Decrypting Escherichia coli DNA polymerase V mutasome ATP regulation

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

DECRYPTING ESCHERICHIA COLI DNA POLYMERASE V MUTASOME ATP
REGULATION

by

Dan Danh Vo
______________________

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
MOLECULAR BIOLOGY

May 2018

ii

For my parents, Xang Vo and Hong Thi Nguyen, for your love and sacrifices.

For my wife, Tien, my daughter, Mai-Han, my son, Minh Hieu, and my unborn child,
you will always be the source of my strength and my inspiration.

iii

Acknowledgements

Thank you, Dr. Myron Goodman, for the opportunity to join your lab and for the freedom
to pursue one of the most challenging, yet intellectually stimulating and scientifically fulfilling
projects in trying to characterize pol V Mutasome. I thank you for teaching me that it’s ok to admit
what I don’t know, to ask for help, and to let go of my so-called stubbornness. I am a better scientist
because of it. Thank you, Dr. Chi Mak, and Dr. Fabien Pinaud, for your invaluable inputs during
our many discussions. Chi, you are probably the nicest professor I’ve had the pleasure to meet.
Thank you for your patience and for your smFRET analysis software, that made all the difference.
Thank you, Dr. Keriann Oertell and Aida Bareghamyan, for your friendship, for distracting
me from my work, and for filling those times with fun and genuine laughter.
I would like to thank my parents, Xang Vo and Hong Nguyen, for your unwavering support,
for your love and sacrifices, and most of all, for being shining examples of wonderful parents.
I thank my wife, Tien, for your love, for your sacrifices while I pursue my academic goals,
and for the wonderful family we have and will continue to build. And to my children, who never
fail to put a smile on my face when I come home, I am forever grateful. Thank you!

iv

Table of Contents

Dedication ii
Acknowledgements iii
List of Figures vi
List of Tables vii
Abbreviations viii
Abstract ix

CHAPTER 1: INTRODUCTION 1
CHAPTER 2: SINGLE-MOLECULE CHARACTERIZATION OF POL V 7
MUT’S DNA BINDING AND ACTIVITY
2.1 Introduction 7
2.2 Activated form of pol V Mut is UmuD’2C-RecA-ATP 10
2.3 Binding of pol V Mut to DNA is ATP Dependent 11
2.4 Discussion 13
CHAPTER 3: CONFORMATIONAL REGULATION OF POL V MUTASOME 16
ACTIVITY
3.1 Introduction 16
3.2 ATP Binding Induces pol V Mut Conformational Change via UV- 17
Crosslinking
3.3 Discussion 19
CHAPTER 4: DEACTIVATED POL V MUTASOME LOSES DNA BINDING 22
CAPACITY
4.1 Deactivated pol V Mutasome does not Bind DNA in the Presence of ATP S 22

v

CHAPTER 5: CONCLUSIONS 26
5.1 Conclusions 26

MATERIALS AND METHODS 29
BIBLIOGRAPHY 34

vi

List of Figures

Figure 1.1. E coli DNA damaged SOS Response. 3
Figure 1.2. ATP requirement for pol V Mut DNA binding and activity. 4
Figure 1.3. Pol V Mut is a DNA-dependent ATPase and ATP hydrolysis 5
results in pol V Mut dissociation from DNA.
Figure 2.1. Pol V Mut DNA binding with single-molecule FRET. 9
Figure 2.2. DNA binding of activated pol V Mut. 12
Figure 2.3. Pol V Mut WT is active with ATP only in the presence of    complex. 15
Figure 3.1. ATP and DNA induces conformational rearrangement of pol V Mut. 18
Figure 3.2. Effects of ATP and DNA binding on pol V Mut RecA-UmuC crosslinking. 19
Figure 3.3. smFRET conformational study of dual-labeled pol V Mut. 21
Figure 4.1. Deactivation/reactivation of pol V Mut E38K/ C17. 23
Figure 4.2. Dynamically deactivated pol V Mut E38K/ C17 does not bind 23
DNA in the presence of ATP S.
Figure 4.3. ATP S stabilizes pol V Mut E38K/ C17 and prevents its deactivation. 25

vii

List of Tables

Table 1. pol V Mut binding to p/t DNA 12
Table 2. DNA sequences 29

viii

Abbreviations

TLS: translesion DNA synthesis
E. coli: Escherichia coli
UV: ultra violet
ssDNA: single-stranded DNA
dsDNA: double-stranded DNA
pol: polymerase
Pol II: DNA polymerase II
Pol III: DNA polymerase III
Pol IV: DNA polymerase IV
Pol V: DNA polymerase V
Pol V Mut: DNA polymerase V Mutasome
ATP: adenosine triphosphate
ATP S: adenosine 5′-O-(3-thiothiphosphate)
HP: hairpin
p/t: primer template
dNTPs: deoxyribonucleoside triphosphates
SDS-PAGE: Sodium dodecyl sulfate polyacrylamide gel electrophoresis
WT: wild-type
PU: Primer Utilization
nt: nucleotide
oh: overhang
Pi: Phosphate
RA: rotational anisotropy
pAzpa: p-azido-L-phenylalanine
pBpa: p-benzoyl-L-phenylalanine
PBP: phosphate-binding protein
ix

ABSTRACT

DNA damage repair in E. coli is an exquisitely-regulated process that is divided into two
phases: error-free and mutagenic. Error-free DNA damage repair is initiated early during DNA
damage in a process that involves base excision repair (BER), nucleotide excision repair (NER),
and RecA mediated chromosomal recombination. The mutagenic phase is initiated later, after
error-free response has failed to restart DNA replication at the damaged sites. Activation of the
mutagenic phase is an act of cellular desperation, and genomic mutation is its price. Genomic
mutation during mutagenic DNA repair is largely due to the activity of translesion synthesis DNA
polymerase, pol V (UmuD’2C). Pol V associates with RecA to form a multi-subunit complex called
pol V Mutasome (pol V Mut, UmuD’2C-RecA). Upon activation, pol V Mut can rescue damaged
cells by copying past DNA lesions, including leaving behind unwanted mutations; therefore, the
expression and activity of pol V is strictly regulated. Four levels of regulation are imposed upon
pol V: 1. Temporal, 2. Spatial, 3. Internal, 4. Conformational. This thesis attempts to further
investigate the internal and conformational regulation of pol V Mut by characterizing pol V Mut’s
ATP regulation by visualizing pol V Mut’s DNA binding at the single-molecule level and
investigating the molecular switch that governs E. coli DNA polymerase V Mutasome activity
through inter-subunit interaction between UmuC, UmuD’ and RecA.

CHAPTER 1: INTRODUCTION
1.1 INTRODUCTION
Copying DNA and maintaining genome stability is an important function of living
organisms. Faithful replication of genomic DNA ensures cell survival and continued existence of
future generation. Living organisms have evolved different ways to ensure faithful replication of
their genome. In E. coli, genome stability is accomplished through implementation of multiple
DNA repair pathways initiated during DNA damage[1-3].
Translesion synthesis, a process whereby cells copy pass DNA lesions in order to restart
stalled replication forks occurring as a result of DNA damage, is an important and necessary
pathway to maintain survival of damaged cells. The mechanism of E. coli’s response to DNA
damage was first proposed by Miroslav Radman in 1974 that outlined two phases of E. coli’s SOS
DNA damage response[4]. The first is dominated by error-free mechanisms that include base
excision repair (BER), nucleotide excision repair (NER), RecA mediated chromosomal
recombination, as well as activation of translesion synthesis polymerases II and IV[5-7]. The
second, later-phase of the SOS response is marked by increased mutation of genomic DNA and is
therefore termed the mutagenic phase. Polymerase V (pol V) is largely responsible for the
increased mutation rate. Due to a more open conformation at the enzyme’s active site, pol V can
accommodate bulkier DNA substrates and copy pass lesions that would otherwise block DNA
synthesis at stalled replication forks[8-11]. Pol V lacks an exonuclease subunit that can increase
fidelity; therefore, it is an error-prone DNA polymerase with an error rate of 10
-3
-10
-4
nucleotides
on undamaged DNA[12]. In fact, pol V alone is responsible for about a 100X increase in mutation
above background during UV induced DNA damage[13-15]. For this reason, pol V expression and
activity is tightly regulated at multiple levels to minimize unwanted DNA mutation.
2

Under normal conditions, pol V expression is under strict transcriptional repression by
LexA protein. LexA binds, with differential binding affinity, to the operator regions of the SOS
operon genes and prevents their expression[16-19]. However, under DNA damage, RecA proteins
assemble on single stranded DNA found during DNA damage and form an activated form of RecA,
RecA nucleoprotein filament, or RecA*. RecA* interacts with LexA, resulting in the autocleavage
of LexA and relieves the repression of more than 40 SOS genes[20, 21]. Derepression of SOS
genes is time dependent, based on the binding affinity of LexA to the 20 base pairs (bp) consensus
sequence on the operator regions. The strength of LexA binding to the operator regions determines
how early or late the SOS genes are expressed after UV-damage induction[16-18, 22, 23]. Within
one minute after DNA damage, error-free phase proteins are expressed and the DNA damage repair
pathways are initiated.
3

Figure 1.1. E. coli DNA damaged SOS response.
Under normal conditions, SOS gene expression is under strict transcriptional repression by LexA protein. However, during UV-
induced DNA damage, RecA proteins bind to ssDNA and form a RecA nucleoprotein filament, RecA*. RecA* causes the
autocleavage of LexA, relieving its repression on SOS genes. DNA polymerase II, IV are induced early and pol V are induced late
after DNA damage. The UmuDC gene operon, coding for pol V, produces the UmuC and UmuD proteins. UmuD is further
processed by RecA* to form UmuD′, which associate with UmuC to generate pol V (UmuD′ 2C).

If error-free repair pathways are insufficient to re-initiate stalled replication fork, RecA*-
mediated derepression of SOS continues and at about 40-45 minutes post DNA damage, the
UmuDC operon is derepressed, resulting in the expression of UmuD and UmuC proteins[24]. Umu
stands for UV-induced mutagenesis. RecA* further induces the SOS response by causing the
autocleavage of UmuD to its active form, UmuD’, by removing the first 24 amino acids[25, 26].
UmuD’ forms a homodimer, UmuD’2, and associates with UmuC protein to form E. coli
polymerase V (Figure 1.1). This occurs on the cell membrane[27, 28]. Pol V in its UmuD’2C form
4

is inactive, not until it further interacts with RecA* and removes a RecA monomer from the 3’ tip
of RecA* will it become active. This form of pol V, which consists of RecA-UmuD’2C, is called
pol V Mutasome (pol V Mut)[29].
Recent data by Erdem et al. indicate that pol V Mut needs ATP in order to bind a DNA
primer template (p/t) and initiate deoxyribonucleoside triphosphate (dNTP) incorporation[30].
Without first binding to ATP, pol V Mut is unable to associate with a p/t DNA, rendering the
polymerase essentially inactive (Figure 1.2).

Figure 1.2. ATP requirement for pol V Mut DNA binding and activity. Binding of pol V Mut WT (1 M) (a) and pol V Mut
E38K/ C17 (0.4 M) (b) to 3nt oh HP as measured through change in rotational anisotropy. Activity gel showing nucleotide
incorporation requires the presence of ATP or ATP S. No DNA binding or activity is observed with pol V Mut WT in the presence
of ATP.

In the presence of ATP, pol V Mut binds to DNA and becomes a DNA-dependent ATPase.
The hydrolysis rate of ATP corresponds to the rate of pol V Mut’s dissociation from the DNA
(Figure 1.3). Through the use of an ATPase deficient RecA double mutant (E38K/K72R), Erdem
5

et al. demonstrated the ATPase activity of pol V Mut is independent of the intrinsic RecA ATPase
activity. Therefore, the association of RecA and UmuD’2C to form pol V Mut creates a new ATP
binding pocket that is different from the canonical Walker A motif found in RecA and other ATP
binding sites[31]. This property, which acts to limit enzyme’s processivity, represents a new
regulatory mechanism employed by E. coli to suppress unwanted mutations induced during SOS
DNA repair.

Figure 1.3. Pol V Mut is a DNA-dependent ATPase and ATP hydrolysis results in pol V Mut dissociation from DNA. ATP hydrolysis
by (a) pol V Mut WT compared to RecA WT and (b) pol V Mut E38K/K72R compared to RecA E38K/K72R, was measured at varying
concentrations (0.1 M. 0.2 M, and 0.4 M) of protein. All reactions were carried out in the presence of 30 nt ssDNA (1 M),
ATP (500 M) and MDCC-PBP (5 M). MDCC-PBP fluorescence increases as Pi is released due to ATP hydrolysis. c. Fluorescence
depolarization of fluorescein-labeled 12 nt oh HP DNA was used to measure the dissociation constant of pol V Mut E38K/ C17 in
the presence of ATP (filled circles) and ATP S (open circles). A stable protein-DNA complex (0.4 M and 0.05 M, respectively)
was pre-formed in the presence of nucleotide followed by the addition of excess (160 times) trap DNA (unlabeled 12 nt oh HP).
The decrease in anisotropy over time was fit to an exponential decay to determine koff (0.053 ± 0.0025 s-1) in the presence of
ATP.

Because pol V is an error-prone DNA polymerase, there are multiple layers of regulations
imposed upon its expression and activity to prevent aberrant dNTP incorporation. The four levels
6

of regulation imposed on polymerase V are 1. Temporal 2. Spatial 3. Internal and 4.
Conformational[32, 33]. Temporal regulation refers to the delay in polymerase V expression after
UV irradiation. At about 40-45 minutes post UV induction, UmuD and UmuC proteins are
expressed, post-translationally processed, and assembled into a polymerase V Mut as described
above[27, 28]. Spatial regulation refers to the sequestering of UmuC on the cell membrane upon
its expression. UmuC is insoluble by itself. Not until UmuD is processed to UmuD’, dimerizes,
and associates with UmuC to form pol V (UmuD’2C) will UmuC gets released from the cell
membrane and enters the cytosol. UmuC remains on the cell membrane in E. coli mutants where
UmuD is unable to be processed to UmuD’[27, 28, 34]. Internal regulation refers to the absolute
requirement of ATP for the binding and synthesis of DNA. Also, the hydrolysis of ATP results in
pol V Mut’s dissociation from DNA[30]. This limits the enzyme’s activity. And finally,
conformational regulation refers to the inter-subunit interactions between the UmuC, UmuD’, and
RecA during the formation of active mutasomal complex and through the process of DNA binding
and synthesis[35].
This thesis attempts to further investigate the internal and conformational regulation of pol
V Mut activity by characterizing pol V Mut’s ATP regulation by visualizing pol V Mut’s DNA
binding at the single-molecule level and investigating the molecular switch that governs E. coli
DNA polymerase V Mutasome activity with respect to the intramolecular interaction between pol
V Mut subunits: UmuC, UmuD’ and RecA. Identifying the molecular control of pol V Mutasome
during DNA synthesis will be an important step in piecing together the multiple levels of complex
control of E. coli DNA polymerase V activity.

7

CHAPTER 2: SINGLE-MOLECULE CHARACTERIZATION OF POL V MUT’S DNA
BINDING AND ACTIVITY

2.1 INTRODUCTION
Until recently, characterization of pol V Mut’s DNA binding and activity has been done
using rotational anisotropy and bulk biochemical assays where a high concentration of active
enzyme is required. These ensemble techniques were instrumental in identifying key mechanistic
and regulatory properties of pol V Mut to date; however, they lack the sensitivity needed to identify
the finer regulatory mechanisms employed by polymerase V Mutasome[15, 29, 30, 36]. Antoine
van Oijen, (University of Wollongong, Australia), recently employed single-molecule live-cell
imaging techniques to demonstrate that the UmuC protein is sequestered on the cell membrane,
and only upon interaction with UmuD’ to form UmuD’2C (pol V), does pol V enter the cytosol
and bind to the DNA replicative complex left behind at stalled replication fork after E. coli DNA
polymerase III has skipped ahead to initiate downstream replication[37-40]. Due to the scarcity
of polymerase V protein in living cells before DNA damage, without the sensitivity of single-
molecule live-cell imaging, we would not be able to identify the spatial regulation employed by E.
coli to suppress pol V Mut’s assembly and activity in order to prevent unwanted DNA mutation.
To investigate the different properties of pol V Mut during DNA synthesis, we have
attempted to visualize pol V Mut’s DNA binding and activity through single-molecule Förster
Resonance Energy Transfer (smFRET) experiments. smFRET takes advantage of the
photophysical properties of two fluorophores, and the efficiency of the energy being transferred
between donor and acceptor fluorophores can be used to determine the distance between labeled
sites on two closely associated proteins or protein subunits[41]. FRET efficiency is calculated
8

using the formula E = IA / (ID+IA) where IA and ID represent acceptor and donor emission
respectively. The distance between the two fluorophores is calculated using the formula E = 1 / [1
+ (R / Ro)
6
], where R represents the distance between the fluorophores and Ro is the Förster critical
distance. Ro is defined as the acceptor-donor separation radius for which the transfer rate equals
the rate of donor decay in the absence of acceptor or when E = 50%[41, 42]. The Ro values differ
for each fluorophore pair. For the fluorophore pair being used in this experiment, Alexa Fluor 555
(Alexa-555) and Alexa Fluor 647 (Alexa-647), the Ro is 51 Angstroms (LifeTechnologies).
To study pol V Mut’s DNA binding properties at the single-molecule level, the FRET
donor, Alexa-555, is attached to the biotinylated DNA primer template on the double stranded
region via an amino-linker 6 bases from the 3’OH. FRET acceptor, Alexa-647, is attached via
click-chemistry to the F21 position of RecA subunit on pol V Mutasome using unnatural amino
acid p-azido-L-phenylalanine (pAzpa)[43].
The experiment setup is as follows: Alexa-555-labeled biotinylated primer template (p/t)
DNA is added to streptavidin coated slides and allowed to bind to the slide surface (Figure 2.1a).
Excess DNA is washed away prior to the addition of Alexa-647 acceptor-labeled pol V Mut in the
presence or absence of ATP/ATP S. The experiment is run for 5 minutes with images taken at 100
ms/frame for ATP and 300 ms/frame for ATP S. FRET binding events are observed as counter-
correlated emission signal of the acceptor fluorophore upon donor excitation (see, for example,
Figure 2.1b). Pol V Mut DNA binding events and residence time data are extracted using home-
built smFRET analysis software obtained through our collaboration with Dr. Chi Mak (USC)
(Figure 2.1).
9

Figure 2.1. Pol V Mut DNA binding with single-molecule FRET. a. Sketch of smFRET experimental setup. Alexa-555 donor-labeled
p/t DNA is attached to slide surface via streptavidin-biotin interaction. Alexa-647 acceptor-labeled pol V Mut is then added and
DNA binding is observed as a rise in acceptor fluorophore emission that counter-correlated with a drop in donor emission. b.
Example smFRET trace showing multiple binding and unbinding events of ATP S activated pol V Mut E38K/ C17 (green = donor,
red = acceptor, blue = FRET efficiency). c. FRET efficiency of ATP S activated pol V Mut E38K/ C17 binding to Alexa-555 p/t DNA.
FRET efficiency is calculated by the formula E = IA / (ID+IA) where IA and ID represent acceptor and donor emission respectively
d. Example smFRET images of pol V Mut DNA binding. Alexa-555 p/t DNA is shown in green, Alexa-647 pol V Mut is shown in red.
Binding events are shown as overlap of pol V Mut and p/t DNA in yellow/orange. Pol V Mut alone do not bind to p/t DNA (first
panel). If ATP S is added to pol V Mut and p/t DNA solution, pol V Mut is activated in real-time by ATP S and bind to p/t DNA
marked by arrows (second panel). If pol V Mut is activated by ATP S prior to addition to p/t DNA, this resulted in multiple rapid
binding events as marked by arrows (third panel). Image shown is data integrated over 1 min. after pol V Mut addition (panel 1)
or first binding events (panel 2&3). Scale bar is 1.6 m.

In addition to using ATP and ATP S, a non-hydrolysable or slowly-hydrolysable form of
ATP, we also test pol V Mut’s DNA binding through the use of two forms of pol V Mutasome:
Pol V Mut made with RecA wild-type and Pol V Mut made with RecA E38K/ C17. RecA
E38K/ C17 is a mutant form of RecA that causes constitutive formation of pol V Mut and genomic
DNA mutation even in the absence of DNA damage. Pol V Mut made from RecA E38K/ C17 is
demonstrated to be much more active compared to pol V Mut made from RecA WT in vitro[30].
10

The increased activity is most likely due to the removal of the negatively charged 17 amino acids
at the C terminus of RecA, this leads to enhanced affinity to DNA resulting in the formation of a
better filament, and overall activity of the enzyme[44].

2.2 ACTIVATED FORM OF POL V MUT IS UMUD’2C-RECA-ATP
Activated pol V Mut was shown to be a multi-subunit protein composing of UmuD’2C-
RecA-ATP[29]. Here, using single-molecule Förster Resonance Energy Transfer (smFRET)
technique, where the RecA subunit of pol V Mut is fluorescently labeled with Alexa Fluor 647
(Alexa-647), we demonstrate that pol V Mut does not associate with Alexa-555 labeled p/t DNA
without the addition of a nucleotide cofactor, ATP or ATP S (Figure 2.1 d, left panel). No DNA
binding is seen upon the addition of pol V Mut alone to DNA attached to the surface of a cover
slip. However, when ATP or ATP S is added to the pol V Mut-DNA solution, pol V Mut gets
activated in real time to form an active complex that binds Alexa-555-labeled p/t DNA (Figure 2.1
d, middle panel). A lag time of 30 seconds is observed prior to the first DNA binding event for
real-time activated pol V Mut complex formation with ATP S. This lag time most likely represents
the time required to activate pol V Mut after ATP/ATP S binding. A stable and active form of pol
V Mut can be obtained by first activating pol V Mut with ATP/ATP S prior to addition onto the
slide. Using this form of the protein results in rapid binding of activated pol V Mut to labeled DNA
within 5 seconds post enzyme addition (Figure 2.1 d, right panel). Using both Alexa-647 labeled-
wild type and E38K/ C17 RecA to form pol V Mut, we determine that ATP or ATP S is an
absolute requirement for pol V Mut DNA binding regardless of the form of RecA used to make
11

pol V Mut. This confirms previously published results demonstrating the requirement of an ATP
cofactor for DNA binding and synthesis by pol V Mut in bulk biochemical assays[30].

2.3 BINDING OF POL V MUT TO DNA IS ATP DEPENDENT
At single-molecule resolution, binding of pol V Mut to DNA is dependent upon the bound
ATP molecule. Using ATP and ATP S as well as two forms of RecA for the experiment, we
determine that the residence time of pol V Mut binding to p/t DNA is 6 seconds when activated
with ATP S and 3 seconds when activated with ATP. The DNA residence time of pol V Mut WT
is similar to that of pol V Mut E38K/ C17 for both ATP and ATP S. However, the specific activity
of E38K/ C17 is 5-10X stronger for pol V Mut E38K/ C17 since 5 times more pol V Mut WT
was added in the reaction to observe similar binding events during smFRET experiments (5 nM
vs 1 nM). Similarly, the specific activity of ATP S activated pol V Mut is stronger than that of
ATP activated pol V Mut since a higher protein concentration is needed to obtain similar binding
activity when pol V Mut is activated with ATP (1 nM vs 5 nM) (Figure 2.2). This result explains
increased activity of pol V Mut in the presence of ATP S compared to ATP seen in bulk
biochemical assays. The presence of dNTPs does not stabilize pol V Mut DNA binding as the
residence time remain similar with or without the addition of dNTPs in the reaction (Table 1).
12

Figure 2.2. DNA binding of activated pol V Mut. Residence time of ATP and ATP S activated pol V Mut on Alexa-555-labeled p/t
DNA. Pol V Mut were made with either Alexa-647-labeled RecA E38K/ C17 (a and b) or Alexa-647-labeled RecA WT (c and d).
1nM pol V Mut E38K/ C17 or 5nM pol V Mut WT was added to p/t DNA attached to slide surface with 500 M ATP/ATP S.
Residence time was obtained by fitting the data to a single exponential decay function.

13

Most importantly, in contrast to the results obtained through bulk anisotropy or
biochemical assays, the binding of pol V Mut WT to p/t DNA in the presence of ATP is clearly
observed at the single-molecule level. The low affinity in which pol V Mut WT binds to DNA in
the presence of ATP explains previously published results in which pol V Mut WT was not seen
to interact with DNA or perform DNA synthesis in the same condition. Bulk biochemical assays
were not sensitive enough to observe the sporadic binding of pol V Mut WT to DNA in the
presence of ATP. In addition, we observe pol V Mut’s dissociation from p/t DNA in the presence
of ATP S, this is in sharp contrast with previously published data obtained through rotational
anisotropy experiments showing pol V Mut remaining stably bound to DNA in the presence of
ATP S for at least 4 minutes[30]. Using single-molecule FRET, we move away from ensemble
averaging and are able to extract pol V Mut-DNA kinetic information at the single-molecule level.
This allows us to observe the intricate differences between pol V Mut made with RecA WT and
constitutively active mutant RecA E38K/ C17 as well as their properties when activated using
ATP vs. ATP S.

2.4 DISCUSSION
Using the highly sensitive single-molecule FRET imaging technique, we definitively
proved the absolute requirement of ATP for the activation of pol V Mut. Activated pol V Mut,
with a bound molecule of ATP, is necessary for DNA binding and DNA synthesis. Unlike bulk
biochemical assays and anisotropy, where binding and activity of pol V Mut WT was not seen in
the presence of ATP, we observed DNA binding of ATP activated pol V Mut WT with similar
residence time to that of the constitutively active mutant, pol V Mut E38K/ C17. The main
14

difference between the two forms of pol V Mut resides in the specific activity of the enzyme when
comparing WT to constitutively active mutant, E38K/ C17. In order to observe similar DNA
binding activity, a higher protein concentration of pol V Mut WT is needed. This provides rational
for the decreased DNA synthesis activity seen with pol V Mut WT in vitro.
Lower specific activity of the enzyme is also observed when comparing ATP vs. ATP S
activated pol V Mut. Not only is a higher concentration of enzyme needed in the presence of ATP,
a shorter residence time is also observed, 3 seconds for ATP and 6 seconds for ATP S. Although
the difference between 3 and 6 seconds may seem insignificant at first glance, the DNA binding
data follow an exponential decay, and 24% of the binding events last longer than 10 seconds for
ATP S compared to 1% for ATP when looking at pol V Mut E38K/ C17. In vitro bulk
biochemical data suggest pol V Mut’s dNTP incorporation rate ranges from 22-60
seconds/nucleotide on a 12 nt overhang hairpin DNA. This might be an over-estimation since the
experiment was done by incubating DNA with pol V and RecA* in the presence of ATP S and
dTTP instead of activated pol V Mut. Based on that unpublished observation, we believe only
longer binding events are productive for dNTP incorporation. This further explains the increased
activity of pol V Mut in the presence of ATP S compared to ATP.
Pol V Mut WT is able to bind DNA and perform nucleotide addition in the presence of
ATP if beta processivity clamp ( ) is first loaded onto the DNA (Figure 2.3)[30]. Experiments are
underway to analyze pol V Mut binding to DNA in the presence of  clamp in the single-molecule
FRET setting. The weak DNA binding properties of pol V Mut WT in the presence of ATP can be
interpreted as a regulatory mechanism to prevent error-prone DNA polymerase V from interacting
with naked DNA in the absence of DNA damage. As shown by recent publications, pol III skips
15

over DNA damage site, leaving behind the  clamp to be occupied by translesion synthesis
polymerases pol II/IV/V[37, 40]. This provides another regulatory mechanism to limit access of
pol V Mut to DNA away of DNA damage sites.

Figure 2.3. Pol V Mut WT is active with ATP only in the presence of    complex. Sketch of the experimental set up is illustrated
above the gels. To prevent the  clamp from sliding off the DNA, a 12nt oh HP was designed containing biotin/streptavidin on
both sides of the 3’ OH. The activity of pol V Mut WT was measured in the presence and absence of   . In the presence of   
complex (right panel) pol V Mut WT is able to extend p/t DNA with ATP, in contrast no DNA synthesis is observed with ATP in the
absence of the    complex (left panel). Adapted from Erdem et. al.

In addition, with smFRET, we observed pol V Mut dissociation from p/t DNA in the
presence of ATP S. Previous data suggest pol V Mut form a stable complex with DNA in the
presence of ATP S and the hydrolysis of ATP causes pol V Mut to dissociate from DNA[30]. Our
single-molecule FRET data suggest there are multiple pathways for DNA dissociation since we
observe multiple binding and release events on a single DNA molecule when incubated with
ATP S activated pol V Mut.
16

CHAPTER 3: CONFORMATIONAL REGULATION OF POL V MUTASOME
ACTIVITY
3.1 INTRODUCTION
Inter-subunit interaction of pol V Mutasome plays an important role in its activity. Results
obtained in 2002, demonstrating the formation of an inactive pol V Mutasome upon the addition
of a RecA monomer from the 5’ tip of RecA* illustrates the importance of surface interaction
between the pol V Mut’s subunits[36]. Active vs. inactive conformation of pol V Mut depends on
spatial interaction between RecA and UmuC. Pol V can also associate with free RecA in solution
to form an inactive complex. Only when pol V strips off a RecA monomer from the 3’ tip of RecA*
will an active pol V Mut form[35, 36]. This special interaction of UmuC, UmuD’, and RecA at the
3’ tip of RecA* is further demonstrated when an S117F RecA mutant[45] is used to form pol V
Mut. The S117 position of RecA is oriented at the end surface of the 3’ tip of RecA nucleoprotein
filament. Formation of pol V Mut using S117F RecA mutant results in a catalytically dead pol V
Mut[29]. Since S117F RecA mutant retains all normal RecA properties and can also be used to
form an active homolog mutasome of pol V without any problem, it is the specific interaction
between pol V and the surface containing the S117 position of RecA that result in the formation
of active pol V Mutasomal complex.
For that reason, we incorporated a crosslinkable ligand at the N113 position of RecA via
unnatural amino acid incorporation of p-benzoyl-L-phenylalanine (pBpa) and performed
conformational analysis of pol V Mut via UV crosslinking and Mass Spectrometry analysis[35,
46].
17

Upon activation of the benzophenone with UV light at wavelength of 365nm, the excited
pBpa reacts with unactivated C-H bonds to form a covalent interaction that can be detected via
Western Blot. Crosslinked products can then be isolated and subjected to Mass Spectrometry
analysis to identify peptide-specific interaction between the RecA and pol V subunits of the
Mutasomal complex.

3.2 ATP BINDING INDUCES POL V MUT CONFORMATIONAL CHANGE VIA UV
CROSSLINKING
Pol V Mut (UmuD’2C-RecA) is unable to bind to DNA without ATP. We hypothesize that
the binding of ATP induces a conformational change in pol V Mut and opens a DNA binding
pocket that allows pol V Mut to associate with DNA. In order to prove this hypothesis, pol V Mut
is made using either RecA WT or E38K/ C17, both genetically modified to encode the
crosslinkable ligand, pBpa, at position N113. The presence of a crosslinkable ligand does not
significantly alter pol V Mut’s activity[35]. With N113pBpa position of RecA acting as the anchor
for our experiment, pol V Mut (UmuD’2C-RecA) shows little interaction between UmuC and
RecA. Upon the addition of ATP or ATP S, a significant increase in the amount of UmuC-RecA
crosslinked product is observed via anti-UmuC Western Blot at around 120 kDa. UmuC is 42 kDa
and RecA is 38 kDa; however, due to UV crosslinking, the UmuC-RecA product does not travel
as a linearized protein through SDS-PAGE and is detected at 120 kDa. This increase in UmuC-
RecA crosslinking signifies a conformational switch upon the binding of ATP/ATP S by pol V
Mut that shifts the N113 position of RecA near the surface of UmuC. A second conformational
switch is observed when DNA is added in the presence of ATP/ATP S, resulting in the
18

disappearance of the UmuC-RecA crosslinking band. The second switch is most noticeable with
ATP S due to the increased affinity of pol V Mut to DNA in the presence of ATP S. Using the
constitutively active form of RecA for pol V Mut formation, where it was previously demonstrated
to bind and synthesize DNA more efficiently in the presence of ATP, we observe a 10% decrease
in UmuC-RecA crosslinking when pol V Mut in incubated with ATP + DNA, providing further
evidence for the second conformational switch in pol V Mut upon binding to DNA. The relatively
weak binding of pol V Mut to DNA in the presence of ATP results in minor disappearance of the
UmuC-RecA crosslinking product (Figure 3.1).

Figure 3.1. ATP and DNA induced conformational rearrangement of pol V Mut. Western blot using anti-UmuC antibody
demonstrating pol V Mut conformational switch upon the binding of ATP/ATP S (500μM) and again upon the binding of DNA
(5μM) in the presence of ATP/ATP S. Pol V Mut was generated using a. RecA WT N113pBpa b. RecA E38K/ C17 N113pBpa.

Pol V Mut conformational rearrangement is also concentration dependent. Titration of
ATP S into reaction solution containing pol V Mut results in increased UmuC-RecA crosslinking,
and the subsequent titration of DNA into pol V Mut + ATP S reaction results in decreased UmuC-
RecA crosslinked product after 30-minute incubation (Figure 3.2).
19

Figure 3.2. Effects of ATP and DNA binding on pol V Mut RecA-UmuC cross-linking. Western blot using anti-UmuC antibody
demonstrating increased UmuC-RecA cross-linking with increasing ATP S concentration and decreasing UmuC-RecA cross-linking
with increasing DNA concentration in the presence of 500 μM ATP S. Pol V Mut was generated using RecA N113pBpa, with ATP S
and hairpin DNA added where indicated. ATP S concentration ranges from 0.8 to 500 μM and primer-template DNA concentration
ranges from 0.01– 5 μM (adapted from Gruber et al.).

3.3 DISCUSSION
Multiple conformational rearrangements are observed within pol V Mut as the enzyme
goes through one round of DNA synthesis. At every step of the way, from the formation of pol V
Mut through the interaction of pol V (UmuD’2C) with a 3’ RecA monomer from RecA*, to ATP-
mediated activation, and DNA binding, the inter-subunit interaction between UmuC, UmuD’, and
RecA plays a critical role in determining enzyme’s activity. It is conceivable that even more
conformational rearrangements are happening as pol V Mut binds dNTPs, incorporates
nucleotides, and moves along the DNA prior to dissociation. With UV crosslinking, we are missing
critical conformational information if the crosslinkable ligand, pBpa, is not in close contact with
other pol V subunits. Mass spectrometry allows us to differentiate contact points within a subunit
20

if a crosslinked product is formed; however, if the conformational shift happens in space, where
pBpa is not in close enough proximity to form a crosslinked product with nearby subunits, we are
unable to visualize the changes with our current technique. In order to obtain more sensitive
measurement of pol V Mut conformational rearrangement, smFRET experiments are underway
using a dual-labeled pol V Mut for conformational studies of pol V Mut subunit interaction during
DNA synthesis. Active pol V Mut is formed using Alexa-555-labeled pol V and Alexa-647-labeled
RecA. The donor fluorophore, Alexa-555, is attached at the F1 position of UmuC and acceptor
fluorophore, Alexa-647 is attached at the F21 position of RecA protein (Figure 3.3 a and b). Having
two labels on pol V Mut does not interfere with its ability to incorporate dNTPs onto a DNA primer
template (Figure 3.3 c). Conformational rearrangement of pol V Mut can then be seen using single-
molecule FRET studies where dual-labeled pol V Mut is attached to the slide surface using a
biotinylated antibody against the 6XHis Tag at the N terminal of UmuC. The subsequent addition
of ATP, DNA, and dNTPs will result in conformational rearrangements as pol V Mut binds each
substrate in order to perform DNA synthesis. Using smFRET, we can calculate the relative
distance between the two labeled sites and observe the whole cycle of DNA replication as pol V
Mut gets activated with ATP, binds DNA, incorporates dNTPs, and dissociates from DNA primer
template.
21

Figure 3.3. smFRET conformational study of dual-labeled pol V Mut. a. Active pol V Mut is formed using Alex-555-labeled UmuC
and Alexa-647-labeled RecA or vice versa. 1. Alexa-555 RecA Alexa-647 pol V Mutasome. 2. Alexa-555 RecA 3. Alexa-647 RecA
Alexa-555 pol V Mutasome 4. Alexa-647 RecA. b. Activity of dual-labeled pol V Mut E38K/ C17 was measured with 3nt oh HP in
the presence of ATP or ATP S (500 M) and dNTPs (500 M).

22

CHAPTER 4: DEACTIVATED POL V MUTASOME LOSES DNA BINDING
CAPACITY
4.1 DEACTIVATED POL V MUTASOME DOES NOT BIND DNA IN THE PRESENCE OF
ATP S
Unlike all other DNA polymerases which can perform multiple rounds of DNA replication,
where activity is only limited by substrate availability, pol V Mut, at 37
o
C, can associate with
DNA and perform only one round of DNA synthesis. Pol V Mut becomes deactivated after a single
round of DNA replication and can only be reactivated by incubating the polymerase with fresh
RecA* (Figure 4.1) (Malgorzata Jaszczur, manuscript in preparation). There are two ways to
deactivate pol V Mut: passive deactivation, where pol V Mut is incubated at 37
o
C without
nucleotide cofactors or DNA, and dynamic deactivation, where pol V Mut is allowed to synthesize
DNA at 37
o
C, this results in more rapid deactivation of pol V Mut. To investigate the possible
mechanism of deactivation and properties of deactivated pol V Mut, we performed single-molecule
DNA binding assays using Alexa-647-labeled pol V Mut E38K/ C17 and demonstrated that
dynamically deactivated pol V Mut loses the ability to bind DNA in the presence of ATP S.
Deactivated enzyme cannot be reactivated by the simple addition of ATP S. This corresponds with
the lack of enzymatic activity of deactivated pol V Mut unless it is reactivated by incubating with
new RecA* (Figure 4.2).
23

Figure 4.1. Deactivation/reactivation of pol V Mut E38K/ C17. Activity of pol V Mut E38K/ C17 (200 nM) was measured with
excess of 12 nt oh HP (1 M) in the presence of saturating concentration of ATP S (500 M) and dNTP’s (mix of dTTP, dCTP, dGTP
500 M each). Deactivated pol V Mut was reactivated by adding RecA* (200 nM).

Figure 4.2. Dynamically deactivated pol V Mut E38K/ C17 does not bind DNA in the presence of ATP S. Pol V Mut was made
using Alexa-647 labeled–RecA E38K/ C17. To deactivate protein, 1 μM pol V Mut is incubated with ATP S (500 μM), DNA (5 μM),
and dNTPS (500 μM) for 4 hrs at 37
o
C. a. Activity assay using
32
P-labeled p/t DNA showing dynamically deactivated pol V Mut
failed to incorporate dNTPs after 4 hr incubation (lane 3). Deactivated pol V Mut can be re-activated using fresh RecA* (lane 4).
Active pol V Mut was incubated for 4 hr on ice prior to activity and smFRET binding assay (lane 2). b. Example smFRET images of
pol V Mut DNA binding. Alexa-555 p/t DNA is shown in green, Alexa-647 pol V Mut is shown in red. Binding events are shown as
overlap of pol V Mut and p/t DNA in yellow/orange marked with arrows. Dynamically deactivated pol V Mut failed to bind Alexa-
555 p/t DNA in the presence of additional ATP S (500 M). Bright red dots are pol V Mut crashing on the slide. These do not
colocalize with any p/t DNA. Scale bar is 1.6 m.

To further investigate the possible mechanism of enzyme deactivation, pol V Mut was
made with RecA E38K/ C17 and passively deactivated without nucleotide cofactors or DNA at
37
o
C for 4 hr. The enzyme was also incubated with ATP S +/- DNA at 37
o
C for 4 hr prior to
24

performing single-molecule FRET DNA binding and activity assay. Similar to dynamically
deactivated pol V Mut, passively deactivated pol V Mut fails to bind Alexa-555 labeled-p/t DNA.
From previous results, we know pol V Mut binding to ATP or ATP S results in a conformational
shift that brings the N113 position of RecA closer to UmuC, this conformational shift leads to the
formation of an active pol V Mut complex that can bind and synthesize DNA (Chapter 3). Data
from pol V Mut deactivation experiment in the presence of ATP S suggest the conformational
shift as a result of ATP S binding also prevent pol V Mut from being deactivated when incubated
at 37
o
C. After 4 hr incubation, ATP S activated pol V Mut retains DNA binding and synthesis
activity. On the other hand, in the presence of ATP S and DNA, a large portion of pol V Mut is
deactivated and we observe fewer DNA binding events as well as decreased pol V Mut activity.
After binding to DNA in the presence of ATP S, a second conformational change is observed,
resulting in the deactivation of pol V Mut. This deactivated form of pol V Mut, similar to
dynamically deactivated pol V Mut, no longer binds to DNA, even in the presence of excess
amount of ATP S. All deactivated pol V Mut can be reactivated by incubating with new RecA*
(Figure 4.3). This RecA exchange results in placing the new RecA subunit of pol V Mut in an
orientation that allows pol V Mut to bind a new molecule of ATP S and become activated. In the
presence of ATP S, DNA, dNTPs, and RecA*, pol V Mut can go through many cycles of being
activated by ATP S, bind and synthesize DNA, gets deactivated, and then reactivated by RecA*
to continue the DNA synthesis cycle until the DNA substrate is exhausted. Enzymatic deactivation
can be seen as yet another way for E. coli to limit pol V Mut’s activity in order to prevent unwanted
mutation.

25

Figure 4.3. ATP S stabilizes pol V Mut E38K/ C17 and prevents its deactivation. Pol V Mut was made using Alexa-647 labeled–
RecA E38K/ C17. To deactivate protein, 1 μM pol V Mut is incubated alone (lane 4), + ATP S (500 μM) (lane 6), or with ATP S
(500 μM) + DNA (5 μM) (lane 8) for 4 hr at 37
o
C prior to performing activity and smFRET DNA binding assay. a. Activity assay using
32
P-labeled 12 nt oh HP DNA demonstrating statically deactivated pol V Mut failed to incorporate dNTPs after 4 hr incubation
(lane 4). ATP S protects pol V Mut from being deactivated (lane 6), and pol V Mut loses ATP S protection after DNA binding in
the presence of ATP S. All deactivated pol V Mut can be re-activated by adding fresh RecA* (lane 5,7,9). Active pol V Mut can
synthesize DNA in the presence of ATP (lane 2) or ATP S (lane 3) after 4 hr incubation on ice. b. Example smFRET images of pol V
Mut DNA binding. Alexa-555 p/t DNA is shown in green, Alexa-647 pol V Mut is shown in red. Binding events are shown as overlap
of pol V Mut and p/t DNA in yellow/orange marked with arrows. Similar to results obtained from activity assay, no DNA binding
was seen when pol V Mut is deactivated alone (panel 2) and minimal binding when pol V Mut is deactivated with ATP S + DNA
(panel 4). More pol V Mut crashed on the slide when deactivated with ATP S + DNA as shown by bright red dots. These do not
colocalize with any Alexa-555 p/t DNA. ATP S protects pol V Mut from deactivating and protein is still able to bind DNA after
“deactivation” for 4 hr at 37
o
C (panel 3). DNA binding of active pol V Mut is shown in the first panel. All smFRET binding assay
was done with additional ATP S (500 μM) added prior to addition to Alexa-555 p/t DNA covered slides. Scale bar is 1.6 m.

26

CHAPTER 5: CONCLUSIONS

5.1 CONCLUSIONS
In E. coli, activation of the mutagenic phase of SOS DNA repair is an act of cellular
desperation, and genomic mutation is its price. Since polymerase V is at the epicenter of mutagenic
DNA repair, its expression and activity are strictly regulated. Extensive research over the years
have identified four levels of regulation imposed upon pol V expression and activity. The four
levels of regulation are temporal, spatial, internal, and conformational.
These are broad concepts that act in concert within the organism to limit unwanted
mutation during DNA damage repair, however, within each regulatory concept are many minor
mechanisms that act to prevent pol V enzyme formation and activity. For example, temporal
regulation describes the delay in UmuD and UmuC protein expression after DNA damage, and
spatial regulation describes the sequestering of UmuC to the cell membrane prior to pol V
formation; however, within that also contains other enzymatic processes that act to limit the
availability of UmuD, UmuD’, and UmuC for pol V assembly.
Lon and ClpXP are proteases that are expressed at the same time as UmuD and UmuC. In
order to activate UmuC to form pol V (UmuD’2C), UmuD is post-translationally modified through
its interaction with RecA* to form UmuD’. UmuD’ then dimerizes to UmuD’2 and binds UmuC
to form pol V (UmuD’2C). However, UmuD can also dimerize to form UmuD2 or interact with
UmuD’ to form UmuD-UmuD’. While UmuD’2 is stable in the presence of Lon and ClpXP, both
UmuD2 and UmuD-UmuD’ can be rapidly degraded by ClpXP protease to prevent pol V
formation[47]. UmuD2 and UmuC are also substrates for Lon protease[48, 49]. This limits the
27

availability of mutagenically active UmuD’2 and UmuC for the formation of pol V. Not only is the
expression of UmuD and UmuC under transcriptional repression by LexA protein, once expressed,
their availability is limited by active proteolytic degradation of its intermediates through Lon and
ClpXP proteases.
While internal regulation of pol V Mut refers to the ATP requirement for DNA binding
and ATP hydrolysis for DNA dissociation, pol V Mut WT has very low affinity to DNA when
activated with ATP. In fact, no DNA binding or enzymatic activity was observed when experiment
is done using bulk biochemical assays. DNA binding of ATP activated pol V Mut WT is observed
using highly sensitive single-molecule FRET and activity is observed only when DNA was preload
with beta processivity clamp. These data suggest the low affinity of ATP activated pol V Mut to
DNA is a deliberate act in the cell to prevent pol V Mut from associating with naked DNA. Live-
cell imaging data from 2015 demonstrate that pol V Mut is recruited to DNA damage sites and
associates with  clamp that was left behind after pol III skips ahead and initiate downstream DNA
synthesis[37, 40]. This ensures pol V Mut’s activity is specifically targeted to the DNA damage
site and not naked DNA where the pol V Mut can only leave behind unwanted mutation without
the desired lesion bypass.
We’ve only scratched the surface in trying to understand pol V Mut activity. With single-
molecule imaging techniques, we can visualize lesion bypass by pol V Mut with and without the
 clamp to understand how pol V Mut incorporates a nucleotide across DNA lesions. Also, with
single-molecule FRET, using a dual-labeled pol V Mut complex, we can understand the molecular
control during the activation and deactivation of pol V Mut, as well as dynamic subunit movement
of pol V Mut during DNA synthesis.
28

Another aspect of pol V Mut regulation that can be studied using highly sensitive single-
molecule imaging techniques is the switching of translesion synthesis DNA polymerase on
damaged DNA. Until recently, the standing question is how translesion synthesis polymerases like
pol V, IV, and II switch place with pol III at DNA damaged site; however, with recent data
demonstrating that pol III skips ahead and initiate DNA synthesis downstream of the lesion,
leaving behind  processivity clamp to be occupied by translesion synthesis polymerase, the
question becomes how does translesion synthesis polymerases switch from the relatively error-
free pol II/IV to mutagenically active pol V[37, 40]. One can study this process by using two sets
of donor-acceptor fluorophores in a single-molecule FRET experiment to track the interaction of
labeled-pol II/IV and labeled-pol V on a single DNA molecule loaded with  processivity
clamp[50]. This will enable us to understand the decision process of when a cell chooses to use
pol II/IV vs. mutagenically active pol V.
Although much work has been done in trying to understand how pol V Mut is regulated
and perform DNA lesion bypass, we now have the proper tools necessary to dive deeper and to
visualize every step of the enzymatic process.

29

MATERIALS AND METHODS

Proteins
His-tagged pol V was purified from E. coli strain RW644 as described by Karata et al. and
RecA WT and RecA E38K/ C17 was purified by standard protocol[35, 51, 52]. In order to
incorporate unnatural amino acid p-azido-L-phenylalanine (pAzpa) and p-benzoyl-L-
phenylalanine (pBpa) into RecA protein, we employ Peter Schultz’s lab method in which an
Amber codon was inserted at the F21 or N113 position on RecA. Protein expressing E. coli strain
(BLR) is then co-transformed with plasmid containing the Amber codon modified RecA along
with a plasmid encoding for the orthogonal tRNA synthethase that recognize the Amber codon
and insert the modified amino acid at the Amber stop site. RecA is induced by shaking with
30

unnatural amino acid p-azido-L-phenylalanine (pAzpa) or p-benzoyl-L-phenylalanine (pBpa)
(Bachem) and L-Arabinose plus IPTG overnight at 16
o
C[43, 46, 53].
pAzpa and pBpa substituted RecA protein was purified using the same standard protocol for
RecA WT and RecA E38K/ C17[52].

Preparing Cyanogen-Bromide Sepharose resin
50 mg Cyanogen-Bromide Sepharose resin (Sigma-Aldrich) is first activated in 5 ml of 1
mM HCl, then washed with ddH20 and equilibrated in coupling buffer (0.1 mM NaHCO3, 0.5 M
NaCl, pH 8.3). 5 nmole 5’NHS-modified 45mer ssDNA is then added and rotated overnight at 4
o
C
to enable resin-ssDNA formation. Resin is then washed with coupling buffer to remove unbound
ssDNA. Unreacted cyanate esters are blocked with 1 M Ethanolamine pH 8.0 (Sigma-Aldrich)
overnight at 4
o
C. Resin is then washed by alternating coupling buffer and acetate buffer (0.1 M
CH3COONa, 0.5 M NaCl, pH 4.0) 5X prior to storing in 1M NaCl. Typically, about 1 nmole of
ssDNA is bound to 50 mg resin.

Assembling of pol V Mutasome
Pol V Mut is assembled first by generating RecA nucleoprotein filament on single stranded
DNA attached to Cyanogen-Bromide Sepharose resin (Sigma-Aldrich). RecA is added to
Sepharose beads containing 5’ NHS-modified 45nt ssDNA in the presence of excess ATP S and
incubated at 37
o
C for 15 min in 1X reaction buffer (20mM Tris-HCl pH 7.5, 25 mM Sodium
Glutamate, 8 mM MgCl2, 5 mM DTT, 4% glycerol, 0.1 mM EDTA). Following incubation,
31

unbound RecA and ATP S is washed extensively via centrifugation of Bio-Rad spin column at 0.1
g for 1 minute. Polymerase V is then added and incubated at 37
o
C for 15 min to allow pol V to
strip a single RecA protein from the 3’ tip of the RecA nucleoprotein filament to form pol V Mut.
The pol V Mut is obtained in the flow through following centrifugation in the form of UmuD’2C-
RecA[30].

Activity Assay
For activity assay, pol V mutasome, UmuD’2C-RecA is incubated in the presence of 5’
32
P-
labeled DNA, dNTPs (500 uM) and either ATP S (500 uM) or ATP (500 uM) for 30 min at 37
o
C
in 1X polymerase reaction buffer (20mM Tris-HCl pH 7.5, 25 mM Sodium Glutamate, 8 mM
MgCl2, 5 mM DTT, 4% glycerol, 0.1 mM EDTA). Stop solution containing 50 mM EDTA and
90% formamide is added to terminate the polymerase reaction. Nucleotide addition is visualized
by resolving
32
P-labeled DNA on 20% polyacrylamide gel prior to detection and quantification of
gel band intensity by phosphor-imaging and IMAGEQUANT software respectively.

Labeling and purification of pAzpa substituted RecA
To attach a fluorescent label onto Azido-phenylalanine modified RecA, F21pAzpa RecA
is incubated with Alexa Fluor dye (ThermoFisher Scientific) at a ratio of 1:5 and rotated at 4
o
C
overnight. Following incubation, dye and protein mixture is loaded onto a Ceramic Hydroxyapatite
(Bio-Rad) column and labeled protein is eluted by running a phosphate gradient from 0 to 0.25M
potassium phosphate. Unbound dye is eluted through the wash, Alexa Fluor labeled RecA is eluted
32

early during the gradient and unlabeled RecA comes out in the later part of the phosphate gradient.
Protein concentration and labeling efficiency is determined via spectrophotometry.

Single-molecule FRET experiment
To perform single-molecule FRET imaging experiments, high precision microscope glass
coverslips (Marienfeld, #1.5, Ø25 mm) were first cleaned by sonicating in ddH20 for 1 minute
followed by further sonication with 100 mM KOH for 20 minutes. Coverslips were then cleaned
in Piranha solution of 1:3 hydrogen peroxide to sulfuric acid for 5 minutes followed by sonication
in ddH20 for 10 min to clean off Piranha residues. 3-Aminopropyltriethoxysilane (Sigma-Aldrich)
is then attached to the slides prior to overnight incubation with Polyethylene glycol (PEG)/Biotin-
PEG-SVA (Laysan Bio). A second round of pegylation was done for 2 hrs using 4-Methyl-PEG-
NHS-Ester (Thermo-Fisher) to reduce background fluorescence[54].
On the day of single molecule FRET experiment, coverslips were washed with ddH20 prior
to incubating with Streptavidin for 10 min in buffer containing 20 mM Tris-HCl PH 7.5 and 50
mM NaCl. Coverslips were coated with 20 pM Alexa 555-labeled primer template DNA for 5
minutes. Unbound DNA were washed in imaging buffer prior to the addition of pol V Mut made
with F21-Alexa 647-labeled RecA WT or E38K/ C17.
Imaging buffer contain 1X reaction buffer + 50ug/ml BSA, 2mM Trolox, 10 mM PCA,
and 100 nM PCD[55].

33

UV crosslinking experiment
For UV crosslinking experiment, WT or E38K/ C17 RecA modified with pBpa at the
N113 position were used for the formation of pol V Mut. Pol V Mut was incubated with ATP/
ATP S and/or DNA substrate for 30 minutes at 37
o
C prior to UV exposure at 365nm wavelength
for 60 minutes with gentle mixing at 15-minute interval. Crosslinked product is boiled in protein
loading dye and separated using 12% PAGE. UmuC-RecA crosslinked bands are detected using
affinity-purified anti-UmuC antibody in standard Western Blot.

34

BIBLIOGRAPHY

1. Witkin, E.M., Ultraviolet mutagenesis and inducible DNA repair in Escherichia coli. Bacteriol Rev,
1976. 40(4): p. 869-907.
2. Devoret, R., et al., Recovery of phage lambda from ultraviolet damage. Basic Life Sci, 1975. 5A: p.
155-71.
3. Wyrick, J.J. and S.A. Roberts, Genomic approaches to DNA repair and mutagenesis. DNA Repair
(Amst), 2015. 36: p. 146-55.
4. Wagner, R.E., Jr. and M. Radman, A mechanism for initiation of genetic recombination. Proc Natl
Acad Sci U S A, 1975. 72(9): p. 3619-22.
5. Friedberg, E.C., A history of the DNA repair and mutagenesis field: The discovery of base excision
repair. DNA Repair (Amst), 2016. 37: p. A35-9.
6. Friedberg, E.C., A history of the DNA repair and mutagenesis field: I. The discovery of enzymatic
photoreactivation. DNA Repair (Amst), 2015. 33: p. 35-42.
7. Friedberg, E.C., The molecular basis of mutagenesis - 40 years of research on genomic integrity. S
Afr Med J, 2013. 103(12 Suppl 1): p. 982-3.
8. Ling, H., et al., Crystal structure of a Y-family DNA polymerase in action: a mechanism for error-
prone and lesion-bypass replication. Cell, 2001. 107(1): p. 91-102.
9. Ohmori, H., et al., The Y-family of DNA polymerases. Mol Cell, 2001. 8(1): p. 7-8.
10. Yang, W., Portraits of a Y-family DNA polymerase. FEBS Lett, 2005. 579(4): p. 868-72.
11. Yang, W. and R. Woodgate, What a difference a decade makes: insights into translesion DNA
synthesis. Proc Natl Acad Sci U S A, 2007. 104(40): p. 15591-8.
12. Tang, M., et al., Roles of E. coli DNA polymerases IV and V in lesion-targeted and untargeted SOS
mutagenesis. Nature, 2000. 404(6781): p. 1014-8.
13. Pham, P., et al., A model for SOS-lesion-targeted mutations in Escherichia coli. Nature, 2001.
409(6818): p. 366-70.
14. Jarosz, D.F., et al., Y-family DNA polymerases in Escherichia coli. Trends Microbiol, 2007. 15(2): p.
70-7.
15. Schlacher, K., et al., DNA polymerase V and RecA protein, a minimal mutasome. Mol Cell, 2005.
17(4): p. 561-72.
16. Little, J.W., Autodigestion of lexA and phage lambda repressors. Proc Natl Acad Sci U S A, 1984.
81(5): p. 1375-9.
17. Little, J.W., et al., Cleavage of the Escherichia coli lexA protein by the recA protease. Proc Natl Acad
Sci U S A, 1980. 77(6): p. 3225-9.
18. Luo, Y., et al., Crystal structure of LexA: a conformational switch for regulation of self-cleavage.
Cell, 2001. 106(5): p. 585-94.
19. Volkert, M.R., D.L. George, and E.M. Witkin, Partial suppression of the LexA phenotype by
mutations (rnm) which restore ultraviolet resistance but not ultraviolet mutability to Escherichia
coli B/r uvr A lexA. Mutat Res, 1976. 36(1): p. 17-28.
20. Courcelle, J., A.K. Ganesan, and P.C. Hanawalt, Therefore, what are recombination proteins there
for? Bioessays, 2001. 23(5): p. 463-70.
21. Fernandez De Henestrosa, A.R., et al., Identification of additional genes belonging to the LexA
regulon in Escherichia coli. Mol Microbiol, 2000. 35(6): p. 1560-72.
35

22. Little, J.W., Isolation of recombinant plasmids and phage carrying the lexA gene of Escherichia coli
K-12. Gene, 1980. 10(3): p. 237-47.
23. Walker, G.C., Mutagenesis and inducible responses to deoxyribonucleic acid damage in Escherichia
coli. Microbiol Rev, 1984. 48(1): p. 60-93.
24. Bagg, A., C.J. Kenyon, and G.C. Walker, Inducibility of a gene product required for UV and chemical
mutagenesis in Escherichia coli. Proc Natl Acad Sci U S A, 1981. 78(9): p. 5749-53.
25. Shinagawa, H., et al., RecA protein-dependent cleavage of UmuD protein and SOS mutagenesis.
Proc Natl Acad Sci U S A, 1988. 85(6): p. 1806-10.
26. Blanco, M., et al., Influence of RecA protein on induced mutagenesis. Biochimie, 1982. 64(8-9): p.
633-6.
27. Robinson, A., et al., Regulation of Mutagenic DNA Polymerase V Activation in Space and Time.
PLoS Genet, 2015. 11(8): p. e1005482.
28. Ghodke, H., H. Ho, and A.M. van Oijen, Single-molecule live-cell imaging of bacterial DNA repair
and damage tolerance. Biochem Soc Trans, 2018. 46(1): p. 23-35.
29. Jiang, Q., et al., The active form of DNA polymerase V is UmuD'(2)C-RecA-ATP. Nature, 2009.
460(7253): p. 359-63.
30. Erdem, A.L., et al., DNA polymerase V activity is autoregulated by a novel intrinsic DNA-dependent
ATPase. Elife, 2014. 3: p. e02384.
31. Walker, J.E., et al., Distantly related sequences in the alpha- and beta-subunits of ATP synthase,
myosin, kinases and other ATP-requiring enzymes and a common nucleotide binding fold. EMBO
J, 1982. 1(8): p. 945-51.
32. Goodman, M.F., et al., Insights into the complex levels of regulation imposed on Escherichia coli
DNA polymerase V. DNA Repair (Amst), 2016. 44: p. 42-50.
33. Jaszczur, M., et al., Mutations for Worse or Better: Low-Fidelity DNA Synthesis by SOS DNA
Polymerase V Is a Tightly Regulated Double-Edged Sword. Biochemistry, 2016. 55(16): p. 2309-18.
34. Koch, W.H., et al., Escherichia coli umuDC mutants: DNA sequence alterations and UmuD cleavage.
Mol Gen Genet, 1992. 233(3): p. 443-8.
35. Gruber, A.J., et al., A RecA protein surface required for activation of DNA polymerase V. PLoS
Genet, 2015. 11(3): p. e1005066.
36. Schlacher, K., et al., RecA acts in trans to allow replication of damaged DNA by DNA polymerase
V. Nature, 2006. 442(7105): p. 883-7.
37. Gabbai, C.B., J.T. Yeeles, and K.J. Marians, Replisome-mediated translesion synthesis and leading
strand template lesion skipping are competing bypass mechanisms. J Biol Chem, 2014. 289(47): p.
32811-23.
38. Heller, R.C. and K.J. Marians, Replisome assembly and the direct restart of stalled replication forks.
Nat Rev Mol Cell Biol, 2006. 7(12): p. 932-43.
39. Marians, K.J., Understanding how the replisome works. Nat Struct Mol Biol, 2008. 15(2): p. 125-7.
40. Yeeles, J.T. and K.J. Marians, Dynamics of leading-strand lesion skipping by the replisome. Mol
Cell, 2013. 52(6): p. 855-65.
41. Roy, R., S. Hohng, and T. Ha, A practical guide to single-molecule FRET. Nat Methods, 2008. 5(6):
p. 507-16.
42. Joo, C. and T. Ha, Single-molecule FRET with total internal reflection microscopy. Cold Spring Harb
Protoc, 2012. 2012(12).
43. Chin, J.W., et al., Addition of p-azido-L-phenylalanine to the genetic code of Escherichia coli. J Am
Chem Soc, 2002. 124(31): p. 9026-7.
44. Lavery, P.E. and S.C. Kowalczykowski, Biochemical basis of the constitutive repressor cleavage
activity of recA730 protein. A comparison to recA441 and recA803 proteins. J Biol Chem, 1992.
267(29): p. 20648-58.
36

45. Dutreix, M., et al., New recA mutations that dissociate the various RecA protein activities in
Escherichia coli provide evidence for an additional role for RecA protein in UV mutagenesis. J
Bacteriol, 1989. 171(5): p. 2415-23.
46. Chin, J.W., et al., Addition of a photocrosslinking amino acid to the genetic code of Escherichiacoli.
Proc Natl Acad Sci U S A, 2002. 99(17): p. 11020-4.
47. Gonzalez, M., et al., Subunit-specific degradation of the UmuD/D' heterodimer by the ClpXP
protease: the role of trans recognition in UmuD' stability. EMBO J, 2000. 19(19): p. 5251-8.
48. Frank, E.G., et al., Regulation of SOS mutagenesis by proteolysis. Proc Natl Acad Sci U S A, 1996.
93(19): p. 10291-6.
49. Gonzalez, M., et al., Lon-mediated proteolysis of the Escherichia coli UmuD mutagenesis protein:
in vitro degradation and identification of residues required for proteolysis. Genes Dev, 1998.
12(24): p. 3889-99.
50. Lee, J., et al., Single-molecule four-color FRET. Angew Chem Int Ed Engl, 2010. 49(51): p. 9922-5.
51. Karata, K., et al., Simple and efficient purification of Escherichia coli DNA polymerase V: cofactor
requirements for optimal activity and processivity in vitro. DNA Repair (Amst), 2012. 11(4): p. 431-
40.
52. Cox, M.M., K. McEntee, and I.R. Lehman, A simple and rapid procedure for the large scale
purification of the recA protein of Escherichia coli. J Biol Chem, 1981. 256(9): p. 4676-8.
53. Young, T.S., et al., An enhanced system for unnatural amino acid mutagenesis in E. coli. J Mol Biol,
2010. 395(2): p. 361-74.
54. Chandradoss, S.D., et al., Surface passivation for single-molecule protein studies. J Vis Exp,
2014(86).
55. Aitken, C.E., R.A. Marshall, and J.D. Puglisi, An oxygen scavenging system for improvement of dye
stability in single-molecule fluorescence experiments. Biophys J, 2008. 94(5): p. 1826-35.

Asset Metadata

Creator Vo, Dan Danh (author)

Core Title Decrypting Escherichia coli DNA polymerase V mutasome ATP regulation

Contributor Electronically uploaded by the author (provenance)

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Molecular Biology

Publication Date 05/01/2018

Defense Date 03/20/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag ATP,E. coli,OAI-PMH Harvest,polymerase V,RecA,single-molecule FRET

Format application/pdf (imt)

Language English

Advisor Goodman, Myron F. (committee chair), Mak, Chi H. (committee member), Pinaud, Fabien F. (committee member)

Creator Email danvo@ucla.edu,danvo84@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-498034

Unique identifier UC11268192

Identifier etd-VoDanDanh-6294.pdf (filename),usctheses-c40-498034 (legacy record id)

Legacy Identifier etd-VoDanDanh-6294.pdf

Dmrecord 498034

Document Type Dissertation

Format application/pdf (imt)

Rights Vo, Dan Danh

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Abstract (if available)

Abstract DNA damage repair in E. coli is an exquisitely‐regulated process that is divided into two phases: error‐free and mutagenic. Error‐free DNA damage repair is initiated early during DNA damage in a process that involves base excision repair (BER), nucleotide excision repair (NER), and RecA mediated chromosomal recombination. The mutagenic phase is initiated later, after error‐free response has failed to restart DNA replication at the damaged sites. Activation of the mutagenic phase is an act of cellular desperation, and genomic mutation is its price. Genomic mutation during mutagenic DNA repair is largely due to the activity of translesion synthesis DNA polymerase, pol V (UmuD’₂C). Pol V associates with RecA to form a multi‐subunit complex called pol V Mutasome (pol V Mut, UmuD’₂C-RecA). Upon activation, pol V Mut can rescue damaged cells by copying past DNA lesions, including leaving behind unwanted mutations