

genes, umuC+ and umuD+, either of which when mutated eliminate UV and

chemical-induced SOS mutagenesis in E. coli [117, 118].

45

Studies of the UmuD protein have indicated that in solution, it exists as a
homodimer [119]. The UmuD homodimer is not active in mutagenesis, however
interaction with RecA nucleoprotein filaments stimulates autodigestion of the N—
terminal 24 amino acids of UmuD [120]. The resulting proteolytic fragment of
UmuD, designated UmuD’, is active in SOS-induced mutagenesis [121, 122].

This activated UmuD’ homodimer binds a monomer of UmuC to form the
UmuD’2C complex [119]. The UmuD’2C complex supports a polymerase activity

and is, therefore, also referred to as DNA polymerase V (Pol V) [123, 124].

It is now recognized that E. coli DNA polymerase V belongs to the special subset
of polymerases that conduct replication past fork-blocking DNA lesions, i.e.
translesion synthesis. Translesion synthesis polymerases are typically much less
faithful than the major replicative polymerases and, in addition, they lack 3’-—>5’
proofreading activity. Therefore translesion synthesis is often associated with a
high mutagenic potential. The idea that UV-induced SOS-dependent
mutagenesis in E. coli is a direct result of error-prone TLS by Pol V was tested
when Tang et al. [125] reconstituted translesion synthesis reactions in vitro using
purified Pol V, and template primers that contained either UV-induced thymine-
thymine cyclobutane pyrimidine dimers (T-T dimers) or thymine-thymine 6-4
pyrimidine-pyrimidone photoproducts (6-4 photoproducts). These investigators
found that Pol V was able to bypass both T-T dimers and 6-4 photoproducts

efficiently. Moreover, they found that the mutation spectrum generated by Pol V

46

in vitro was extremely similar to that which results from SOS-dependent UV-
induced mutagenesis in vivo, confirming the role of error-prone TLS by Pol V in

SOS-induced mutagenesis [125].

As described above, initial studies of the umuDC gene products indicated that
only the proteolytic fragment of UmuD, i.e. UmuD’, participates in mutagenic
TLS, suggesting that UmuD is nothing more than an inactive precursor of UmuD’
(reviewed in [1]). However, more recently Graham Walker’s laboratory [126]
demonstrated that expression of a non-cleavable form of the UmuD protein,
along with UmuC, increases cell survival following exposure to UV radiation,
despite the fact that UmuD is unable to participate in TLS. What is more, they
showed that expression of these proteins causes a UV-induced delay in DNA
replication. Taken together, these results suggest a role for UmuD along with

UmuC in regulation of DNA damage-induced checkpoints.

Based on these more recent results, the Walker laboratory has proposed an
intriguing DNA damage tolerance model where the products of the umuDC
operon promote increased cell survival by two distinct and sequential
mechanisms. Upon exposure to DNA damaging agents, RecA mediated
cleavage of the LexA repressor results in derepression of the umuDC operon,
resulting in an increase in the level of the UmuD and UmuC proteins. Together,
the UmuD and UmuC proteins act to delay DNA synthesis and cell growth,

allowing additional time for DNA repair to occur and ultimately increasing cell

47

survival. If DNA lesions persist, RecA promotes autocleavage of UmuD to form
UmuD’. Formation of UmuD’ switches the function of the umuDC gene products

to their second role, which is translesion synthesis. Translesion synthesis past

the remaining DNA lesions by the UmuD’zC complex (Pol V) further promotes

cell survival by allowing DNA replication to resume. This sequential response to
DNA damage allows for an added measure of control by preferentially promoting

accurate DNA repair before potentially mutagenic TLS [126].

b. DNA Polymerase IV (DinB)

Induction of the SOS response system contributes to increased mutagenesis in
two ways termed targeted and untargeted mutagenesis. Targeted mutagenesis
occurs as a result of an SOS-dependent increase in the frequency of error-prone
replication of DNA lesions, i.e. TLS, and requires the function of Pol V and RecA
as discussed above. On the other hand, SOS-dependent untargeted
mutagenesis occurs in the absence of DNA damage and consists of two distinct
pathways. One pathway of untargeted mutagenesis can be observed when an
undamaged lambda phage is transfected into a host previously exposed to a
DNA damaging agent. In such lambda phage untargeted mutagenesis, the
number of mutations that occur in the undamaged phage increase as a
consequence of being replicated in a host previously exposed to a DNA
damaging agent. In other words, the lambda phage is replicated in an error-prone
manner despite the absence of DNA lesions. In E. coli, untargeted mutagenesis

of lambda phage is dependent upon the function of the dinB gene.

48

The dinB gene was originally identified in a screen for genes induced by DNA
damage [127]. This same locus was subsequently identified by the Ohmori
laboratory (and named dinP) as a putative SOS-inducible gene during an E. coli
genome sequencing project [128]. Consistent with the requirement for DinB in
untargeted mutagenesis, overexpression of the DinB protein in E. coli causes a
dramatic increase in the frequency of spontaneous - 1 frameshift mutations

(referred to as the DinB mutator phenotype).

Subsequent studies demonstrated that purified DinB protein possesses a
polymerase activity [129]. Moreover, eliminating the polymerase activity of DinB
by mutation, abolishes the DinB mutator phenotype in E. coli, indicating that the
polymerase activity of DinB is required for its mutagenic function [129]. Based on
these studies, as well as its homology to the UmuC family of proteins, the DinB
protein was determined to be an E. coli TLS polymerase, which is named DNA

polymerase IV [129].

Most TLS polymerases are characterized by their relatively low fidelity when
replicating undamaged DNA, both as a result of misincorporation of nucleotides
as well as their lack of 3’—*5’ exonuclease (proofreading) activity. Like other TLS
polymerases, Pol IV is devoid of 3’—->5’ exonuclease activity [129]. However,
steady state kinetic analysis of the frequency of misinsertions by Pol IV on

undamaged DNA indicates that it is only four to fivefold less accurate when

49

replicating undamaged DNA than is E. coli Pol III, the major replicative DNA
polymerase [125, 130]. These data indicate that, although Pol IV has been
implicated in untargeted and spontaneous mutagenesis, it is a relatively faithful
enzyme, at least for the insertion step, of DNA synthesis [131]. In contrast, Pol IV
is much less stringent when extending primer termini. Pol IV is capable of
elongating misaligned primer/template structures [129]. Because an increase in
the frequency of -1 frameshifts is observed in both dinB-dependent untargeted
mutagenesis, as well as in the DinB mutator phenotype, it is thought that Pol IV
likely contributes to mutagenesis by extending misaligned primer termini [131-

133].

Although the primary role of Pol lV seems to be for mutagenesis in the absence
of DNA damage, Pol IV is also capable of performing TLS across a number of
different types of DNA lesions in vitro, albeit with widely varying efficiencies.
Specifically, Pol IV has been shown to be able to bypass 8-oxoguanine [131], 06-
methylguanine [131], abasic sites [125, 134], N-2-acetylaminofiuorene and N—2-
aminofluorene adducted guanines [135], cyclobutane pyrimidine dimers and 6-4
pyrimidine-pyrimidone photoproducts [125], cisplatinated guanine adducts [136],
and benzo[a]pyrene diol epoxide adducted guanines (BPDE-G) [137]. However,
whether or not the specificity of Pol IV in vitro reflects its in vivo action remains
unclear because, although in vivo studies support a role for Pol IV in bypassing
oxidative damages [138], alkylating DNA lesions [139] and BPDE-G adducts

[137, 140], Pol IV does not seem to be required in vivo for TLS past abasic sites

50

[134]. Moreover, dinB mutant strains exhibit no clear phenotype with regard to
mutagenesis following treatment with DNA damaging agents, therefore the role

of Pol IV in targeted damage-induced mutagenesis is difficult to reconcile [131].

In addition to its roles in untargeted mutagenesis and TLS, Pol IV has also been
implicated in other cellular functions. First, Pol IV has been shown to be
important for adaptive mutagenesis. Adaptive mutagenesis refers to the ability of
non-dividing cells to acquire advantageous mutations during periods of nutrient
stress, which allow them to resume growth despite continued selective pressure.
Pol IV is required for ~80% of adaptive mutations which, are typically -1
frameshift mutations, a hallmark of Pol IV-dependent mutagenesis [141]. A
further requirement for Pol IV, along with Pol V, is in long-term survival and
fitness. Interestingly, both E. coli TLS polymerases Pol IV and Pol V are
upregulated during stationary phase even in the absence of DNA damage [142,
143]. More importantly, it has been demonstrated that cells lacking either Pol IV
or Pol V are much less efficient at competing for energy resources than wild-type
bacteria in stationary phase [142], indicating that these polymerases play an

important role in promoting genetic diversity [144].

3. Damage Avoidance in Prokaryotes

In the early 1960’s, Paul Howard-Flanders and his colleagues observed that uvrA

recA double mutants of E. coli, which are defective in both nucleotide excision

51

repair and in recombination, are much more sensitive to UV radiation than
either of the single mutants [145], reviewed in [146]. These observations were
the first indication that the survival of cells exposed to DNA damaging agents is
affected by cellular responses other than lesion repair. Shortly after this initial
discovery, the Howard-Flanders laboratory also demonstrated that in E. coli
defective in excision repair (uvr cells), DNA replication results in the production
of newly synthesized daughter strands that are initially significantly smaller than
those generated in unirradiated cells. They further observed that, after
increasing lengths of time post-irradiation, these daughter strands are
extended, ultimately reaching the length of those synthesized in unirradiated
cells [147]. These data were interpreted to mean that, following UV radiation
exposure, DNA replication occurs normally until a fork-blocking lesion is
encountered, at which point, replication is temporarily delayed and then
resumes somewhere beyond the lesion. This aborted DNA synthesis results in
the production of a DNA gap in the newly synthesized daughter strand. The fact
that DNA replication generates two identical sister molecules, led to the
hypothesis that the daughter strand gap generated by aborted DNA synthesis
could be filled by a recombination mechanism. This hypothesis was further
supported by experiments from the same laboratory, which used density labels
to provide direct evidence of exchanges between DNA strands during post-

replicative gap filling [148].

52

Although this model describes a process often referred to as postreplicational
repair, this mechanism does not result in the removal of DNA lesions, and
therefore, does not constitute a mechanism of repair, but rather a mechanism
of tolerance of the presence of DNA damage. This same process is also
sometimes referred to as daughter strand gap repair, which is an appropriate
designation with respect to the gaps themselves that are repaired by this

mechanism (reviewed in [1]).

The mechanistic details and many of the proteins involved in prokaryotic
damage avoidance remain uncertain. However, at least some of the proteins
involved have been identified and are summarized in a recent review [146]. The
function of the recA gene is absolutely required, and is likely involved in the
pairing of the single-stranded DNA, generated as a result of aborted DNA
synthesis, to its homologous DNA strand. About half of the DNA daughter
strand gaps are repaired by a process that is dependent on the function of the
recombination protein F (recF) gene, which most likely works in conjunction
with the RecO and RecR proteins to direct loading of the RecA protein onto the
gapped DNA. The remaining 50% of daughter strand gaps are filled by a
process that is both recF and recBC-independent. The recF-independent
pathway requires the 5'——>3’ exonuclease activity of DNA polymerase I. Because
DNA polymerase I is known to be involved in joining Okazaki fragments that are
formed during lagging strand synthesis, it has been suggested that daughter

strand gaps generated in the lagging strand are selectively repaired in the

53

recF-independent pathway, whereas daughter strand gaps formed in the

leading strand may be repaired by the recF-independent pathway.

Damage tolerance mechanisms involving recombination provide an alternative
error-free strategy to the potentially mutagenic TLS discussed above. However,
the fact that the RecA protein is required for both DNA daughter strand gap
repair and for TLS by Pol V in bacteria, suggests that these two processes may
not be completely independent of one another. Although advances have been
made in understanding the mechanisms of damage tolerance in prokaryotes,

much more remains to be learned.

B. Damage Tolerance Mechanisms in Eukaryotes

The damage tolerance responses employed by eukaryotes are very similar in
strategy to those of prokaryotes. That is to say, in each case either translesion
synthesis (TLS) or damage avoidance (DA) mechanisms are employed to allow
replication past fork-blocking DNA lesions that would othenrvise result in cell
death. However, whereas in prokaryotes regulation of damage tolerance
processes is primarily controlled by RecA, in eukaryotes the damage tolerance
processes are controlled by a more sophisticated strategy of protein conjugation

(reviewed in [149]).

54

Much of what is now known about the eukaryotic DNA damage tolerance
mechanisms was originally determined using the budding yeast, Saccharomyces
cerevisiae, as a model system. However, almost all of the proteins involved in
damage tolerance in yeast have functional homologs in higher eukaryotes,
suggesting that the mechanisms of DNA damage tolerance are conserved from
lower to higher eukaryotes. In yeast, genetic analysis has led to the generally
accepted view that the two pathways of eukaryotic damage tolerance, i.e. TLS
and DA, are regulated by the function of the RAD6 and RAD18 genes.
Consequently, mutation of either gene results in the most severe DNA damage-
induced sensitivities of all the genes whose functions are known to be required

for damage tolerance (reviewed in [149]).

The Rad6 protein is a ubiquitin-conjugating enzyme that forms a heterodimer
with Rad18, which is a single-stranded binding protein with ATPase activity [150-
152]. Recent studies have indicated that the target of ubiquitination by the Rad6-
Rad18 complex is the processivity factor, proliferating cell nuclear antigen
(PCNA) [153]. The current model for Rad6-Rad18-dependent regulation of DNA
damage tolerance through covalent modification of PCNA is as follows: Rad18
recruits Rad6 to sites of single-stranded DNA gaps (sites of aborted DNA
synthesis). The Rad6-Rad18 complex mono-ubiquitinates PCNA on lysine164,
which promotes TLS [154]. Further ubiquitination of PCNA (conducted by the
MmsZ-ch13-Rad5 complex discussed below) promotes error-free DA (reviewed

in [9]). Although this model suggests an elegant mechanism by which the

55

eukaryotic damage tolerance processes are regulated, more recent studies have
challenged these currently accepted views. In particular, a 2006 study
demonstrated, in vitro, that mono-ubiquitination of PCNA had no affect on the
affinity of the TLS polymerases for PCNA, nor did it alter activity of the TLS
polymerases examined [155]. Based on results such as these, it seems that

there is still much more to learn about regulation of eukaryotic damage tolerance.

1. Translesion Synthesis in Eukaryotes

Genetic studies in E. coli clearly demonstrate that damage-induced mutagenesis
is an active process that relies upon the induction of key components of the SOS
regulatory system, including RecA and UmuDC (see above). Initially, it was
thought that RecA and UmuDC were merely accessory proteins that were able to
enhance the processivity of the high fidelity E. coli replicative polymerase III,
thereby eliminating inhibition of DNA synthesis opposite a DNA lesion. The
notion that cells contain specific specialized DNA polymerases, whose function is
to carryout TLS was first suggested in 1996, when a heterodimer of the S.
cerevisiae Rev3 and Rev7 proteins (both long implicated in mutagenesis) was
shown to possess a polymerase activity, and to be capable of bypassing a UV-
induced thymine-thymine cyclobutane pyrimidine dimer (T-T dimer) [156]. At
nearly the same time, the S. cerevisiae Rev1 protein was found to possess a
deoxycytidyl transferase activity that could promote DNA synthesis past an

abasic site [157]. Although the UmuDC gene is homologous to the Rev1 gene,

56

because neither was related to any of the known DNA polymerases (including
the newly discovered heterodimer of Rev3 and Rev7 called Pol C), the idea that

the E. coli UmuDC gene might encode a DNA polymerase was overlooked until

1999, when the UmuD’zc complex was purified to homogeneity and found to

demonstrate DNA polymerase activity [123, 124]. Shortly thereafter, a multitude

of proteins homologous to UmuD’2C and Rev1 were discovered and also found

to be polymerases capable of DNA synthesis past replication-blocking DNA
lesions. These translesion synthesis polymerases were initially referred to as the
UmuC/DinB/Rev1/Rad30 superfamily, but have now, with the exception of Pol C,
collectively come to be known as the Y-family of DNA polymerases [158]. To
date, over 300 proteins that share homology to the Y-family of polymerases have
been identified in eukaryotes and prokaryotes, including archaea [159]. In many
cases each organism encodes more than one TLS polymerase. For example, E.
colihave both Pol IV and Pol V (see above), whereas humans possess four Y-
family polymerases, Pol Tl. Pol L, Pol x and Rev1 along with the B-family TLS

polymerase, Pol g [158, 159].

Overall, the properties of polymerases involved in TLS differ quite dramatically
from those of the classical replicative polymerases. For example, the major
replicative polymerases are typically highly processive and highly accurate
enzymes with active sites that have evolved with strict requirements for correct
Watson-Crick base pairing. In addition, many of the classical replicative

polymerases contain an intrinsic 3’—>5’ exonuclease activity that allows them to

57

remove and resynthesize any mispairs they may generate. In contrast, the TLS
polymerases (particularly those from the Y-family) are generally poorly
processive enzymes that copy DNA with low fidelity, both as a result of their less
restrictive active sites and their lack of 3’-—>5’ exonuclease activity (for review see

[160]).

Perhaps the most striking difference between the classical and TLS polymerases
is the property by which these enzymes were first characterized, their ability to
synthesize DNA opposite lesions. The classical high-fidelity replicative
polymerases are predominantly blocked by the presence of DNA lesions, as a
consequence of their restrictive active sites. However, the open active sites of
TLS polymerases are able to accommodate altered bases and as a result, these
polymerases are highly specialized to synthesize DNA opposite distorting lesions
[161-164]. Because DNA is constantly subject to damage from both endogenous
as well as environmental sources, some of which escape the traditional repair
processes, the TLS polymerases can provide the particularly useful advantage of
overcoming lesion blocks by conducting replicational bypass across from such
DNA lesions. Nevertheless, because of the poor fidelity of the TLS polymerases,
the protection of cells from premature replication arrest sometimes comes at the

cost of generating mutations (reviewed in [160]).

a. DNA Polymerase I;

DNA polymerase C (Pol C) consists of two subunits, a catalytic subunit called

58

Rev3, as well as an accessory subunit called Rev7. The REV3 gene, designated
rev for “reversionless”, was initially identified in yeast during a screen for mutants
defective in UV—induoed reversion of the highly UV—revertible, ochre suppressible
arg4-17 allele [165]. Several years later, a similar type of UV-induced reversion

study was used to identify RE V7 in yeast [166].

The S. cerevisiae REV3 gene was cloned in 1989 and nucleotide sequence
analysis revealed that, despite being non-essential for yeast viability, the REV3
gene encodes a predicted protein with sequence similarity to the B—family of DNA
polymerases, which also includes the replicative polymerases or, 6, and s [167]. It
was somewhat surprising therefore that when assayed for polymerase activity,
purified Rev3 protein demonstrated only a very weak and unstable activity in vitro
[156]. The apparent lack of polymerase activity of Rev3 alone was however,
thought to be indicative of a requirement for additional subunits to provide
enhanced activity and stability to the Rev3 polymerase [156]. Yeast two-hybrid
assays demonstrated that Rev7 was capable of associating with Rev3,
suggesting that Rev7 was a subunit of a Rev3-containing DNA polymerase [156].
Indeed, upon addition of Rev7, the polymerase activity of Rev3 increased 20 to
30-fold [156]. Because the Rev3/Rev7 complex was the sixth eukaryotic
polymerase to be discovered, it was named DNA polymerase i; [156]. It should
be noted that native Pol 2; has never been directly purified from S. cerevisiae or
from any other species. Therefore, It remains unknown whether or not additional

subunits are required for maximum polymerase stability or activity. [168].

59

Compared to other B-family polymerases, such as Pol 6 or Pol a, yeast Pol 2; is a
poorly processive enzyme [156, 169]. Polymerase 2; also lacks 3’—>5’
exonuclease activity, meaning that, unlike Pol 6 or Pol s, Pol 7; is unable to
proofread any mispairs generated during DNA synthesis [156]. Overall, on

undamaged DNA, yeast Pol C incorporates nucleotides with moderate fidelity.
That is to say, Pol 1;, which generates errors at rates of 1.3 x 10-3, is much less

accurate when copying undamaged DNA than related eukaryotic B-family

polymerases 6, s and or (which generate errors at rates of 1.3 x 10's, 2 x 10's,

and 9.6 x 10'5 respectivelY). but is substantially more accurate than TLS
polymerases from the Y-family, such as Pol Tl or Pol x (which generate errors at

rates of 3.5 x 10'2 and 5.8 x 10'3 respectively) [170]. Although lack of 3’—>5’

exonuclease activity no doubt partially contributes to the reduced fidelity of Pol C,
even exonuclease deficient B-family polymerases, such as Pol or, incorporate
nucleotides with a higher fidelity (~9.6 x 105) than Pol i; Therefore the ability of
Pol C to discriminate between nucleotides during DNA synthesis, although

substantially better than that of Y-family polymerases, is intrinsically poorer than

that of related B—family polymerases [170].

Perhaps one of the most remarkable characteristics of Pol g is the unusual

propensity with which it extends from DNA mispairs [169, 171]. Compared to Pol

60

or, which is also a B-family polymerase lacking 3’—>5’ exonuclease activity, yeast
Pol Z; is anywhere from twofold to more than 1,200—fold more efficient at
extending from terminal mispairs, depending on the sequence context and type
of mismatch extended [169]. In fact, in some cases Pol C is nearly as efficient at
extending from terminal mispairs as it is in extending from correctly paired termini
[169]. Based on these data, as well as the fact that Pol Z; is quite adept at
extending from abnormal primer ends resulting from the presence of DNA lesions
(discussed below), it has been suggested that the primary function of Pol t; in
TLS is not to insert nucleotides opposite the lesions, but rather to extend from

atypical primer termini generated by other TLS polymerases [171].

In keeping with the relaxed fidelity and with the unique ability of Pol ‘g to extend
from DNA mispairs, this polymerase is responsible for the majority of
spontaneous mutations in yeast. The function of Pol ?; was first implicated in
spontaneous mutagenesis when a yeast antimutator strain was isolated and the
antimutator locus was determined to be a REV3 allele [172]. Based on data from
that study, as well as a study by Roche et al. [173], it has been estimated that Pol
z is responsible for between 50% to 75% of all spontaneous mutations that arise
In yeast [172, 174]. Furthermore, Pol C has been implicated in spontaneous
mutagenesis that arises from a variety of circumstances, including those that
result from transcription [175], double-strand break repair [176], and deficiency in

excision repair [173].

61

In addition to its involvement in spontaneous mutagenesis, Pol C is also unique
among B-family polymerases in that it is capable of replicating past fork-blocking
DNA lesions. The ability of Pol C to perform such translesion synthesis was first
demonstrated in 1996, when the Lawrence laboratory showed that Pol C can
replicate past a UV-induced thymine-thymine cyclobutane pyrimidine dimer (T -T
dimer) in primer extension assays in vitro [156]. However, the ability of Pol C to
replicate past a T-T dimer in vitro apparently does not translate to an in vivo
requirement for this polymerase in TLS past such lesions, because subsequent
work by the same laboratory demonstrated that the frequency of TLS past T-T
dimers was unaltered in yeast strains lacking Rev3, the catalytic subunit of Pol C
[177]. It is now clear that DNA Pol Tl. not Pol C, is chiefly responsible for bypass

of T-T dimers in vivo (see below).

Despite the fact that Pol C is not required for bypass of T-T dimers, yeast rev
mutants nevertheless show reduced levels of UV-induced mutagenesis,
suggesting a role for Pol C in error—prone TLS past UV-induced DNA lesions.
Therefore, the role of yeast Pol C in bypassing the less abundant, but more
mutagenic UV-induced thymine-thymine pyrimidine-pyrimidone (6—4)
photoproduct (6-4 photoproduct) has also been examined extensively. Results
from studies of the ability of yeast Pol C to perform TLS past 6-4 photoproducts in
vitro have been somewhat conflicting. A report by Johnson et al. indicated that
Pol C is unable to incorporate nucleotides opposite the 3’T of a 6-4 photoproduct,

but efficiently and accurately extends from nucleotides inserted opposite the 3’T

62

by Pol Tl (which preferentially misinserts a G residue) [178]. These results
suggest that Pol C contributes to UV-induced mutagenesis only by extending
from a mispaired primer terminus generated by another TLS polymerase [178].
However, in a study by Guo et al., Pol C was determined to be efficient at
nucleotide insertion opposite both residues of a 6-4 photoproduct, inserting an A,
T and more rarely a G opposite the 3’T and primarily inserting the correct A
opposite the 5’T [179]. Results from this study, therefore, suggest that Pol C is
directly responsible for generating mutations opposite 6—4 photoproducts by
inserting the incorrect nucleotides opposite the 3’T of this UV-induced DNA

lesion.

The contribution of Pol C, as well as Pol n, to TLS past 6-4 photoproducts in vivo
is perhaps best demonstrated in experiments where yeast strains lacking one or
the other of such polymerases are transformed with plasmid constructs
containing a single defined lesion and then assayed for the frequency with which
they complete lesion bypass. Unfortunately, like with in vitro studies, the results
of these in vivo studies are equally conflicting. For example, a study by Bresson
and Fuchs demonstrated that the mutagenic bypass of a 6-4 photoproduct
resulted exclusively from incorporation of a G opposite the 3’T and furthermore,
that mutagenic bypass of this lesion was essentially suppressed in yeast strains
lacking Pol Tl (the role of Pol C was not tested) [180]. These results support the in
vitro studies performed by Johnson et al., which indicate that Pol TI is responsible

for mistakes opposite 6—4 photoproducts and that Pol C contributes to UV-induced

63

mutagenesis only by extending from such mistakes [178]. In contrast, a similar
study conducted in the Lawrence laboratory demonstrated that only 15% of the
total mutagenic TLS past a 6-4 photoproduct resulted from misincorporation
opposite the 3’T of this lesion and that C, T and G were all misincorporated
(although G was misincorporated most frequently) [181]. In this study, Pol Tl
deficiency only reduced the mutagenic bypass of the 6-4 photoproduct by 7%,
indicating a much more limited role for Pol Tl in the mutagenic bypass of this DNA
lesion [181]. More recently, Abdulovic and Jinks-Robertson conducted an elegant
study in which yeast mutant strains deficient in Pol C, Pol n, or both polymerases
were used to examine the relative contribution of each of these enzymes to UV-
induced cell killing and mutagenesis. What is more, these studies were
conducted in the presence and in the absence of a photolyase that specifically
removes cyclobutane pyrimidine dimers (CPDs), allowing them to distinguish
between the contributions of CPDs and 6—4 photoproducts to overall survival and
mutagenesis [182]. Results from these experiments indicate a role for Pol C, but
not Pol Tl in TLS past UV-induced 6-4 photoproducts. Such results are consistent
with the studies conducted by the Lawrence laboratory and in direct contradiction
to those conducted by Bresson and Fuchs, which suggested a major role for Pol

n in the bypass of this lesion [182].

Although the exact function of Pol C in TLS past 6-4 photoproducts, i.e. whether it

is required for the insertion step, the extension step, or both steps, is still a matter

for debate, a clear role for Pol C as a TLS extender has been indicated in

64

replication past abasic sites. Independent studies, conducted by Nelson et al.
[157], Yuan et al. [183], and Haracska et al. [184], each demonstrated that, in
vitro, Pol C is unable to insert nucleotides opposite an abasic site. However, all
three studies demonstrated that Pol C is able to extend from a nucleotide inserted
opposite this lesion by another polymerase. Although the polymerase performing
the insertion step differed in each study (Rev1, Pol *1 or Pol 6 respectively), all
three groups agreed that the requirement for Pol C in TLS past abasic sites is
strictly a reflection of the ability of this enzyme to extend from atypical primer

termini that present a block to other polymerases.

The ability of yeast Pol C to bypass a variety of DNA lesions that result from
chemical mutagens has also been examined. For example, in vivo experiments
using rev mutant strains transformed with plasmids containing an N-2-
acetylaminofluorene adducted guanine (AAF-G) indicate an absolute requirement
for Pol C in TLS past this lesion. In vitro studies, however, demonstrated that
yeast Pol C is unable to insert nucleotides opposite an AAF-G, but is able to
efficiently extend from an AAF-G overlapped by a primer [179]. Taken together,
these data indicate that, as with abasic sites, Pol C is required only for the
extension step of TLS past AAF-induced DNA lesions and that the insertion step

is likely performed by another TLS polymerase.

The role of yeast Pol C in bypassing benzo[a]pyrene adducted guanine (BPDE-G)

has also been examined both in vitro and in vivo. Reports of the ability of Pol C to

65

bypass BPDE-G in vitro vary from complete blockage [185] to unassisted
accurate bypass [186]. However, the latter results may not be relevant with
respect to in vivo conditions, because the reactions were conducted with high
levels of Pol C and the time allowed for lesion bypass was greater than 90 min
[169]. Experiments using site specifically modified plasmids and a yeast rev3
mutant strain demonstrated that the frequency of TLS in such cells is reduced to
16% of that in wild-type cells, indicating a requirement for Pol C in TLS past
BPDE-adducted guanines in vivo. Nevertheless, the exact function of Pol C in

such TLS remains unclear [187].

In addition to being deficient in mutagenesis induced by UV, the yeast rev3 and
rev7 mutant strains are also deficient in mutagenesis induced by a variety of
other DNA damaging agents, including ionizing radiation [188, 189], 4-
nitroquinoline-1-oxide [190], methyl methane sulfonate [191], and ethyl methane
sulfonate [192]. The fact that the rev mutant strains exhibit similar phenotypes
when they are exposed to these damaging agents as they do when exposed to
UV, suggests that Pol C is required for TLS past DNA lesions induced by these

damaging agents as well.

Homologs of the yeast REV3 gene have been identified in many eukaryotes,
including Drosophila melanogaster [193], Arabidopsis thaliana [194], mouse
[195], chicken [196], and, of particular interest, humans [197, 198]. The human

homolog of the S. cerevisiae REV3 gene encodes a predicted protein of 3,130

66

amino acid residues, which is twice the size of the yeast Rev3 protein (1,504
amino acids). The predicted human Rev3 protein shares three regions of
sequence homology with the yeast Rev3 protein, a 340 residue amino-terminal
domain that is 29% identical to the yeast protein, a small, central 55 residue
Rev7 binding domain that is 29% identical, and a carboxy-terminal 850 residue
DNA polymerase domain that shares 39% identity with yeast Rev3 [197]. It has
been suggested that the large, nonhomologus regions of the hRev3 protein are
indicative of additional or more diverse functions for the human protein compared
to the yeast counterpart [197]. The fact that in yeast, Rev3 is not essential for
viability, but that disruption of Rev3 in mice results in embryonic lethality [199-
201], supports the idea that Rev3 has additional, crucial roles for in higher

eukaryotes.

Similar to the yeast Rev3 protein, the human Rev3 protein contains a polymerase
domain placing it in the B-family [197]. However, to date, the hRev3 protein has
never been successfully expressed or purified, presumably as a result of its large
size (353 kDa) and low cellular levels. Therefore, formal demonstration of Rev3
polymerase activity has not yet been achieved, and in vitro primer extension
assays using reconstituted hPol C are lacking. Nevertheless, both human and
mouse cells expressing high levels of hREV3 antisense RNA demonstrate a
lower frequency of UV-induced mutations than their appropriate control strains
[202, 203]. In addition, by using photolyases, the Yasui laboratory demonstrated

that reduction of hREV3 mRNA renders nucleotide excision repair deficient

67

human cells sensitive to UV-induced 6-4 photoproducts, but not to CPDs [204].
Together, these data indicate that, as in yeast, Rev3 is required for the
mutagenic TLS of UV-induced 6-4 photoproducts in human cells, and suggest
that the functions of Pol C are conserved from yeast to higher eukaryotes [197,

202,205]

The human homolog of the yeast Rev7 protein was initially identified using a
yeast two-hybrid assay with a fragment of hRev3 as bait [206]. Human Rev7 is a
211 amino acid protein with a predicted molecular weight of 24 kDa. At the amino
acid level, hRev7 shares 23% identity and 53% similarity to the yeast Rev7
protein. The Rev7 protein contains a moderately conserved domain (HORMA)
that has also been found in hMAD2, a mitotic spindle assembly checkpoint
protein and in scHop1, a protein involved in meiotic-synaptonemal-complex
assembly [206, 207]. Although the exact function of this domain is unknown, it
has been suggested that the HORMA domain is involved in recognizing
chromatin states resulting from the presence of DNA adducts, double-strand
breaks, or failure of DNA to attach to the spindle, and that it acts as an adaptor

that recruits proteins involved in DNA repair [207].

The hRev7 protein also shares 23% identity to and 54% similarity with the mitotic
spindle assembly checkpoint protein, hMad2 [206]. Rev7 has been shown, in
Xenopus extracts, to associate with activators of the anaphase-promoting

complex, CDH1 and CDC20, and to inhibit cell progression from metaphase to

68

anaphase [208, 209]. What is more, in addition to interacting with hRev3, a
strong interaction has also been demonstrated between hRev7 and hMADZ,
suggesting that hRev7 plays a role in the mitotic spindle assembly checkpoint
[206]. However, overexpression of hRev7 in human osteosarcoma cells did not
lead to cell cycle arrest [206] and decreasing the level of hRev7 protein in
nasopharyngeal carcinoma cells had no obvious effects on cell proliferation, cell
cycle distribution, or mitotic checkpoint control [210]. Therefore, if a role for

hRev7 in mitotic checkpoint control exists, it remains uncharacterized.

Although the most well characterized function of Pol C is in TLS, Pol C has been
implicated in at least two additional cellular processes. First, a requirement for
Pol C in somatic hypermutation [211] has been demonstrated both in human and
mouse cells [212, 213]. Somatic hypermutation refers to the programmed
process whereby a high level of mutations are introduced within the gene
segment that encodes the variable region of an antibody, and is one mechanism
through which antibody diversity is generated (for review see [214]). In the first
phase of SHM, activation-induced deaminase (AID) catalyzes targeted
deamination of deoxycytidine residues in DNA, generating U:G mispairs.
Subsequent excision of uracil, by a uracil DNA glycosylase, generates abasic
sites, which ultimately lead to nucleotide substitution mutations. It is thought that
the role of Pol C in SHM is in processing these abasic sites. As described above,
although Pol C is unable to insert nucleotides opposite abasic sites, it is required

from the extension of nucleotides inserted opposite these DNA lesions by other

69

enzymes [157, 183, 184]. In particular, it is believed that Pol C contributes to the
base substitution mutations of C:G base, pairs by extending from nucleotides

inserted opposite abasic sites by Rev1 [214].

Finally, increasing evidence has indicated that Pol C is involved in homologous
recombination (HR)-mediated repair of double-strand breaks. Polymerase C was
first associated with double-strand break repair in yeast, when it was found to be
responsible for elevated rates of mutations near sites of recombinational DNA
double-strand break repair [176, 215]. More recently, chromatin
immunoprecipitation experiments demonstrated that Pol C, together with Rev1, is
enriched at areas surrounding a double-strand break generated by homothallic
endonuclease [216]. Recruitment of Rev1, and presumably the Pol C/Rev1
complex, was determined to be dependent on the kinase activity of the yeast
ATR homolog, MecI, suggesting a model whereby the Pol C/Rev1 complex is
recruited to the vicinity of a double-strand break by proteins phosphorylated in a

Mec1-dependent fashion [216].

Data from vertebrate cells are also consistent with a role for Pol C in HR-
mediated double—strand break repair. Chicken B lymphocyte DT40 cells with
disrupted REV3 alleles (REV3-/-) exhibit a threefold higher frequency of
spontaneous chromosomal aberrations compared to wild-type cells and

demonstrate reduced frequencies of gene targeting [196]. In addition, REV3-/-

cells were not only more sensitive to ionizing radiation than wild-type cells in G1

70

and early S-phase (likely due to impaired TLS), but were also more sensitive to

ionizing radiation than wild-type cells in G2 phase, and were slightly more

sensitive than a RAD54-/- clone defective in HR [196]. In addition, studies of
mouse cell lines lacking Rev3 suggest that, similar to chicken, Rev3 is employed
in double-strand break repair in mice. Disruption of Rev3 in mice causes
embryonic lethality [199-201]. Moreover cells from REV3 (-/-) embryos
demonstrated a significant increase in double-stranded DNA breaks as well as

chromatid and chromosome aberrations [217].

b. Rev1

Rev1, like Rev3 and Rev7, was discovered during a search for yeast mutants
defective in UV-induced mutagenesis [165]. The REV1 gene was cloned in 1989
and sequence analysis showed that an amino-terminal portion of Rev1 was 25%
identical and 42% similar to UmuC, the catalytic subunit of E. coli Pol V [218]. At
the time the function of the UmuC protein was still largely uncharacterized.
However, these two proteins are now both known to be founding members of the

Y-family of polymerases [158].

The yeast Rev1 protein has at least two distinct cellular functions. The first
function is as a template-dependent deoxycytidyl transferase [157]. During the
original biochemical study, the yeast Rev1 protein was found to preferentially
insert a dCMP (but not dAMP, dTMP or dGMP) in a template dependent manner,

i.e. opposite a template G. Even more surprisingly, Rev1 was found to be more

71

efficient at inserting a dCMP opposite a template abasic site than opposite a
template G [219]. This subsequent finding was particularly relevant in light of the
fact that in S. cerevisiae a C is incorporated opposite 60%-85% of abasic sites

that are bypassed in vivo [220].

Structural analysis of the polymerase domain of S. cerevisiae Rev1, in complex
with a template primer and an incoming dCTP has provided interesting insight
into the mechanism by which Rev1 incorporates strictly C residues but does so in
a template-dependent manner [221]. In the structure of this ternary complex, the
templating G is flipped out of the DNA helix by a Ieucine residue that protrudes
into the DNA. The templating G forms two hydrogen bonds with the main-chain
amides of nearby methionine and glycine residues. This pattern of hydrogen
bonding is such that any templating base other than a G would pose unfavorable
steric constraints. The evicted template G is replaced by an adjoining arginine
residue that can form hydrogen bonds with an incoming dCTP. Again, the pattern
of hydrogen bonding is such that binding to any other dNTP would be
unfavorable. Thus the mechanism for DNA polymerization by Rev1 is one in
which the specificity for both the templating base and the incoming nucleotide are

governed by the protein and not by the DNA [221].

Evidence of the existence of a second function of Rev1 in TLS, that is distinct

from the deoxycytidyl transferase activity, was first demonstrated in a study by

Nelson et al. [177], where they showed in vivo that the function of yeast Rev1 is

72

required for bypass of a UV-induced 6-4 photoproduct. A requirement for Rev1 in
TLS past a 6-4 photoproduct is not particularly surprising in light of the fact that
the REV1 gene was initially identified in yeast as a loci involved in UV-induced
mutagenesis. However, because Rev1 possesses only a deoxycytidyl
transferase activity, and because incorporation of a C occurs only very rarely
across from 6-4 photoproducts, these data also suggest the existence of a

second, non-catalytic function for Rev1 in TLS.

Further evidence for the existence of a second function of Rev1 in TLS, was
demonstrated when Otsuka et al. assayed Rev1 for the ability to perform TLS
past an abasic site in vivo [222]. In these experiments yeast rev1 mutant strains
carrying a cyc1-31 mutation were transformed with lesion-containing
oligonucleotides that have the CYC1+ sequence spanning the mutation site.
Transformants are only obtained if the oligonucleotide is incorporated into DNA
and then used as a template for TLS. Therefore these assays allow for a
measurement of the frequency of TLS past a single defined lesion at a specified
genomic location. As expected based on in vitro studies [219], these experiments
showed that Rev1 is essential for bypass of an abasic site. However, when
similar studies were conducted in a rev1 mutant strain that produces a protein
lacking deoxycytidyl transferase activity, TLS past an abasic site still occurred
with about 50% of the frequency in the wild-type strain. These data suggest that

although the transferase activity of Rev1 contributes to bypass of abasic sites, a

73

function of Rev1 distinct from its transferase activity is also important for TLS

[223].

The human homolog of the yeast RE V1 gene has also been identified [219, 224].
The hREV1 gene encodes a 1,251 amino acid protein that is similar in size to the
yeast Rev1 protein (985 amino acids). The human Rev1 protein shares four
regions of high sequence homology with the yeast protein. Two amino-terminal
domains, including a BRCA1 C-terminal (BRCT) domain that is 35% identical,
and an internally conserved region of unknown function that is 29% identical to
the yeast Rev1 protein, a polymerase domain that shares 31% identity with that
of the yeast protein, and a carboxy-terminal ubiquitin binding motif that shares

21% identity with yeast Rev1 [224, 225].

In addition to possessing significant sequence homology, the cellular functions of
the yeast and human Rev1 homologs also appear to be highly conserved. Like
its homolog in S. cerevisiae, the human Rev1 protein, possesses a deoxycytidyl
transferase activity that is highly specific and template dependent [224].
Moreover, similar to its yeast counterpart, hRev1 is able to utilize this
deoxycytidyl transferase activity to bypass abasic sites in vitro [224]. However,
the physiological relevance of this reaction in humans is questionable because
unlike in yeast, where dCMP is predominantly incorporated opposite abasic sites,

in human cells dAMP is the most commonly incorporated nucleotide [226].

74

The enigmatic second, so-called non-catalytic function of Rev1 also seems to be
conserved from yeast to humans. In a study using chicken DT40 cells lacking
Rev1, the deoxycytidyl transferase domain (but not the C-terminal domain) of
human Rev1 was determined to be dispensable for DNA damage tolerance
[227]. In addition, human fibroblasts expressing high levels of hREV1 antisense
RNA exhibited greatly reduced levels of UV—induced mutagenesis compared to
control cells, indicating that Rev1 is required for most mutations induced by UV
radiation in human cells [219]. Like yeast Rev1, human Rev1 is unable to
perform TLS opposite either of the two most common UV-induced DNA lesions in
vitro and furthermore, incorporation of a C opposite these DNA lesions occurs
very rarely. These data suggest that, as in yeast, a second transferase-

independent role for Rev1 in TLS exists in humans.

Rev1 contains a highly conserved BRCT domain in its N—terminal region that
appears to be of some importance for the non-catalytic function of this enzyme.
Studies using cells obtained from mice with a targeted deletion in the Rev1
BRCT domain demonstrated that deletion of this domain results in a complete
loss of UV-induced mutations in the HPRT gene [228]. The reduction of UV-
induced mutagenesis was accompanied by loss of transversions at T-T dimer
sites, indicating the importance of the Rev1 BRCT domain in TLS past UV-
induced T-T dimers. Interestingly, however, a deletion and point mutation
analysis of the Rev1 protein indicated that the BRCT domain is dispensable for

transferase activity in vitro [229]. Together, these data suggest that the

75

requirement for the Rev1 BRCT domain in TLS is unrelated to the catalytic

function of this enzyme.

More recent evidence suggests that the BRCT domain is required for interaction
between Rev1 and PCNA. As described above, if DNA damage occurs that
blocks replication, mono-ubiquitination of PCNA can be used as a mechanism to
activate TLS [153, 154]. Studies by Guo et al. [230], demonstrated that Rev1 can
bind directly to PCNA and that this interaction is enhanced by mono-
ubiquitination of PCNA. Moreover, they found that deletion or inactivation of the
Rev1 BRCT domain by mutation abolishes the interaction between Rev1 and
PCNA. Interestingly, deletion or inactivation of the Rev1 BRCT domain also
eliminates targeting of Rev1 to replication foci in undamaged cells, but only
slightly reduces such targeting in cells that are UV-irradiated. Similar studies
conducted in the same laboratory showed that the two conserved ubiquitin-
binding motifs located at the C-terminus of Rev1 also mediate enhanced
association with mono-ubiquitinated PCNA [231]. In contrast to results obtained
by deleting the Rev1 BRCT domain, deletion of the ubiquitin-binding motifs had
no effect on constitutive nuclear targeting to Rev1, but completely abolished
nuclear targeting in the presence of UV radiation. These data suggest that
localization of Rev1 to sites of replication-blocking DNA lesions is most efficient

when both the BRCT domain and the ubiquitin-binding motifs are fully functional.

76

In addition to interacting with mono-ubiquitinated PCNA, both mouse and human
Rev1 have been shown to interact with Pol n, Pol L, Pol x, and the accessory
subunit of Pol C, Rev7, through the same carboxy-terminal 100 amino acid region
[232, 233]. However, this interaction, at least with respect to Pol x, had no effect
on the polymerase activity of either enzyme. These data suggest that competitive
binding to Rev1 plays an important regulatory role for TLS in mammals. Taken
together with the fact that Rev1 can be localized to sites of DNA damage through
interaction with mono-ubiquitinated PCNA, it is tempting to speculate that the
second, as of yet undefined function of Rev1 in TLS is as a structural scaffold

that coordinates other TLS polymerases at sites of DNA damage.

c. DNA Polymerase Tl

Shortly after identification of the Rev1 protein in eukaryotes, a search of the S.
cerevisiae genome identified a gene with significant homology to the eukaryotic
Rev1 protein. When this gene was disrupted in yeast, they were rendered more
sensitive to UV radiation and as a result, this new gene was designated RAD30
[234]. Subsequent studies of the Rad30 protein demonstrated that unlike Rev1,
which possesses only a deoxynucleotidyl transferase activity, Rad30 is a bona
fide polymerase, able to incorporate all four dNTPs onto a primer in a template-
specific manner [235]. Because Rad30 was the seventh eukaryotic polymerase

to be discovered it was renamed DNA polymerase Tl (Pol n) [235].

77

Studies of the yeast rad30 mutant strain demonstrated that deletion of RAD30 in
yeast results in a 10-fold increase in UV-induced mutagenesis. This was the first
indication that the product of the RAD30 gene, Pol 1], might participate in an
error-free mechanism of DNA repair [234]. This hypothesis was further refined
when the purified Pol Tl protein was shown to be both efficient and accurate at
bypassing thymine-thymine cyclobutane pyrimidine dimers (T -T dimers) in vitro,
indicating that Pol Tl participates in tolerance of UV-induced DNA lesions by

performing error-free TLS past T-T dimers [235].

Cells from patients with the variant form of the disease xeroderma pigmentosum
(XP-V) are proficient in nucleotide excision repair (unlike cells from patients with
the classical forms of xeroderma pigmentosum), but are defective in replication of
DNA containing UV-induced lesions [236]. Pol Tl possesses the ability to carry
out error-free TLS past UV—induced T-T dimers in vitro. This finding raised the
possibility that lack of the human homolog of Pol TI is the cause of the disease
xeroderma pigmentosum variant in humans [235]. This prediction was soon
confirmed by two independent groups who demonstrated that XP-V cell lines
harbor mutations in the POLH gene, which encodes human Pol *1 [237, 238]. The
discovery that mutations in the POLH gene, that encodes Pol Tl. are responsible
for the variant form of the disease xeroderma pigmentosum added further

significance to the discovery of Pol n.

78

Structural studies of the catalytic core of S. cerevisiae Pol Tl have proven to be
particularly useful for understanding the mechanism by which TLS polymerases
are able to incorporate nucleotides opposite bulky DNA lesions [163]. Although
there is very little similarity between the amino acid sequence of Pol n and those
of the classical polymerases, Pol Tl retains the characteristic right hand
architecture of a DNA polymerase, including the palm, fingers and thumb
domains. Similar to the classical polymerases, the palm domain of Pol n harbors
the catalytic triad, viz., Asp, Asp-Glu active site residues. In fact, the core of the
palm domain of Pol n is nearly superimposable upon the core of the palm domain
of classical DNA polymerases. The fingers domain, which is typically thought to
mediate nucleotide selectivity, is small and stubby, compared to other DNA
polymerases. In addition, Pol Tl lacks the equivalent of helices “O” and “O1” that
are thought to play a role in fidelity enhancement. The thumb domain, that
mediates DNA binding and polymerase processivity, is similarly small and stubby
compared to the thumb domain of other polymerases. Unlike the classical
polymerases, Pol 11 contains a fourth domain, called the polymerase-associated
domain, which sits alongside the fingers domain [163]. This additional domain
increases the DNA binding surface area of Pol n. Although the crystal structure of
Pol Tl was determined in the absence of DNA, the similarity between the palm
domain of Pol n and that of other DNA polymerases, allowed both a template-
primer and an incoming nucleotide to be modeled into the Pol n DNA binding
site. The active site of Pol n was found to be much more open then the active

sites of other DNA polymerases. Taken together, these data suggest that TLS

79

polymerases are able to accommodate bulky DNA lesions into their DNA binding
pockets because they possess more open and extended active sites than the

classical high-fidelity polymerases [163].

Polymerase n is most well know for its remarkable ability to catalyze error-free
TLS past UV—induced T-T dimers. Structural and kinetic studies have suggested,
surprisingly, that Pol n is able to incorporate both residues of a T-T dimer into its
active site simultaneously, making it particularly efficient at conducting TLS past
this distorting DNA lesion [163, 239]. The fact that Pol Tl possesses such an open
active site provides an explanation for the low fidelity of this polymerase when
replicating undamaged DNA (discussed further below), but seems inconsistent
with the error-free nature of Pol Tl when replicating past UV-induced T-T dimers.
It is currently thought that unlike the classical replicative polymerases that govern
fidelity primarily by the geometric complementarity of the incoming nucleotide,
Pol Tl relies on Watson-Crick-type hydrogen bonds to achieve accurate DNA
synthesis. The effect of hydrogen bonds on the replication fidelity of Pol n was
directly tested in vitro by using a thymine analog, diflourotoluene, that is nearly
identical to thymine in shape, but lacks the ability to form hydrogen bonds with
adenine [240]. Pol Tl was unable to insert any nucleotide opposite a template
containing diflourotoluene, whereas insertion by the E. coli Klenow enzyme was
unhindered. Thus, unlike the classical replicative polymerases, the fidelity of Pol

n (and likely Y-family polymerases in general) is governed largely based on

80

hydrogen bonds between the templating base and the incoming nucleotide rather

than by its size or shape [240].

Although the ability of Pol TI to bypass UV-induced T-T dimers accurately and
efficiently has been well established [239, 241, 242], the ability of Pol 7] to bypass
a variety of other types of DNA lesions has also been studied extensively in vitro.
Unlike with T-T dimers, Pol TI is unable to replicate past a UV-induced, 6-4
photoproduct. However, Pol TI is able to catalyze misinsertion of a G opposite the
3’T of a 6-4 photoproduct with limited efficiency [241]. Because 6-4
photoproducts are formed more commonly at 5’-TC-3’ and 5’-CC-3’ sites, it is
thought that the ability of Pol TI to insert a G opposite the 3’ nucleotide of a
photoproduct might contribute to the overall error-free bypass of 6—4
photoproducts [243]. Pol n is also significantly inhibited by both abasic sites [184,
244, 245] and N-2-acetylaminofluorene adducted guanines in vitro [244]. In
contrast, Pol TI is readily able to bypass an 8-oxoguanine lesion, although reports
of the insertion specificity by Pol *1 across from this lesion are conflicting. In one
study, Pol n was reported to insert both A and C residues with similar efficiencies
[245]. However, in another study Pol Tl was found to replicate past an 8-oxoG
lesion in a primarily error-free manner, misinserting an A only a small proportion
of the time [246]. It is likely that these discrepancies result from the influence of
sequence context on insertion specificities. Finally, with some DNA lesions, Pol Tl
promotes primarily error-prone TLS. For example, Pol Tl has been shown to

predominantly insert an A residue opposite benzo[a]pyrene adducted guanines

81

[245, 247]. Because G—aT is the most common type of base substitution induced
by BPDE in mammalian cells, it has been speculated that error-prone TLS by Pol

n is the cause of this base substitution mutation [245, 247].

When compared to the classical replicative polymerases, both yeast and human
Pol Tl have reduced fidelity when copying undamaged DNA and in addition, both
yeast and human Pol 11 lack 3’—>5’ proofreading exonuclease activity [239, 248,
249]. As a result, Pol TI is at least 100-times less accurate than the major
replicative polymerases during the nucleotide incorporation step DNA synthesis
[250]. Although it seems counterintuitive that the benefits of possessing Pol n,
i.e. lesion bypass, outweigh the mutagenic potential of this polymerase, human
Pol n is much more processive when synthesizing DNA opposite T-T dimers than
it is when copying undamaged DNA [242]. This suggests a mechanism whereby
Pol Tl switches to a less processive mode when it encounters an undamaged
template, presumably excluding this polymerase from active DNA synthesis

[242].

In addition to its function in TLS, Pol n has also been shown to play a role in
other cellular processes, including somatic hypermutation (SHM) [211] and
homologous recombination. Somatic hypermutation, an important process
through which antibody diversity is achieved by hypermutation of the antibody
variable region, can be divided into two phases, a first phase where base

substitution mutations are induced in C:G base pairs, and a second phase where

82

A:T base pairs are the primary target for mutation. It is during this second phase,
predominated by mutagenesis of A:T base pairs, that Pol Tl plays a critical role.
Clear evidence for the requirement of Pol 71 in SHM was demonstrated by studies
of humans and mice lacking this polymerase, which showed a striking reduction
of the number of mutations at A and T sites within antibody variable genes [251,
252]. The exact mechanism by which Pol Tl generates mutations at A:T sites

during SHM remains to be determined [214].

Polymerase Tl has also been reported to be involved in homologous
recombination (HR). In a study conducted in the West laboratory [253], they
demonstrated that, in vitro, Pol Tl (but not Pol 6) was able to extend DNA from D-
loop recombination intermediates where the invading strand serves as a primer.
Moreover, these investigators showed that this D-loop extension activity is
diminished in extracts from XP-V cells, which lack Pol Tl- Using a completely
different approach, Shunichi Takeda’s laboratory evaluated HR-dependent
double-strand break repair by introducing an artificial SCneo HR substrate DNA
construct carrying an l-Scel restriction site into chicken DT40 cells that lack Pol
n. Double-strand breaks induced by transient expression of l-Scel are repaired
by gene conversion from an upstream donor homologous sequence,
generating a functional neomycin resistance gene. Therefore, the number of
HR events can be determined by counting the number of neomycin-resistant
colonies. When cells lacking Pol TI were assayed for the efficiency with which I-

SceI-induced double-strand breaks were repaired by HR they were shown to be

83

10-fold less efficient than wild-type cells. Furthermore, the gene conversion
deficiencies exhibited by cells lacking Pol Tl could be rescued by transfection of
wild-type Pol n, but not by transfection of a catalytically dead mutant of Pol 1],
indicating that the activity of this polymerase is required for these gene

conversion events [254].

d. DNA Polymerase 1.

Only a few months after the discovery of human Pol n, a second homolog of the
S. cerevisiae RAD30 gene, initially designated RAD3OB, was identified in
humans [255]. Like Pol Tl. the Rad30B protein was also shown to possess DNA
polymerase activity, and was therefore designated DNA polymerase I. (Pol l.)
[256]. Unlike Pol n, which is distributed ubiquitously throughout eukaryotes, Pol L
is found only in higher eukaryotes, beginning with Drosophila. Interestingly, the
enzymatic properties of Drosophila Pol r are much more comparable to those of
human and yeast Pol 1] than they are to human Pol r. [257]. Based on these
observations, it is thought that Pol r. likely arose as a genetic duplication of Pol 1],
just prior to the emergence of insects, and that further selective pressures have

resulted in the evolution of two distinct polymerases in higher eukaryotes [258].

Similar to other TLS polymerases, Pol L is a distributive enzyme that lacks
exonuclease activity and is highly error-prone when copying undamaged DNA
[256]. However, the misincorporation frequency and specificity of Pol L is

unparalleled even with respect to other TLS polymerases [250]. The fidelity of Pol

84

r is uniquely template dependent [256]. On primed single stranded DNA in vitro,

Pol L exhibits the highest efficiency and fidelity opposite a template A,
incorporating the incorrect nucleotide at a reasonable frequency of ~1 x 10'4
[256]. In contrast, Pol L is extremely inaccurate when synthesizing DNA opposite

a template T, exhibiting a much higher error rate of ~7 x 10'1 [256]. In fact, within

certain sequence contexts, Pol l. incorrectly inserts a G opposite a template T up

to 11 times more frequently than it inserts the correct A residue, meaning the

fidelity of Pol I. can vary up to 105-fold merely as a result of which templating

base is being replicated [159, 171, 256, 259]. VIfith undamaged DNA, Pol L is just
as efficient at extending from a mispaired primer terminus as it is at generating a
mispair [260], making Pol I. one of the most error-prone DNA polymerases

discovered to date [256].

Overall, the structure of Pol l. is similar to that of other polymerases, typified by a
right hand architecture consisting of palm, fingers and thumb domains. Like other
Y—family polymerases thus far characterized, Pol r. contains the unique
polymerase associated domain, and thumb and finger domains that are shorter
than those of the classical polymerases. However, in contrast to all other known
polymerases, including fellow members of the Y-family, during DNA synthesis,
the active site of Pol L employs Hoogsteen base pairing, rather than the typical
Watson-Crick base pairing [261]. In such Hoogsteen base pairing, the templating

base is driven to the syn conformation by bulky residues in the fingers domain,

85

forcing the incoming nucleotide to form hydrogen bonds with the Hoogsteen

edge, rather than the Watson-Crick edge, of the templating base [261].

The revelation that Pol u. employs Hoogsteen base pairing, rather that the typical
Watson-Crick, has provided an interesting explanation for the ability of Pol L to be
relatively accurate when synthesizing DNA opposite a template A, but be highly
error-prone when inserting opposite a template T. Opposite a template A in syn
conformation, only an incoming anti T can make two favorable hydrogen bonds.
In contrast, a template T lacks a Hoogsteen edge, which would prevent hydrogen

bonding, thus making base pairing ambiguous in this context [261].

DNA polymerase L, like its paralog Pol n, is also proficient at bypassing fork-
blocking DNA lesions. However, the lesion bypass specificity of Pol L is
somewhat different than that of its ancestor Pol 1]. Perhaps the most notable
difference between the two enzymes concerns the proficiency and fidelity with
which they are able to bypass UV-induced DNA lesions. Unlike Pol n, which is
characterized by its ability to bypass T-T dimers both efficiently and accurately,
Pol L bypasses T-T dimers in a more limited and inaccurate manner [262, 263].
With regard to a 6-4 photoproduct, Pol L performs relatively efficient but
inaccurate nucleotide incorporation (although it is unable to perform the
subsequent extension steps), whereas Pol TI is primarily blocked by this UV-
induced DNA lesion [262, 263]. More recently, convincing evidence of a role for

Pol L in mutagenic TLS past UV-induced DNA lesions has been indicated from

86

studies using mice carrying a naturally occurring nonsense mutation in the Poli.
These mice are characterized by elevated UV-induced mutation frequencies and
are more susceptible to certain types of UV-induced cancers [264, 265]. In fact,
studies have indicated that in xeroderma pigmentosum variant cells, which lack
Pol n, mutagenic TLS of UV-induced DNA lesions by Pol L is responsible for the
elevated frequency of UV-induced mutations that ultimately leads to their

malignant transformation [266].

In addition to DNA lesions induced by UV, the efficiency and accuracy with which
Pol L is able to replicate past a variety of other types of DNA lesions has also
been examined in vitro. It is generally agreed upon that Pol L is able to efficiently
insert any of the four nucleotides opposite an abasic site, but is not able to
perform the subsequent extension reaction [171, 258, 267]. Likewise, it is
generally accepted that Pol L is relatively efficient at incorporating a C opposite
an N-2-acetylamlnofluorene adducted guanine, but further extension is negligible
[258, 267]. However, reports of the ability of Pol L to bypass 8-oxoguanine lesions
in primer templates varies somewhat between laboratories. In a study by Zhang
et al., they found that 8-oxoguanine presented a severe block to TLS by Pol L,
which could only be overcome by adding increasing amounts of Pol L to the
reaction [267]. Whereas a study by Vaisman et al., demonstrated only a very
modest reduction in TLS past an 8-oxoguanine adduct [258]. Both groups found
that on the whole, Pol L favors insertion of the correct C residue opposite an 8-

oxoguanine lesion [258, 267]. Although benzo[a]pyrene adducted guanines

87

present a strong block to TLS by Pol L, insertion opposite benzo[a]pyrene and
benzo[a]phenanthrene adducted adenines is both efficient and accurate
(although the subsequent extension requires another polymerase). Because Pol L
is also very accurate at replication of an undamaged template A, it has been
suggested that Pol L has evolved to protect organisms from mutations induced by
damaged adenosines in a manner similar to which Pol Tl protects us from

mutations induced by damaged thymines [250].

Interestingly, Pol L has also been shown to possess an intrinsic 5’
deoxyribosephosphate lyase activity. In other words, Pol L is able to remove a 5’
deoxyribose phosphate group from DNA [268]. In reconstituted reactions
containing uracil-DNA glycosylase, apurinic/apyrimidinic endonuclease, and DNA
ligase l, Pol L can use its dRP lyase and polymerase activities to repair G:U and
A:U pairs in DNA [268]. Furthermore, in cell extracts devoid of Pol 8, addition of
purified Pol L restores base excision repair of substrates containing uracil [269],
suggesting that Pol L has a role in a specialized form of base excision repair of

uracil-containing substrates.

e. DNA Polymerase 1:

DNA polymerase x (Pol x) was originally identified in humans as a homolog of
the E. coli DinB (Pol IV) protein, which was discovered based on its involvement
in the UV-induced, untargeted mutagenesis of bacteriophage lambda, and later

shown to be a specialized DNA polymerase [270-272]. The DinB subfamily is the

88

most widely distributed, found in eukaryotes, prokaryotes, as well as archaea.
However, it is clearly not essential for life as it is conspicuously absent from the

genomes of Saccharomyces cerevisiae and Drosophila melanogaster [158].

Polymerase K, like other TLS polymerases, lacks 3’—>5’ proofreading
exonuclease activity [273-275], and similarly, is much less accurate than the
classical replicative polymerases when replicating undamaged DNA. For

example, in an M13 mutagenesis fonrvard assay, DNA synthesis by Pol x

produced 760 errors per 105 nucleotides, whereas synthesis by Pol 6 produced

only three errors per 105 nucleotides [274]. Polymerase x is also less accurate
than the S. cerevisiae B-family TLS polymerase C, which generated 130 errors
per 105 nucleotides [170], but is the most accurate of the human Y-family
polymerases, as Pol Tl [248] and Pol L [276] produced 3,700 and 72,000 errors
per 105 nucleotides, respectively. Based on structural analysis, it has been

suggested that the higher fidelity of Pol x compared to other Y-family

polymerases results from a more constrained active site [277].

In addition to being relatively inaccurate during DNA synthesis, Pol x is also quite
proficient at extending from mispaired primer termini. In fact, Pol 1c extends from
each of the 12 possible DNA mismatches more efficiently than it produces the
mispairs. These results suggest that Pol x, like Pol C, contributes to spontaneous

mutagenesis by extending from mispairs generated by other DNA polymerases

89

[275, 278]. In keeping with this characteristic, upregulation of Pol x has been
shown to elevate the frequency of spontaneous mutations tenfold in mouse cells

and seven to ninefold in human cells [270, 279].

The specificity of lesion bypass of human Pol x has been studied extensively in
vitro using primer extension assays. These experiments have demonstrated that,
Pol x is not able to bypass either UV-induced T-T dimers or 6-4 photoproducts
[272, 280, 281]. Additionally, Pol x is unable to bypass cisplatin adducts [273,
281] and is either inefficient or unable to bypass abasic sites [272, 280, 281].
Polymerase x is, however, able to bypass DNA adducts induced by
acetylaminofluorene, as well as 8-oxoguanine adducts, but does so primarily in

an error-prone manner [135, 273, 280, 281].

Although Pol x has the potential to introduce mutations during error-prone
translesion synthesis, it has been shown, also in vitro, to bypass thymine glycol,
a biologically important form of oxidative base damage, in an error-free manner
by inserting the correct base, adenine across from this lesion [282]. Moreover,
Pol x bypasses lesions resulting from benzo[a]pyrene diol epoxide (BPDE), an
active metabolite of the environmental carcinogen benzo[a]pyrene, in an error-

free manner [280, 283, 284].

In addition to a requirement for incorporation of nucleotides opposite DNA

damage, Pol x has also been implicated in the extension step of TLS. For

90

example, although Pol x cannot insert nucleotides opposite the 3’T of a T-T
dimer, it has been shown to efficiently extend from a G placed opposite the 3’T of

this lesion [278]. Polymerase x has also been shown to efficiently extend from a
variety of nucleotides placed opposite an 06-methylguanine lesion or an 8-

oxoguanine lesion by Pol 6 [285]. What is more, Pol x efficiently extends from an
A residue misincorporated opposite a BPDE adducted guanine by Pol TI. which is

unable to extend from this lesion-containing terminus alone [286].

There have been fewer studies reported which examine the role of Pol x in vivo.
The Ohmori laboratory developed mouse embryonic stem cells defective in Pol x
and found that such cells were not only hypersensitive to the cytotoxic effects of
BPDE, they also accumulated more BPDE-induced mutations than the parental
cell strains [287]. In addition, when mouse embryonic fibroblasts (MEFs)
generated in the Livneh laboratory were assayed for their ability to bypass a site
specific BPDE lesion in a plasmid based TLS assay, they found that Pol x was
responsible for approximately two-thirds of lesion bypass events and that such
events occurred in primarily an error-free manner [288]. Together these data
indicate, in accordance with in vitro data, that Pol x plays an important role in the

accurate bypass of DNA lesions induced by benzo[a]pyrene.

Both the mouse and human POLK genes contain two xenobiotic responsive
elements (XRE) in their promoters [289, 290]. Xenobiotic responsive elements

are binding sites for polycyclic aromatic hydrocarbons (PAH), such as

91

benzo[a]pyrene, that when bound result in upregulation of specific PAH-inducible
genes. One such gene is the CYP1A1 gene, which encodes cytochrome P450.
Cytochrome P450 is a protein that metabolizes benzo[a]pyrene and other PAHs
into compounds, which are more easily excreted from cells. However,
metabolism can also inadvertently result in the activation of some PAHs into
electrophilic forms, such as benzo[a]pyrene diol epoxide, that are highly reactive
to DNA. The presence of these XRE sequences in the POLK gene may also be

indicative of a requirement for Pol x in TLS past DNA lesions induced by PAHs.

Interestingly, the bacterial homolog of Pol K, Pol V is also able to bypass BPDE
adducted guanines in an error-free manner in vitro, indicating that the ability to
perform TLS past BPDE-induced DNA lesions is conserved from E. coli to
mammals [137]. However, such a role in E. coli cells is likely not physiologically
relevant because E. coli do not posses an enzyme that activates benzo[a]pyrene
into BPDE. Moreover, it is unlikely that mammalian Pol x is conserved strictly for
bypassing BPDE-induced DNA lesions [291]. More recently, it was found that Pol
x is able to bypass estrogen-derived DNA adducts [292]. Furthermore, Pol x is
highly expressed in human testis and ovaries [271] and the adrenal cortex of
mice [293] where steroid hormones are produced. Based on these data, it has
been suggested that the cognate lesions for TLS by Pol x are steroid-derived

DNA adducts, especially those that are structurally similar to BPDE and are

formed at the N2 position of guanine [291].

92

In vivo experiments with mouse embryonic stem cells lacking Pol x also
demonstrated that such cells are moderately sensitive to UV radiation [287]. This
finding was particularly intriguing in light of the fact that in vitro, Pol x is unable to
bypass T—T dimers or 6-4 photoproducts [272, 280, 281]. Polymerase x is,
however, able to extend from a G or A placed opposite the 3’T of a T-T dimer
[278]. Initially the increased sensitivity of cells lacking Pol x to UV radiation was
attributed to a role for Pol x in the extension step of TLS past UV-induced T-T
dimers [287]. However, more recent evidence has indicated that such an
increase in UV—induced sensitivity might be attributed to a role for Pol K in
nucleotide excision repair (NER). In a study by Alan Lehmann’s group [294], they
showed that both recovery of RNA synthesis after UV irradiation (associated with
transcription-coupled NER) and unscheduled DNA synthesis (a measure of
global genome NER), are substantially reduced in MEFs lacking Pol x. In
addition they demonstrated that DNA repair synthesis and photoproduct removal
are reduced in cells lacking Pol x compared to wild-type. These data suggest that
Pol x is involved in the repair synthesis step of NER. However, it remains unclear
how employing a polymerase with such a reduced fidelity would be beneficial in a

process that should be primarily error-free.

2. Damage Avoidance in Eukaryotes

The eukaryotic damage avoidance pathway is much less well understood than

eukaryotic TLS. Relatively recently, however, some of the key factors involved in

93

this pathway have been identified in S. cerevisiae. One such factor is Mms2. The
MMSZ gene was first identified in a yeast mutant strain sensitive to methyl
methane sulfonate, and functional complementation analyses placed the product
of the MMSZ gene in the RAD6 pathway (discussed above) [295, 296]. Yeast
mmsZ mutants are much less sensitive to DNA damaging agents then either the
rad6 or rad18 mutants are, therefore it is commonly accepted that, the function of
the MMSZ gene is required for only one of the two RAD6-mediated damage
tolerance pathways [296]. Based on the fact that the mmsZ mutant demonstrates
a high frequency of spontaneous mutations that can be completely abolished by
inactivating the REV3-mediated TLS pathway, and that a rev3/mm32 double
mutant (but neither of the single mutants) is as sensitive to DNA damaging
agents as a rad18 single mutant is, the function of the MMSZ gene has been

implicated in the error-free DA pathway [296, 297].

The MMSZ gene encodes a protein that is homologous to a ubiquitin conjugating
enzyme, but lacks the active site cysteine residue required for ubiquitin
conjugation [296]. Such ubiquitin enzyme variants have previously been
hypothesized to function as dominant-negative regulators of ubiquitin
conjugation. Based on the epistatic relationship between mmsz and rad6, it was
believed for sometime that the function of MmsZ was to modulate the ubiquitin
conjugating activity of Rad6. However, direct evidence to support such a theory
has not yet been established (reviewed in [149]). On the other hand, co-

purification assays and yeast two-hybrid screens have demonstrated that MmsZ

94

forms a stable complex with the ubiquitin conjugating enzyme, ch13 [298, 299].
Moreover, the ubc13 yeast mutant is phenotypically indistinguishable from the
mmsZ mutant. Based on these data, MmsZ is now believed to modulate the
activity of ch13 in the DA pathway [300]. The Mms2-ch13 complex has been
shown to catalyze the formation of lysine63-linked poly-ubiquitin chains [298].
Studies have further demonstrated that the cognate E3 for MmsZ-ch13 is Rad5,
a DNA helicase protein with ATPase activity [301, 302]. Rad5 interacts with both
ch13 and Rad18 and thus provides a physical link between the RAD6 and DA

pathways [303].

In a recent review by Andersen et al., a model demonstrating how the functions
of the Mms2 and ch13 proteins might be used for initiation of DA, was proposed
[9]. In this model, Rad18 recruits Rad6 to segments of single-stranded DNA that
are bound by PCNA, such as sites of unrepaired DNA damage. The DNA-bound
PCNA is then mono-ubiquitinated by the Rad6-Rad18 complex, which either
directly or indirectly, promotes TLS. The Mms2-ch13 complex is recruited to
DNA-bound PCNA through interaction with Rad5. The presence of the Mms2-
ch13-Rad5 complex results in further addition of ubiquitin residues linked
through Iysine63. Finally, such poly-ubiquitination of PCNA promotes the error-

free damage avoidance pathway.

Although significant advances have been made concerning the processes by

which DA is initiated, the actual mechanism of error-free DA is much less well

95

understood. Thus far, two possible mechanisms of DA have been suggested,
namely replication fork regression and template switching [9]. Replication fork
regression, refers to a mechanism whereby the replication fork migrates
backward until the original template strands are re-annealed and the newly
synthesized daughter strand is expelled. Such a process can result in the
generation of a so-called “chicken foot structure” that can serve as a homologous
recombination intermediate. Evidence supporting the use of a fork regression
mechanism in DA comes from studies by Blastyak et al. [304], who showed that

yeast Rad5 exhibits a DNA helicase activity that can facilitate fork regression.

Despite some evidence of Rad5-mediated fork regression, it is most commonly
accepted that the mechanism employed by the error-free DA pathway is template
switching (reviewed in [9]). In template switching, when a high fidelity replicative
polymerase is blocked by a DNA lesion, this polymerase temporarily uses a
segment of DNA homologous to that containing the lesion (for example the newly
replicated daughter strand) as a template to replicate around the damaged DNA.
The template switch mechanism of DA is supported by the studies of Li et al.
[305], who demonstrated that human fibroblasts strains with reduced levels of the
human homolog of the yeast Mms2 gene, hMms2, were virtually unable to use
an allelic gene copy as a template for replication past UV-induced DNA lesions.
These results are consistent with the notion that error-free damage avoidance

occurs through a mechanism template switching.

96

Although the mechanism through which error-free damage avoidance is achieved
is still unclear, this process is apparently highly conserved throughout all
eukaryotes. Homologs of all the proteins known to be involved in DA, including
Mms2, chIS and Rad5, have been found in plants, mammals and other higher
eukaryotes (see [9] and references therein). Such a high level of conservation is

likely a reflection of the importance of error-free damage avoidance process.

97

REFERENCES:

1. Friedberg, E. C., Walker, G.C., Siede, W., Wood, R.D., Schultz, RA, and
Ellenberger, T (2006). DNA Repair and Mutagenesis, 2nd edition,
American Society of Microbiology Press, Washginton, DC.

2. Lindahl, T. (1979). DNA glycosylases, endonucleases for
apurinic/apyrimidinic sites, and base excision-repair. Prog. Nucleic Acid
Res. Mol. Biol. 22, pp. 135-192.

3. Lindahl, T., and Karlstrom, O. (1973). Heat-induced depyrimidination of
deoxyribonucleic acid in neutral solution. Biochemistry 12, pp. 5151-5154.

4. Nakamura, J., Walker, V. E., Upton, P. B., Chiang, S. Y., Kow, Y. W., and
Swenberg, J. A. (1998). Highly sensitive apurinic/apyrimidinic site assay
can detect spontaneous and chemically induced depurination under
physiological conditions. Cancer Res. 58, pp. 222-225.

5. Knrvawicz, J., Arczewska, K. D., Speina, E., Maciejewska, A., and
Grzesiuk, E. (2007). Bacterial DNA repair genes and their eukaryotic
homologues: 1. Mutations in genes involved in base excision repair (BER)
and DNA-end processors and their implication in mutagenesis and human
disease. Acta. Biochim. Pol. 54, pp. 413-434.

6. Bjelland, S., and Seeberg, E. (2003). Mutagenicity, toxicity and repair of
DNA base damage induced by oxidation. Mutat. Res. 531, pp. 37-80.

7. Altieri, F., Grillo, C., Maceroni, M., and Chichiarelli, S. (2008). DNA
damage and repair: from molecular mechanisms to health implications.
Antioxid. Redox Signal 10, pp. 891-937.

8. Hakem, R. (2008). DNA-damage repair; the good, the bad, and the ugly.
EMBO J. 27, pp. 589-605.

9. Andersen, P. L., Xu, F., and Xiao, W. (2008). Eukaryotic DNA damage
tolerance and translesion synthesis through covalent modifications of
PCNA. Cell Res. 18. pp. 162-173.

98

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

Fromme, J. C., and Verdine, G. L. (2004). Base excision repair. Adv.
Protein Chem. 69, pp. 1-41.

Zharkov, D. O. (2008). Base excision DNA repair. Cell. Mol. Life Sci.

Hazra, T. K., lzumi, T., Boldogh, l., lmhoff, B., Kow, Y. W., Jaruga, P.,
Dizdaroglu, M., and Mitra, S. (2002). Identification and characterization of
a human DNA glycosylase for repair of modified bases in oxidatively
damaged DNA. Proc. Natl. Acad. Sci. USA 99, pp. 3523-3528.

Hazra, T. K., Kow, Y. W., Hatahet, Z., lmhoff, B., Boldogh, |., Mokkapati,
S. K., Mitra, S., and lzumi, T. (2002). Identification and characterization of
a novel human DNA glycosylase for repair of cytosine-derived lesions. J.
Biol. Chem. 277, pp. 30417-30420.

Bandaru, V., Sunkara, 8., Wallace, S. S., and Bond, J. P. (2002). A novel
human DNA glycosylase that removes oxidative DNA damage and is
homologous to Escherichia coli endonuclease VIII. DNA Repair (Amst) 1,
pp. 517-529.

Nash, H. M., Bruner, S. D., Scharer, O. D., Kawate, T., Addona, T. A.,
Spooner, E., Lane, W. S., and Verdine, G. L. (1996). Cloning of a yeast 8-
oxoguanine DNA glycosylase reveals the existence of a base-excision
DNA-repair protein superfamily. Curr. Biol. 6, pp. 968-980.

Engelward, B. P., Boosalis, M. S., Chen, B. J., Deng, Z., Siciliano, M. J.,
and Samson, L. D. (1993). Cloning and characterization of a mouse 3-
methyladenine/7-methyl-guanine/3-methylguanine DNA glycosylase cDNA
whose gene maps to chromosome 11. Carcinogenesis 14, pp. 175-181.

Dodson, M. L., Schrock, R. D., 3rd, and Lloyd, R. S. (1993). Evidence for
an imino intermediate in the T4 endonuclease V reaction. Biochemistry 32,
pp. 8284-8290.

Weiss, B. (1976). Endonuclease II of Escherichia coli is exonuclease Ill. J.
Biol. Chem. 251, pp. 1896-1901.

Demple, B., Herman, T., and Chen, D. S. (1991). Cloning and expression
of APE, the cDNA encoding the major human apurinic endonuclease:

99

20.

21.

22.

23.

24.

25.

26.

27.

28.

definition of a family of DNA repair enzymes. Proc. Natl. Acad. Sci. USA
88. pp. 11450-11454.

Hadi, M. Z., and Wilson, D. M., 3rd. (2000). Second human protein with
homology to the Escherichia coli abasic endonuclease exonuclease Ill.
Environ. Mol. Mutagen 36, pp. 312-324.

Ljungquist, S., Lindahl, T., and Howard-Flanders, P. (1976). Methyl
methane sulfonate-sensitive mutant of Escherichia coli deficient in an
endonuclease specific for apurinic sites in deoxyribonucleic acid. J.
Bacteriol. 126, pp. 646-653.

Ljungquist, S. (1977). A new endonuclease from Escherichia coli acting at
apurinic sites in DNA. J. Biol. Chem. 252, pp. 2808-2814.

Popoff, S. C., Spira, A. L, Johnson, A. W., and Demple, B. (1990). Yeast
structural gene (APN1) for the major apurinic endonuclease: homology to
Escherichia coli endonuclease IV. Proc. Natl. Acad. Sci. USA 87, pp.
4193-4197.

Avery, A. M., Kaur, B., Taylor, J. S., Mello, J. A., Essigmann, J. M., and
Doetsch, P. W. (1999). Substrate specificity of ultraviolet DNA
endonuclease (UVDE/Uve1p) from Schizosaccharomyces pombe. Nucleic
Acids Res. 27, pp. 2256-2264.

Kanno, S., lwai, S., Takao, M., and Yasui, A. (1999). Repair of
apurinic/apyrimidinic sites by UV damage endonuclease; a repair protein
for UV and oxidative damage. Nucleic Acids Res. 27, pp. 3096-3103.

Kanno, S., Kuzuoka, H., Sasao, 8., Hong, Z., Lan, L., Nakajima, S., and
Yasui, A. (2007). A novel human AP endonuclease with conserved zinc-
finger-like motifs involved in DNA strand break responses. EMBO J. 26,
pp. 2094-2103.

Dianov, G., and Lindahl, T. (1994). Reconstitution of the DNA base
excision-repair pathway. Curr. Biol. 4, pp. 1069-1076.

Dianov, G., Sedgwick, B., Daly, G., Olsson, M., Lovett, S., and Lindahl, T.
(1994). Release of 5'-tenninal deoxyribose-phosphate residues from

100

29.

30.

31.

32.

33.

34.

35.

36.

37.

incised abasic sites in DNA by the Escherichia coli RecJ protein. Nucleic
Acids Res. 22, pp. 993-998.

Sandigursky, M., and Franklin, W. A. (1992). DNA
deoxyribophosphodiesterase of Escherichia coli is associated with
exonuclease I. Nucleic Acids Res. 20, pp. 4699-4703.

Graves, R. J., Felzenszwalb, |., Laval, J., and O'Connor, T. R. (1992).
Excision of 5'-terminal deoxyribose phosphate from damaged DNA is
catalyzed by the Fpg protein of Escherichia coli. J. Biol. Chem. 267, pp.
14429-14435.

Allinson, S. L., Dianova, II, and Dianov, G. L. (2001). DNA polymerase
beta is the major dRP lyase involved in repair of oxidative base lesions in
DNA by mammalian cell extracts. EMBO J. 20, pp. 6919-6926.

Sobol, R. W., Prasad, R., Evenski, A., Baker, A., Yang, X. P., Horton, J.
K., and Wilson, S. H. (2000). The lyase activity of the DNA repair protein
beta-polymerase protects from DNA-damage-induced cytotoxicity. Nature
405, pp. 807-810.

Pawl, G., Taylor, R., Minton, K., and Friedberg, E. C. (1976). Enzymes
involved in thymine dimer excision in bacteriophage T4—infected
Escherichia coli. J. Mol. Biol. 108, pp. 99-109.

Klungland, A., and Lindahl, T. (1997). Second pathway for completion of
human DNA base excision-repair: reconstitution with purified proteins and
requirement for DNase IV (FEN1). EMBO J. 16, pp. 3341-3348.

Boyce, R. P., and Howard-Flanders, P. (1964). Release of Ultraviolet
Light-Induced Thymine Dimers from DNA in E. Coli K-12. Proc. Natl.
Acad. Sci. USA 51, pp. 293-300.

Setlow, R. B., and Carrier, W. L. (1964). The Disappearance of Thymine
Dimers from DNA: An Error-Correcting Mechanism. Proc. Natl. Acad. Sci.
USA 51 , pp. 226-231.

Fousteri, M., and Mullenders, L. H. (2008). Transcription-coupled
nucleotide excision repair in mammalian cells: molecular mechanisms and
biological effects. Cell Res. 18, pp. 73-84.

101

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

Sancar, A., and Reardon, J. T. (2004). Nucleotide excision repair in E. coli
and man. Adv. Protein Chem. 69, pp. 43-71.

Howard-Flanders, P., Boyce, R. P., and Theriot, L. (1966). Three loci in
Escherichia coli K-12 that control the excision of pyrimidine dimers and
certain other mutagen products from DNA. Genetics 53, pp. 1119-1136.

Truglio, J. J., Croteau, D. L., Van Houten, B., and Kisker, C. (2006).
Prokaryotic nucleotide excision repair: the UvrABC system. Chem. Rev.
106. pp. 233-252.

Mellon, I., and Hanawalt, P. C. (1989). Induction of the Escherichia coli
lactose operon selectively increases repair of its transcribed DNA strand.
Nature 342, pp. 95-98.

Sancar, A., Lindsey-Boltz, L. A., UnsaI-Kacmaz, K., and Linn, S. (2004).
Molecular mechanisms of mammalian DNA repair and the DNA damage
checkpoints. Annu. Rev. Biochem. 73, pp. 39-85.

Shuck, S. 0., Short, E. A., and Turchi, J. J. (2008). Eukaryotic nucleotide
excision repair: from understanding mechanisms to influencing biology.
Cell Res. 18, pp. 64-72.

Wakasugi, M., Kawashima, A., Morioka, H., Linn, S., Sancar, A., Mori, T.,
Nikaido, O., and Matsunaga, T. (2002). DDB accumulates at DNA damage
sites immediately after UV irradiation and directly stimulates nucleotide
excision repair. J. Biol. Chem. 277, pp. 1637-1640.

Maddukuri, L., Dudzinska, D., and Tudek, B. (2007). Bacterial DNA repair
genes and their eukaryotic homologues: 4. The role of nucleotide excision
DNA repair (NER) system in mammalian cells. Acta Biochim. Pol. 54, pp.
469-482.

Lehmann, A. R. (1982). Three complementation groups in Cockayne
syndrome. Mutat. Res. 106, pp. 347-356.

Van den Boom, V., Citterio, E., Hoogstraten, D., Zotter, A., Egly, J. M., van
Cappellen, W. A., Hoeijmakers, J. H., Houtsmuller, A. B., and Vermeulen,
W. (2004). DNA damage stabilizes interaction of CSB with the
transcription elongation machinery. J. Cell Biol. 166, pp. 27-36.

102

48.

49.

50.

51.

52.

53.

54.

55.

Fousteri, M., Vermeulen, W., van Zeeland, A. A., and Mullenders, L. H.
(2006). Cockayne syndrome A and B proteins differentially regulate
recruitment of chromatin remodeling and repair factors to stalled RNA
polymerase II in vivo. Mol. Cell 23, pp. 471-482.

Kamiuchi, S., Saijo, M., Citterio, E., de Jager, M., Hoeijmakers, J. H., and
Tanaka, K. (2002). Translocation of Cockayne syndrome group A protein
to the nuclear matrix: possible relevance to transcription-coupled DNA
repair. Proc. Natl. Acad. Sci. USA 99, pp. 201-206.

Groisman, R., Polanowska, J., Kuraoka, l., Sawada, J., Saijo, M., Drapkin,
R., Kisselev, A. F., Tanaka, K., and Nakatani, Y. (2003). The ubiquitin
ligase activity in the DDBZ and CSA complexes is differentially regulated
by the COP9 signalosome in response to DNA damage. Cell 113, pp. 357-
367.

Nakatsu, Y., Asahina, H., Citterio, E., Rademakers, S., Vermeulen, W.,
Kamiuchi, S., Yeo, J. P., Khaw, M. C., Saijo, M., Kodo, N., Matsuda, T.,
Hoeijmakers, J. H., and Tanaka, K. (2000). XAB2, a novel
tetratricopeptide repeat protein involved in transcription-coupled DNA
repair and transcription. J. Biol. Chem. 275, pp. 34931-34937.

Smerdon, M. J., and Conconi, A. (1999). Modulation of DNA damage and
DNA repair in chromatin. Prog. Nucleic Acid Res. Mol. Biol. 62, pp. 227-
255.

Teng, Y., Yu, Y., Ferreiro, J. A., and Waters, R. (2005). Histone
acetylation, chromatin remodelling, transcription and nucleotide excision
repair in S. cerevisiae: studies with two model genes. DNA Repair (Amst)
4, pp. 870-883.

Tremeau-Bravard, A., Riedl, T., Egly, J. M., and Dahmus, M. E. (2004).
Fate of RNA polymerase II stalled at a cisplatin lesion. J. Biol. Chem. 279,
pp. 7751-7759.

Sarker, A. H., Tsutakawa, S. E., Kostek, 8., N9, C., Shin, D. S., Peris, M.,
Campeau, E., Tainer, J. A., Nogales, E., and Cooper, P. K. (2005).
Recognition of RNA polymerase II and transcription bubbles by XPG,
CSB, and TFIIH: insights for transcription-coupled repair and Cockayne
Syndrome. Mol. Cell 20, pp. 187-198.

103

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

Jung, Y., and Lippard, S. J. (2006). RNA polymerase II blockage by
cisplatin-damaged DNA. Stability and polyubiquitylation of stalled
polymerase. J. Biol. Chem. 281, pp. 1361-1370.

Laine, J. P., and Egly, J. M. (2006). Initiation of DNA repair mediated by a
stalled RNA polymerase IIO. EMBO J. 25, pp. 387-397.

Nudler, E. (1999). Transcription elongation: structural basis and
mechanisms. J. Mol. Biol. 288, pp. 1-12.

Cleaver, J. E. (1968). Defective repair replication of DNA in xeroderma
pigmentosum. Nature 218, pp. 652-656.

Cockayne, E. A. (1936). Dwarfism with retinal atrophy and deafness. Arch.
Dis. Child 11, pp. 1-8.

Mayne, L. V., and Lehmann, A. R. (1982). Failure of RNA synthesis to
recover after UV irradiation: an early defect in cells from individuals with

Cockayne's syndrome and xeroderma pigmentosum. Cancer Res. 42, pp.
1473-1478.

Lehmann, A. (1987). Cockayne's syndrome and trichothiodystrophy:
defective repair without cancer. Cancer Rev. 7, pp. 82-103.

Dubaele, S., Proietti De Santis, L., Bienstock, R. J., Keriel, A., Stefanini,
M., Van Houten, B., and Egly, J. M. (2003). Basal transcription defect
discriminates between xeroderma pigmentosum and trichothiodystrophy in
XPD patients. Mol. Cell 11 , pp. 1635-1646.

Iyer, R. R., Pluciennik, A., Burdett, V., and Modrich, P. L. (2006). DNA
mismatch repair: functions and mechanisms. Chem. Rev. 106, pp. 302-
323.

Li, G. M. (2008). Mechanisms and functions of DNA mismatch repair. Cell
Res. 18, pp. 85-98.

Pukkila, P. J., Peterson, J., Herman, G., Modrich, P., and Meselson, M.
(1983). Effects of high levels of DNA adenine methylation on methyl-
directed mismatch repair in Escherichia coli. Genetics 104, pp. 571-582.

104

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

Su, S. S., Lahue, R. 8., Au, K. G., and Modrich, P. (1988). Mispair
specificity of methyl-directed DNA mismatch correction in vitro. J. Biol.
Chem. 263, Pp. 6829-6835.

Parker, B. O., and Marinus, M. G. (1992). Repair of DNA heteroduplexes
containing small heterologous sequences in Escherichia coli. Proc. Natl.
Acad. Sci. USA 89, pp. 1730-1734.

Grilley, M., Welsh, K. M., Su, S. S., and Modrich, P. (1989). Isolation and
characterization of the Escherichia coli mutL gene product. J. Biol. Chem.
264, pp. 1000-1004.

Welsh, K. M., Lu, A. L., Clark, 8., and Modrich, P. (1987). Isolation and
characterization of the Escherichia coli mutH gene product. J. Biol. Chem.
262, PP. 15624-15629.

Au, K. G., Welsh, K., and Modrich, P. (1992). Initiation of methyl-directed
mismatch repair. J. Biol. Chem. 267, pp. 12142-12148.

Dao, V., and Modrich, P. (1998). Mismatch-, MutS-, MutL-, and helicase Il-
dependent unwinding from the single-strand break of an incised
heteroduplex. J. Biol. Chem. 273, pp. 9202-9207.

Cooper, D. L., Lahue, R. S., and Modrich, P. (1993). Methyl-directed
mismatch repair is bidirectional. J. Biol. Chem. 268, pp. 11823-11829.

Grilley, M., Griffith, J., and Modrich, P. (1993). Bidirectional excision in
methyl-directed mismatch repair. J. Biol. Chem. 268, pp. 11830-11837.

Lahue, R. 8., Au, K. G., and Modrich, P. (1989). DNA mismatch correction
in a defined system. Science 245, pp. 160-164.

Shinya, E., and Shimada, T. (1994). Identification of two initiator elements
in the bidirectional promoter of the human dihydrofolate reductase and
mismatch repair protein 1 genes. Nucleic Acids Res. 22, pp. 2143-2149.

Bronner, C. E., Baker, S. M., Morrison, P. T., Warren, G., Smith, L. G.,
Lescoe, M. K., Kane, M., Earabino, C., Lipford, J., Lindblom, A., and et al.
(1994). Mutation in the DNA mismatch repair gene homologue hMLH1 is

105

78.

79.

80.

81.

82.

83.

84.

85.

associated with hereditary non-polyposis colon cancer. Nature 368, pp.
258-261.

Papadopoulos, N., Nicolaides, N. C., Wei, Y. F., Ruben, S. M., Carter, K.
C., Rosen, C. A., Haseltine, W. A., Fleischmann, R. D., Fraser, C. M.,
Adams, M. D., and et al. (1994). Mutation of a mutL homolog in hereditary
colon cancer. Science 263, pp. 1625-1629.

Nicolaides, N. C., Papadopoulos, M, Liu, 8., Wei, Y. F., Carter, K. C.,
Ruben, S. M., Rosen, C. A., Haseltine, W. A., Fleischmann, R. 0., Fraser,
C. M., and et al. (1994). Mutations of two PMS homologues in hereditary
nonpolyposis colon cancer. Nature 371, pp. 75-80.

Lipkin, S. M., Wang, V., Jacoby, R., Banerjee—Basu, S., Baxevanis, A. D.,
Lynch, H. T., Elliott, R. M., and Collins, F. S. (2000). MLH3: a DNA
mismatch repair gene associated with mammalian microsatellite instability.
Nat. Genet. 24, pp. 27-35.

Fishel, R., Lescoe, M. K., Rao, M. R., Copeland, N. G., Jenkins, N. A.,
Garber, J., Kane, M., and Kolodner, R. (1993). The human mutator gene

homolog MSH2 and its association with hereditary nonpolyposis colon
cancer. Cell 75, pp. 1027-1038.

Leach, F. S., Nicolaides, N. C., Papadopoulos, N., Liu, B., Jen, J.,
Parsons, R., Peltomaki, P., Sistonen, P., Aaltonen, L. A., Nystrom-Lahti,
M., and et al. (1993). Mutations of a mutS homolog in hereditary
nonpolyposis colorectal cancer. Cell 75, pp. 1215-1225.

Paquis-Flucklinger, V., Santucci-Darmanin, 8., Paul, R., Saunieres, A.,
Turc-Carel, C., and Desnuelle, C. (1997). Cloning and expression analysis
of a meiosis-specific MutS homolog: the human MSH4 gene. Genomics
44, pp. 188-194.

Her, C., and Doggett, N. A. (1998). Cloning, structural characterization,
and chromosomal localization of the human orthologue of Saccharomyces
cerevisiae MSH5 gene. Genomics 52, pp. 50-61.

Papadopoulos, N., Nicolaides, N. C., Liu, B., Parsons, R., Lengauer, C.,
Palombo, F., D'Arrigo, A., Markowitz, S., erlson, J. K., Kinzler, K. W., and
et al. (1995). Mutations of GTBP in genetically unstable cells. Science
268, pp. 1915-1917.

106

86.

87.

88.

89.

90.

91.

92.

93.

94.

Drummond, J. T., Li, G. M., Longley, M. J., and Modrich, P. (1995).
Isolation of an hMSH2-p160 heterodimer that restores DNA mismatch
repair to tumor cells. Science 268, pp. 1909-1912.

Palombo, F., Gallinari, P., laccarino, I., Lettieri, T., Hughes, M., D'Arrigo,
A., Truong, O., Hsuan, J. J., and Jiricny, J. (1995). GTBP, a 160-kilodalton
protein essential for mismatch-binding activity in human cells. Science
268, pp. 1912-1914.

Palombo, F., laccarino, I., Nakajima, E., lkejima, M., Shimada, T., and
Jiricny, J. (1996). hMutSbeta, a heterodimer of hMSH2 and hMSH3, binds
to insertion/deletion loops in DNA. Curr. Biol. 6, pp. 1181-1184.

Genschel, J., Littman, S. J., Drummond, J. T., and Modrich, P. (1998).
Isolation of MutSbeta from human cells and comparison of the mismatch
repair specificities of MutSbeta and MutSalpha. J. Biol. Chem. 273, pp.
19895-19901.

Li, G. M., and Modrich, P. (1995). Restoration of mismatch repair to
nuclear extracts of H6 colorectal tumor cells by a heterodimer of human
MutL homologs. Proc. Natl. Acad. Sci. USA 92, pp. 1950-1954.

Schmutte, C., Marinescu, R. C., Sadoff, M. M., Guerrette, S., Overhauser,
J., and Fishel, R. (1998). Human exonuclease l interacts with the
mismatch repair protein hMSH2. Cancer Res. 58, pp. 4537-4542.

Tishkoff, D. X., Amin, N. S., \fiars, C. S., Arden, K. C., and Kolodner, R. D.
(1998). Identification of a human gene encoding a homologue of
Saccharomyces cerevisiae EXO1, an exonuclease implicated in mismatch
repair and recombination. Cancer Res. 58, pp. 5027-5031.

Lin, Y. L., Shivji, M. K., Chen, C., Kolodner, R., Wood, R. D., and Dutta, A.
(1998). The evolutionarily conserved zinc finger motif in the largest subunit
of human replication protein A is required for DNA replication and
mismatch repair but not for nucleotide excision repair. J. Biol. Chem. 273,
pp. 1453-1461.

Ramilo, C., Gu, L., Guo, 8., Zhang, X., Patrick, S. M., Turchi, J. J., and Li,
G. M. (2002). Partial reconstitution of human DNA mismatch repair in vitro:
characterization of the role of human replication protein A. Mol. Cell. Biol.
22, pp. 2037-2046.

107

95.

96.

97.

98.

99.

100.

101.

102.

103.

Umar, A., Buermeyer, A. B., Simon, J. A., Thomas, D. C., Clark, A. B.,
Liskay, R. M., and Kunkel, T. A. (1996). Requirement for PCNA in DNA
mismatch repair at a step preceding DNA resynthesis. Cell 87, pp. 65-73.

Johnson, R. E., Kowali, G. K., Guzder, S. N., Amin, N. S., Holm, C.,
Habraken, Y., Sung, P., Prakash, L., and Prakash, S. (1996). Evidence for
involvement of yeast proliferating cell nuclear antigen in DNA mismatch
repair. J. Biol. Chem. 271, pp. 27987-27990.

Gu, L., Hong, Y., McCulloch, S., Watanabe, H., and Li, G. M. (1998). ATP-
dependent interaction of human mismatch repair proteins and dual role of
PCNA in mismatch repair. Nucleic Acids Res. 26, pp. 1173-1178.

Dzantiev, L., Constantin, N., Genschel, J., Iyer, R. R., Burgers, P. M., and
Modrich, P. (2004). A defined human system that supports bidirectional
mismatch-provoked excision. Mol. Cell 15, pp. 31-41.

Zhang, Y., Yuan, F., Presnell, S. R., Tian, K., Gao, Y., Tomkinson, A. E.,
Gu, L., and Li, G. M. (2005). Reconstitution of 5'-directed human mismatch
repair in a purified system. Cell 122, pp. 693-705.

Longley, M. J., Pierce, A. J., and Modrich, P. (1997). DNA polymerase
delta is required for human mismatch repair in vitro. J. Biol. Chem. 272,
pp. 10917-10921.

Raschle, M., Dufner, P., Marra, G., and Jiricny, J. (2002). Mutations within
the hMLH1 and hPMSZ subunits of the human MutLalpha mismatch repair
factor affect its ATPase activity, but not its ability to interact with
hMutSalpha. J. Biol. Chem. 277, pp. 21810-21820.

Blackwell, L. J., Bjornson, K. P., and Modrich, P. (1998). DNA—dependent
activation of the hMutSalpha ATPase. J. Biol. Chem. 273, pp. 32049-
32054.

Blackwell, L. J., Martik, D., Bjornson, K. P., Bjornson, E. S., and Modrich,
P. (1998). Nucleotide-promoted release of hMutSalpha from heteroduplex
DNA is consistent with an ATP-dependent translocation mechanism. J.
Biol. Chem. 273, pp. 32055-32062.

108

104.

105.

106.

107.

108.

109.

110.

111.

112.

113.

Gradia, S., Subramanian, D., leon, T., Acharya, 8., Makhov, A., Griffith,
J., and Fishel, R. (1999). hMSH2-hMSH6 forms a hydrolysis-independent
sliding clamp on mismatched DNA. Mol. Cell 3, pp. 255-261.

Holmes, J., Jr., Clark, 8., and Modrich, P. (1990). Strand-specific
mismatch correction in nuclear extracts of human and Drosophila
melanogaster cell lines. Proc. Natl. Acad. Sci. USA 87, pp. 5837-5841.

Thomas, D. C., Roberts, J. D., and Kunkel, T. A. (1991). Heteroduplex
repair in extracts of human HeLa cells. J. Biol. Chem. 266, pp. 3744-3751.

Bennett, S. E., Umar, A., Oshima, J., Monnat, R. J., Jr., and Kunkel, T. A.
(1997). Mismatch repair in extracts of Werner syndrome cell lines. Cancer
Res. 57, pp. 2956-2960.

Langland, G., Kordich, J., Creaney, J., Goss, K. H., Lillard-Wetherell, K.,
Bebenek, K., Kunkel, T. A., and Groden, J. (2001). The Bloom's syndrome
protein (BLM) interacts with MLH1 but is not required for DNA mismatch
repair. J. Biol. Chem. 276, pp. 30031-30035.

Pedrazzi, G., Perrera, C., Blaser, H., Kuster, P., Marra, G., Davies, S. L.,
Ryu, G. H., Freire, R., Hickson, I. D., Jiricny, J., and Stagljar, I. (2001).
Direct association of Bloom's syndrome gene product with the human
mismatch repair protein MLH1. Nucleic Acids Res. 29, pp. 4378-4386.

Rustgi, A. K. (2007). The genetics of hereditary colon cancer. Genes Dev.
21, pp. 2525-2538.

Lynch, H. T. (1974). Familial cancer prevalence spanning eight years.
Family N. Arch. lntem. Med. 134, pp. 931-938.

Lehmann, A. R., Niimi, A., Ogi, T., Brown, S., Sabbioneda, S., VIfing, J. F.,
Kannouche, P. L., and Green, C. M. (2007). Translesion synthesis: Y-
family polymerases and the polymerase switch. DNA Repair (Amst) 6, pp.
891-899.

Radman, M. (1974). Phenomenology of an inducible mutagenic DNA
repair pathway in E. coli: SOS repair hypothesis. in Molecular and
environmental aspects of mutagenesis (6th Rochester Conference on
Environmental Toxicology) (Prakash, L., Sherman, F., Miller, MW,

109

114.

115.

116.

117.

118.

119.

120.

121.

122.

Lawrence, CW, and Taber, H.W., Ed.), pp 128-142, Charles C. Thomas,
Springfield, Illinois.

Goodman, M. F., and Tippin, B. (2000). Sloppier copier DNA polymerases
involved in genome repair. Curr. Opin. Genet. Dev. 10, pp. 162-168.

Kato, T., and Shinoura, Y. (1977). Isolation and characterization of
mutants of Escherichia coli deficient in induction of mutations by ultraviolet
light. Mol. Gen. Genet. 156, pp. 121-131.

Steinborn, G. (1978). Uvm mutants of Escherichia coli K12 deficient in UV
mutagenesis. I. Isolation of uvm mutants and their phenotypical
characterization in DNA repair and mutagenesis. Mol. Gen. Genet. 165,
pp. 87-93.

Shinagawa, H., Kato, T., lse, T., Makino, K., and Nakata, A. (1983).
Cloning and characterization of the umu operon responsible for inducible
mutagenesis in Escherichia coli. Gene 23, pp. 167-174.

Elledge, S. J., and Walker, G. C. (1983). Proteins required for ultraviolet
light and chemical mutagenesis. Identification of the products of the umuC
locus of Escherichia coli. J. Mol. Biol. 164, pp. 175-192.

Woodgate, R., Rajagopalan, M., Lu, 0., and Echols, H. (1989). UmuC
mutagenesis protein of Escherichia coli: purification and interaction with
UmuD and UmuD’. Proc. Natl. Acad. Sci. USA 86, pp. 7301-7305.

Burckhardt, S. E., Woodgate, R., Scheuennann, R. H., and Echols, H.
(1988). UmuD mutagenesis protein of Escherichia coli: overproduction,
purification, and cleavage by RecA. Proc. Natl. Acad. Sci. USA 85, pp.
1811-1815.

Nohmi, T., Battista, J. R., Dodson, L. A., and Walker, G. C. (1988). RecA-
mediated cleavage activates UmuD for mutagenesis: mechanistic
relationship between transcriptional derepression and posttranslational
activation. Proc. Natl. Acad. Sci. USA 85, pp. 1816-1820.

Shinagawa, H., Iwasaki, H., Kato, T., and Nakata, A. (1988). RecA
protein-dependent cleavage of UmuD protein and SOS mutagenesis.
Proc. Natl. Acad. Sci. USA 85, pp. 1806-1810.

110

123.

124.

125.

126.

127.

128.

129.

130.

131.

Tang, M., Shen, X., Frank, E. G., O'Donnell, M., Woodgate, R., and
Goodman, M. F. (1999). UmuD'(2)C is an error-prone DNA polymerase,
Escherichia coli pol V. Proc. Natl. Acad. Sci. USA 96, pp. 8919-8924.

Reuven, N. B., Arad, G., Maor-Shoshani, A., and Livneh, Z. (1999). The
mutagenesis protein UmuC is a DNA polymerase activated by UmuD',
RecA, and SSB and is specialized for translesion replication. J. Biol.
Chem. 274, pp. 31763-31766.

Tang, M., Pham, P., Shen, X., Taylor, J. S., O'Donnell, M., Woodgate, R.,
and Goodman, M. F. (2000). Roles of E. coli DNA polymerases IV and V
in lesion-targeted and untargeted SOS mutagenesis. Nature 404, pp.
1014-1018.

Opperman, T., Murli, S., Smith, B. T., and Walker, G. C. (1999). A model
for a umuDC-dependent prokaryotic DNA damage checkpoint. Proc. Natl.
Acad. Sci. USA 96, pp. 9218-9223.

Kenyon, C. J., and Walker, G. C. (1980). DNA-damaging agents stimulate
gene expression at specific loci in Escherichia coli. Proc. Natl. Acad. Sci.
USA 77, pp. 2819-2823.

Ohmori, H., Hatada, E., Qiao, Y., Tsuji, M., and Fukuda, R. (1995). dinP, a
new gene in Escherichia coli, whose product shows similarities to UmuC
and its homologues. Mutat. Res. 347, pp. 1-7.

Wagner, J., Gruz, P., Kim, S. R., Yamada, M., Matsui, K., Fuchs, R. P.,
and Nohmi, T. (1999). The dinB gene encodes a novel E. coli DNA
polymerase, DNA pol IV, involved in mutagenesis. Mol. Cell 4, pp. 281-
286.

Kobayashi, S., Valentine, M. R., Pham, P., O'Donnell, M., and Goodman,
M. F. (2002). Fidelity of Escherichia coli DNA polymerase IV. Preferential
generation of small deletion mutations by dNTP-stabilized misalignment.
J. Biol. Chem. 277, pp. 34198-34207.

Fuchs, R. P., Fujii, S., and Wagner, J. (2004). Properties and functions of
Escherichia coli: Pol IV and Pol V. Adv. Protein Chem. 69, pp. 229-264.

111

132.

133.

134.

135.

136.

137.

138.

139.

Wood, R. D., and Hutchinson, F. (1984). Non-targeted mutagenesis of
unirradiated lambda phage in Escherichia coli host cells irradiated with
ultraviolet light. J. Mol. Biol. 173, pp. 293-305.

Kim, S. R., Maenhaut-Michel, G., Yamada, M., Yamamoto, Y., Matsui, K.,
Sofuni, T., Nohmi, T., and Ohmori, H. (1997). Multiple pathways for SOS-
induced mutagenesis in Escherichia coli: an overexpression of dinB/dinP
results in strongly enhancing mutagenesis in the absence of any
exogenous treatment to damage DNA. Proc. Natl. Acad. Sci. USA 94, pp.
13792-13797.

Maor-Shoshani, A., Hayashi, K., Ohmori, H., and Livneh, Z. (2003).
Analysis of translesion replication across an abasic site by DNA
polymerase IV of Escherichia coli. DNA Repair (Amst) 2, pp. 1227-1238.

Suzuki, M, Ohashi, E., Hayashi, K., Ohmori, H., Grollman, A. P., and
Shibutani, S. (2001). Translesional synthesis past acetylaminofluorene—
derived DNA adducts catalyzed by human DNA polymerase kappa and
Escherichia coli DNA polymerase IV. Biochemistry 40, pp. 15176-15183.

Boudsocq, F., lwai, S., Hanaoka, F., and Woodgate, R. (2001). Sulfolobus
solfataricus P2 DNA polymerase IV (Dpo4): an archaeal DinB-like DNA
polymerase with lesion-bypass properties akin to eukaryotic poleta.
Nucleic. Acids Res. 29, pp. 4607-4616.

Shen, X., Sayer, J. M., Kroth, H., Ponten, l., O'Donnell, M., Woodgate, R.,
Jerina, D. M., and Goodman, M. F. (2002). Efficiency and accuracy of
SOS-induced DNA polymerases replicating benzo[a]pyrene-7,8-diol 9,10-
epoxide A and G adducts. J. Biol. Chem. 277, pp. 5265-5274.

Wagner, J., Etienne, H., Janel-Bintz, R., and Fuchs, R. P. (2002).
Genetics of mutagenesis in E. coli: various combinations of translesion
polymerases (Pol II, IV and V) deal with lesion/sequence context diversity.
DNA Repair (Amst) 1, pp. 159-167.

Bjedov, l., Dasgupta, C. N., Slade, D., Le Blastier, S., Selva, M., and
Matic, l. (2007). Involvement of Escherichia coli DNA polymerase IV in

tolerance of cytotoxic alkylating DNA lesions in vivo. Genetics 176, pp.
1431-1440.

112

140.

141.

142.

143.

144.

145.

146.

147.

148.

149.

Napolitano, R., JaneI-Bintz, R., Wagner, J., and Fuchs, R. P. (2000). All
three SOS-inducible DNA polymerases (Pol II, Pol IV and Pol V) are
involved in induced mutagenesis. EMBO J. 19, pp. 6259-6265.

McKenzie, G. J., Lee, P. L., Lombardo, M. J., Hastings, P. J., and
Rosenberg, S. M. (2001). SOS mutator DNA polymerase lV functions in
adaptive mutation and not adaptive amplification. Mol. Cell 7, pp. 571-579.

Yeiser, B., Pepper, E. D., Goodman, M. F., and Finkel, S. E. (2002). SOS-
induced DNA polymerases enhance Iong-terrn survival and evolutionary
fitness. Proc. Natl. Acad. Sci. USA 99, pp. 8737-8741.

Layton, J. C., and Foster, P. L. (2003). Error-prone DNA polymerase IV is
controlled by the stress-response sigma factor, RpoS, in Escherichia coli.
Mol. Microbiol. 50, pp. 549-561.

Schlacher, K., Pham, P., Cox, M. M., and Goodman, M. F. (2006). Roles
of DNA polymerase V and RecA protein in SOS damage-induced
mutation. Chem. Rev. 106, pp. 406-419.

Howard-Flanders, P., and Boyce, R. P. (1966). DNA repair and genetic
recombination: studies on mutants of Escherichia coli defective in these
processes. Radiat. Res. Suppl 6:156+.

Smith, K. C. (2004). Recombinational DNA repair: the ignored repair
systems. Bioessays 26, pp. 1322-1326.

RUpp, W. D., and Howard-Flanders, P. (1968). Discontinuities in the DNA
synthesized in an excision-defective strain of Escherichia coli following
ultraviolet irradiation. J. Mol. Biol. 31, pp. 291-304.

RUpp, W. D., Wilde, C. E., 3rd, Reno, D. L., and Howard-Flanders, P.
(1971). Exchanges between DNA strands in ultraviolet-irradiated
Escherichia coli. J. Mol. Biol. 61, pp. 25-44.

Pastushok, L., and Xiao, W. (2004). DNA postreplication repair modulated
by ubiquitination and sumoylation. Adv. Protein Chem. 69, pp. 279-306.

113

150.

151.

152.

153.

154.

155.

156.

157.

158.

Jentsch, S., McGrath, J. P., and Varshavsky, A. (1987). The yeast DNA
repair gene RAD6 encodes a ubiquitin-conjugating enzyme. Nature 329,
pp. 131-134.

Bailly, V., Lamb, J., Sung, P., Prakash, S., and Prakash, L. (1994).
Specific complex formation between yeast RAD6 and RAD18 proteins: a
potential mechanism for targeting RADG ubiquitin-conjugating activity to
DNA damage sites. Genes Dev. 8, pp. 811-820.

Bailly, V., Lauder, S., Prakash, S., and Prakash, L. (1997). Yeast DNA
repair proteins Rad6 and Rad18 form a heterodimer that has ubiquitin
conjugating, DNA binding, and ATP hydrolytic activities. J. Biol. Chem.
272, pp. 23360-23365.

Hoege, C., Pfander, B., Moldovan, G. L., Pyrowolakis, G., and Jentsch, S.
(2002). RAD6-dependent DNA repair is linked to modification of PCNA by
ubiquitin and SUMO. Nature 419, pp. 135-141.

Stelter, P., and Ulrich, H. D. (2003). Control of spontaneous and damage-
induced mutagenesis by SUMO and ubiquitin conjugation. Nature 425, pp.
188-191.

Haracska, L., Unk, l., Prakash, L., and Prakash, S. (2006). Ubiquitylation
of yeast proliferating cell nuclear antigen and its implications for
translesion DNA synthesis. Proc. Natl. Acad. Sci. USA 103, pp. 6477-
6482.

Nelson, J. R., Lawrence, C. W., and Hinkle, D. C. (1996). Thymine-
thymine dimer bypass by yeast DNA polymerase zeta. Science 272, pp.
1646-1649.

Nelson, J. R., Lawrence, C. W., and Hinkle, D. C. (1996). Deoxycytidyl
transferase activity of yeast REV1 protein. Nature 382, pp. 729-731.

Ohmori, H., Friedberg, E. C., Fuchs, R. P., Goodman, M. F., Hanaoka, F.,
Hinkle, D., Kunkel, T. A., Lawrence, C. W., Livneh, Z., Nohmi, T., Prakash,
L., Prakash, S., Todo, T., Walker, G. C., Wang, Z., and Woodgate, R.
(2001). The Y-family of DNA polymerases. Mol. Cell 8, pp. 7-8.

114

159.

160.

161.

162.

163.

164.

165.

166.

167.

168.

Yang, W., and Woodgate, R. (2007). What a difference a decade makes:
insights into translesion DNA synthesis. Proc. Natl. Acad. Sci. USA 104,
pp. 15591-15598.

McCulloch, S. D., and Kunkel, T. A. (2008). The fidelity of DNA synthesis
by eukaryotic replicative and translesion synthesis polymerases. Cell Res.
18. pp. 148-161.

Ling, H., Boudsocq, F., Woodgate, R., and Yang, W. (2001). Crystal
structure of a Y-family DNA polymerase in action: a mechanism for error-
prone and lesion-bypass replication. Cell 107, pp. 91-102.

Silvian, L. F., Toth, E. A., Pham, P., Goodman, M. F., and Ellenberger, T.
(2001). Crystal structure of a DinB family error-prone DNA polymerase
from Sulfolobus solfataricus. Nat. Struct. Biol. 8, pp. 984-989.

Trincao, J., Johnson, R. E., Escalante, C. R., Prakash, S., Prakash, L.,
and Aggarwal, A. K. (2001). Structure of the catalytic core of S. cerevisiae
DNA polymerase eta: implications for translesion DNA synthesis. Mol. Cell
8, pp. 417-426.

Zhou, B. L., Pata, J. D., and Steitz, T. A. (2001). Crystal structure of a
DinB lesion bypass DNA polymerase catalytic fragment reveals a classic
polymerase catalytic domain. Mol. Cell 8, pp. 427-437.

Lemontt, J. F. (1971). Mutants of yeast defective in mutation induced by
ultraviolet light. Genetics 68, pp. 21-33.

Lawrence, C. W., Das, G., and Christensen, R. B. (1985). REV7, a new
gene concerned with UV mutagenesis in yeast. Mol. Gen. Genet. 200, pp.
80-85.

Morrison, A., Christensen, R. B., Alley, J., Beck, A. K., Bernstine, E. G.,
Lemontt, J. F., and Lawrence, C. W. (1989). REV3, a Saccharomyces
cerevisiae gene whose function is required for induced mutagenesis, is
predicted to encode a nonessential DNA polymerase. J. Bacteriol. 171,
pp. 5659-5667.

Lawrence, C. W. (2002). Cellular roles of DNA polymerase zeta and Rev1
protein. DNA Repair (Amst) 1, pp. 425—435.

115

169.

170.

171.

172.

173.

174.

175.

176.

177.

178.

Lawrence, C. W. (2004). Cellular functions of DNA polymerase zeta and
Rev1 protein. Adv. Protein Chem. 69, pp. 167-203.

Zhong, X., Garg, P., Stith, C. M., Nick McElhinny, S. A., Kissling, G. E.,
Burgers, P. M., and Kunkel, T. A. (2006). The fidelity of DNA synthesis by
yeast DNA polymerase zeta alone and with accessory proteins. Nucleic
Acids Res. 34, pp. 4731-4742.

Johnson, R. E., Washington, M. T., Haracska, L., Prakash, S., and
Prakash, L. (2000). Eukaryotic polymerases iota and zeta act sequentially
to bypass DNA lesions. Nature 406, pp. 1015-1019.

Quah, S. K., von Borstel, R. C., and Hastings, P. J. (1980). The origin of
spontaneous mutation in Saccharomyces cerevisiae. Genetics 96, pp.
819-839.

Roche, H., Gietz, R. D., and Kunz, B. A. (1994). Specificity of the yeast
rev3 delta antimutator and REV3 dependency of the mutator resulting
from a defect (rad1 delta) in nucleotide excision repair. Genetics 137, pp.
637-646.

Lawrence, C. W., and Maher, V. M. (2001). Mutagenesis in eukaryotes
dependent on DNA polymerase zeta and Rev1p. Phil. Trans. R. Soc.
Lond. B 356. pp. 41-45.

Datta, A., and Jinks-Robertson, S. (1995). Association of increased
spontaneous mutation rates with high levels of transcription in yeast.
Science 268, pp. 1616-1619.

Holbeck, S. L., and Strathern, J. N. (1997). A role for REV3 in
mutagenesis during double-strand break repair in Saccharomyces
cerevisiae. Genetics 147, pp. 1017-1024.

Nelson, J. R., Gibbs, P. E., Nowicka, A. M., Hinkle, D. C., and Lawrence,
C. W. (2000). Evidence for a second function for Saccharomyces
cerevisiae Rev1 p. Mol. Microbiol. 37, pp. 549-554.

Johnson, R. E., Haracska, L., Prakash, S., and Prakash, L. (2001). Role of
DNA polymerase zeta in the bypass of a (6-4) TT photoproduct. Mol. Cell.
Biol. 21 , pp. 3558-3563.

116

179.

180.

181.

182.

183.

184.

185.

186.

187.

Guo, 0., Wu, X., Rajpal, D. K., Taylor, J. S., and Wang, Z. (2001).
Translesion synthesis by yeast DNA polymerase zeta from templates
containing lesions of ultraviolet radiation and acetylaminofluorene. Nucleic
Acids Res. 29, pp. 2875-2883.

Bresson, A., and Fuchs, R. P. (2002). Lesion bypass in yeast cells: Pol eta
participates in a multi-DNA polymerase process. EMBO J. 21, pp. 3881-
3887.

Gibbs, P. E., McDonald, J., Woodgate, R., and Lawrence, C. W. (2005).
The relative roles in vivo of Saccharomyces cerevisiae Pol eta, Pol zeta,
Rev1 protein and Pol32 in the bypass and mutation induction of an abasic
site, T-T (6-4) photoadduct and T-T cis-syn cyclobutane dimer. Genetics
169, pp. 575-582.

Abdulovic, A. L., and Jinks-Robertson, S. (2006). The in vivo
characterization of translesion synthesis across UV-induced lesions in
Saccharomyces cerevisiae: insights into Pol zeta- and Pol eta-dependent
frameshift mutagenesis. Genetics 172, pp. 1487-1498.

Yuan, F., Zhang, Y., Rajpal, D. K., Wu, X., Guo, D., Wang, M., Taylor, J.
S., and Wang, Z. (2000). Specificity of DNA lesion bypass by the yeast
DNA polymerase eta. J. Biol. Chem. 275, pp. 8233-8239.

Haracska, L., Washington, M. T., Prakash, S., and Prakash, L. (2001).
Inefficient bypass of an abasic site by DNA polymerase eta. J. Biol. Chem.
276, pp. 6861-6866.

Rechkoblit, 0., Zhang, Y., Guo, D., Wang, Z., Amin, S., Krzeminsky, J.,
Louneva, N., and Geacintov, N. E. (2002). trans-Lesion synthesis past
bulky benzo[a]pyrene diol epoxide N2-dG and N6-dA lesions catalyzed by
DNA bypass polymerases. J. Biol. Chem. 277, pp. 30488-30494.

Simhadri, S., Kramata, P., Zajc, B., Sayer, J. M., Jerina, D. M., Hinkle, D.
C., and Wei, C. S. (2002). Benzo[a]pyrene diol epoxide-deoxyguanosine
adducts are accurately bypassed by yeast DNA polymerase zeta in vitro.
Mutat. Res. 508, pp. 137-145.

Zhao, B., Wang, J., Geacintov, N. E., and Wang, Z. (2006). Poleta,
Polzeta and Rev1 together are required for G to T transversion mutations

117

188.

189.

190.

191.

192.

193.

194.

195.

induced by the (+)- and (-)-trans-anti-BPDE-N2-dG DNA adducts in yeast
cells. Nucleic Acids Res. 34, pp. 417-425.

McKee, R. H., and Lawrence, C. W. (1979). Genetic analysis of gamma-
ray mutagenesis in yeast. ll. Allele-specific control of mutagenesis.
Genetics 93, pp. 375-381.

McKee, R. H., and Lawrence, C. W. (1979). Genetic analysis of gamma-
ray mutagenesis in yeast. I. Reversion in radiation-sensitive strains.
Genetics 93, pp. 361-373.

Prakash, L. (1976). Effect of genes controlling radiation sensitivity on
chemically inducedmutations in Saccharomyces cerevisiae. Genetics 83,
pp. 285-301.

Lawrence, C. W., Nisson, P. E., and Christensen, R. B. (1985). UV and
chemical mutagenesis in rev7 mutants of yeast. Mol. Gen. Genet. 200, pp.
86-91.

Lemontt, J. F. (1972). Induction of forward mutations in mutationally
defective yeast. Mol. Gen. Genet. 119, pp. 27-42.

Eeken, J. C., Romeijn, R. J., de Jong, A. W., Pastink, A., and Lohman, P.
H. (2001). Isolation and genetic characterisation of the Drosophila
homologue of (SCE)REV3, encoding the catalytic subunit of DNA
polymerase zeta. Mutat. Res. 485, pp. 237-253.

Sakamoto, A., Lan, V. T., Hase, Y., Shikazono, N., Matsunaga, T., and
Tanaka, A. (2003). Disruption of the AtREV3 gene causes hypersensitivity
to ultraviolet B light and gamma-rays in Arabidopsis: implication of the
presence of a translesion synthesis mechanism in plants. Plant Cell 15,
pp. 2042-2057.

Kajiwara, K., J, O. W., Sakurai, T., Yamashita, S., Tanaka, M., Sato, M.,
Tagawa, M., Sugaya, E., Nakamura, K., Nakao, K., Katsuki, M., and
Kimura, M. (2001). Sez4 gene encoding an elongation subunit of DNA
polymerase zeta is required for normal embryogenesis. Genes Cells 6, pp.
99-106.

118

196.

197.

198.

199.

200.

201.

202.

203.

204.

Sonoda, E., Okada, T., Zhao, G. Y., Tateishi, S., Araki, K., Yamaizumi, M.,
Yagi, T., Verkaik, N. S., van Gent, D. C., Takata, M., and Takeda, S.
(2003). Multiple roles of Rev3, the catalytic subunit of polzeta in
maintaining genome stability in vertebrates. EMBO J. 22, pp. 3188-3197.

Gibbs, P. E., McGregor, W. G., Maher, V. M., Nisson, P., and Lawrence,
C. W. (1998). A human homolog of the Saccharomyces cerevisiae REV3
gene, which encodes the catalytic subunit of DNA polymerase zeta. Proc.
Natl. Acad. Sci. USA 95, pp. 6876-6880.

Morelli, C., Mungall, A. J., Negrini, M., Barbanti-Brodano, G., and Croce,
C. M. (1998). Alternative splicing, genomic structure, and fine
chromosome localization of REV3L. Cytogenet. Cell Genet. 83, pp. 18-20.

Bemark, M., Khamlichi, A. A., Davies, S. L., and Neuberger, M. S. (2000).
Disruption of mouse polymerase zeta (Rev3) leads to embryonic lethality
and impairs blastocyst development in vitro. Curr. Biol. 10, pp. 1213-1216.

Esposito, G., Godindagger, l., Klein, U., Yaspo, M. L., Cumano, A., and
Rajewsky, K. (2000). Disruption of the Rev3I—encoded catalytic subunit of
polymerase zeta in mice results in early embryonic lethality. Curr. Biol. 10,
pp. 1221-1224.

VIfittschieben, J., Shivji, M. K., Lalani, E., Jacobs, M. A., Marini, F.,
Gearhart, P. J., Rosewell, l., Stamp, G., and Wood, R. D. (2000).
Disruption of the developmentally regulated Rev3! gene causes embryonic
lethality. Curr. Biol. 10, pp. 1217-1220.

Li, Z., Zhang, H., McManus, T. P., McCormick, J. J., Lawrence, C. W., and
Maher, V. M. (2002). hREV3 is essential for error-prone translesion

synthesis past UV or benzo[a]pyrene diol epoxide-induced DNA lesions in
human fibroblasts. Mutat. Res. 510, pp. 71-80.

Diaz, M., Verkoczy, L. K., Flajnik, M. F., and Klinman, N. R. (2001).
Decreased frequency of somatic hypermutation and impaired affinity
maturation but intact germinal center formation in mice expressing
antisense RNA to DNA polymerase zeta. J. Immunol. 167, pp. 327-335.

Nakajima, S., Lan, L., Kanno, S., Takao, M., Yamamoto, K., Eker, A. P.,
and Yasui, A. (2004). UV light-induced DNA damage and tolerance for the

119

205.

206.

207.

208.

209.

210.

211.

212.

survival of nucleotide excision repair-deficient human cells. J. Biol. Chem.
279. pp. 46674-46677.

Diaz, M., Watson, N. B., Turkington, G., Verkoczy, L. K., Klinman, N. R.,
and McGregor, W. G. (2003). Decreased frequency and highly aberrant
spectrum of ultraviolet-induced mutations in the hprt gene of mouse
fibroblasts expressing antisense RNA to DNA polymerase zeta. Mol.
Cancer Res. 1, pp. 836-847.

Murakumo, Y., Roth, T., Ishii, H., Rasio, D., Numata, S., Croce, C. M., and
Fishel, R. (2000). A human REV7 homolog that interacts with the
polymerase zeta catalytic subunit hREV3 and the spindle assembly
checkpoint protein hMADZ. J. Biol. Chem. 275, pp. 4391-4397.

Aravind, L., and Koonin, E. V. (1998). The HORMA domain: a common
structural denominator in mitotic checkpoints, chromosome synapsis and
DNA repair. Trends Biochem. Sci. 23, pp. 284-286.

Pfleger, C. M., Salic, A., Lee, E., and Kirschner, M. W. (2001). Inhibition of
th1-APC by the MADZ-related protein MADZLZ: a novel mechanism for
regulating th1. Genes Dev. 15, pp. 1759-1764.

Chen, J., and Fang, G. (2001). MADZB is an inhibitor of the anaphase-
promoting complex. Genes Dev. 15, pp. 1765-1770.

Cheung, H. W., Chun, A. C., Wang, Q., Deng, W., Hu, L., Guan, X. Y.,
Nicholls, J. M., Ling, M. T., Chuan Wong, Y., Tsao, S. W., Jin, D. Y., and
Wang, X. (2006). Inactivation of human MADZB in nasopharyngeal
carcinoma cells leads to chemosensitization to DNA-damaging agents.
Cancer Res. 66, pp. 4357-4367.

Wrttschieben, J. P., Reshmi, S. C., Gollin, S. M., and Wood, R. D. (2006).
Loss of DNA polymerase zeta causes chromosomal instability in
mammalian cells. Cancer Res. 66, pp. 134-142.

Zan, H., Komori, A., Li, Z., Cerutti, A., Schaffer, A., Flajnik, M. F., Diaz, M.,
and Casali, P. (2001). The translesion DNA polymerase zeta plays a
major role in lg and bcl-6 somatic hypermutation. Immunity 14, pp. 643-
653.

120

213.

214.

215.

216.

217.

218.

219.

220.

221.

222.

Diaz, M., Flajnik, M. F., and Klinman, N. (2001). Evolution and the
molecular basis of somatic hypermutation of antigen receptor genes.
Philos. Trans. R. Soc. Land. B Biol. Sci. 356, pp. 67-72.

Di Noia, J. M., and Neuberger, M. S. (2007). Molecular mechanisms of
antibody somatic hypermutation. Annu. Rev. Biochem. 76, pp. 1-22.

Rattray, A. J., Shafer, B. K., McGill, C. B., and Strathern, J. N. (2002). The
roles of REV3 and RAD57 in doubIe-strand-break-repair-induced
mutagenesis of Saccharomyces cerevisiae. Genetics 162, pp. 1063-1077.

Hirano, Y., and Sugimoto, K. (2006). ATR homolog Mec1 controls
association of DNA polymerase zeta-Rev1 complex with regions near a
double-strand break. Curr. Biol. 16, pp. 586-590.

Van Sloun, P. P., Varlet, l., Sonneveld, E., Boei, J. J., Romeijn, R. J.,
Eeken, J. C., and De VIfind, N. (2002). Involvement of mouse Rev3 in
tolerance of endogenous and exogenous DNA damage. Mol. Cell. Biol.
22, pp. 2159-2169.

Larimer, F. W., Perry, J. R., and Hardigree, A. A. (1989). The REV1 gene
of Saccharomyces cerevisiae: isolation, sequence, and functional
analysis. J. Bacteriol. 171 , pp. 230-237.

Gibbs, P. B, Wang, X. 0., Li, Z., McManus, T. P., McGregor, W. G.,
Lawrence, C. W., and Maher, V. M. (2000). The function of the human
homolog of Saccharomyces cerevisiae REV1 is required for mutagenesis
induced by UV light. Proc. Natl. Acad. Sci. USA 97, pp. 4186-4191.

Gibbs, P. E., and Lawrence, C. W. (1995). Novel mutagenic properties of
abasic sites in Saccharomyces cerevisiae. J. Mol. Biol. 251, pp. 229-236.

Nair, D. T., Johnson, R. E., Prakash, L., Prakash, S., and Aggarwal, A. K.
(2005). Rev1 employs a novel mechanism of DNA synthesis using a
protein template. Science 309, pp. 2219-2222.

Otsuka, C., Sanadai, S., Hata, Y., Okuto, H., Noskov, V. N., Loakes, D.,
and Negishi, K. (2002). Difference between deoxyribose- and
tetrahydrofuran-type abasic sites in the in viva mutagenic responses in
yeast. Nucleic Acids Res. 30, pp. 5129-5135.

121

223.

224.

225.

226.

227.

228.

229.

230.

231.

Otsuka, C., Loakes, D., and Negishi, K. (2002). The role of deoxycytidyl
transferase activity of yeast Rev1 protein in the bypass of abasic sites.
Nucleic Acids Res. Suppl. 87-88.

Lin, W., Xin, H., Zhang, Y., Wu, X., Yuan, F., and Wang, Z. (1999). The
human REV1 gene codes for a DNA template-dependent dCMP
transferase. Nucleic Acids Res. 27, pp. 4468-4475.

Gan, G. N., Wrttschieben, J. P., Wrttschieben, B. O., and Wood, R. D.
(2008). DNA polymerase zeta (pol zeta) in higher eukaryotes. Cell Res.
18’ pp- 174-183.

Avkin, S., Adar, S., Blander, G., and Livneh, Z. (2002). Quantitative
measurement of translesion replication in human cells: evidence for
bypass of abasic sites by a replicative DNA polymerase. Proc. Natl. Acad.
Sci. USA 99, pp. 3764-3769.

Ross, A. L., Simpson, L. J., and Sale, J. E. (2005). Vertebrate DNA
damage tolerance requires the C-terminus but not BRCT or transferase
domains of REV1. Nucleic Acids Res. 33, pp. 1280-1289.

Jansen, J. G., Tsaalbi-Shtylik, A., Langerak, P., Calleja, F., Meijers, C. M.,
Jacobs, H., and de Wrnd, N. (2005). The BRCT domain of mammalian
Rev1 is involved in regulating DNA translesion synthesis. Nucleic Acids
Res. 33, pp. 356-365.

Masuda, Y., Takahashi, M., Tsunekuni, N., Minami, T., Sumii, M.,
Miyagawa, K., and Kamiya, K. (2001). Deoxycytidyl transferase activity of
the human REV1 protein is closely associated with the conserved
polymerase domain. J. Biol. Chem. 276, pp. 15051-15058.

Guo, C., Sonoda, E., Tang, T. S., Parker, J. L., Bielen, A. B., Takeda, S.,
Ulrich, H. D., and Friedberg, E. C. (2006). REV1 protein interacts with
PCNA: significance of the REV1 BRCT domain in vitro and in viva. Mal.
Cell 23, pp. 265-271.

Guo, C., Tang, T. S., Bienko, M., Parker, J. L., Bielen, A. B., Sonoda, E.,
Takeda, S., Ulrich, H. D., Dikic, |., and Friedberg, E. C. (2006). Ubiquitin-
binding motifs in REV1 protein are required for its role in the tolerance of
DNA damage. Mol. Cell. Biol. 26, pp. 8892-8900.

122

232.

233.

234.

235.

236.

237.

238.

239.

240.

Guo, C., Fischhaber, P. L., Luk-Paszyc, M. J., Masuda, Y., Zhou, J.,
Kamiya, K., Kisker, C., and Friedberg, E. C. (2003). Mouse Rev1 protein
interacts with multiple DNA polymerases involved in translesion DNA
synthesis. EMBO J. 22, pp. 6621-6630.

Ohashi, E., Murakumo, Y., Kanjo, N., Akagi, J., Masutani, C., Hanaoka, F.,
and Ohmori, H. (2004). Interaction of hREV1 with three human Y-family
DNA polymerases. Genes Cells 9, pp. 523-531.

McDonald, J. R, Levine, A. S., and Woodgate, R. (1997). The
Saccharomyces cerevisiae RAD30 gene, a homologue of Escherichia coli
dinB and umuC, is DNA damage inducible and functions in a novel error-
free postreplication repair mechanism. Genetics 147, pp. 1557-1568.

Johnson, R. E., Prakash, S., and Prakash, L. (1999). Efficient bypass of a
thymine-thymine dimer by yeast DNA polymerase, Poleta. Science 283,
pp. 1001-1004.

Lehmann, A. R., Kirk-Bell, S., Arlett, C. F., Paterson, M. C., Lohman, P.
H., de Weerd-Kastelein, E. A., and Bootsma, D. (1975). Xeroderma
pigmentosum cells with normal levels of excision repair have a defect in
DNA synthesis after UV-irradiation. Proc. Natl. Acad. Sci. USA 72, pp.
219-223.

Johnson, R. E., Kondratick, C. M., Prakash, S., and Prakash, L. (1999).
hRAD30 mutations in the variant form of xeroderma pigmentosum.
Science 285, pp. 263-265.

Masutani, C., Kusumoto, R., Yamada, A., Dohmae, N., Yokoi, M., Yuasa,
M., Araki, M., lwai, S., Takio, K., and Hanaoka, F. (1999). The XPV
(xeroderma pigmentosum variant) gene encodes human DNA polymerase
eta. Nature 399, pp. 700-704.

Johnson, R. E., Washington, M. T., Prakash, S., and Prakash, L. (2000).
Fidelity of human DNA polymerase eta. J. Biol. Chem. 275, pp. 7447-
7450.

Washington, M. T., Helquist, S. A., Kool, E. T., Prakash, L., and Prakash,
S. (2003). Requirement of Watson-Crick hydrogen bonding for DNA
synthesis by yeast DNA polymerase eta. Mal. Cell. Biol. 23, pp. 5107-
5112.

123

241.

242.

243.

244.

245.

246.

247.

248.

249.

250.

Masutani, C., Kusumoto, R., lwai, S., and Hanaoka, F. (2000).
Mechanisms of accurate translesion synthesis by human DNA polymerase
eta. EMBO J. 19, pp. 3100-3109.

McCulloch, S. D., Kokoska, R. J., Masutani, C., lwai, S., Hanaoka, F., and
Kunkel, T. A. (2004). Preferential cis-syn thymine dimer bypass by DNA
polymerase eta occurs with biased fidelity. Nature 428, pp. 97-100.

Prakash, S., Johnson, R. E., and Prakash, L. (2004). Eukaryotic
Translesion Synthesis DNA Polymerases: Specificity of Structure and
Function. Annu. Rev. Biochem.

Kusumoto, R., Masutani, C., lwai, S., and Hanaoka, F. (2002). Translesion
synthesis by human DNA polymerase eta across thymine glycol lesions.
Biochemistry 41, pp. 6090-6099.

Zhang, Y., Yuan, F., Wu, X., Rechkoblit, 0., Taylor, J. S., Geacintov, N.
E., and Wang, Z. (2000). Error-prone lesion bypass by human DNA
polymerase eta. Nucleic Acids Res. 28, pp. 4717-4724.

Haracska, L., Yu, S. L., Johnson, R. E., Prakash, L., and Prakash, S.
(2000). Efficient and accurate replication in the presence of 7,8-dihydro-8-
oxoguanine by DNA polymerase eta. Nat. Genet. 25, pp. 458-461.

Chiapperino, D., Kroth, H., Kramarczuk, I. H., Sayer, J. M., Masutani, C.,
Hanaoka, F., Jerina, D. M., and Cheh, A. M. (2002). Preferential
misincorporation of purine nucleotides by human DNA polymerase eta
opposite benzo[a]pyrene 7,8-diol 9,10-epoxide deoxyguanosine adducts.
J. Biol. Chem. 277, pp. 11765-11771.

Matsuda, T., Bebenek, K., Masutani, C., Hanaoka, F., and Kunkel, T. A.
(2000). Low fidelity DNA synthesis by human DNA polymerase-eta.
Nature 404, pp. 1011-1013.

Washington, M. T., Johnson, R. E., Prakash, S., and Prakash, L. (1999).
Fidelity and processivity of Saccharomyces cerevisiae DNA polymerase
eta. J. Biol. Chem. 274, pp. 36835-36838.

Vaisman, A., Lehmann, A. R., and Woodgate, R. (2004). DNA
polymerases n and L. Adv. Protein Chem. 69, pp. 205-228.

124

251.

252.

253.

254.

255.

256.

257.

258.

Zeng, X., Winter, D. B., Kasmer, C., Kraemer, K. H., Lehmann, A. R., and
Gearhart, P. J. (2001). DNA polymerase eta is an A-T mutator in somatic
hypermutation of immunoglobulin variable genes. Nat. Immunol. 2, pp.
537-541.

Delbos, F., De Smet, A., Faili, A., Aoufouchi, S., Weill, J. C., and
Reynaud, C. A. (2005). Contribution of DNA polymerase eta to
immunoglobulin gene hypermutation in the mouse. J. Exp. Med. 201, pp.
1191-1196.

Mcllwraith, M. J., Vaisman, A., Liu, Y., Fanning, E., Woodgate, R., and
West, S. C. (2005). Human DNA polymerase eta promotes DNA synthesis
from strand invasion intermediates of homologous recombination. Mol.
Cell 20, pp. 783-792.

Kawamoto, T., Araki, K., Sonoda, E., Yamashita, Y. M., Harada, K.,
Kikuchi, K., Masutani, C., Hanaoka, F., Nozaki, K., Hashimoto, N., and
Takeda, S. (2005). Dual roles for DNA polymerase eta in homologous
DNA recombination and translesion DNA synthesis. Mol. Cell 20, pp. 793-
799.

McDonald, J. P., Rapic-Otrin, V., Epstein, J. A., Broughton, B. C., Wang,
X., Lehmann, A. R., Wolgemuth, D. J., and Woodgate, R. (1999). Novel
human and mouse homologs of Saccharomyces cerevisiae DNA
polymerase eta. Genomics 60, pp. 20-30.

Tissier, A., McDonald, J. R, Frank, E. G., and Woodgate, R. (2000).
poliota, a remarkably error-prone human DNA polymerase. Genes Dev.
14, pp. 1642-1650.

Ishikawa, T., Uematsu, N., Mizukoshi, T., lwai, S., Iwasaki, H., Masutani,
C., Hanaoka, F., Ueda, R., Ohmori, H., and Todo, T. (2001). Mutagenic
and nonmutagenic bypass of DNA lesions by Drosophila DNA
polymerases dpoleta and dpoliota. J. Biol. Chem. 276, pp. 15155-15163.

Vaisman, A., Frank, E. G., McDonald, J. P., Tissier, A., and Woodgate, R.
(2002). poliota-dependent lesion bypass in vitro. Mutat. Res. 510, pp. 9-
22.

125

259.

260.

261.

262.

263.

264.

265.

266.

Vaisman, A., and Woodgate, R. (2001). Unique misinsertion specificity of
poliota may decrease the mutagenic potential of deaminated cytosines.
EMBO J. 20, pp. 6520-6529.

Vaisman, A., Tissier, A., Frank, E. G., Goodman, M. F., and Woodgate, R.
(2001). Human DNA polymerase iota promiscuous mismatch extension. J.
Biol. Chem. 276, pp. 30615-30622.

Nair, D. T., Johnson, R. E., Prakash, S., Prakash, L., and Aggarwal, A. K.
(2004). Replication by human DNA polymerase-iota occurs by Hoogsteen
base-pairing. Nature 430, pp. 377-380.

Tissier, A., Frank, E. G., McDonald, J. P., lwai, S., Hanaoka, F., and
Woodgate, R. (2000). Misinsertion and bypass of thymine-thymine dimers
by human DNA polymerase iota. EMBO J. 19, pp. 5259-5266.

Vaisman, A., Frank, E. G., lwai, S., Ohashi, E., Ohmori, H., Hanaoka, F.,
and Woodgate, R. (2003). Sequence context-dependent replication of
DNA templates containing UV-induced lesions by human DNA polymerase
iota. DNA Repair (Amst) 2, pp. 991-1006.

Dumstorf, C. A., Clark, A. B., Lin, 0., Kissling, G. E., Yuan, T.,
Kucherlapati, R., McGregor, W. G., and Kunkel, T. A. (2006). Participation
of mouse DNA polymerase iota in strand-biased mutagenic bypass of UV
photoproducts and suppression of skin cancer. Proc. Natl. Acad. Sci. USA
103, pp. 18083-18088.

Ohkumo, T., Kondo, Y., Yokoi, M., Tsukamoto, T., Yamada, A., Sugimoto,
T., Kanao, R., Higashi, Y., Kondoh, H., Tatematsu, M., Masutani, C., and
Hanaoka, F. (2006). UV-B radiation induces epithelial tumors in mice
lacking DNA polymerase eta and mesenchymal tumors in mice deficient
for DNA polymerase iota. Mol. Cell. Biol. 26, pp. 7696-7706.

Wang, Y., Woodgate, R., McManus, T. P., Mead, S., McCormick, J. J.,
and Maher, V. M. (2007). Evidence that in xeroderma pigmentosum
variant cells, which lack DNA polymerase h, DNA polymerase i causes the
very high frequency and unique spectrum of UV-induced mutations.
Cancer Res. 67, pp.

126

267.

268.

269.

270.

271.

272.

273.

274.

275.

Zhang, Y., Yuan, F., Wu, X., Taylor, J. S., and Wang, Z. (2001). Response
of human DNA polymerase iota to DNA lesions. Nucleic Acids Res. 29,
pp. 928-935.

Bebenek, K., Tissier, A., Frank, E. G., McDonald, J. P., Prasad, R.,
Wilson, S. H., Woodgate, R., and Kunkel, T. A. (2001). 5'-Deoxyribose
phosphate lyase activity of human DNA polymerase iota in vitro. Science
291, pp. 2156-2159.

Prasad, R., Bebenek, K., Hou, E., Shock, D. D., Beard, W. A., Woodgate,
R., Kunkel, T. A., and Wilson, S. H. (2003). Localization of the
deoxyribose phosphate lyase active site in human DNA polymerase iota
by controlled proteolysis. J. Biol. Chem. 278, pp. 29649-29654.

Ogi, T., Kato, T., Jr., Kato, T., and Ohmori, H. (1999). Mutation
enhancement by DINB1, a mammalian homologue of the Escherichia coli
mutagenesis protein dinB. Genes Cells 4, pp. 607-618.

Gerlach, V. L., Aravind, L., Gotway, G., Schultz, R. A., Koonin, E. V., and
Friedberg, E. C. (1999). Human and mouse homologs of Escherichia coli
DinB (DNA polymerase IV), members of the UmuC/DinB superfamily.
Proc. Natl. Acad. Sci. USA 96, pp. 11922-11927.

Johnson, R. E., Prakash, S., and Prakash, L. (2000). The human DlNBi
gene encodes the DNA polymerase Poltheta. Proc. Natl. Acad. Sci. USA
97, pp. 3838-3843.

Gerlach, V. L., Feaver, W. J., Fischhaber, P. L., and Friedberg, E. C.
(2001). Purification and characterization of pol kappa, a DNA polymerase
encoded by the human DlNB1 gene. J. Biol. Chem. 276, pp. 92-98.

Ohashi, E., Bebenek, K., Matsuda, T., Feaver, W. J., Gerlach, V. L.,
Friedberg, E. C., Ohmori, H., and Kunkel, T. A. (2000). Fidelity and
processivity of DNA synthesis by DNA polymerase kappa, the product of
the human DINB1 gene. J. Biol. Chem. 275, pp. 39678-39684.

Zhang, Y., Yuan, F., Xin, H., Wu, X., Rajpal, D. K., Yang, 0., and Wang, Z.
(2000). Human DNA polymerase kappa synthesizes DNA with
extraordinarily low fidelity. Nucleic Acids Res. 28, pp. 4147-4156.

127

276.

277.

278.

279.

280.

281.

282.

283.

284.

Bavoux, C., Hoffmann, J. S., and Cazaux, C. (2005). Adaptation to DNA
damage and stimulation of genetic instability: the double-edged sword
mammalian DNA polymerase kappa. Biochimie 87, pp. 637-646.

Uljon, S. M, Johnson, R. E., Edwards, T. A., Prakash, S., Prakash, L., and
Agganrval, A. K. (2004). Crystal structure of the catalytic core of human
DNA polymerase kappa. Structure (Camb) 12, pp. 1395-1404.

Washington, M. T., Johnson, R. E., Prakash, L., and Prakash, S. (2002).
Human DlNB1-encoded DNA polymerase kappa is a promiscuous
extender of mispaired primer termini. Proc. Natl. Acad. Sci. USA 99, pp.
1910-1914.

Bergoglio, V., Bavoux, C., Verbiest, V., Hoffmann, J. S., and Cazaux, C.
(2002). Localisation of human DNA polymerase kappa to replication foci.
J. Cell. Sci. 115. pp. 4413-4418.

Zhang, Y., Yuan, F., Wu, X., Wang, M., Rechkoblit, 0., Taylor, J. S.,
Geacintov, N. E., and Wang, Z. (2000). Error-free and error-prone lesion
bypass by human DNA polymerase kappa in vitro. Nucleic Acids Res. 28,
pp. 4138-4146.

Ohashi, E., Ogi, T., Kusumoto, R, lwai, S., Masutani, C., Hanaoka, F.,
and Ohmori, H. (2000). Error-prone bypass of certain DNA lesions by the
human DNA polymerase kappa. Genes Dev. 14, pp. 1589-1594.

Fischhaber, P. L., Gerlach, V. L., Feaver, W. J., Hatahet, 2., Wallace, 8.
S., and Friedberg, E. C. (2002). Human DNA polymerase kappa bypasses
and extends beyond thymine glycols during translesion synthesis in vitro,
preferentially incorporating correct nucleotides. J. Biol. Chem. 277, pp.
37604-3761 1 .

Suzuki, M, Ohashi, E, Kolbanovskiy, A., Geacintov, N. E., Grollman, A.
P., Ohmori, H., and Shibutani, S. (2002). Translesion synthesis by human
DNA polymerase kappa on a DNA template containing a single
stereoisomer of dG-(+)- or dG-(-)-anti-N(2)-BPDE (7,8-dihydroxy-anti-9,10-
epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene). Biochemistry 41, pp. 6100-
6106.

Zhang, Y., Wu, X., Guo, D., Rechkoblit, O., and Wang, Z. (2002). Activities
of human DNA polymerase kappa in response to the major

128

285.

286.

287.

288.

289.

290.

291.

292.

benzo[a]pyrene DNA adduct: error-free lesion bypass and extension
synthesis from opposite the lesion. DNA Repair (Amst) 1, pp. 559-569.

Haracska, L., Prakash, L., and Prakash, S. (2002). Role of human DNA
polymerase kappa as an extender in translesion synthesis. Proc. Natl.
Acad. Sci. USA 99. pp. 16000-16005.

Zhang, Y., Wu, X., Guo, D., Rechkoblit, O., Geacintov, N. E., and Wang,
Z. (2002). Two-step error-prone bypass of the (+)- and (-)-trans-anti-
BPDE-N2-dG adducts by human DNA polymerases eta and kappa. Mutat.
Res. 510, pp. 23-35.

Ogi, T., Shinkai, Y., Tanaka, K., and Ohmori, H. (2002). Polkappa protects
mammalian cells against the lethal and mutagenic effects of
benzo[a]pyrene. Proc. Natl. Acad. Sci. USA 99, pp. 15548-15553.

Avkin, S., Goldsmith, M., Velasco-Miguel, S., Geacintov, N., Friedberg, E.
C., and Livneh, Z. (2004). Quantitative analysis of translesion DNA
synthesis across a benzo[a]pyrene-guanine adduct in mammalian cells:
the role of DNA polymerase kappa. J. Biol. Chem. 279, pp. 53298-53305.

Ogi, T., Mimura, J., Hikida, M., Fujimoto, H., Fujii-Kuriyama, Y., and
Ohmori, H. (2001). Expression of human and mouse genes encoding
polkappa: testis-specific developmental regulation and AhR-dependent
inducible transcription. Genes Cells 6, pp. 943-953.

O-Wang, J., Kawamura, K., Tada, Y., Ohmori, H., Kimura, H., Sakiyama,
S., and Tagawa, M. (2001). DNA polymerase kappa, implicated in
spontaneous and DNA damage-induced mutagenesis, is overexpressed in
lung cancer. Cancer Res. 61, pp. 5366-5369.

Ohmori, H., Ohashi, E., and Ogi, T. (2004). Mammalian Pol kappa:
regulation of its expression and lesion substrates. Adv. Protein Chem. 69,
pp. 265-278.

Suzuki, N., ltoh, S., Poon, K., Masutani, C., Hanaoka, F., Ohmori, H.,
Yoshizawa, I., and Shibutani, S. (2004). Translesion synthesis past
estrogen-derived DNA adducts by human DNA polymerases eta and
kappa. Biochemistry 43, pp. 6304-6311.

129

293.

294.

295.

296.

297.

298.

299.

300.

301.

Velasco-Miguel, 8., Richardson, J. A., Gerlach, V. L., Lai, W. C., Gao, T.,
Russell, L. D., Hladik, C. L., White, C. L., and Friedberg, E. C. (2003).
Constitutive and regulated expression of the mouse Dinb (Polkappa) gene
encoding DNA polymerase kappa. DNA Repair (Amst) 2, pp. 91-106.

Ogi, T., and Lehmann, A. R. (2006). The Y-family DNA polymerase kappa
(pol kappa) functions in mammalian nucleotide-excision repair. Nat. Cell
Biol. 8, pp. 640-642.

Prakash, L., and Prakash, S. (1977). Isolation and characterization of
MMS-sensitive mutants of Saccharomyces cerevisiae. Genetics 86, pp.
33-55.

Broomfield, 8., Chow, B. L., and Xiao, W. (1998). MMSZ, encoding a
ubiquitin-conjugating-enzyme-Iike protein, is a member of the yeast error-
free postreplication repair pathway. Proc. Natl. Acad. Sci. USA 95, pp.
5678-5683.

Xiao, W., Chow, B. L., Fontanie, T., Ma, L., Bacchetti, 8., Hryciw, T., and
Broomfield, S. (1999). Genetic interactions between error-prone and error-

free postreplication repair pathways in Saccharomyces cerevisiae. Mutat.
Res. 435, pp. 1-11.

Hofmann, R. M., and Pickart, C. M. (1999). Noncanonical MMS2-encoded
ubiquitin-conjugating enzyme functions in assembly of novel polyubiquitin
chains for DNA repair. Cell 96, pp. 645-653.

Brown, M., Zhu, Y., Hemmingsen, S. M., and Xiao, W. (2002). Structural
and functional conservation of error-free DNA postreplication repair in
Schizosaccharomyces pombe. DNA Repair (Amst) 1, pp. 869-880.

Brusky, J., Zhu, Y., and Xiao, W. (2000). UBC13, a DNA-damage-
inducible gene, is a member of the error-free postreplication repair
pathway in Saccharomyces cerevisiae. Curr. Genet. 37, pp. 168-174.

Johnson, R. E., Henderson, S. T., Petes, T. D., Prakash, 8., Bankmann,
M., and Prakash, L. (1992). Saccharomyces cerevisiae RADS—encoded
DNA repair protein contains DNA helicase and zinc-binding sequence
motifs and affects the stability of simple repetitive sequences in the
genome. Mal. Cell. Biol. 12, pp. 3807-3818.

130

302.

303.

304.

305.

Johnson, R. E., Prakash, 8., and Prakash, L. (1994). Yeast DNA repair
protein RAD5 that promotes instability of simple repetitive sequences is a
DNA-dependent ATPase. J. Biol. Chem. 269, pp. 28259-28262.

Ulrich, H. D., and Jentsch, S. (2000). Two RING finger proteins mediate
cooperation between ubiquitin-conjugating enzymes in DNA repair. EMBO
J. 19, pp. 3388-3397.

Blastyak, A., Pinter, L., Unk, l., Prakash, L., Prakash, 8., and Haracska, L.
(2007). Yeast Rad5 protein required for postreplication repair has a DNA
helicase activity specific for replication fork regression. Mol. Cell 28, pp.
167-175.

Li, Z., Xiao, W., McCormick, J. J., and Maher, V. M. (2002). Identification
of a protein essential for a major pathway used by human cells to avoid
UV- induced DNA damage. Proc. Natl. Acad. Sci. USA 99, pp. 4459-4464.

131

CHAPTER II

hRev7, putative subunit of hPolC, plays a critical role in survival, induction
of mutations, and progression through S-phase, of UV(254nm)-irradiated

human fibroblasts

Kristin McNally, Jessica A. Neal, Terrence P. McManus, J. Justin

McCormick, and Veronica M. Maher*

Carcinogenesis Laboratory, Cell and Molecular Biology Program, Department of
Microbiology & Molecular Genetics, and Department of Biochemistry &

Molecular Biology, Michigan State University, East Lansing, MI 48824-1302

*Carrespanding author at. Veronica M. Maher, Carcinogenesis Laboratory, Food
Safety and Toxicology Bldg., Michigan State University, East Lansing, MI 48824-
1302, USA; Ph: (517) 353-7785; Fax (517) 353-9004; E-mail address:

maher@msu.edu

132

ABSTRACT:

Translesion synthesis (TLS) refers to mechanisms by which specialized DNA
polymerases incorporate nucleotides opposite fork-blocking lesions and extend
replication until standard replicative polymerases take over. The first eukaryotic
TLS polymerase discovered, S. cerevisiae PolC, consists of catalytic subunit
Rev3 and non-catalytic subunit Rev7. Human homologs of these two proteins

have been identified. Studies by Lawrence, Maher, and colleagues comparing

UV(254nm)-irradiated human fibroblast cell strains expressing high levels of

hRev3 antisense to their normal parental strains demonstrated that there was no
difference in cell survival, but that the frequency of UV-induced mutations in the
derivative strains was 10-fold lower than that of the parental strains, indicating
that hRev3 plays a critical role in such mutagenesis. To examine the role of
hRev7 in TLS, we generated human fibroblasts expressing hRev7 siRNA,
identified two derivative cell strains with significantly reduced levels of hRev7,
and compared them to their parental strain and a vector control for cell survival,
induction of mutations, and ability to traverse the cell cycle following exposure to
UV radiation. Cells with reduced hRev7 were ~2-times more sensitive to UV-
induced cytotoxicity than the controls, indicating that unlike hRev3, hRev7 plays
a protective role for cells exposed to UV radiation. When these cell strains were
assayed for the frequency of mutations induced by UV in their HPRT gene, cell
stains with reduced hRev7 were 5-times less sensitive to UV-induced

mutagenesis than control strains. In addition, when these four strains were

133

synchronized at the G1/S border, released from the block, UV-irradiated, and
allowed to traverse the cell cycle, the rate of progression through S-phase of the
cell strains with reduced hRev7 was significantly slower than that of the control
strains. These data strongly support the hypothesis that hRev7 is required for

TLS past UV-photoproducts, and together with hRev3, comprise hPoIC.

134

INTRODUCTION:

Human cells are continually exposed to endogenous and exogenous DNA
damaging agents, many of which create fork-blocking lesions. If DNA replication
past such lesions cannot take place, this can lead to cell death, nevertheless
replication past such lesions can result in mutations. Because mutations play a
crucial causal role in the development of cancer, it is important to examine

processes that produce them.

Human cells have efficient, error-free repair pathways for excising DNA fork-
blocking lesions from either strand of their DNA. They also possess cell cycle
checkpoints [1], some of which, when activated, provide additional time for
excision repair to occur before the replicative polymerases encounter fork-
blocking lesions, such as UV-induced pyrimidine dimers. In spite of these
protective processes, replication forks still encounter lesions. Cells have evolved
damage tolerance mechanisms to cope with such lesions, viz., translesion
synthesis and damage avoidance pathways. Such methods of dealing with fork-
blocking damage have been, and continue to be actively examined. Overviews

summarizing in detail such areas of research can be found in reference [2].

Translesion synthesis in both prokaryotes and eukaryotes involves specialized

DNA polymerases capable of incorporating nucleotides directly across from fork-

blocking DNA lesions. This insertion step can be error-free or error-prone,

135

depending upon 1) the type of DNA lesion encountered, 2) the specialized
polymerases involved, and 3) the sequence context surrounding the site of the
damage. Insertion of a nucleotide or nucleotides by one or other such
polymerases is followed by extension, i.e., the addition of nucleotides beyond the
site of the blocking lesion. This latter step also involves TLS DNA polymerases.
Such extension beyond the damage is necessary if the high fidelity replicative
DNA polymerases are to resume their function. Thus, TLS is a two-step process
whereby specialized DNA polymerases, with relaxed fidelity, incorporate and/or
extend nucleotides at sites of fork-blocking DNA damage, allowing DNA

replication to continue, but often introducing mutations.

Reports and summaries of the discovery of many translesion synthesis
polymerases, first in S. cerevisiae, and later in mammalian cells, can be found in
the cited references (see for example, [3-6]). However, many aspects still remain
to be clarified. PolC was found using S. cerevisiae cells whose specific mutated
phenotypes could not be reverted to wild type by exposure to mutagenic agents.
Genes that complemented the deficiencies in such strains of S. cerevisiae, i.e.,
allowed them to revert, were identified and subsequently shown to code for
proteins that allow replication past fork-blocking DNA damage [7, 8]. For
example, the yeast Rev3 protein was found to exhibit polymerase activity in
primer extension assays in vitro. The addition of yeast Rev7 to such assays
enhanced the polymerase activity of Rev3 over 20-fold. Together, Rev3 and

Rev7 were recognized as constituting yeast PolC [9].

136

Genes coding for the human homologs of yeast Rev3 [10-13] and Rev7 [14] were
subsequently identified. By using antisense directed against hRev3 mRNA,
Lawrence, Maher, and their colleagues [10, 15] demonstrated that hRev3, the
putative catalytic subunit of hPolC, is critically involved in generating UV-induced
mutations in diploid human fibroblasts. These results indicate that hRev3 is
essential for a mutagenic process involving DNA lesions that interfere with
replication, just as yeast Rev3 is. The hRev3 protein of human cells, a predicted
353 kDa molecule [10], has not yet been isolated, but the non-catalytic subunit,

hRev7, a much smaller molecule, has been isolated [14].

The present study was carried out to test the hypothesis that hRev7, the putative
noncatalytic subunit of hPolC, is also involved in human cell mutagenesis. For
such a study, an approach similar to that used for investigating the role of hRev3
was employed, but instead of using antisense RNA to block expression of the
target protein, siRNA against hRev7 was used to reduce the level of this protein
in human fibroblasts. The fact that antibodies capable of detecting very low levels
of hRev7 protein were available allowed us to identify independent cell strains in
which the level of hRev7 protein had been greatly reduced by siRNA. Comparing
the results obtained using these cell strains with those obtained using their
parental human fibroblasts and a vector control strain allowed us to demonstrate
that hRev7, the noncatalytic subunit of hPolC, plays a role in the survival of UV-

irradiated human cells, and has a significant role in UV-induoed mutagenesis.

137

Using these human cell strains, we also demonstrated that reduction in the

expression of hRev7 impedes the cells’ ability to progress through S-phase.

138

MATERIALS AND METHODS:

Cell culture— Cells were grown in Eagle’s minimum essential medium,
supplemented with 0.2 mM L-aspartic acid, 0.2 mM L-serine, 1 mM sodium
pyruvate, 10% supplemented calf serum (HyClone), 100 units/ml penicillin, 100

uglml streptomycin, 1 jig/ml hydrocortisone and 1 jig/ml tetracycline.

Cell strains-- The parental human cell strain used for these studies, designated
MSU-1.2.9N.58, was derived from the cell strain MSU-1.2, a spontaneous
derivative of the infinite life span cell strain MSU-1.1, whose origin from the
foreskin-derived from a normal neonate and subsequent acquisition of an
unlimited life span in culture has been described [16]. MSU-1.2 cells are near-
diploid, chromosomally-stable, and grow vigorously as a result of expressing their

endogenous gene for platelet-derived growth factor.

Derivation of cell strains with reduced hRev7-- Oligonucleotides designed to
target hRev7 mRNA were annealed to a complementary oligonucleotide
according to the manufacturer’s protocol (Ambion). Using T4 DNA ligase (New
England Biolabs), annealed-oligonucleotides were ligated into the p8ilencer3.1
vector (Ambion), which includes the gene coding for puromycin resistance, and
purified. The parental M8U-1.2.9N.58 cells were transfected with such siRNA
vectors, using Lipofectamine (lnvitrogen) according to the manufacturer’s

protocol, and stable transfectants were selected and maintained in medium

139

containing 1 jig/ml puromycin.

Preparation of nuclear protein extracts and Western blot analysis-
Subconfluent monolayers of cells were washed with ice-cold phosphate buffered
saline (PBS), scraped from the 150-mm-diameter dishes in 1 ml of lysis buffer A
(10 mM HEPES, 10 mM KCI, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM DTT, 0.5 mM
PMSF), and incubated on ice for 15 min. 10% NP-40 (62 ul) was added to each
sample of Iysed cells, and the samples were vortexed for 10 sec, and centrifuged
for 30 sec at 10,000 RPM, 4°C. The supernatant containing cytoplasmic proteins
was removed, and the nuclear pellet was washed once in 1 ml of buffer A
containing 10% NP-40. Nuclear proteins were extracted by disruption of the
nuclei in 40 pl of lysis buffer C (20 mM HEPES, 0.4 M NaCl, 1 mM EDTA, 1 mM
EGTA, 1 mM DTT, 1 mM sodium orthovanadate, 1 mM PMSF) and incubated on
ice for 15 min with vortexing every 5 min. Nuclear extracts were centrifuged at for
5 min at 16,000 RPM, 4°C. The supernatants, which contained the nuclear
proteins, were saved. Protein was quantified using the Bradford method (Pierce).
Protein Iysates were subjected to gel electrophoresis using 14% SDS-
polyacrylamide, transferred to a PDVF Immobilon membrane (Millipore), and
probed with a 1:600 dilution of a custom-made (Bethyl) rabbit polyclonal antibody
raised against the C-terminal 19 amino acids of the human Rev7 protein. The
membrane was probed with a 1:7500 dilution of goat anti-rabbit secondary
antibody (Sigma) and visualized using SuperSignal chemiluminescent detection

reagent (Pierce). Equal protein loading was confirmed by probing with a 1:10,000

140

dilution of a rabbit Ku80 antibody (Santa Cruz) and a 1:10,000 dilution of anti-

rabbit secondary antibody (Santa Cruz).

Assay of UV cytotoxicity— The cytotoxic effect of UV(254nm) radiation was

determined by assaying the colony-forrning ability of the cells as described [17].
Briefly, cells in exponential growth were detached from the dishes with trypsin,
plated at cloning densities (100-600 cells per 100-mm-diameter dish), and
allowed 12 h for attachment. The cells were rinsed twice with PBS, the excess

PBS was removed, and the cells were irradiated with the designated doses of

UV(254nm) as described [17]. Immediately after irradiation, the cells were given

fresh culture medium. The culture medium was renewed 24 h after irradiation
and again after 7 days. After 14 days, the resulting clones were stained with
crystal violet. Cell survival was determined by comparing the cloning efficiency of
the irradiated cells with that of the sham-irradiated control cells. Cell survival at
each dose was expressed as a percent of the cloning efficiency of the

unirradiated control cells for each cell strain.

Assay for frequency of UV-induced mutations-- The mutagenic effect of UV
radiation in each strain was determined from the freqUency of cells that lost

expression of the HPRT gene and, therefore, were resistant to 6-thioguanine. As

described [17], sufficient sets of cells plated at densities of 0.5-1.5 x 106 cells per

150-mm-diameter dish, were used, to ensure at least 1 x 106 surviving target

141

cells per dose. Cells were allowed 12 h for attachment, then rinsed twice with
PBS, UV-irradiated at the designated doses, and immediately covered with fresh
culture medium. The culture medium was renewed 24 h after irradiation, and the

cells were allowed to replicate for 4 days. Cells were then detached using trypsin,

pooled, plated at densities of 0.5-1.0 x 106 cells per 150-mm-diameter dish, and

allowed to grow exponentially for 4 additional days in order to deplete the pre-

existing wiId-type HPRT protein. Cells were then detached using trypsin, and

plated at a density of 500 cells/cm2 in medium containing 40 [M 6-thioguanine

(TG) to select for cells lacking functional HPRT protein. At the same time, a
portion of cells from each population was plated in non-selective medium at a
density of 100 cells per 100-mm-diameter dish to assay the colony-forming ability
of the cells at the time of selection. The medium on these cells was renewed
after seven days. After 14 days, the colonies that had formed were stained with
crystal violet, and the frequency of 6-TG-resistant colonies was calculated using
the cloning efficiency of the cells at the time of selection. The induced
frequencies for each cell strain were calculated by subtracting the background
frequencies in the sham-irradiated control populations that accompanied each

experiment.

Hypoxanthine phosphoribosyltransferase (HPRT) mutation spectrum
analysis-- HPRT-defective colonies were obtained essentially as described
above for the UV-induced mutation frequency protocol, except that populations

were kept independent to avoid sibling mutations. The TG-resistant mutant

142

clones were isolated by treating with trypsin, and each independent mutant was
subjected to reverse transcription, followed by two rounds of PCR to amplify the
HPRT coding region. PCR products were purified (Qiagen) and sequenced at the
MSU macromolecular structure facility to determine the specific mutation in the
coding region of HPRT. Only base substitutions at adjacent pyrimidines that

resulted in an amino acid change were considered UV-induced mutations.

Cell synchronization and flow cytometry analysis- Each cell strain was

plated at a density of 0.2 x 106 cells per 100-mm-diameter dish and allowed 12 h

for attachment. Cells were then re-fed with complete culture medium containing
Iovastatin at a final concentration of 60 [M to synchronize the cells in early G1
phase. After 12 additional hours of incubation, the medium containing Iovastatin
was removed, the cells were washed twice with PBS, and culture medium
containing aphidicolin at a final concentration of 2 jig/ml and mevalonic acid at a
final concentration of 6 mM was added to the dishes for 12 h to synchronize the
cells at the G1/S border. The cells were released from the aphidicolin/mevalonic
acid block by rinsing twice with PBS, and immediately irradiated with the
designated doses of UV as described [17]. At the designated times past-
irradiation, cells were detached using trypsin, fixed in 80% ethanol, and stained
with a propidium iodide solution (PBS, 1 mg/ml propidium iodide, 10% Triton X-
100, 0.5 mM EDTA, 10 mg/ml RNase A) for cell cycle analysis by flow cytometry.

Asynchronously-growing cells were assayed in parallel experiments.

143

Statistical methods -- To compare the slopes of the curves in Fig. 2A and 2B, a
“regression model was used. The slopes indicate the effect of UV on the survival
(A) of the cells’ colony-forming ability and on their frequency of UV-induced
mutants (B). The data were taken from a series of experiments that were treated

as blocks in the regression analysis.

For comparing the types of mutations induced (see Table 1), the sparse
categories, i.e., C-)G and T-)G, were collapsed for the chi-square analysis.
They showed no statistically significant difference between the clones and the
comparison group (P-value = 0.60). Other categories were collapsed to reduce

the number of table cells analyzed with small expected frequencies.

144

RESULTS:

Efficient reduction of hRev7 protein using siRNA- The parental cell strain,
designated M8U-1.2.9N.58, was transfected with a vector expressing an siRNA
targeted against hRev7 and also carrying the gene for puromycin resistance to
allow selection of transfectants. As a control, we similarly transfected the parental
cells with a vector containing this selectable marker and an siRNA that has no
significant homology to any human gene sequence. Puromycin-resistant clones
were isolated and expanded. Nuclear protein extracts from the parental cells,
vector control transfectants, and from candidate transfectants that received the
siRNA against hRev7 were prepared and analyzed by Western blotting for their
level of expression of hRev7. Figure 1 shows a representative Western blot.
Lane 1 shows hRev7 protein from the parental cell strain (P), migrating as
expected for a 24 kDa protein. Lane 2 shows hRev7 protein from the vector
control cell strain (VC). Lanes 3 and 4 were loaded with protein from two clones,
designated 2-2 and 2-6, that had been transfected with a vector expressing
hRev7 siRNA. No hRev7 protein could be observed in the latter two clones (Fig.
1). However, when the Western blots were allowed 2 24 h of exposure, a very
low level of hRev7 protein could be detected on the blots (data not shown). The
morphology of clones 2-2 and 2-6 did not differ from that of the parental or the
vector control cells. Neither did the rate of growth in culture of these two cell

strains differ from that of their parent or the vector control cells (data not shown).

145

Fig. 1 - Western blot analysis of hRev7 protein levels in the cell strains. A
polyclonal antibody against hRev7 was used to analyze the level of hRev7
protein in nuclear Iysates extracted from the parental cells (P), the vector control
cells (VC), and two derivative clones expressing a transfected siRNA targeted
against hRev7, viz., clones 2-2 and 2-6. The arrowhead on the right indicates the

location of a 22 kDa marker. Ku80 was used as the loading control.

146

 

hRev7” - -

 

 

 

KUBO- —--—-~~

 

147

Flg. 1

Effect of reduction of the level of hRev7 protein on the sensitivity of the cell

strains to the cytotoxic effect of UV(254nm) radiation— To examine the

sensitivity of these cell strains to the cytotoxic effect of UV, we irradiated the
parental cells, the vector control cells, and the two cell clones virtually devoid of
hRev7 protein and assayed them for survival of colony-fanning ability. As shown

in Fig. 2A, the parental cell strain and the vector control strain exhibited identical

sensitivity to the cytotoxic effect of UV(254nm) radiation. The dose required to

reduce their survival to 37% that of unirradiated cells was 12 Jlmz. The survival

of clones 2-2 and 2-6 were identical to each other, but both were 1.7 times more

sensitive to UV-induced cell killing than the control cells. The dose required to

reduce their survival to 37% that of unirradiated cells was only 8.5 J/m2. These

differences were shown to be statistically significant. The data indicate that

hRev7 plays a protective role in survival of cells exposed to UV irradiation.

Effect of reduced hRev7 protein on the frequency of UV-induced mutations-
The frequency of mutations induced by UV in the HPRT gene of these four cell
strains was determined using resistance to 6-thioguanine as an indicator of cells
with a mutation in their HPRT gene. As shown in Fig. 28, the frequency of

mutations induced by UV in the parental cells and vector control cells expressing

hRev7 were identical. The dose that reduced their survival to 37%, i.e., 12 J/m2,

induced an HPRT mutation frequency of ~135 x 106. In contrast, the two clones

148

Fig. 2 - Effect of reduced expression of hRev7 on the survival of cell

strains exposed to UV(245nm) radiation and on the frequency of UV-induced

mutations. (A) Clones 2-2 and 2-6 (closed symbols), which have greatly
reduced levels of expression of hRev7 protein, along with their parental strain
and a vector control strain (open symbols) were UV-irradiated and assayed for
cell survival as determined by colony forming ability. Some data points have
been offset slightly to make them visible. The lines represent least squares lines.
(B) The frequency of UV-induced mutations in the HPRT gene of these four cell
strains was determined by resistance to 6-thioguanine. The frequency of 6-
thioguanine—resistant cells was calculated using the cloning efficiency of cells at
the time of selection, which averaged ~44% for the parent and the vector control
cells, and ~25% for clones 2-2 and 2-6. Induced frequencies were calculated by

subtracting the background frequencies observed in sham-irradiated populations.

For the vector and parental control cells, these values were always <11 x 106.
For clone 2-2, they ranged from 0-18 x 10's, with the majority being 0-4 x 106.

For clone 2-6, they ranged from 8-25 x 10's, with the majority being 8-11 x 106.

Some data points have been offset slightly to make them visible. The lines

represent least squares lines.

149

100

80'

Percent Survival

‘8

160

120

Induced TG' Cells per 10‘ Clonable Cells

40'

20'

T U

      
 
 

4 6
\ji

    

s; A
‘ \
O-l, ..

V t " fir I V '

8101214 16

 

 

 

G
O

S

 

2

 

 

 

6UV (Jim!)

150

o 12 14 16
Flg.2

with greatly reduced expression of hRev7 protein showed a statistically

significant decrease in the frequency of UV(254nm)-induced mutations (P-value
for comparison of slopes). The frequency of mutants induced by 12 J/m2 in
clones 2-2 and 2-6 was only 32 x 10'6 clonable cells. At a dose that reduced their
survival to 37%, i.e., 8.5 J/mz, the frequency of induced mutants was ~20 x 10-6,

significantly lower than the 135 x 10.6 seen in the two control cell strains. This

significant reduction in the frequency of UV-induced mutations in cells with
reduced hRev7 indicates that hRev7 plays an Important role in UV-induced

mutagenesis in human fibroblasts.

Effect of decreased hRev7 protein expression on kinds of UV-induced
mutations— The kinds of base substitutions induced by UV in the coding region
of the HPRT gene of cells expressing or not expressing hRev7, as determined by
nucleotide sequencing, are shown in Table 1. The data report kinds of mutations
induced in the two clones extremely deficient in expression of hRev7 (clones 2-2
and 2-6) and those induced in the comparison group (parent and vector control).

[A table detailing the kinds and showing the context of 95 independent

UV(254nm)-induced mutations in HPRT of these two sets of human skin

fibroblasts can be found in the Supplement]. There was no statistically significant
difference in the kinds of mutations induced in the parental and vector control

cells compared to those induced in the two derivative cell strains, virtually devoid

151

Table 1 — Effect of reduced expression of hRev7 on the types of base
substitution mutations induced by UVmsnm) radiation. Cell clones with
reduced hRev7 (clone 2-2 and 2-6), their parental strain and a vector control
were treated with UV and the types of base substitutions induced in the HPRT
genes of these cell strains was determined by subjecting 6-thioguanine resistant
colonies to reverse transcription PCR and then sequencing the resulting cDNA.
The number of each of the six possible types of UV-induced base substitution
mutations that occur at sites at pyrimidines (also expressed as a percentage of
the total number of base substitutions) is indicated. Only independent base
substitutions at adjacent pyrimidines that resulted in an amino acid change were

considered UV-induced mutations.

TABLE 1. Kinds of UV-induced base substitutions in the HPRT
gene of cells with normal or greatly decreased levels of hRev7

 

 

 

Base changes vzgflgxd ”21:32.3“
C97 25 (47.2%) 19 (45.2%)
T->c 6 (11.3%) 2 (4.8%)
79“ 8 (15.1%) 11 (26.2%)
09" 10 (18.9%) 7 (16.7%)
cac 4 (7.5%) 1 (2.4%)
T79 0 2 (4.8%)
T0131 __ _____ _—___53 “710070) " ‘ 42'" "(160%)“ '

 

153

of hRev7 protein. These data indicate that reducing the level of hRev7 results in
a decrease in the frequency of UV-induced mutations without altering the types of

base substitutions generated.

Evidence that reduction of expression of hRev7 protein in human
fibroblasts results in a UV-induced delay in traversing S-phase- To assess
potential effects of decreased hRev7 on cell cycle progression after UV
irradiation, the parental and vector control cell strains and the two clones with

greatly reduced hRev7 were synchronized at the G1/S border as described,

released from the block, and immediately UV-irradiated with 12 Jlmz, the dose

determined to reduce the survival of normal cells to ~37% and that of the hRev7-
deficient cells to ~20% (see Fig. 2A). Independent synchronized populations of
these UV-irradiated cells were assayed by flow cytometry immediately following
release from the G1/S block (0 h), or after 10 h or 16 h. As shown in Fig. 3A, at
the time of release from the replication block (0 h), all four strains were
synchronized at the G1/S border. Ten h later, the parental (P) and vector control
cells (VC) were predominantly in S-phase and G2 phase. In contrast, after 10 h,
cell strains 2-2 and 2-6 were still predominantly in G1 and S-phase, i.e., their cell
cycle progression was greatly delayed, compared to that of the two control
strains. The data from cells assayed 16 h following UV irradiation show that the
cell strains with reduced hRev7 were still delayed in S-phase, compared to their

parental strain and the vector control cell strain.

154

Fig. 3 - Flow cytometry analysis of UV-irradiated cells. (A) Cell strains with
reduced levels of hRev7 (2-2 and 2-6), their parental strain (P), and a vector
control transfectant (VC) were synchronized at the G1/S border as described,
and UV-irradiated immediately after release from synchrony. The cell strains
were assayed by flow cytometry for DNA content at 0 h, 10 h, and 16 h post-
irradiation. (B) These four cell strains were UV-irradiated while growing
asynchronously and were assayed by flow cytometry for DNA content at 0 h and
10 h post-irradiation. (C) As a control, clones 2-2, 2-6, and their parental strain
were synchronized at the G1/8 border, as above, released from synchrony, but
not exposed to UV, and analyzed by flow cytometry at 0 h, 4 h, and 6 h after

release from the G1/8 block.

155

r p 2r r
#121: ”mm. E-

  

52:.

E.- e

 

These four cell strains were similarly irradiated with 12 J/m2 while growling

asynchronously and assayed by flow cytometry. As shown in Fig. 3B, UV-
irradiation of asynchronously-growing cells with decreased expression of hRev7
also delayed their progression through S-phase compared to control cells. As a
control, we synchronized the parental and the derivative strains with reduced
hRev7, released them from the block, as described above, but did not expose
them to UV radiation. The subsequent flow cytometric analysis (Fig. 3C)
demonstrated that all three unirradiated cell strains proceeded through the cell
cycle at an equal rate. Taken together, the data in Fig. 3 indicate that cells with
reduced hRev7 replicate their DNA more slowly in the presence of UV-induced

DNA damage.

157

DISCUSSION:

A role for hRev7 as the non-catalytic subunit of PoIC was suggested by its
homology to the yeast Rev7 protein, as well as its physical interaction with
human Rev3 in a yeast-two-hybrid assay [14]. The results of our study establish
a functional role for hRev7 in UV-induced mutagenesis, and what is more, a role
in the survival of colony-forming ability. The fact that at every dose of UV
radiation, the cells with greatly decreased levels of hRev7 protein exhibited ~5-
fold lower frequency of induced mutations than the control cell strains is strong
evidence that in human cells the hRev7 protein plays a role in error-prone TL8

past UV-induced DNA photoproducts, as does yeast Rev7.

The mutagenesis data from the present study, using human fibroblasts that have
greatly reduced levels of hRev7 protein, confirm and greatly strengthen the
results obtained previously in this laboratory using similar cell lines and antisense
against hRev3 [10, 15]. The ~5-fold decrease in UV-induced mutation frequency
observed in cells virtually devoid of hRev7 is similar to the decrease in frequency
observed previously in this laboratory using human fibroblasts expressing
antisense hRev3 [10, 15]. These data strongly support the hypothesis that hRev3
and hRev7 function in the same pathway, most likely as catalytic and non-

catalytic subunits, respectively, of human polymerase C.

158

Our results demonstrating that human fibroblasts with reduced hRev7 are more
sensitive to the cytotoxic effects of UV-radiation than the control strains (Fig. 2A)
support the results obtained by Cheung et al., which demonstrated that
nasopharyngeal carcinoma cell lines with reduced hRev7 are also more sensitive
to the cytotoxic effects of specific DNA damaging anticancer drugs than the
control cells [18]. The fact that reduction of hRev7 results in increased sensitivity
to the cytotoxic effects of various DNA damaging agents in both normal and
cancer cells, underscores the importance of hRev7 for protecting human cells

from DNA damage.

In addition to sharing a high degree of amino acid sequence similarity to the
yeast Rev7 protein, hRev7 also has a high degree of similarity to the mitotic
checkpoint protein hMADZ, and thus is also referred to as hMADZB [14, 19]. In
fact, hRev7 has been shown to interact with hMAD2 and furthermore, to inhibit
the anaphase-promoting complex by binding to activators th1 and Cdc20 in
Xenopus extracts, suggesting a role for hRev7 in regulating the mitotic
checkpoint [14, 20, 21]. One might hypothesize that the sensitivity to the
cytotoxic effects of UV we observe in our human fibroblasts reflects interference
with the mitotic checkpoint response of UV-irradiated cells. However, Cheung et
al., [18] found that reducing hRev7 in nasopharyngeal carcinoma cells had no
effect on their mitotic checkpoint response. Therefore, we consider it unlikely that
the increased sensitivity to the cytotoxic effects of UV-radiation that we observe

in human fibroblasts with reduced hRev7 results from an aberrant mitotic

159

checkpoint response. Nevertheless, further experimentation could be conducted
to specifically determine whether or not hRev7 affects the mitotic checkpoint of

human fibroblast cells in the presence of DNA damage.

In addition to decreasing the survival of UV-irradiated human fibroblast strains,
reduced levels of hRev7 also resulted in a marked decrease in their rate of
progression through the DNA synthesis phase of the cell cycle (Fig. 3). This
suggests that for human cells with reduced expression of hRev7, and therefore
with reduced hPolC, the presence of DNA damage during S-phase presents a
major problem for DNA replication. If cells with decreased expression of hRev7
were simply unable to resume DNA replication as a result of impaired TLS, this
situation would be expected to result in replication fork breakdown and ultimately
lead to cell death. It is always possible that replication arrest due to impaired TLS
accounts for the increase in sensitivity to the cytotoxic effects of UV that we
observe when cell strains with reduced hRev7 are irradiated. However, previous
results in our laboratory using human cells expressing antisense against hRev3,
the catalytic subunit of hPolC, demonstrated that such cells did not differ from the
parental cells in sensitivity to the cytotoxic effects of UV [15]. Taken together,
these results suggest that in our clones with reduced hRev7, factors other than
impaired TLS contribute to the observed increase in sensitivity to the cytotoxic

effects of UV.

In 2005, a study by Bi et al. [22], demonstrated mouse embryonic fibroblasts

160

lacking POIK were unable to recover from a benzo[a]pyrene diol epoxide (BPDE)-
induced S-phase checkpoint and, in addition, were more sensitive to the
cytotoxic effects of BPDE than were the wild-type cells. The fact that we found a
clear indication of a UV-induced delay in S-phase, and an increase in sensitivity
to the cytotoxic effects of UV in cell strains with reduced hRev7, suggests that
there is a similar requirement for hRev7 in recovery from a UV-induced S-phase

checkpoint as there is for Pair: in a BPDE-induced S-phase checkpoint.

In summary, our mutagenesis data demonstrate that hRev7, like hRev3, is
required for TLS past UV photoproducts and is causally involved in producing the
mutations that result from such TLS. These data strongly support the hypothesis
that hRev7, together with hRev3, comprise hPolC. In addition, our data
demonstrate a requirement for hRev7 in protecting human fibroblasts from the
cytotoxic effects of UV-induced DNA damage that was not found in human
fibroblast cells expressing antisense against hRev3. This suggests that the
protective role of hRev7 for cells exposed to UV radiation is independent of the

requirement for hRev7 in hPolC-dependent TLS.

161

ACKNOWLEDGEMENTS:

We thank Dr. Christopher Lawrence of the Department of Biochemistry and
Biophysics at the University of Rochester for his helpful advice, insight, and
constant interest in our research. We also thank Dr. Katheryn Meek, Department
of Pathobiology & Diagnostic Investigation at Michigan State University, for her
advice and encouragement during this research, and Dr. Dennis Gilliland, Co-
director of the Center for Statistical Training and Consulting at Michigan State
University for statistical analysis of the data. A special note of thanks to Andrew
McCoy and James Reinhart for their assistance with experiments. This research
was supported by Grant CA91490 awarded to VMM from the National Cancer

Institute of NIH.

162

REFERENCES:

10.

Zhou, B. B., and Elledge, 8. J. (2000). The DNA damage response:
putting checkpoints in perspective. Nature 408, pp. 433-439.

Yang, W. (2004). Advances in Protein Chemistry. Vol. 69, DNA Repair
and Replication ed., Elsevier Academic Press, San Diego, CA, USA.

Holmquist, G. P., and Maher, V. M. (2002). Vol. 510, Elsevier Science,
New York, NY, USA.

Lawrence, C. W. (2004). Cellular functions of DNA polymerase zeta and
Rev1 protein. Adv. Protein Chem. 69, pp. 167-203.

Ohmori, H., Ohashi, E., and Ogi, T. (2004). Mammalian Pol kappa:
regulation of its expression and lesion substrates. Adv. Protein Chem. 69,
pp. 265-278.

Vaisman, A., Lehmann, A. R., and Woodgate, R. (2004). DNA
polymerases n and L. Adv. Protein Chem. 69, pp. 205-228.

Lemontt, J. F. (1971). Pathways of ultraviolet mutability in Saccharomyces
cerevisiae. I. Some properties of double mutants involving uvs9 and rev.
Mutat. Res. 13, pp. 311-317.

Lawrence, C. W., Das, G., and Christensen, R. B. (1985). REV7, a new
gene concerned with UV mutagenesis in yeast. Mal. Gen. Genet. 200, pp.
80-85.

Nelson, J. R., Lawrence, C. W., and Hinkle, D. C. (1996). Thymine-
thymine dimer bypass by yeast DNA polymerase zeta. Science 272, pp.
1646-1649.

Gibbs, P. E., McGregor, W. G., Maher, V. M., Nissan, P., and Lawrence,
C. W. (1998). A human homolog of the Saccharomyces cerevisiae REV3
gene, which encodes the catalytic subunit of DNA polymerase zeta. Proc.

163

11.

12.

13.

14.

15.

16.

17.

18.

Natl. Acad. Sci. USA 95. pp. 6876-6880.

Lin, W., Wu, X., and Wang, Z. (1999). A full-length cDNA of hREV3 is
predicted to encode DNA polymerase zeta for damage-induced
mutagenesis in humans. Mutat. Res. 433, pp. 89-98.

Morelli, C., Mungall, A. J., Negrini, M., Barbanti-Brodano, G., and Croce,
C. M. (1998). Alternative splicing, genomic structure, and fine
chromosome localization of REV3L. Cytogenet. Cell Genet. 83, pp. 18-20.

Xiao, W., Lechler, T., Chow, B. L., Fontanie, T., Agustus, M., Carter, K. C.,
and Wei, Y. F. (1998). Identification, chromosomal mapping and tissue-
specific expression of hREV3 encoding a putative human DNA
polymerase zeta. Carcinogenesis 19, pp. 945-949.

Murakumo, Y., Roth, T., Ishii, H., Rasio, D., Numata, 8., Croce, C. M., and
Fishel, R. (2000). A human REV7 homolog that interacts with the
polymerase zeta catalytic subunit hREV3 and the spindle assembly
checkpoint protein hMAD2. J. Biol. Chem. 275, pp. 4391-4397.

Li, Z., Zhang, H., McManus, T. P., McCormick, J. J., Lawrence, C. W., and
Maher, V. M. (2002). hREV3 is essential for error-prone translesion
synthesis past UV or benzo[a]pyrene diol epoxide-induced DNA lesions in
human fibroblasts. Mutat. Res. 510, pp. 71-80.

Morgan, T. L., Yang, D. J., Fry, D. G., Hurlin, P. J., Kohler, S. K., Maher,
V. M., and McCormick, J. J. (1991). Characteristics of an infinite life span
diploid human fibroblast cell strain and a near-diploid strain arising from a
clone of cells expressing a transfected v-myc oncogene. Exp. Cell Res.
197, pp. 125-136.

Maher, V. M., and McCormick, J. J. (1996). The HPRT gene as a model
system for mutations analysis. in Technologies for Detection of DNA
Damage and Mutations (Pfeifer, G. P., Ed.), pp 381-390, Plenum Press,
New York.

Cheung, H. W., Chun, A. C., Wang, Q., Deng, W., Hu, L., Guan, X. Y.,
Nicholls, J. M., Ling, M. T., Chuan Wang, Y., Tsao, S. W., Jin, D. Y., and
Wang, X. (2006). Inactivation of human MADZB in nasopharyngeal
carcinoma cells leads to chemosensitization to DNA-damaging agents.
Cancer Res. 66, pp. 4357-4367.

164

19.

20.

21.

22.

Cahill, D. P., da Costa, L. T., Carson-Walter, E. B., Kinzler, K. W.,
Vogelstein, B., and Lengauer, C. (1999). Characterization of MAD2B and
other mitotic spindle checkpoint genes. Genomics 58, pp. 181-187.

Chen, J., and Fang, G. (2001). MADZB is an inhibitor of the anaphase-
promoting complex. Genes Dev. 15, pp. 1765-1770.

Pfleger, C. M., Salic, A., Lee, E., and Kirschner, M. W. (2001). Inhibition of
th1-APC by the MADZ-related protein MADZL2: a novel mechanism for
regulating th1. Genes Dev. 15, pp. 1759-1764.

Bi, X., Slater, D. M., Ohmori, H., and Vaziri, C. (2005). DNA polymerase
Kappa is specifically required for recovery from the benzo[a]pyrene-Di-
hydrodiol epoxide (BPDE)-induced S-phase checkpoint. J. Biol. Chem.

165

CHAPTER III

hRev7, the accessory subunit of hPol C, protects human fibroblasts from the
cytotoxic effect of benzo[a]pyrene diol epoxide, but is not essential for

translesion synthesis past DNA damage induced by this agent.

Jessica A. Neal, Kathryn L. Fletcher, J. Justin McCormick, and Veronica M.

Maher*

Carcinogenesis Laboratory, Cell and Molecular Biology Program, Department of
Microbiology & Molecular Genetics, and Department of Biochemistry &

Molecular Biology, Michigan State University, East Lansing, MI 48824-1302

*Carrespanding author at: Veronica M. Maher, Carcinogenesis Laboratory, Food
Safety and Toxicology Bldg, Michigan State University, East Lansing, MI 48824-
1302, USA; Ph: (517) 353-7785; Fax (517) 353-9004; E-mail address:

maher@msu.edu

166

ABSTRACT:

DNA polymerase zeta (Pol C), comprised of hRev3 and hRev7, is a specialized
DNA polymerases that, unlike the classical replicative polymerases, is able to
replicate past fork-blocking DNA lesions (i.e. to carry out translesion synthesis).
The catalytic subunit of hPol C, hRev3, has been shown to play a critical role in
the induced mutagenesis of human cells. However, much less is known about
the role of hRev7, the accessory subunit of hPol C, in such mutagenesis.
Recently, we generated human fibroblasts with significantly reduced levels of

hRev7 protein and used such cells to demonstrate that hRev7 is required to

protect cells from the cytotoxic and mutagenic effects of UV(254nm) radiation. To

determine the effect of hRev7 on the biological effects of benzo[a]pyrene diol
epoxide (BPDE) cell strains with reduced levels of hRev7 were compared to their
parental strain and a vector control for cell survival, induction of mutations, and
ability to progress through the cell cycle following exposure to BPDE. We
determined that like with UV, cell strains with reduced hRev7 were more
sensitive to the cytotoxic effects of BPDE than the control strains, and that these
cell strains progressed through S-phase at a much slower rate than the control
cells following BPDE treatment. But, our surprise, the frequency of BPDE-
induced mutations in the cells with reduced hRev7 did not differ from those
induced in their parental and vector control strains, indicating the hRev7 is not
required for translesion synthesis past BPDE-induced DNA damage. We also

found that cell strains with reduced hRev7 were more sensitive than their

167

parental and vector cantrol strains to the cytotoxic effect of ionizing radiation,
cisplatin, and N-methyI-N-nitrosourea, indicating that hRev7 also protects cells
from these DNA damaging agents and suggesting that hRev7 is generally
required to protect cells from the cytotoxic effect of DNA damaging agents.
Taken together, these results indicate that the protective role hRev7 plays for the
survival of cells exposed to DNA damaging agents is independent of its role in

translesion synthesis.

168

INTRODUCTION:

Human cells undergo countless rounds of DNA replication, which must be very
accurate to preserve critical genetic information. To maintain such a significant
level of accuracy, the classical replicative polymerases have evolved highly
selective active sites that only accommodate nucleotides when they are correctly
paired to the DNA template. In addition, many of these DNA polymerases
possess 3’-—>5’ proofreading exonuclease activity, which removes nucleotides
that are incorrectly incorporated during replication, allowing an additional attempt
at accurate DNA synthesis. As a consequence of their stringency, the classical
replicative polymerases cannot tolerate fluctuations in the DNA structure,
including those that result from DNA damage. Nevertheless, DNA is continually
subjected to a variety of insults, from both endogenous and environmental
agents that generate DNA damage. Much of this damage is excised by DNA
repair mechanisms before replication occurs. However, if repair is slow or the
DNA damage is extensive, DNA lesions may persist during replication. If the high
fidelity replicative polymerase complex encounters a DNA lesion that blocks

elongation, potentially fatal stalling or arrest of replication will occur.

To avoid replication arrest, mechanisms may be employed that enable DNA
lesions to be tolerated without their physical removal. One such mechanism is
translesion synthesis (TLS). Translesion synthesis involves the use of

specialized DNA polymerases to replicate past the DNA lesions until the high

169

fidelity replicative polymerases are able to resume DNA synthesis. Several novel
DNA polymerases have been discovered, whose primary function is likely to
carry out TLS. These TLS polymerases typically contain active sites that are less
restrictive, making them better able to accommodate distortions in DNA (see for
example [1-4]). Although TLS polymerases demonstrate the unique ability to
synthesize past replication-blocking DNA lesions, enabling the cell to survive
such DNA damage, they are also characterized by relaxed nucleotide selectivity
and lack of 3’-—> 5’ proofreading exonuclease activity. As a result, protection of
cells from replication arrest can come at the cost of introducing mutations in

DNA, which can result in the development of cancer.

More than 300 polymerases involved in TLS have been discovered in
eukaryotes, bacteria and archaea [5]. The first TLS polymerase to be identified in
eukaryotes was DNA polymerase zeta (Pol C) [6]. DNA polymerase C was initially
characterized in the budding yeast, Saccharomyces cerevisiae, and is composed
of two subunits, a catalytic subunit, called Rev3, as well as an accessory subunit,
Rev7 [6]. Studies using yeast rev mutant strains have demonstrated that Pol C is
responsible for the majority of both spontaneous [7, 8] and DNA damage-induced
mutations that occur in this organism [9-14], suggesting that this polymerase
participates in error-prone TLS past a extensive array of DNA lesions (reviewed

in [15]).

170

Human homologs of the yeast REV genes have been identified [16, 17]. The
transcript of the human REV3 gene encodes a 353 kDa protein, which is about
twice the size of the yeast protein [18]. Presumably because of the large size and
low cellular levels of hRev3, the protein has never been expressed or isolated
and therefore, in vitro studies using human Pol C are lacking [19]. However,
human cells expressing high levels of hREV3 antisense RNA have been reported
to demonstrate a lower frequency of both ultraviolet (UV) and benzo[a]pyrene
diol epoxide (BPDE)-induced mutations than the control cells, indicating that, as
in yeast, hRev3 is required for induced mutagenesis and suggesting that the

functions of Pol C are conserved from yeast to humans [18, 20].

To investigate the role of hRev7 in TLS, we recently generated two human
fibroblast cell strains in which the levels of hRev7 protein were significantly
reduced by small interfering RNA (siRNA) [21]. When cell strains with reduced
hRev7 were UV-irradiated, their rate of progression through S-phase was
considerably slower, and their cell survival was significantly reduced, compared
to control strains. In addition, the frequency of UV-induced mutations in cell
strains with reduced hRev7 was five times lower than normal. These data
showed that like hRev3, hRev7, presumably as a part of human Pol C, plays a

role in UV-induoed mutagenesis of human cells.

To determine whether hRev7 is similarly involved in the tolerance of mutations

induced by benzo[a]pyrene diol epoxide, the reactive form of the widespread

171

environmental carcinogen benzo[a]pyrene, cells strains with reduced levels of
hRev7 were compared to their parental strain and a vector control for their
response to the biological effects of BPDE. Our results show that cell strains with
reduced hRev7 progress through the cell cycle at a slower rate than control
strains after exposure to BPDE, and are also more sensitive to its cytotoxic
effect. These data indicate that hRev7 is required for protecting cells from BPDE-
induced killing. We also found that cell strains with reduced hRev7 are more
sensitive to the cytotoxic effects of ionizing radiation (IR), cisplatin, and N-methyl-
N-nitrosourea (MNU), suggesting that hRev7 has a generalized role in protecting

cells from an assortment of DNA damaging agents.

Our results further demonstrate, that the frequency of mutations induced by
BPDE in cells with reduced levels of hRev7 does not differ from the frequency
induced in the control cell strains, indicating that hRev7 is not required for TLS
past BPDE-induced DNA lesions. These results are particularly surprising in light
of the fact that previous studies in our laboratory [20] demonstrated that hRev3,
the catalytic subunit of hPol C, is required for mutagenic TLS past BPDE-induced
DNA lesions. However, after careful re-evaluation of our previous data, we now
believe that neither hRev3 nor hRev7 are required TLS past BPDE-induced

mutagenesis.

172

MATERIALS AND METHODS:

Cell culture-- Cells were routinely cultured in Eagle’s minimal essential medium
supplemented with L-aspartic acid (0.2 mM), L-serine (0.2 mM), sodium pyruvate
(1 mM), supplemented calf serum (SCS) (Hyclone, Logan, UT) (10% (v/v)),
penicillin (100 U/ml), streptomycin (100 pg/ml), tetracycline (1 pg/ml) and

hydrocortisone (1 jig/ml) (culture medium). The cells were cultured in a 37°C

humidified incubator with 5% C02, 95% air.

Cell strain -- The human fibroblast cell strains used as the parental strains in
this study, designated MSU-1.2.9N or MSU-1.2.9N.58 (9N and 9N.58 for short),
were derived from the infinite life span, telomerase positive, near-diploid,
karyotypically stable, MSU-1.2 lineage of cells established in the Carcinogenesis

Laboratory [22].

Cell strains 2-2 and 2-6, which have significantly reduced levels of hRev7 protein,
as well as the vector control strain, VCA, were derived from the parental strain,
9N.58, by McNalIy et al., as described [21]. The two additional cell strains with
reduced hRev7 that were used in this study, designated 2.5 and 3.2, as well as
the vector control, V1.1, were derived from the same parental strain, i.e. 9N.58,
using the methods previously described [21]. Briefly, the p8ilencer3.1/hRev7
vector, which expresses siRNA targeted to hRev7 and contains puromycin as a

selectable marker, or the appropriate vector control, p8ilencer3.1, were

173

transfected into the parental strain 9N.58 using LipofectamineTM (Life

Technologies, GibcoBRL) according to the manufacturer’s instruction. Forty-eight
h post-transfection, the cells were selected in culture medium containing 1 [Lg/ml
puromycin. This medium was renewed every 5-7 days for 2 weeks. When
puromycin-resistant colonies formed, they were isolated, expanded, and
analyzed by Western blot for the level of expression of hRev7 protein. Cell
strains 2-2, 2-6, 2.5 and 3.2 were determined to express significantly reduced

levels of hRev7 protein (Figure 1 and [21]).

Cell strains expressing high levels of hRev3 antisense (6| and 12C) were
generated by Li et al., as described in reference [20]. In short, the pKS-2 plasmid
[18], which expresses hRev3 antisense RNA under control of the TetP promoter
(Tet off system) and contains puromycin as a selectable marker, was transfected
into the parental strain using LipofectAMlNE according to the manufacturer’s
instructions (Life Technologies, GibcoBRL). Forty-eight h after the transfection,
cells were selected with puromycin (1 jig/ml). When puromycin resistant colonies
formed, they were isolated and expanded. Clones 6| and 120 were determined

to express high levels of hRev3 antisense RNA by Northern blot analysis [20].

Western blot analysis— Western blot analysis was performed as described [21].
Briefly, to obtain nuclear protein extracts, subconfluent monolayers of cells were
washed with ice cold phosphate buffered saline (PBS), and cells were removed

from 150 mm-diameter dishes by scraping into 1 ml of Lysis Buffer A (10 mM

174

HEPES, 10 mM KCI, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM DTT, 0.5 mM PMSF).
Cells were incubated on ice for 15 min and then lysed by adding 62 MI of 10%
Nonidet P-40 and vortexing for 10 sec. Cell lysates were centrifuged for 30 sec at
10,000 RPM, 4°C and the supernatant, containing the cytoplasmic proteins, was
removed. Cell pellets were then washed once with 1 ml of Lysis Buffer A
containing Nonidet P-40. Nuclear protein extracts were obtained by incubating
cell pellets in 40 III of Lysis Buffer C (20 mM HEPES, 0.4 M NaCl, 1 mM EDTA, 1
mM EGTA, 1 mM DTT, 1 mM sodium orthovanadate, 1 mM PMSF) and vortexing
every 5 min for a total of 15 min. The nuclear extracts were centrifuged for 5 min
at 16,000 RPM, 4°C and the supernatants, containing the nuclear proteins, were
saved. The protein concentration was measured using a Coomassie (Bradford)
Protein Assay Kit (Pierce). For Western analysis, 75 pg of nuclear protein was
separated by gel electrophoresis using 14% SDS-polyacrylamide gels and
electrotransferred to a polyvinylidene fluoride membrane (Millipore). The
membrane was blocked for 1 h at room temperature in 5% non-fat milk diluted in
tris-buffered saline containing 0.1% Tween-20, and incubated with primary
antibodies against hRev7 (1:600, Bethyl) or Ku80 (1:10,000, Santa Cruz) at 4°C
overnight. The following day, the membrane was incubated for 1 h at room
temperature with horseradish peroxidase—conjugated, goat anti-rabbit secondary
antibody (1:10,000, Sigma) and signals were visualized using the SuperSignal

Chemiluminescent substrate (Pierce).

175

Exposure of cells to chemical mutagens-- When treating with chemical
mutagens, the number of DNA lesions generated is dependent upon the cell
density at the time of treatment. Therefore, both for survival and for mutagenesis
studies, cells in exponential growth were detached from the dishes using trypsin
and plated in 150 mm-diameter dishes approximately 16 h prior to treatment,

such that the density of cells at the time of treatment would be as near 10,000

cells/cm2 as possible. Following the 16 h attachment period, the culture medium

was removed from each dish, cells were rinsed twice with (PBS), and then
covered in Eagle’s minimal essential medium [for MNU treatments, Eagle’s
minimal essential medium buffered with 15mM HEPES (pH 7.2)]. Immediately
prior to treatment, BPDE (Midwest Research Institute) or MNU (Sigma) were
dissolved in anhydrous dimethylsulfoxide (DMSO), and the designated doses
were delivered by micropipette. To ensure that all cells were exposed to the
same concentration of DMSO, regardless of the dose of BPDE or MNU,
appropriate amounts of DMSO were added to dishes (including the untreated
control cells) to equal the total amount of DMSO delivered to cells treated with
the highest dose. Cisplatin (American Pharmaceutical Partners Inc.), which was
supplied in an aqueous saline solution (1 mg/ml), was delivered to the cells

directly by micropipette. Cells were exposed to BPDE or cisplatin for 1 h or to

MNU for 30 min, at 37°C in a humidified 5% CO2 incubator. At the end of the

exposure period the medium containing the mutagen was removed, cells were
rinsed twice with PBS, and covered with fresh culture medium. Induced

cytotoxicity and mutagenesis assays were performed as described below.

176

Exposure of cells to ultraviolet radiation- The method for exposing cells to

UV(254 nm) radiation differs somewhat from the method for exposure to chemical

agents because, unlike what happens with chemical treatments, the number of
DNA lesions generated by exposure to UV radiation is not dependent on cell
density. For UV radiation, 12 h prior to UV exposure, cells in exponential growth

were trypsinized and plated at cloning densities (100-500 cells per 100 mm-

diameter dish) for assaying cell survival, or at 1x106 cells per 150 mm-diameter

dish for mutagenesis assays. Immediately prior to UV exposure, the culture
medium was removed from each dish and the cells were washed twice with PBS.
The cells were then UV-irradiated for the appropriate amount of time to result in
the designated doses of UV as described previously [23] and replenished with
fresh culture medium. The culture medium was renewed the following day and
UV-induced cytotoxicity and mutagenesis assays were conducted as described

below.

Exposure of cells to ionizing radiation— On the day of treatment, exponentially
growing cells were detached from dishes with trypsin and diluted to 200,000
cells/ml in culture medium containing 2% supplemented calf serum. Cells were

irradiated as described [24] in 50 ml polypropylene tubes on ice using a US.

Nuclear 60Co variable flux, sealed source irradiator with a dose rate of 1.378

Gy/min.

177

Cell survival assay-- The procedures for determining the cytotoxic effects of
DNA damaging agents by colony forming ability differ slightly based upon the

particular type of DNA damaging agent used. For chemical mutagens, the cells

were exposed to the appropriate agent at a density of 10,000 cells/cm2 as

described above. Immediately following treatment, the cells were rinsed with
PBS, and detached from the dishes using trypsin. Trypsinized cells were then
diluted and plated into four 100 mm-diameter dishes for each dose at cloning
densities (i.e. the densities necessary to obtain approximately 50 surviving
colonies per 100 mm-diameter dish depending on the expected cytotoxicity).
After 7 days cells were provided with fresh culture medium, and after 14 days
they were stained with crystal violet. To determine sensitivity to the cytotoxic
effect of a particular agent (expressed as percent survival), the cloning
efficiencies of cells exposed to the mutagen were normalized to the cloning

efficiency of the untreated control cells.

For UV, cells were plated at cloning densities 12 h prior to exposure and then
UV-irradiated the as described. Cells were provided with fresh culture medium
after 1 day after UV-irradiation and again after 7 days. Fourteen days after UV-
irradiation cells were stained with crystal violet. As with chemical mutagens, the
survival of colony-forming ability of cells exposed to UV was determined by
normalizing the cloning efficiency of treated cells to the cloning efficiency of the

untreated cells.

178

To determine the cytotoxicity induced by ionizing radiation, cells were detached
using trypsin and irradiated in suspension as described earlier. Immediately after
irradiation, each cell suspension was diluted appropriately into fresh culture
medium and plated into 4, 100 mm-diameter dishes for each dose at cloning
densities. The culture medium was renewed after 7 days and the cells were
stained after 14 days. The survival of the irradiated cells was calculated by

normalizing the cloning efficiency of the irradiated cells to that of the control cells.

Mutagenesis assay— To determine the frequency of induced mutations in the
hypoxanthine phaspharibasyl transferase (HPR‘l) gene, assays were performed
as described [23]. In short, a sufficient number of target cells were plated into
150 mm-diameter dishes to ensure that after treatment, the number of surviving
cells was large enough to result in at least 40, 6-thioguanine (TG) resistant

clones. Because exposure to either BPDE or UV causes a high frequency of

induced mutants, it is sufficient to have approximately 0.8x106 surviving cells.

Following exposure to the mutagenic agent, cells were maintained in exponential
growth for an 8-day expression period to allow any wild-type HPRT protein to be
depleted. After the 8-day expression period, cells were trypsinized and diluted to
2,500 cells/ml in culture medium. To determine the cloning efficiency of the cells
at the time of selection, a small portion of the cell suspension (2 ml) was diluted
further and plated into 4, 100 mm-diameter dishes at cloning densities. To assay

for TG-resistance, the remainder of the cell suspension was selected with TG, at

179

a final concentration of 40 (M, and then cells were plated at a density of 25,000
cells per 100 mm-diameter dish. All dishes were supplied with fresh culture
medium (with or without TG as appropriate) after 7 days, and stained with crystal
violet after 14 days. The observed mutation frequency was corrected by the
cloning efficiency of the unselected cells. The induced mutation frequency was
calculated by subtracting corrected frequencies observed in untreated control

cells from the corrected mutant frequencies of the treated cells.

Analysis of the cell cycle progression of BPDE ar UV treated cells by flow

cytometry— Cells in exponential growth were detached from the dishes using

trypsin, and plated in 100 mm-diameter dishes such that the density of G1/S-

synchronized cells at the time of treatment would be as near 10,000 cells/cm2 as

possible. For the cell strains analyzed, approximately 140,000 cells were
required for each 100 mm-diameter dish. Cells were allowed 16 h to attach and
then arrested in G1 phase using Iovastatin (Sigma) at a final concentration of 60
[1M Twelve h later, cells were rinsed twice with PBS and replenished with fresh

culture medium containing aphidicolin (Sigma) at a final concentration of 2 jig/ml

and mevalonic acid (Sigma) at a final concentration of 6 mM to synchronize them

at the G1/S border. After an additional 12 h, cells were washed twice with PBS to

release them from synchrony and immediately treated with BPDE or UV as
described above. Every 4 h for the first 24 h post-treatment, a set of cells were

detached using trypsin, harvested by centrifugation and fixed in 80% ethanol.

180

Prior to flow cytometry, cells were washed in PBS containing 1% SCS and then
incubated in a propidium iodide solution (PBS, 1 mg/ml propidium iodide, 10%
Triton X-100, 0.5 mM EDTA, 10 mg/ml RNase A) for 1 h. Cells were analyzed for
DNA content by flow cytometry at the Flow Cytometry Core Facility at Michigan

State University.

CHAT selection-- To eliminate the background frequency of 6-thioguanine
resistant cells, they were plated in culture medium containing CHAT (20 uM
deoxycytidine, 100 (M hypoxanthine, 0.4 [J.M aminopterin, 30 [M thymidine) and
maintained in exponential growth for 3 weeks. Following the selection period,
cells were grown for several days in medium containing CHT (CHAT medium
excluding aminopterin) and then for several days in the appropriate culture

medium.

181

RESULTS:

Effect of reduced hRev7 protein on the survival of human fibroblasts
exposed to BPDE and an the frequency of BPDE-induced mutationsu Using
siRNA, we previously generated two derivatives of the human fibroblast strain
9N.58 with significantly reduced levels of hRev7 protein [21]. These derivative
strains were designated 2-2 and 2-6. Those two cell strains, along with two
additional 9N.58-derived cell strains with reduced hRev7 protein (designated 2.5
and 3.2), as well as their appropriate controls, were used in our current study
(Figure 1). To determine whether reducing the level of hRev7 protein alters the
response of human fibroblast cells to the cytotoxic or mutagenic effects of BPDE,
two cell strains with significantly reduced levels of hRev7 protein, (clones 2-2 and
2-6) were assayed along with their parental strain and a vector control
transfectant for their sensitivity to the cytotoxic effect of BPDE as measured by
survival of colony-forming ability (Figure 2A). These data show that, whereas the
vector control strain demonstrated a BPDE-induced cytotoxicity that was very
similar to that of the parent strain, each of the cell strains with reduced hRev7
protein was considerably more sensitive to BPDE-induced cytotoxicity.
Specifically, 80% of the parent and vector control cells survived after being
exposed to 0.07 (M BPDE, but only 50% of cells with reduced hRev7 survived
following exposure to the same dose. These results, which show that cells with

reduced hRev7 are more sensitive to the cytotoxic effect of BPDE than control

182

Figure 1 - Western blot analysis of the level of hRev7 protein. The level of
hRev7 protein in nuclear extracts obtained from cell strains transfected with a
vector expressing hRev7 siRNA (clones 2-2, 2-6, 2.5 and 3.2), their parental
strain (P), or cell strains transfected with a vector control expressing an siRNA
with limited homology to known sequences in the human genome (VCA and
V1.1) was analyzed by Western blotting. Ku80 was used as a loading control.
Note that the level of hRev7 protein in cell strains expressing hRev7 siRNA is
significantly lower than the level of hRev7 protein in the parent or vector control

strains.

2-2 2-6 VCA P V1.1 2.5 3.2

hRev7—b - - -

Ku80—P ‘-'w 5 -- Q * '

184

cells, indicate that the hRev7 protein plays a protective role for cells following

exposure to BPDE.

Because cell strains with reduced hRev7 protein were more sensitive to the
cytotoxic effect of BPDE, we were interested to determine whether the frequency
of BPDE-induced mutations in these cell strains also differed from those of the
control strains. Therefore, cell strains with reduced hRev7 were compared to
their parental cell strain as well as to the vector control strain for the mutagenic
effects of BPDE. To our surprise, the BPDE-induced mutation frequencies of cell
strains with reduced expression of hRev7 protein did not differ significantly from
those of their parental cell strain or from those of the vector control cell strain
(Figure 28). These data suggest that hRev7 is not essential for TLS past BPDE-

induced DNA lesions.

Effect of reduced hRev7 on cell cycle progression following BPDE
treatment-- Our laboratory previously demonstrated that decreasing the level of
hRev7 protein in human fibroblasts rendered them more sensitive to the cytotoxic
effect of UV and resulted in impaired progression through S-phase following UV-
irradiation [21]. We hypothesized that this UV-induced delay in cell cycle
progression contributed to the increased UV-induced cytotoxicity that we
observed. In the present study we found that, like with UV, cell strains with

reduced hRev7 are more sensitive to the cytotoxic effect of BPDE than control

185

Figure 2 - Reducing the level of hRev7 in human fibroblasts renders them
more sensitive to the cytotoxic effect of BPDE, but does not affect their
frequency of BPDE-induced mutations. Cell strains with reduced hRev7,
designated clones 2-2 and 2-6 (closed symbols), the parental strain (open
circles) and the vector control strain (open triangles) were treated with BPDE and
assayed for (A) cell survival and (B) the frequency of mutations induced in the

HPRT gene. The solid lines represent the least squares regression for the data.

186

0.00 0.03 0.06 0.09 0.12 0.15 0.18

V I

    
  

 

I l I V I r I t 1 U

hem

Ileii

Percent Survival
8

U
C

, 7
9i i.
:96".

L l l I l l l L l I L llllll

and

 

 

0)
83

I38

501-

Induced TGr Mutants per 10‘ Clonable Cells

 

B 1
.00 0.03 0.06 0.09 0.12. 0.151018
BPDEIuM)

 

 

 

187

strains. Therefore, we also examined the ability of these cell strains to progress

through the cell cycle following BPDE treatment.

To determine the effect of decreased expression of hRev7 protein on the rate of
cell cycle progression of BPDE-treated cells, the cell strains with reduced hRev7

(2-2 and 2-6), their parental cell strain (P) and the vector control strain (VC) were

synchronized at the G1/S border, released from synchrony, and exposed to

BPDE for 1 h. At the end of BPDE exposure, populations of each of the four cell
strains were harvested every 4 h for 24 h and analyzed by flow cytometry to
determine the percentage of cells in each stage of the cell cycle. Figure 3 shows
the resulting DNA histograms obtained from the sets of cells harvested at various
times after BPDE treatment. Immediately after release from synchrony and just

prior to BPDE treatment (0 h), the majority of cells in each of the four cell strains

were at the G1/S border, indicating that each of the four cell strains synchronized

equally well. After 4 h, most of the cells previously synchronized at the G1/S

border had progressed into S-phase regardless of the level of hRev7 protein,
indicating that cells released from the block and entered S-phase at similar rates

post-BPDE treatment. Eight h after BPDE treatment, many of the synchronized

control cells had moved through 8-phase and entered into the 62 phase,

whereas significantly fewer of the cells with reduced hRev7 had completed 8-

phase, indicating that following BPDE treatment, cell strains with reduced hRev7

188

Figure 3 - Reducing the level of hRev7 in human fibroblasts results in a
BPDE-induced delay in progression through the cell cycle. Cell strains with

reduced levels of hRev7 (2-2 and 2-6), their parental cell strain (P), and a vector

control transfectant (VC) were synchronized at the G1/S border and treated with
BPDE for one hour immediately after release from synchrony. Cells were
harvested, fixed, and stained with propidium iodide for analysis of DNA content

by flow cytometry. The distribution of cells in each phase in the cell cycle 0 h, 4 h,

8 h, 12 h, 16 h, 20 h, and 24 h post-BPDE treatment are depicted.

189

Elk Dec-kirk

.Prrrrbbrrrrbmpr

progressed through S-phase at a slower rate than the two control strains. After

12 h, the control cells had divided and cycled back into G1. In contrast, cells with

reduced hRev7 were still primarily in S and G2 phase i.e., compared to control

cells, cells with reduced hRev7 were still delayed in cell cycle progression.

Sixteen h post-BPDE treatment, the control strains had progressed through G1

and back into S-phase, whereas the cell strains with reduced hRev7 were

primarily in G2 and G1, a distribution similar to that of the control cells 4 h earlier.

After 20 h, although only a proportion of the cells remained synchronized, the

synchronized control cells were moving through S-phase and into G2, but the

cells with reduced hRev7 were delayed in moving into S phase and G2 phase,

indicating that, even after entering a second cell cycle, cells with reduced hRev7
progressed somewhat slower than the control cells. Twenty-four h post-BPDE
treatment, the synchrony of all four cell strains was lost. Taken together, the data
presented in Figure 3 indicate that, after BPDE treatment, cells with reduced

hRev7 progress more slowly through the cell cycle than the control strains.

Effect of reduced hRev7 an the survival of cells exposed to DNA damaging
agents— Cell strains with reduced hRev7 are sensitive to the cytotoxic effects of
both UV and BPDE, which generate structurally distinct types of DNA lesions.
Therefore, we examined whether the cell strains with reduced hRev7 also

differed from the control strains in their response to the cytotoxic effects of other

191

Figure 4 - Reducing the level of hRev7 in human fibroblast cells renders
them more sensitive to the cytotoxic effects of a variety of different types
of DNA damaging agents. Cell strains 2.5 and 3.2, which have significantly
reduced hRev7 protein, (closed symbols) were compared to their parental strain
and the vector control strain (open symbols) for their sensitivity to the cytotoxic
effects of (A) the DNA cross-linking agent cisplatin, (B) ionizing radiation (IR),
and (C) the alkylating agent N-methyl-N-hydroxyurea (MNU). The solid lines

represent the least squares regression for the data.

192

 

 

 

 

 

3233.0. acacia.

Cisplatin (11M)

 

 

 

 

 

.2523 £86..

In (G!)

 

‘I‘ dil‘

 

20 30
MNU (1.1M)

10

 

 

 

193

types of DNA lesions by exposing them to ionizing radiation; the DNA cross-
Iinking agent, cisplatin; and the alkylating agent, MNU. As shown in Figure 4, cell
strains with reduced hRev7 were more sensitive to cell killing by each of these
three DNA damaging agents, indicating that hRev7 is required to protect human

fibroblasts from a variety of distinct types of DNA damage.

Effect of hRev3 antisense expression on BPDE-induced cytotoxicity and
mutagenicity-- A previous study in our laboratory [20], using human fibroblasts
expressing hREV3 antisense RNA, appeared to demonstrate that hRev3, the
catalytic subunit of hPol C, is required for the mutagenic bypass of BPDE-induced
DNA lesions [20]. Therefore, it was quite unexpected that reducing hRev7, the
accessory subunit of hPol C, had no effect on the frequency of BPDE-induced
mutations. Recently, the reasons for this discrepancy have become clear. In our
previous study, cell strains expressing high levels of hRev3 antisense were
derived from the human fibroblast strain designated 9N. During the course of this
study, the parental strain, 9N, began to accumulate a large number of TG-
resistant cells. In an effort to eliminate this high TG-resistant background
frequency, a new strain, designated 9N.58, was cloned from 9N and used as the
control strain for this study. Unfortunately, we have now learned that, 9N.58 has
a much higher BPDE-induced mutation frequency than 9N does. As a result,
9N.58 is not a suitable comparison for cell strains derived from‘ 9N. To accurately
determine the effect of high levels of expression of hRev3 antisense on the

survival of cells exposed to BPDE and on the frequency of BPDE-induced

194

Figure 5 - Expression of high levels of hRev3 antisense does not affect the
cell survival or induced mutation frequency of human fibroblasts treated
with BPDE. Cell strains expressing high levels of hRev3 antisense, designated
clones 6| and 12C (closed symbols), and their parental strain, 9N, (open circles)
were treated with BPDE and assayed for (A) cell survival and (B) induced
mutation frequency. The solid lines represent the least squares regression for the

data.

195

0.16 0.20

0.12

0.00

0.04 0.08

 

 

 

 

mm m

32.55 «ceased

 

 

 

.00 0.04 0.08 0.12 0.16 0.20

 

 

300

m.m.w mm
2 2 1 4|

2.8 8326.0 .2 .8. 2.8 .9 88.2..

0
0

BPDE (pl!)

196

mutations, the original parental strain, 9N, was CHAT selected to eliminate the
TG-resistant background. This newly selected, low-background parental strain
was then compared to derivative strains 6| and 12C, which express high levels of
hRev3 antisense, for BPDE-induced cytotoxicity and mutagenicity. As shown in
Figure 5A, neither of the cell strains expressing high levels of hRev3 antisense
differed from their parental strain in sensitivity to the cytotoxic effect of BPDE.
These results are consistent with data obtained previously using 9N.58 as the
control strain, and indicate that hRev3 is not required to protect cells from BPDE-
induced cell killing. However, when cell strains expressing high levels of hRev3
antisense were compared to their appropriate parental strain for their frequency
of BPDE-induced mutations (Figure SB), in contrast to results obtained previously
by Li et al., there were no significant differences in the number of BPDE-induced
mutations. These results indicate that hRev3 is not required for BPDE-induced

mutagenesis in human fibroblasts.

Effect of hRev3 antisense expression on UV-induced cytotoxicity and
mutagenicity-- In our 2002 study, cell strains expressing high levels of hRev3
antisense were also compared to the 9N.58 for their sensitivity to the cytotoxic
and mutagenic effects of UV. Because 9N is the more appropriate control for
such studies, they too were repeated using 9N as the parental strain. Consistent
with our results from 2002, the data presented in Figure 6A show that the hRev3

antisense expressing strains (6| and 120) did not differ from their parental strain

197

Figure 6 - Expression of high levels of hRev3 antisense does not affect the
survival of human fibroblasts exposed to UV radiation, but does cause a
significant reduction in UV-induced mutations. Cell strains expressing hRev3
antisense, designated clones 6| and 12C (closed symbols), and their parental
strain, 9N, (open circles) were treated with BPDE and assayed for (A) cell
survival and (B) induced mutation frequency. The solid lines represent the least

squares regression for the data.

198

14

 

 

 

 

 

 

 

 

 

2
.1
0
.1
.8
.6
.4
.2
o P P P P r F P P P D b P o
0 0 0 0
0% 7 M 3 2 M W m M M 0
1 2 1 1

32:3 .82.... 2.8 8.85.0 .2 .8 385.... .0» 88.2..

UV (Jim?)

199

(9N) in sensitivity to the cytotoxic effect of UV, indicating that hRev3 is not
required to protect cells from UV-induced killing. When these same cell strains
were assayed for their frequency of UV-induced mutations (Figure 63), cell
strains 6| and 12C exhibited considerably lower UV-induced mutation
frequencies than their parental strain. These results are consistent with the
results obtained in our previous study and suggest that, unlike with BPDE, hRev3

is required for mutagenesis induced by UV in human fibroblast cells.

Effect of hRev3 antisense expression on cell cycle progression following
UV or BPDE treatment-- Cell strains with reduced hRev7 progress through the
cell cycle at a considerably slower rate than control cells following treatment with
either BPDE or UV. Consequently, we were interested to determine how
treatment with these agents would affect the cell cycle progression of strains
expressing hRev3 antisense. Therefore, cell strains 6| and 12C, which express

high levels of hRev3 antisense, as well as their parental strain, 9N, were

synchronized at the G1/8 border, released from synchrony, and exposed to

either BPDE or UV. To determine the distribution of cells in each stage of the cell
cycle, populations of each of the three cell strains were harvested every 4 h for
12 h following BPDE or UV exposure, and analyzed by flow cytometry. As shown
in Figure 7, expression of hRev3 antisense had no effect on the progression of
human fibroblasts through the cell cycle after either BPDE treatment (Figure 7A)

or UV treatment (Figure 7B). These results indicate that expressing high hRev3

200

Figure 7 - Expression of high levels of hRev3 antisense has no affect on the
progression of human fibroblasts through the cell cycle after BPDE or UV

treatment. Cell strains expressing high levels of hRev3 antisense (6| and 120)

and their parental cell strain (P) were synchronized at the G1/S border and

treated with BPDE or UV immediately after release from synchrony. Cells were
harvested, fixed, and stained with propidium iodide for analysis of DNA content
by flow cytometry. The distribution of cells in each phase in the cell cycle 0 h, 4 h,

8 h, and 12 h post-BPDE treatment are depicted.

e
WERE

e
WEEE

antisense has no affect on the progression of human fibroblasts through the cell

cycle following BPDE or UV treatment.

203

DISCUSSION:

The results presented in Figure 2A show that cells with reduced expression of
hRev7 protein are more sensitive than the control cell strains to the cytotoxic
effect of BPDE, indicating that the hRev7 protein protects human cells from
BPDE-induced killing. This finding is especially interesting because our
mutagenesis data demonstrate that reducing the level of hRev7 in human
fibroblasts has no affect on the frequency of BPDE-induced mutations in such
strains (Figure 2B), indicating hRev7 is not required for TLS past BPDE-induced
DNA lesions. Taken together, these results indicate that although hRev7 acts to
protect cells from BPDE-induced cell killing, this function is independent of the

role of hRev7 in TLS.

We also found that cell strains with reduced hRev7 are more sensitive to the
cytotoxic effects of cisplatin, ionizing radiation and MNU (Figure 4A-C). What is
more, our laboratory previously demonstrated that reducing the level of hRev7
renders cells more sensitive to the cytotoxic effect of UV radiation [21]. These
results suggest that hRev7 is generally required to protect human fibroblasts
cells from cell killing induced by exposure to DNA damaging agents. Similar
results were reported by Cheung et al. [25], who showed that downregulation of
hRev7 (also referred to as MADZB) in nasopharyngeal carcinoma cells rendered
such cells generally more sensitive to DNA damaging agents, but not to cytotoxic

agents that have different mechanisms of action, such as anti-metabolite or

204

microtubule-disrupting agents. These results indicate that hRev7 plays a role in

preventing cell death caused by DNA damage in both normal and cancer cells.

Analysis of cell cycle progression by flow cytometry revealed that following
exposure to BPDE, cell strains with reduced hRev7 progress through S-phase at
a slower rate than control strains (Figure 3). A similar effect was demonstrated by
previous studies in our laboratory, which showed that human fibroblast cells with
reduced hRev7 also progressed through S-phase more slowly than control
strains following UV-irradiation [21]. Interestingly, expressing high levels of
hRev3 antisense did not alter the rate of cell cycle progression of such strains
following treatment by either UV or BPDE (Figure 7A and 73), indicating that the
requirement for hRev7 in completing S-phase following exposure to DNA
damage is independent of its role as an accessory subunit of hPol C. Moreover,
because hRev7 was not required for TLS past BPDE-induced DNA lesions, we
conclude that it is unlikely that the delay in S-phase we observe in the human
cells with decreased expression of hRev7 resulted from impaired ability to carry

out TLS.

Interestingly, a similar effect has been demonstrated in mouse embryonic
fibroblasts (MEFs) lacking Pol K [26]. Such cells are more sensitive to UV-
induced cytotoxicity than the wild-type cells. In addition, MEFs lacking Pol x
undergo a sustained replication block following treatment with UV [27]. Because

Pol x plays a very small role, if any, in TLS past UV-induced DNA lesions it was

205

unclear why cells lacking this polymerase demonstrate such phenotypes in
response to UV radiation. However, more recent experiments by Ogi and
Lehmann [28], demonstrate that the UV—induced sensitivity of MEFs lacking Pol it
can be attributed to a substantial reduction of nucleotide excision repair activity in
such cells, indicating an unexpected additional role for Pol x in nucleotide
excision repair. Perhaps, similar to what has been demonstrated with Pol x, the
fact that hRev7 is dispensable for TLS past BPDE-induced DNA lesions, but is
required both for efficient progression through S-phase and for survival of BPDE
treated cells, indicates of an additional, as yet uncharacterized, function of

hRev7.

The fact that we found that the accessory subunit of hPol C (hRev7) is
dispensable for BPDE-induced mutagenesis in human fibroblasts was initially
surprising, given that previous studies conducted in our laboratory [20] indicated
a requirement for the catalytic subunit of hPol C (hRev3) in such mutagenesis.
However, as explained above, for that study the control strain used was not the
direct parent of the hRev3 antisense expressing strains. Instead 9N.58 was a
subclone of their actual parental strain, 9N. For reasons yet unknown to us, cell
strain 9N.58 exhibits a much higher frequency of BPDE-induced mutations than
cell strain 9N does (Figure 6B). As a result, when the hRev3 antisense
expressing strains were compared to 9N.58 in the previous experiments [20], the

results were misinterpreted as showing that the frequency of BPDE-induced

206

mutations in cell strains expressing hRev3 antisense were lower than in the

control cell strain.

In our current study, we compared cell strains expressing high levels of hRev3
antisense, 6| and 12C, to the more appropriate control strain, 9N, for the
frequency of BPDE-induced mutations, and found that expressing hRev3
antisense had no affect on the frequency of mutations induced by BPDE in these
cell strains (Figure SB). Although it is still not possible to demonstrate directly that
our hRev3 antisense expressing strains have reduced levels of hRev3 protein
(because of a lack of a suitable antibody), these two cell strains demonstrate
significantly lower UV-induced mutation frequencies than their parental strain
(Figure 7B). Such a result is consistent with having reduced levels of Rev3 [18,
29,30]. Taken together, these data provide evidence that hRev3 is not involved
in TLS past BPDE-induced DNA lesions. Combined with the fact that in the
present study, we also found that hRev7 is not required for TLS past BPDE-
induced DNA lesions, we conclude that hPol C is not required for TLS past

BPDE-induced DNA lesions in human fibroblasts.

207

ACKNOWLEDGEMENTS:

We thank Dr. Kristin McNally of the Laboratory of Persistent Viral Diseases at the
Rocky Mountain Laboratories for generating cells strains 2-2 and 2-6 with
reduced levels of hRev7 protein expresssion. We also thank Dr. Katheryn Meek,
Department of Pathobiology & Diagnostic Investigation at Michigan State
University, for her helpful advice and careful reading of this manuscript. This
research was supported by Grant CA91490 awarded to VMM from the National

Cancer Institute of NIH.

208

REFERENCES:

1. Ling, H., Boudsocq, F., Woodgate, R., and Yang, W. (2001). Crystal
structure of a Y-family DNA polymerase in action: a mechanism for error-
prone and lesion-bypass replication. Cell 107, pp. 91-102.

2. Silvian, L. F., Toth, E. A., Pham, P., Goodman, M. F., and Ellenberger, T.
(2001). Crystal structure of a DinB family error-prone DNA polymerase
from Sulfolobus solfataricus. Nat. Struct. Biol. 8, pp. 984-989.

3. Trincao, J., Johnson, R. E., Escalante, C. R., Prakash, 8., Prakash, L.,
and Aggarwal, A. K. (2001). Structure of the catalytic core of S. cerevisiae
DNA polymerase eta: implications for translesion DNA synthesis. Mol. Cell
8, pp. 417-426.

4. Zhou, B. L., Pata, J. D., and Steitz, T. A. (2001). Crystal structure of a
DinB lesion bypass DNA polymerase catalytic fragment reveals a classic
polymerase catalytic domain. Mal. Cell 8, pp. 427-437.

5. Yang, W., and Woodgate, R. (2007). What a difference a decade makes:
insights into translesion DNA synthesis. Proc. Natl. Acad. Sci. USA 104,
pp. 15591-15598.

6. Nelson, J. R., Lawrence, C. W., and Hinkle, D. C. (1996). Thymine-
thymine dimer bypass by yeast DNA polymerase zeta. Science 272, pp.
1646-1649.

7. Quah, S. K., van Borstel, R. C., and Hastings, P. J. (1980). The origin of
spontaneous mutation in Saccharomyces cerevisiae. Genetics 96, pp.
819-839.

8. Roche, H., Gietz, R. D., and Kunz, B. A. (1994). Specificity of the yeast
rev3 delta antimutator and REV3 dependency of the mutator resulting

from a defect (rad1 delta) in nucleotide excision repair. Genetics 137, pp.
637-646.

9. McKee, R. H., and Lawrence, C. W. (1979). Genetic analysis of gamma-
ray mutagenesis in yeast. I. Reversion in radiation-sensitive strains.
Genetics 93, pp. 361-373.

209

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

McKee, R. H., and Lawrence, C. W. (1979). Genetic analysis of gamma-
ray mutagenesis in yeast. II. Allele-specific control of mutagenesis.
Genetics 93, pp. 375-381.

Prakash, L. (1976). Effect of genes controlling radiation sensitivity on
chemically inducedmutations in Saccharomyces cerevisiae. Genetics 83,
pp. 285—301.

Lawrence, C. W., Nissan, P. E., and Christensen, R. B. (1985). UV and
chemical mutagenesis in rev7 mutants of yeast. Mol. Gen. Genet. 200, pp.
86-91.

Lemontt, J. F. (1972). Induction of forward mutations in mutationally
defective yeast. Mol. Gen. Genet. 119, pp. 27-42.

Ruhland, A., and Brendel, M. . (1979). Mutagenesis by cytostatic
alkylating agents in yeast strains differing in repair capacities. Genetics
92, pp. 83-97.

Lawrence, C. W., and Maher, V. M. (2001). Mutagenesis in eukaryotes
dependent on DNA polymerase zeta and Rev1p. Phil. Trans. R. Soc.
Land. 3 356. Pp. 41-45.

Lawrence, C. W., Das, G., and Christensen, R. B. (1985). REV7, a new
gene concerned with UV mutagenesis in yeast. Mol. Gen. Genet. 200, pp.
80-85.

Murakumo, Y., Roth, T., Ishii, H., Rasio, D., Numata, 8., Croce, C. M., and
Fishel, R. (2000). A human REV7 homolog that interacts with the
polymerase zeta catalytic subunit hREV3 and the spindle assembly
checkpoint protein hMADZ. J. Biol. Chem. 275, pp. 4391-4397.

Gibbs, P. E., McGregor, W. G., Maher, V. M., Nisson, P., and Lawrence,
C. W. (1998). A human homolog of the Saccharomyces cerevisiae REV3
gene, which encodes the catalytic subunit of DNA polymerase zeta. Proc.
Natl. Acad. Sci. USA 95, pp. 6876-6880.

Lawrence, C. W. (2002). Cellular roles of DNA polymerase zeta and Rev1
protein. DNA Repair (Amst) 1, pp. 425-435.

210

20.

21.

22.

23.

24.

25.

26.

27.

Li, Z., Zhang, H., McManus, T. P., McCormick, J. J., Lawrence, C. W., and
Maher, V. M. (2002). hREV3 is essential for error-prone translesion

synthesis past UV or benzo[a]pyrene diol epoxide-induced DNA lesions in
human fibroblasts. Mutat. Res. 510, pp. 71-80.

McNally, K., Neal, J. A., McManus, T. P., McCormick, J. J., and Maher, V.
M. (2008). hRev7, putative subunit of hPolzeta, plays a critical role in
survival, induction of mutations, and progression through S-phase, of
UV(254nm)-irradiated human fibroblasts. DNA Repair (Amst) 7, pp. 597-
604.

Morgan, T. L., Yang, D. J., Fry, D. G., Hurlin, P. J., Kohler, S. K., Maher,
V. M., and McCormick, J. J. (1991). Characteristics of an infinite life span
diploid human fibroblast cell strain and a near-diploid strain arising from a

clone of cells expressing a transfected v-myc oncogene. Exp. Cell Res.
197, pp. 125-136.

Maher, V. M., and McCormick, J. J. (1996). The HPRT gene as a model
system for mutations analysis. in Technologies for Detection of DNA
Damage and Mutations (Pfeifer, G. P., Ed.), pp 381-390, Plenum Press,
New York. .

Wang, Y., Parks, W. C., VIfigle, J. C., Maher, V. M., and McCormick, J. J.
(1986). Fibroblasts from patients with inherited predisposition to
retinoblastoma exhibit normal sensitivity to the mutagenic effects of
ionizing radiation. Mutat. Res. 175, pp. 107-114.

Cheung, H. W., Chun, A. C., Wang, Q., Deng, W., Hu, L., Guan, X. Y.,
Nicholls, J. M., Ling, M. T., Chuan Wang, Y., Tsao, S. W., Jin, D. Y., and
Wang, X. (2006). Inactivation of human MAD2B in nasopharyngeal
carcinoma cells leads to chemosensitization to DNA-damaging agents.
Cancer Res. 66, pp. 4357-4367.

Ogi, T., Shinkai, Y., Tanaka, K., and Ohmori, H. (2002). Polkappa protects
mammalian cells against the lethal and mutagenic effects of
benzo[a]pyrene. Proc. Natl. Acad. Sci. USA 99, pp. 15548-15553.

Bi, X., Slater, D. M., Ohmori, H., and Vaziri, C. (2005). DNA polymerase
Kappa is specifically required for recovery from the benzo[a]pyrene-Di-
hydrodiol epoxide (BPDE)-induced S-phase checkpoint. J. Biol. Chem.

211

28.

29.

30.

Ogi, T., and Lehmann, A. R. (2006). The Y-family DNA polymerase kappa
(pol kappa) functions in mammalian nucleotide-excision repair. Nat. Cell
Biol. 8, pp. 640-642.

Lemontt, J. F. (1971). Mutants of yeast defective in mutation induced by
ultraviolet light. Genetics 68, pp. 21-33.

Diaz, M., Watson, N. B., Turkington, G., Verkoczy, L. K., Klinman, N. R.,
and McGregor, W. G. (2003). Decreased frequency and highly aberrant
spectrum of ultraviolet-induced mutations in the hprt gene of mouse
fibroblasts expressing antisense RNA to DNA polymerase zeta. Mal.
Cancer Res. 1, pp. 836-847.

212

”'lllilljllljljljljl[IljjjlllllllllllllllllllES

62 4583