University of Southern California Dissertations and Theses

Investigating the E. Coli DNA polymerase V mutasome

(USC Thesis Other)

Investigating the E. Coli DNA polymerase V mutasome

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 1
Investigating the E. Coli DNA Polymerase V Mutasome

by

Meghna Patel
———————————

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

May 2013

Copyright 2013 Meghna Patel

2

For Sumit,
Everything.
3
ACKNOWLEDGEMENTS
I would like to thank the members of my committee, Dr. Fabien Pinaud, Dr.
Hanna Reisler, and Dr. Myron Goodman for serving on my committee and overseeing
my pursuit of some very exciting science. Myron, I especially thank you for instilling in
me the belief that I am a peer, always. Your guidance has been vital to my development. I
am forever thankful to Drs. Jeff Bertram, Malgorzata Jaszczur, and Katharina Schlacher,
who taught me what reason, intellect, and brilliance look like in practice. I am grateful for
having been mentored by you. I would like to thank Dr. Thomas Upton, for sharing in my
vision for this project and for your constant optimism and excitement about the work.
Working with you taught me much about what productive, collaborative work should be.
I would like to thank Aysen Erdem for the immense amount of work you put into this
project.
I would like to thank Drs. Michael Cox and Roger Woodgate. Your breadth of
knowledge in your respective fields deepened my understanding of the science and
enabled me to think about the problem from different perspectives. I am greatly indebted
to Drs. Antoine van Oijen and Andrew Robinson. Working with you has been a genuine
pleasure and has solidified in my mind what great science is.
I would like to thank my family for supporting me through this journey. I couldn’t
have gotten through it without your smiles or love. You are all a constant reminder of
what is truly important in my life. I am very thankful to my mother and father for all the
sacrifices they made so that I would be afforded the opportunities I have been. I will
continue to honor you by living a life that, hopefully, will make you proud. I am grateful
4
to my wonderful female scientist friends: Drs. Rebecca Nugent, Laura Sanders, Alison
Kraigsley, and Oriana Bretschinger. I am privileged to know you. You ladies give me
hope for the future of women in science. I would especially like to thank one of my
dearest friends, Dr. Qingfei Jiang, for feeding me, housing me, and for her overall
support. You enlighten me every time we’re together and I am in constant amazement at
your generosity of self and spirit. Lastly I would like to thank my husband, Sumit Yadav.
I would not be here without you. Thank you for your un-ending and un-conditional
support and love. If I become a little more like you with each passing day, at the end of
days, I will consider myself truly endowed.

5
TABLE OF CONTENTS
Dedication 2
Acknowledgements 3
List of Tables 7
List of Figures 8
Abbreviations 10
Abstract 12

CHAPTER 1: INTRODUCTION 13
1.0 Introduction 14

CHAPTER 2: MODEL TESTING VIA UN-NATURAL AMINO ACID
INCORPORATION FOR CROSSLINKING AND FLUORESCENT STUDIES 25
2.0 Introduction 26
2.1 Results 32
2.1.1 Photocroslinking Studies 32
2.1.2 Fluorescence Studies 40
2.2 Discussion and Future Directions 43
2.3 Experimental Procedures 46
2.3.1 Cloning, Expression, and Purification 46
2.3.2 Fluorescent Labeling 48
2.3.3 Pol V Mut 49
2.3.4 Photocroslinking 50
2.3.5 FRET 51

CHAPTER 3: SINGLE MOLECULE IN VIVO VISUALIZATION OF REPLICATING
CELLS WITH DNA DAMAGE 52
3.0 Introduction 53
3.1 Results 55
3.1.1 Disappearance of the ε Subunit of Polymerase III upon Induction of
Polymerase V 55
3.1.2 Pol V Location in Live Cells 59
3.2 Discussion and Future Directions 63
3.3 Experimental Procedures 65
3.3.1 Strain Construction and Single Molecule Imaging 65
3.3.2 In Vitro Exonuclease Protection Assay 66
3.3.3 Rifampicin Mutagenesis 66

6
CHAPTER 4: TOWARDS A POLYMERASE V CRYSTAL STRUCTURE 68
4.0 Introduction 69
4.1 Results/Methods 71
4.1.1 Improvements on Purification to Increase Yield 71
4.1.2 Crystal Screens 74
4.2 Discussion and Future Directions 76

BIBLIOGRAPHY 77
7

TABLES
Table 2.1 RecA p-Benzoyl-Phenylalanine Substitution Crosslinking Results 39
Table 4.1 Co-Concentration Conditions For Pol V 73

8
FIGURES
Figure 1.1 Early Translesion Synthesis (TLS) Models 16
Figure 1.2 Kinetics of Primer/Template Annealing 19
Figure 1.3 Kinetics of Pol V Transactivation by RecA–ssDNA 20
Figure 1.4 The Pol V Mutasome 22
Figure 1.5 Model of Pol V Mut Function 23
Figure 2.1 Un-natural Amino Acid Incorporation 28
Figure 2.2 RecA-RecA Crosslinking 33
Figure 2.3 Enzymatic Activity of p-Benzoyl-Phenylalanine (pBF) RecA
Substitutions 34

Figure 2.4 Crosslinking with RecA
K106pBF
* 35
Figure 2.5 Crosslinking with RecA
K317pBF
36
Figure 2.6 Crosslinking with RecA
F21pBF
37
Figure 2.7 RecA
S117pBF
Activity and Crosslinking 38
Figure 2.8 Emission Curves for Pol V Mut 42
Figure 3.1 Polymerase III Epsilon Disappearance Upon UV Induction of
Pol V 56

Figure 3.2 Inhibition of Exonuclease Activity of Pol III Core By UmuD
2

and UmuD′
2
57

Figure 3.3 Epsilon Degradation is Dependent on Lys97 in UmuD/D′ 58
Figure 3.4 UV Mutagenesis of Strains 59
Figure 3.5 UV-induced Pol V Foci Localize at the Cell Membrane 60
Figure 3.6 UmuC-mKate2 Localization in Different Genetic Backgrounds 61

9
Figure 3.7 UmuC-mKate2 Fluorescent Spots Characterized by Location
in Cell 62

Figure 4.1 Crystals and Diffraction Patterns 75

10
ABBREVIATIONS USED
ATP: adenosine triphosphate
ATP S: adenosine 5’-O-(3-thiothiphosphate)
dNTP: deoxynucleotide
DTT: dithiothreitol
dsDNA: double-stranded DNA
E. coli: Escherichia. Coli
FRET: Fӧrster resonance energy transfer
HAP: hydroxyapatite resin
hp: Hairpin DNA substrate
LB: Luria-Broth
nt: nucleotide
MALS: Multi-angle light scattering
mer: oligomer
oh: overhang
p/t: primer/template
Pol III: DNA polymerase III
Pol IV: DNA polymerase IV
Pol V: DNA polymerase V
Pol V Mut: DNA polymerase V mutasome
RecA*: RecA nucleoprotein filaments
Rif: Rifampicin
11
sm: single molecule
ssDNA: single-stranded DNA
SSB: single-stranded DNA binding protein
TLS: translesion DNA synthesis
unAA: un-natural amino acid
UV: ultraviolet
WT: wild-type

12

ABSTRACT
In Escherichia Coli, cell survival and genomic stability after UV radiation or other DNA
damaging agents depends on repair mechanisms induced as part of the SOS response.
The SOS response involves upregulation of over 40 genes, which are normally repressed
by the LexA protein under non-damage conditions. The early phase of the SOS response
is centered around error-free DNA repair mechanisms. These include base excision repair
(BER), nucleotide excision repair (NER), and recombinational repair. If these
mechanisms are insufficient to surmount the replicational impedance brought about by
DNA damage, the bacterial cell advances into the later part of the SOS response. This
later phase of the response helps ensure cell survival albeit at the cost of faithful DNA
repair. Known as SOS mutagenesis, this phase is characterized by elevated mutation
levels, which are a direct result of error-prone DNA replication. SOS mutagenesis is the
result of the low fidelity DNA polymerase V. DNA polymerase V (Pol V) is a
heterotrimeric protein composed of the catalytically active UmuC and a dimer of UmuD 
.
The heterotrimer is essentially inactive, however, and must strip a RecA monomer and an
ATP cofactor from the 3  tip of a RecA-DNA nucleoprotein filament to form an active
Pol V mutasome composed of UmuC:UmuD 
2
:RecA:ATP. This dissertation focuses on
investigating the Pol V mutasome including deducing protein-protein interactions in the
active, inactive, and de-activated states of the mutasome, detecting a conformational
change in the different activity states of the mutasome, visualizing the SOS response in
live cells, and attempting to ascertain the crystal structure of Pol V.
13

CHAPTER 1
INTRODUCTION
14
Chapter 1.0 Introduction
The beginnings of the SOS response system were gleaned in the early 1950’s
when Jean Weigle observed that if UV-irradiated bacteriophage  infected UV-irradiated
E. coli, this increased the survival rates of the phage (Weigle, 1953). This survival was
not observed, however, if photoreactivation had already occurred in the host cells prior to
phage infection. This implied that photoproducts in the host DNA were necessary for the
survival of the irradiated phage. Weigle also noticed that a large number of the
reactivated bacteriophage contained mutations in their DNA. This was the first
suggestion that cells were able to mount a physiological mechanism to respond to DNA
damage.
In the late 1960’s, Evelyn Witkin proposed a hypothesis that this response was the
result of a damage-inducible system in E.coli (Witkin, 1967). This idea was based on
observations of a few key similarities between the processes of  phage induction and
filament formation in bacterial cells. After exposure to DNA-damaging agents, E. coli
cells continue to elongate, but fail to septate. This occurs as the result of expression of the
SulA protein, which inhibits septation (ref). Phenotypically, the bacterial cells present as
long filaments. Witkin noted that both phage induction and filament formation were mass
effects, that they were initiated by a variety of DNA-damaging agents, that the effects
were reduced by photoreactivation after exposure to UV light at 254 nm, and that they
were both dependent on active protein synthesis (Friedberg et al., 2006). She theorized
that the synthesis of SulA might be regulated in the bacterial cell by repressors similar to
15
 phage repressors and respond to the same inducer, one that is induced in response to
DNA damage.
Miroslav Radman first formally postulated the SOS response in the mid-1970’s
(George, Devoret, & Radman, 1974). He circulated a memorandum to colleagues in the
field suggesting the presence of an inducible “mutation-prone” replication mechanism.
He posited that this function was dependent on the lexA and recA gene products and was
responsible for the UV-induced mutations of  phage and E. coli. He termed this
mechanism “SOS repair” to emphasize that this function was in response to the cell’s
distress signal when encountering DNA damage.
Mutations in the regulatory genes lexA and recA, as well as in the umuDC genes
were found to be essential components in SOS mutagenesis (Kato & Shinoura, 1977;
Steinborn, 1978). LexA was found to be a transcriptional repressor which binds to the
operators of genes involved in the SOS response. The LexA repressor binds, with
differential binding affinities, to a 20bp consensus sequence in the operator region (Little,
Edmiston, Pacelli, & Mount, 1980). The strength of binding determines how early or late
in the SOS response the repressed genes are expressed. Genes with a LexA binding site
closer to the consensus sequence bind the LexA repressor tightly and are therefore
expressed later in the response and, conversely, genes with a sequence further away from
the consensus bind LexA weakly and are induced earlier in the response. Genes that are
induced early tend to be genes involved in error-free repair processes including genes
involved in nucleotide excision repair, such as the uvr genes, and recombinational repair,
like recA (Fernandez de Henestrosa et al., 2000; Little et al., 1980).
16
RecA is induced almost immediately after SOS induction, <1 minute. Following
DNA damage, there are regions of single-stranded DNA associated with DNA lesions or
stalled replication forks (Heller & Marians, 2006; Marians, 2008). RecA forms a
nucleoprotein filament in the presence of ATP (termed RecA*) in these single stranded
regions. RecA* further induces SOS by facilitating the auto-cleavage of the LexA
repressor. Under physiological conditions, this auto-cleavage reaction cannot occur in the
absence of RecA*. LexA binds in the helical groove of the nucleoprotein filament and is
cleaved rapidly upon contact with it into two inactive fragments (Little et al., 1980; Yu &
Egelman, 1993).

Figure 1.1. Early translesion synthesis (TLS) models. A. The Bridges–Woodgate two-step model. In this model, translesion
synthesis (TLS) is catalyzed by DNA polymerase (pol) III. DNA pol III (which requires RecA in a first step) inserts a nucleotide
opposite a template lesion (for example, A opposite X) and, in a second step, also copies past the lesion, which requires the UV
mutagenesis gene products UmuDC. B. The Echols mutasome model. A multiprotein complex that includes UmuC, UmuD′ and DNA
pol III holoenzyme is recruited to a DNA lesion by a RecA nucleoprotein filament. The mutasome complex enables replication to take
place across the lesion, which results in mutations.

Since LexA and RecA are regulators of the SOS response, it is reasonable that
mutations in these genes would abolish mutagenesis because they would be unable to
initiate the response. The umu gene products, however, only affect SOS mutagenesis and
17
not other earlier aspects of the SOS response. Early models proposed that these proteins
were accessory factors necessary for the replicative polymerase, pol III, to bypass lesions
(Figure 1.1). Direct biochemical characterization of these proteins was made difficult due
to the extreme insolubility of the umuC gene product. The protein forms inclusion bodies
and the earliest protocols involved purifying the protein under harsh denaturing
conditions and then refolding it with chaperones (Woodgate, Rajagopalan, Lu, & Echols,
1989). A more efficient method of purification was formulated where the protein was
expressed and purified as a complex with dimeric UmuD′, thereby rendering the protein
soluble as a heterotrimeric complex of UmuD′
2
C (Bruck, Woodgate, McEntee, &
Goodman, 1996). After this technical breakthrough, it was possible to reconstitute the
TLS system in vitro. This led to the important discovery that UmuD′
2
C, in the presence
of RecA, was a stand-alone polymerase able to catalyze DNA in the absence of pol III
(Reuven, Arad, Maor-Shoshani, & Livneh, 1999; Tang et al., 1999). Pol V is able to
synthesize past many different DNA lesions (Tang et al., 2000). Pol V lacks the 3′  5′
exonuclease proofreading activity and the 5´  3′ nick translation activity. It is weakly
processive and has low fidelity, explaining the increase in mutations observed during
SOS mutagenesis. Pol V makes base substitution errors with a frequency of 10
-2
-10
-3
on
undamaged DNA (Tang et al., 2000). Consistent with mutations observed in vivo
(LeClerc, Borden, & Lawrence, 1991; Szekeres, Woodgate, & Lawrence, 1996), Pol V
preferentially mis-incorporates dG opposite of the 3′ T of a TT-(6,4) photoproduct in
vitro (Tang et al., 2000).
18
RecA has multiple roles during the SOS response. Some of these were known: as
discussed earlier, RecA is required for the induction of the response by facilitating the
auto-cleavage of the LexA repressor. In a similar reaction, RecA also mediates the self-
cleavage of the umuD gene product into its mutagenically active form, UmuD 
(Burckhardt, Woodgate, Scheuermann, & Echols, 1988; Shinagawa, Iwasaki, Kato, &
Nakata, 1988). UmuC is insoluble without UmuD . However, RecA was shown to have a
third, more direct role in SOS mutagenesis (Dutreix et al., 1989; Nohmi, Battista,
Dodson, & Walker, 1988; Sweasy, Witkin, Sinha, & Roegner-Maniscalco, 1990).
Raymond Devoret isolated a missense mutant of RecA, recAS117F (reca1730), that
when overexpressed in a lexA(def) background, in the presence of the umuD C genes on a
plasmid, was still defective for SOS mutagenesis. This implied a yet un-identified role for
RecA in mutagenesis.
Early models posited the RecA in a cis filament downstream of the blocked
replication fork. This was based on the known properties of the RecA protein, in which
the RecA nucleoprotein filament most readily forms in single-stranded DNA gaps and is
not known to have any function when not part of a nucleoprotein filament. However,
placing the filament in cis, downstream of the pol V, presented a structural impedance for
Pol V. Not only would Pol V have to interact with the downstream RecA* while
performing TLS, but it would also have to displace it. Pol V’s ability to displace a RecA*
ahead of the replication fork was called into question when substantial movement of the
polymerase was inhibited in the presence of cis filaments (Pham, Bertram, O'Donnell,
Woodgate, & Goodman, 2001).
19

Figure 1.2. Kinetics of primer/template annealing. When forming primer/template DNA molecules by mixing single-stranded DNA
primers and templates and then allowing them to anneal in a bimolecular reaction, the presence of ssDNA on which a RecA filament
can assemble cannot be avoided. There will always be un-annealed template and primer DNA, and therefore an indeterminate amount
of single stranded DNA in all experiments utilizing annealed primer/template substrates. A simple way to eliminate excess ssDNA is
to form a stable hairpin structure so that almost all of the DNA anneals in the form of a hairpin, in a unimolecular reaction, with a very
small proportion of non-hairpin ssDNA remaining. By having a short template overhang (3 nt), RecA filament assembly on the hairpin
primer/template DNA is eliminated, because the footprint of a single RecA monomer is 3 nt.

The effects of RecA* on Pol V activity remained unclear due to an unappreciated
ambiguity. All the reactions were performed with artificial primer/template systems.
When RecA was added to this system it was impossible to know whether RecA* was
acting in cis or in trans because annealing of the primer and template will always yield
some unannealed DNA. It is unavoidable, and RecA* can form on the unannealed DNA,
in trans, or in the single stranded region of the template downstream of the
primer/template junction, in cis (Patel, Jiang, Woodgate, Cox, & Goodman, 2010)
(Figure 1.2). Once this was realized, a simple alteration of the experimental procedure
showed that optimal DNA synthesis with Pol V occurred when RecA* was present in
trans (Schlacher, Cox, Woodgate, & Goodman, 2006). Hairpin DNA, with a very high
melting temperature and a three nucleotide overhang was used as the primer/template,
eliminating the possibility of excess un-controllable amounts of single stranded DNA in
the reaction. The three nucleotide overhang prevents formation of a RecA filament in cis
and trans RecA* can be formed in a separate tube on linear single-stranded DNA
(ssDNA) in the presence of ATP or one of its analogues. When the hairpin DNA and
20
trans RecA* were mixed together with Pol V, full primer utilization was observed for the
first time (Schlacher, Cox, et al., 2006) (Figure 1.3, A). The results were clear: RecA*
was indeed necessary for Pol V activity and it was necessary in trans. The improvements
in reaction efficiency with trans RecA* were dramatic. The reaction exhibited second-
order kinetics and the reaction velocity increased linearly with the addition of trans
RecA* (Schlacher, Cox, et al., 2006) (Figure 1.3, B). Another interesting requirement
was that in order for strong transactivation to occur, it was necessary for Pol V to be
exposed to the 3′-tip of the RecA*. Reactions in which the 5′-tip was exposed to the
polymerase, and the 3′-tip made unavailable showed very weak to no transactivation
(Schlacher, Cox, et al., 2006) (Figure 1.3, C). This implied a conformational requirement
for RecA* transactivation of Pol V, but exactly what this requirement was remained
unknown.

Figure 1.3. Kinetics of pol V transactivation by RecA–ssDNA. A, Pol-V catalyzed primer utilization when copying a hairpin
containing a 3-nucleotide-long template overhang was measured in the presence of ATPγS and RecA (2 μM) at varying trans ssDNA
(36-mer) concentrations (in nM). The specific activity of pol V is measured in pmol DNA per μg pol V per min. The arrowhead
indicates the position of the full-length product. B. Primer extension velocity depends on trans DNA concentration. C. RecA protein
(~4 μM) immobilized on a DNA–bead complex (~1 μM of DNA molecules bound to the bead) with exposed 3′ -proximal RecA ends
causes strong transactivation of pol V. A filament with exposed 5′ ends shows weak transactivation. Adapted from Schlacher, at al
2006

C
21
Further refinement of the experimental procedure helped clarify this role. Stable
RecA* filaments were formed on a biotinylated ssDNA bound to streptavidin coated
agarose beads in the presence of the slowly-hydrolyzable ATP analog, ATPγS. Unbound
RecA* was removed. Pol V was incubated with the RecA* that remained bound to the
resin in the absence of any primer/template DNA. The RecA* was then removed by
centrifugation and the supernatant was incubated with primer/template DNA and dNTPs
to initiate DNA synthesis. Efficient primer extension was observed. When the
supernatant was run on a protein gel, it was found to contain UmuC:UmuD′
2
:RecA in a
1:1:1 ratio. ATP was also present in the complex, stoichiometric with the three protein
components (Jiang, Karata, Woodgate, Cox, & Goodman, 2009)(Figure 1.4, A). This
complex, active for DNA synthesis in the absence of RecA*, was termed Pol V
Mutasome, or Pol V Mut for short (Figure 1.4, B). Pol V Mut was further verified as a
protein complex using laser multi-angle light scattering (MALS) coupled with size
exclusion chromatography (Figure 1.4, C, D). This method allowed for an independent
measurement of the absolute molar mass of the protein complex. The size exclusion peak
that included RecA, as well as UmuD′
2
C, had a molar mass of 113 kDa, in good
agreement with the value predicted for UmuD′
2
C•RecA, 110 kDa. A second peak that did
not include RecA had a molar mass of 73 kDa, consistent with the predicted molecular
weight of UmuD′
2
C alone, 72 kDa (Figure 1.4, C). The direct role of RecA* in activating
Pol V during SOS mutagenesis was finally determined: RecA* donated a RecA and ATP
molecule from the 3′-tip of the filament to activate Pol V for DNA synthesis. In concord
with previous observations, the 5′-tip was not able to activate Pol V. There was a protein
22
complex formed, but it was not active for DNA synthesis. Devoret’s recA1730 mutant
was also able to donate a RecA monomer from the 3′-tip, but in agreement with previous
genetic data, it was inactive for DNA synthesis (Jiang et al., 2009) (Figure 1.4, E, F).

Figure 1.4. The Pol V Mutasome. A. Activated polVMut–(wild-type RecA) complex contains UmuC, UmuD9 and wild-type RecA,
in a 1:1:1 ratio. B. DNA synthesis by Pol V Mut–(RecA E38K/ΔC17). C. After wild-type-RecA*-mediated transactivation of
UmuD′ 2C and removal of wild-type RecA*, the mixture of polVMut–(wildtype RecA) and non-activated Pol V was resolved by size-
exclusion chromatography (upper trace), and the molecular mass corresponding to each peak was measured by MALS. Non-activated
pol V run separately on the silica gel elutes at 18.4 min (lower trace). D. Silver-stained SDS–polyacrylamide gel showing the protein
composition from the two peaks contained in panel c (upper trace). E. The amount of RecA transferred to Pol V from exposed 3′- tips
of RecA* are similar for wild-type RecA*, RecA E38K/ΔC17* and RecA F117S*. F. DNA synthesis catalyzed by Pol V Mut–(wild-
type RecA) or pol V Mut–(RecA E38KDC17) occurs when RecA is transferred from the 3′-RecA* tip. Mutagenically inactive RecA
F117S* fails to activate Pol V. adapted from Jiang et al 2009

These experiments and others led to the current model of Pol V Mut (Figure 1.5).
UmuD′
2
C is inactive in the absence of RecA*. The specific role of RecA* in SOS
mutagenesis is to transfer a RecA•ATP from its 3′-proximal end to pol V (UmuD′
2
C),
thereby activating it for mutagenesis (Jiang et al., 2009). This active species of Pol V,
named Pol V Mut, contains UmuD′
2
C•RecA•ATP. Pol V Mut can now perform TLS in
the absence of RecA* (Jiang et al., 2009).
A
C
D
E
F
B
23

Figure 1.5. Model for pol V Mut function. Pol V is UmuD′2C, and is minimally active on its own. Transfer of an ATP-bound RecA
subunit from RecA* creates the active pol V Mut. Pol V Mut can migrate to a template-primer site where its activity is required.
There, it will extend the primer and insert nucleotides opposite any lesion encountered (called translesion synthesis or TLS). Upon
dissociation, pol V Mut is inactivated. There is also a slow inactivation of pol V Mut in solution, without carrying out TLS.
Deactivated pol V Mut can be reactivated by interaction with another RecA* filament, with transfer of a new RecAATP subunit and
displacement of the old RecA-ATP subunit.

Deactivation of the Pol V Mut complex in the absence of RecA* is proposed to
happen through two independent pathways, either with or without DNA synthesis. In the
absence of DNA substrate, the protein complex undergoes a slow deactivation at 37°C.
When DNA substrate is present, rapid deactivation of Pol V Mut occurs. Unlike other
DNA polymerases, Pol V Mut cannot cycle on primer/template DNA. Each active pol V
Mut complex can promote one and only one round of DNA synthesis (Jiang et al., 2009)
(Figure 1.5). Regardless of the mode of deactivation for Pol V Mut, full activity can be
restored to the complex by interaction with a new RecA* (Jiang et al., 2009). In this way,
active pol V Mut will disappear from the cell soon after the SOS response is switched off,
and the RecA* filaments that sustain it are no longer present.
24
Surprisingly, Pol V Mut deactivation is not a result of dissociation of RecA or
ATP from the active complex. Affinity chromatography showed that both remained
bound to Pol V Mut in its deactivated state. It was proposed that deactivation of the
active enzyme was instead a result of a conformational change within the enzyme.
Previously we have shown that there are two distinct binding modes of RecA to PolV
(Schlacher et al., 2005). Binding of RecA to Pol V occurs in the presence and absence of
DNA and ATP (or one of its analogues). However, UmuD′
2
is unable to bind RecA
unless both DNA and ATP are present. Therefore, we believe that when DNA and ATP
are present, binding of Pol V to RecA occurs through the UmuD′
2
dimer, and binding in
the absence of DNA and ATP must occur through the UmuC subunit. Although Pol V
can stably bind RecA in the absence of DNA and ATP, this mode of binding is not active
for DNA synthesis. We proposed a model in which in the active conformation of Pol V
Mut, RecA is bound to UmuD′
2
and in the inactive conformation, RecA remains bound to
Pol V, but has now moved to the UmuC subunit, rendering it de-activated. It is important
to note that although Pol V Mut is de-activated and can no longer perform DNA
synthesis, it is not inactive. It is fully re-activatable with addition of new RecA*. The
RecA bound to the de-activated enzyme is replaced with a new RecA from the 3′-tip of
an incoming filament. We believe the movement and replacement of RecA on Pol V
constitutes a unique on/off toggle switch to activate and deactivate the enzyme, thereby
tightly regulating the error-prone polymerase so that it does not overwhelm the cell with
mutations. This thesis is aimed at gleaning structural insight into this conformational
switch: where is RecA in the active and de-activated states relative to Pol V?
25

CHAPTER 2
MODEL TESTING VIA UN-NATURAL AMINO ACID
INCORPORATION FOR CROSSLINKING AND FLUORESCENT
STUDIES

26
Chapter 2.0 Introduction
2.0.1 Un-Natural Amino Acids and the pEVOL Method for Incorporation
To investigate conformational changes in Pol V Mut related to the activation/de-
activation cycle, we employed the use of un-natural amino acids incorporated into
specific sites of our protein (Xie & Schultz, 2005). The specific un-natural amino acids
we chose are used for photo-crosslinking and for click chemistry to site-specifically
“click on” a fluorescent tag. The photo-crosslinking proteins are used to examine the
binding states of the various proteins in the active, deactivated, and inactive states of the
mutasome. The fluorescently tagged proteins are used for Förster resonance energy
transfer (FRET) to examine the precise distances between the proteins during the
activation/deactivation cycle.
Non-canonical amino acid incorporation allows for a very controlled
manipulation of proteins for precise biochemical characterization. The ability to
determine the specific site you wish to manipulate for further investigation allows for
detailed, high-resolution information to be obtained. We used the pEVOL system
developed by Peter Schultz and colleagues for incorporation of non-canonical amino
acids into the proteins we wished to investigate (Young & Schultz, 2010). The method
first involves site directed mutagenesis to introduce an amber codon (TAG) at the residue
you wish to replace with the non-canonical amino acid. The amber codon is a good
choice because it is the rarest of the stop codons in E. coli, and its usage frequency in the
bacterium is very low overall. Then, the pEVOL vector containing the orthogonal
aminoacyl-tRNA synthetase/suppressor tRNA (aaRS/tRNA
CUA
) pairs is introduced into
27
the host strain along with the vector containing the coding sequence for the protein with
the amber mutation. The transformed strain is induced and the protein is expressed in the
presence of the un-natural amino acid. Protein purification using standard protocols is
carried out to isolate the protein containing the un-natural amino acid (Figure 2.1, A).
The aaRS/tRNA
CUA
pair is very specific such that the aaRS aminoacylates its cognate
tRNA but not any of the expression strain’s endogenous tRNAs, and likewise, the tRNA
introduced is not aminoacylated by any of the strain’s endogenous aaRS. This keeps non-
specific reactivity low and ensures that the un-natural amino acid is incorporated with
high fidelity and only in response to the nonsense codon (Young & Schultz, 2010). We
chose two un-natural amino acids to incorporate into our proteins: the photo-crosslinker
p-benzoylphenylalanine, pBpF (Chin, Martin, King, Wang, & Schultz, 2002), and p-
azidophenylalanine, pAzF (Chin, Santoro, et al., 2002) for “clicking on” fluorescent tags
(Figure 2.1, B).

28

Figure 2.1. Un-natural Amino Acid Incorporation. a. Scheme for the incorporation of un-natural amino acids into proteins of
interest. An amber mutation is introduced at the site in the protein where the un-natural amino acid will be incorporated. The plasmid
containing the altered gene along with another plasmid containing the orthogonal aminoacyl-tRNA synthetase/suppressor tRNA pair
are co-transformed into the host cell. The strain carrying the plasmids is grown and induced in the presence of the un-natural amino
acid. The cells are harvested, lysed, and the protein is purified using standard protocols. b. Structure of un-natural amino acids used in
this study. P-azido-phenylalanine (pAzF) is used for click chemistry to fluorescently label our proteins of interest. P-benzoyl-
phenylalanine (pBF) is used as a photo-crosslinker to probe for protein-protein interactions.

Benzophenomes are extremely useful in mapping protein-protein interactions.
Unlike other cross-linking compounds, they are chemically stable and can be worked
with in ambient light. They are excited optimally in the long wave UV range, between
350-365nm. Upon excitation, they react with unactivated C–H bonds (Galardy, Craig, &
Printz, 1973) to form a covalent bond. Benzophenomes can be re-excited if they do not
find a suitable C–H bond to react with and so, unlike other photo-crosslinkers, can be
reused. Early studies showed that the photocrosslinking amino acid p-benzoyl-L-
phenylalanine (pBpa) could be site specifically incorporated into synthetic peptides by
solid-phase peptide synthesis (Kauer, Erickson-Viitanen, Wolfe, & DeGrado, 1986).
Since then, the un-natural amino acid has been used extensively to probe peptide-protein
29
interactions in vitro (Dorman & Prestwich, 1994). The advent of the orthogonal
aaRS/tRNA pair allows for the in vivo incorporation of pBpa into proteins to probe for
protein-protein interaction in vitro and in vivo (Chin, Martin, et al., 2002).
The second un-natural amino acid we decided to incorporate into our proteins is
p-azido-L-phenylalanine (pAzF). This un-natural amino acid contains a functional azido
(N3) group which will react with an alkyne. The un-natural amino acid can be used as a
reactive “handle” to site-specifically label proteins with informative probes containing
alkynes, like fluorescent tags (Sakmar, 2011). P-azido-L-phenylalanine is also photo-
reactive (Chin, Santoro, et al., 2002) and can be used in protein-protein crosslinking.
Once pAzF is incorporated into the protein, click chemistry is performed to tag the
protein with a probe. Click chemistry can be done under copper assisted (Wang et al.,
2003) or copper free methods (Jewett & Bertozzi, 2010; Ning, Guo, Wolfert, & Boons,
2008). Although the chemistry was optimized with copper as a catalyst, copper is
cytotoxic, quenches fluorescence and kills the biological activity of most proteins. Since
we wish to examine active states of our enzyme, the copper free method is preferred.
2.0.2 Incorporation Sites
We initially chose to incorporate the un-natural amino acids into the RecA and
UmuC proteins. Since the UmuD′ protein is dimeric, labeling this subunit would yield a
dually labeled Pol V and Pol V Mut and raise ambiguity in the results obtained. Instead,
for Pol V labeling, we chose to label the UmuC subunit. We were initially hesitant to
incorporate un-natural amino acids into UmuC because of how notoriously difficult the
gene is to recombineer. Since the UmuC subunit of our Pol V contains a 6-Histidine tag
30
on its N-terminal end, we chose to first label the protein using trisNTA conjugated
fluorophores (Lata, Gavutis, Tampe, & Piehler, 2006). Eventually though, we optimized
a method for introducing the un-natural amino acids into UmuC and chose to target sites
Phe1 and Phe18. This gave us a range of options to label with fluorescent probes and also
opened up the potential to use UmuC to determine interacting partners through photo-
crosslinking.
Due to its prevalence in almost all classes of organisms and its importance in
many cellular processes, RecA is a heavily studied protein. There is a crystal structure
available for the RecA nucleo-protein filament (Chen, Yang, & Pavletich, 2008). There
are also many papers investigating mutations in the protein (McGrew & Knight, 2003)
and their effect on the function of RecA. RecA’s involvement in mutagenic processes is
also well studies (Dutreix, Burnett, Bailone, Radding, & Devoret, 1992; Dutreix et al.,
1989; Sweasy et al., 1990). Using all of this known information, we chose five sites on
the RecA protein to incorporate with un-natural amino acids: Phe21, Lys106, Ser117,
Phe255, and Lys317 (herein referred to as RecA
F21AzF
, et al). Lys106 is near the proposed
region of interaction with Pol V, amino acids 112-117, on the surface containing the 3′-
tip of RecA* (Boudsocq, Campbell, Devoret, & Bailone, 1997; Sommer, Boudsocq,
Devoret, & Bailone, 1998). The structural information shows that Lys317 is close to the
C-terminal of RecA and on the surface of the filament (Scrima et al., 2008). Labeling at
this residue was also unlikely to eliminate protein function since it had been shown that a
mutation at this site has essentially no effect on UV sensitivity compared to wt RecA
protein in vivo (Kurumizaka & Shibata, 1996). The Ser117 site on RecA is the residue
31
that is mutated in the classic RecA mutant, recA1730: Ser117  Phe. This mutant was
pivotal in the discovery of RecA’s distinct role in SOS mutagenesis. We believe this to be
the direct site of interaction with Pol V and a residue necessary for mutagenic activity.
These sites along with Phe21 and Phe255 give us good spatial range within the protein to
perform mapping of sites of interaction between RecA and the subunits of Pol V via
photo-crosslinking and fluorescent studies including Fӧrster Resonance Energy Transfer
(FRET).

32
2.1 Results
2.1.1 Photocroslinking Studies
Photocrosslinking experiments were carried out with RecA
F21pBF
, RecA
K106pBF
,
RecA
S117pBF
, and RecA
K317pBF
. Activated Pol V Mut was made using streptavidin coated
agarose beads. Inactivated Pol V Mut was made by incubating Pol V with RecA in a 1:1
ratio in the presence of ATPγS. We have shown in previous studies that this results in
interaction/binding between the proteins with a K
d
=197nM, however this is not sufficient
to activate Pol V for DNA synthesis (Schlacher et al., 2005). Therefore, we consider this
the inactive complex. In some of the reactions, Pol V was also incubated with RecA* for
crosslinking studies. Pol V can bind to RecA* at either filament end or within the helical
groove of the filament (Frank et al., 2000). The reactions were exposed to UV light at
365nm for the indicated period of time on ice. The reactions were quenched with a
combination of formamide, EDTA, and protein loading dye. The data was visualized by
running a denaturing, SDS-polyacrylamide gel. The immobilized proteins were then
transferred to a PVDF membrane and probed with antibodies against UmuC, UmuD′, and
RecA for crosslinked products.
33
Figure 2.2. RecA-RecA crosslinking. Crosslinking occurs with p-benzoyl-phenylalanine substituted RecA proteins. RecA pBF was
incubated with ssDNA and ATPγS or alone. RecA is known to form higher order oligomers in both cases. When run on a gel and
probed with anti-RecA antibodies, the crosslinked oligomers are viewed as higher bands. A. RecA S117pBF B. RecA K317pBF C.
RecAK 106pBF.

In every case, RecA-RecA crosslinking was observed (Figure 2.2). This was a
positive confirmation for the existence of crosslinking with these substitutions. RecA
forms higher order oligomers in the absence of DNA and a nucleotide cofactor (Brenner,
Zlotnick, & Griffith, 1988; Story, Weber, & Steitz, 1992) and binds cooperatively to
itself to form a nucleoprotein filament in the presence of DNA and a nucleotide cofactor
(Egelman & Stasiak, 1986; Heuser & Griffith, 1989). All of the substitutions, with the
exception of the RecA
S117pBF
, also yielded an active RecA enzyme, as demonstrated by
RecA
pBF
* transactivation of Pol V (Figure 2.3). The RecA
S117pBF
substitution is very
similar to the classic RecA mutant, RecA1730 (S117F). Instead of replacing the serine
with a phenylalanine, we are replacing it with a phenylalanine analog. Since the original
substitution is dead for mutagenic activity by Pol V (Dutreix et al., 1992; Sweasy et al.,
1990), we would not expect this substitution to have any activity either.
34

Figure 2.3. Enzymatic Activity of p-benzoyl-phenylalanine (pBF) RecA substitutions. A. Pol V DNA synthesis reactions were
performed with a hairpin DNA with a 3nt overhang. 1μM Pol V was incubated with 1μM RecA pBF*, formed by incubating 1μM 30nt
ssDNA, 10 μM RecA pBF, and 0.5mM ATPγS. B. Lane 1: primer alone. Lane 2: primer incubated with RecA to detect any
contaminating polymerase activity. Lane 3: 1μM RecA F21pBF* incubated with Pol V. The mutant is able to transactivate Pol V for full
DNA synthesis, visualized as higher bands. Lane 4: transactivation of Pol V by 1μM wt RecA*. C. Lane 1: primer alone. Lane 2: Pol
V transactivation by 1μM wt RecA*. Lane 3: Pol V transactivation by 1μM RecA K106pBF*. D. Lane 1: primer alone. Lane 2: Pol V
transactivation by 1μM wt RecA*. Lane 3: Pol V transactivation by 1μM RecA K317pBF*.

RecA
K106pBF
showed crosslinking with UmuC when Pol V was incubated with
RecA
K106pBF
* (Figure 2.4, A). There was an increase in the crosslinked product visible
over a time of 60 minutes. This substitution showed no crosslinking product with UmuD′
whether Pol V was incubated in an equimolar ratio with RecA
K106pBF
or with
RecA
K106pBF
* (Figure 2.4, B). There were faint bands visible in some of the Western blots
against UmuD/D′ due to the cross-reactivity of the UmuD′ antibody with RecA and
UmuC (data not shown). The Westerns against RecA revealed crosslinking products in
three separate experiments: RecA
K106pBF
and Pol V in equimolar ratios, RecA
K106pBF

alone, and Pol V incubated with RecA
K106pBF
* (Figure 2.4, C shows Pol V with
RecA
K106pBF
*). This is in agreement with RecA forming higher order oligomers and proof
of incorporation and activity of the p-benzoyl-phenylalanine. Interestingly, the
35
experiments that included Pol V showed more RecA
K106pBF
-RecA
K106pBF
crosslinking
than the one that did not.

Figure 2.4. Crosslinking with RecAK106 pBF*. Crosslinking reactions were carried out as described. Westerns were performed
probing with the antibody indicated. This set of figures represents the reaction in which Pol V was incubated with RecAK106 pBF*. The
reaction was exposed to UV light 365nm for one hour with aliquots taken at times indicated. A. In the Western blot probing for
UmuC, there is a visible crosslinked product that appears after 30 minutes. B. There are no croslinked products visible when probing
for UmuD′. C. There are crosslinked products visible in the reaction when probing for RecA. These products can be a product of
RecA-RecA crosslinking and/or RecA-UmuC crosslinking.

With RecA
K317pBF
, there is crosslinking observed between UmuC and RecA
K317pBF

when RecA
K317pBF
and Pol V are incubated together in an inactive conformation (Figure
2.5, A). The crosslinking product is present on the Western blots with antibodies against
UmuC and RecA. Probing the reactions with UmuD/D′ antibody reveals crosslinked
bands, but upon scrutiny, these bands are most likely RecA/UmuC cross-reactivity with
UmuD antibody. There is also no unique crosslinking observed with RecA
K317pBF
with
either UmuC or UmuD′ within Pol V Mut (Figure 2.5, C).

36

Figure 2.5. Crosslinking with RecA K317pBF. Crosslinking reactions were performed with active Pol V Mut (Reaction 1) or with the
inactive conformation of Pol V and RecA K317pBF, when both are incubated in equimolar ratios (Reaction 2). Crosslinking reactions
were carried out for 30 minutes. A. Probing with UmuC antibody reveals crosslinked product when Pol V is incubated with RecA in
an inactive conformation. B. UmuD antibody reveals some higher bands in both reactions 1 and 2. However, some of these bands are
present in the Pol V alone lane, demonstrating cross-reactivity of this antibody with other proteins. The unique higher band in
Reaction 2 is most likely cross-reactivity of this antibody with RecA K317pBF that has formed higher order oligomers. C. The Western
against RecA reveals crosslinked bands when Pol V and RecA are incubated in equimolar ratios.

For the RecA
F21pBF
, when Pol V is incubated with RecA
F21pBF
in equimolar ratios
in the inactive conformation, there is a unique crosslinked product visualized with the
UmuD′ antibody at 50-60 kDa (Figure 2.6, B). Since the size of this product is greater
than a RecA monomer (38kDa) and smaller than the predicted size of two RecA
molecules (76kDa), we believe this to be a real crosslinked product of UmuD′ and not
simply cross-reactivity between the RecA protein and UmuD′ antibody. We do not
believe it to be cross-reactivity of the antibody with UmuC, because the band is not
present at the zero time point, but only present after one hour exposure to UV light. The
predicted size of a RecA+UmuD′ complex is 50kDa. When Pol V Mut is made, there is a
unique crosslinked product between UmuC and RecA
F21pBF
between 80-100kDa, in close
accord with the predicted size of a UmuC+RecA complex: 86kDa (Figure 2.6, A).
37

Figure 2.6. Crosslinking with RecA F21pBF. Crosslinking reactions were performed with active Pol V Mut (Reaction 1) or with the
inactive conformation of Pol V and RecA F21pBF, when both are incubated in equimolar ratios (Reaction 2). Crosslinking reactions were
carried out for 60 minutes. A. Probing with UmuC antibody reveals a small crosslinked product with Pol V and RecA F21pBF in the
active Mutasome conformation after 1 hour. B. UmuD antibody reveals a unique higher band in Reaction 2, when Pol V is incubated
in an inactive conformation with RecA F21pBF. Pol V and RecA loading controls are also included in this Western, and some of the
bands can be eliminated as cross-reactivity of the antibody, which is also seen in the 0 minute time point, which has not been exposed
to UV. C. Probing with RecA antibody reveals many crosslinked products. Reaction 3 is RecA F21pBF incubated without any Pol V.

The results for RecA
S177pBF
are perhaps the most relevant to historical SOS
mutagenesis. This substitution mimics Devoret’s mutagenically dead RecA1730, and as
expected it is also dead for DNA synthesis in both transactivation by RecA
S117pBF
* and in
Pol V Mut
S117pBF
(Figure 2.7, A). With this substitution, we see crosslinked products
between UmuC and RecA
S117pBF
when they are incubated in equimolar ratio (Figure 2.7,
B), as an inactive complex. There is a corresponding, unique crosslinked product when
the same reaction is probed with RecA antibody. There is absolutely no crosslinking
between the two proteins when Pol V Mut is made. While there are bands in the Western
against UmuD′ when RecA
S117pBF
is incubated with Pol V, these bands cannot be ruled
out as cross-reactivity between RecA and the UmuD′ antibody.
38

Figure 2.7. RecA S117pBF Activity and Crosslinking. A. DNA primer synthesis reactions. RecA S117pBF is unable to activate Pol V either
as a mutasome complex or by transactivation. RecA F21pBF is active in both cases. B. Antibody against UmuC reveals faint crosslinking
bands in Reaction 2 where Pol V and RecA S117pBF are incubated in an inactive conformation. There is no crosslinking observed in Pol
V Mut, Reaction 1. C. The RecA antibody reveals crosslinked products with Pol V Mut (Reaction 1), Inactive Pol V and RecA
(Reaction 2), and RecA alone (Reaction 3).

The crosslinking results are summarized in Table 2.1. Although further mapping
of RecA protein on Pol V needs to be done, it is clear that RecA does interact with
different parts of Pol V depending on the active state it is in. For example, if we look at
the RecA
F21pBF
data independently, this residue on RecA crosslinks to UmuC when in an
active conformation, as Pol V Mut, and crosslinks to UmuD′ when incubated in an
inactive conformation (Figure 2.6). This is evidence of the same residue on RecA
crosslinking to different subunits of Pol V correlating with different modes of activity
and demonstrating differential protein-protein interactions by RecA. The S117pBF
substitution is also informative. No crosslinking at all is observed when the enzyme is in
its active form as Pol V Mut implying that this substitution is unable to make active
contacts with Pol V.

39

RecA Amino acid
substitution
Active Pol V Mut
Crosslinking
Inactive Pol V: RecA,
1:1
Pol V: RecA*
F21pBF UmuC Crosslinking UmuD′ Crosslinking N/A
K106pBF N/A No Crosslinknig UmuC Crosslinking
S117pBF No Crosslinking UmuC Crosslinking N/A
K317pBF No Crosslinking UmuC Crosslinking N/A
Table 2.1. RecA p-benzoylphenylalanine Substitution Crosslinking Results.
It is important to note that different faces of the RecA protein may be interacting
with different subunits of Pol V at the same time. That is, it is quite possible for the RecA
protein to interact with both UmuC and UmuD′
2
at the same time. These residue
substitutions give very specific location information, and if one particular substitution
shows crosslinking to different subunits depending on context, that is evidence of the
RecA protein being able to change its conformation on Pol V. Many substitutions taken
together will be able to provide us with more detailed mapping information of the
location of RecA on Pol V in different activity states.

40
2.1.2 Fluorescence Studies
To investigate if there was a change in conformation of Pol V Mut corresponding
to deactivation of the protein, we performed steady-state ensemble FRET using
fluorescently labeled proteins Pol V (UmuC subunit) and RecA. We measured FRET
distance between RecA and UmuC in Pol V Mut in the active conformation, and the same
distance between the two subunits after DNA synthesis had taken place, presumably in a
deactivated conformation. We used Cy3 labeled UmuC
F1AzF
and Cy5 labeled RecA
F21AzF
.
We incubated Pol V and RecA in an inactive conformation, without exposure to a RecA*
filament, to determine the Fӧrster distance, R
0
, between Cy3-UmuC and Cy5-RecA. We
used Photon Technology International (PTI) software to calculate the Fӧrster distance
based on an emission spectrum of the donor and an excitation spectrum for the acceptor.
The orientation factor was assumed to be 2/3 and 15% was used for the donor quantum
yield. The Fӧrster distance was determined to be 53.546Å, assuming a fast orientation
factor, κ=2/3. The FRET efficiency and distance were calculated using PTI’s FRET
calculator software based strictly on donor fluorescence quenching. The efficiency of
FRET and the distance separating the fluorophores was determined to be 0.288 and 62.27
Å respectively for Pol V and RecA in an inactive conformation.
Next we calculated the FRET efficiency for active Pol V Mut. The efficiency of
energy transfer for Pol V Mut in the active conformation is 0.599 and the inter-dye
distance is 49.951Å (Figure 2.8, A). We have previously shown that Pol V Mut cannot
cycle on primer/template DNA (Jiang 2009). That is the enzyme can perform only one
round of DNA synthesis and cannot reinitiate on another primer/template. To test
41
whether this is due to a change in conformation of the active complex, an excess amount
of hairpin DNA and dNTPs was added to the active complex to initiate DNA synthesis
and allow deactivation of the enzyme to take place. FRET was calculated immediately
after the addition of DNA and an hour after the addition. The FRET efficiency value and
dye distance immediately after the addition of dNTPs and hairpin was very close to the
active Pol V Mut complex: 0.603 and 50.554Å respectively (Figure 2.8, B). After an
hour, the efficiency of FRET had dropped by a large amount to 0.045 and the FRET
distance was 89.491Å. There was also no visible donor quenching (Figure 2.8, C). To test
whether this was a result of the RecA dissociating from the active Pol V Mut complex or
staying associated but having changed enough in conformation such that FRET signal is
no longer detectable, we employed the use of affinity chromatography. There is a 6-
histidine tag on the N-terminal end of UmuC which binds specifically to nickel resin. We
repeated the experiment and bound Pol V Mut to Ni-NTA resin in its active conformation
before the addition of hairpin primer/template and dNTPs and one hour after the addition;
presumably after all possible DNA synthesis had taken place. The enzyme was incubated
with the resin for 20 minutes at room temperature. The flowthrough was collected and the
resin was washed twice with buffer. The bound complex was then eluted with buffer
containing 500mM imidazole. All samples were visualized on an SDS-PAGE mini-gel by
silver staining. What we saw was that RecA remained bound to Pol V after DNA
synthesis took place (Figure 2.8, D). The loss of FRET signal was not due to dissociation
of the RecA.
42

Figure 2.8. Emission Curves For Pol V Mut. Emission curves for Pol V Mut with donor alone (higher curve) and Pol V Mut with
both donor and acceptor (lower curve) are shown. A. Pol V Mut in active conformation before the addition of DNA shows donor
quenching. B. Emission curves immediately after the addition of excess DNA substrate and dNTPS. C. Emission curves one hour after
the addition of substrate DNA and dNTPs. There is no visible donor quenching. D. Affinity chromatography for Pol V Mut before and
after DNA synthesis. Lane 1 represents Pol V Mut sample eluted from a Ni-NTA column before the addition of hairpin DNA and
dNTPs, analogous to the emission trace in Panel A. Lane 2 represents Pol V Mut eluted from the affinity column after DNA synthesis
has occurred. RecA remains bound to Pol V even though there is no donor quenching observed in Panel C.

43
2.2 Discussion and Future Directions
The incorporation of un-natural amino acids into our proteins gives us the
opportunity to study them by different methods. The photo-crosslinking results provide
some evidence to validate our current Pol V model (Figure 1.5). One particular residue is
able to bind to one location on Pol V when in an activated state, and to a different
location when in an inactive state. The mutagenically dead RecA mutant, RecA1730, is
also shown not to make any contact with Pol V when we attempt to make activated Pol V
Mut. Other classic RecA mutants RecA730 or the double mutant RecA730/ΔC17 can be
investigated by photo-crosslinking. These mutants are constitutive for SOS mutagenesis.
A pBF substitution can be made on these RecA mutants, potentially at the Phe21 site
since this substitution retains a wildtype level of activity. It would be interesting to see
where these mutagenically “hyperactive” mutants would interact on Pol V in the active
state of the enzyme. Having a larger number of photo-crosslinkable substitutions in RecA
would also allow us the opportunity to roughly map RecA onto Pol V in the different
activity states of the enzyme. Increasing the number of substitutions would also increase
the likelihood of finding one with greater crosslinking efficiency than the ones we
currently have.
Although FRET studies will be extremely helpful in determining fine
conformational differences in Pol V Mut in its active, de-activated, and inactive states,
the ensemble FRET measurements have proven difficult thus far for a number of reasons.
First, our degree of fluorescent donor or acceptor labeling has never exceeded 20%. On
average we get 10% UmuC labeling and 10% RecA labeling. Because of how Pol V Mut
44
is made, this gives us a very small percentage of donor and acceptor labeled Pol V Mut,
greatly diminishing our signal. In the ensemble Cy3Cy5-Pol V Mut population there exist
some enzymes with only donor label (Cy3-Pol V Mut), some with only acceptor label
(Cy5-Pol V Mut), very few with both donor and acceptor label (Cy3Cy5-Pol V Mut), and
mostly unlabeled enzyme. This greatly affects the signal obtained. Second, the current
method for making Pol V Mut employs the use of biotinylated ssDNA bound to
streptavidin agarose beads for RecA*. Heating biotin/streptavidin releases the bond
(Holmberg Uhlen, 2005). Therefore, this method is not ideal for ensemble FRET
purposes for us because there is always a small level of RecA* contamination in the Pol
V Mut enzyme. Since our RecA* is labeled with acceptor molecule, this results in a huge
background acceptor signal contamination for Pol V Mut. Each RecA* molecule has in
the ideal situation 10 fold more RecA molecules than Pol V. Besides the active
association of Pol V with RecA* at the 3′-filament end, it can also interact with the
filament at the 5′-filament end and within the helical groove of the filament (Frank et al,
2000). In the bulk studies, the fluorescence we see is an average of these three
interactions as well as any intramolecular interactions within an active or de-activated,
labeled Cy3Cy5-Pol V Mut. For these reasons, it is best to move to a method of making
Pol V Mut in which the ssDNA stays covalently bound to the resin irrespective of
temperature, pH, or other buffer conditions. I have developed cyanogen bromide resin
with covalently bound 30nt oligomers for use in making Pol V Mut. The FRET reactions
for active Pol V Mut and de-activated Pol V Mut in this study were performed with Pol V
Mut made using this resin. Any further testing of the model with bulk FRET should be
45
done with resin that avoids even a small amount of contaminating RecA*. With single
molecule FRET measurements we should be able to bypass both of these factors. We can
focus on single molecules we know are labeled with donor and acceptor, we can wash
away contaminants, and perhaps most importantly, we can watch a counter-correlated
change between donor and acceptor dye on a single molecule.
We’ve only begun to scratch the surface of what is possible for FRET studies,
bulk or single molecule. We have four different AzF substitutions on RecA, and they are
all active for Pol V transactivation. These substitutions will allow us to map different
residues of RecA onto Pol V in the different activity states. We also have substitutions
with RecA730 and RecA1730 (RecA730
F21AzF
and RecA1730
F21AzF
). We can finally
begin to investigate whether the in vivo and in vitro levels of activity of these mutants is
related to conformation of the protein. There are also several studies that can be done
with labeled DNA substrate.

46
2.3 Experimental Procedures
2.3.1 Cloning, Expression, and Purification
For cloning of RecA, we used the pAIR79 plasmid which was a gift from Michael
Cox at the University of Wisconsin, Madison. The plasmid contains the recA gene cloned
into Novagen’s pET21D vector between the NcoI and HindIII sites. The plasmid confers
Ampicillin resistance (AmpR) to the host strain and is IPTG inducible behind a T7
promoter. The strain used for RecA protein expression was Novagen’s BLR(DE3). BLR
is a recA- derivative of the E. coli BL21 strain. DE3 indicates that the host is a lysogen of
λDE3 and carries a chromosomal copy of the T7 RNA polymerase gene under control of
the lacUV5 promoter. This makes the strain ideal for production of RecA protein cloned
into the IPTG inducible pET vector. The pEVOL-pAzF and pEVOL-pBpF were gifts
from the Peter Schultz lab at The Scripps Research Institute in San Diego, CA (Young
2010). Site-directed mutagenesis was performed on pAIR79 using Stratagene’s Pfu Ultra
polymerase for the five sites mentioned above. The sequences were confirmed and
competent BLR cells were co-transformed with the mutated pAIR79 and the appropriate
pEVOL vector. pEVOL is Chloramphenicol resistant (CamR) and arabinose inducible.
The cells were grown at 37ºC in LB media to an optical density at 600nm of
approximately 0.4. They were then induced with 100mM IPTG and 2g/L of Arabinose.
One g/L of the appropriate unnatural amino acid was also added at time of induction to
the cells. The unnatural amino acids were purchased from Bachem. The cells were grown
for five hours after induction and then harvested. RecA protein was purified by a standard
protocol (Cox et al, 1981).
47
For substitutions in the UmuC subunit, of Pol V, site directed mutagenesis proved
to be an successful strategy for introducing the amber mutation into our umuC plasmid,
pHUC25 (gift from Roger Woodgate at NIH). This is a large, low-copy plasmid, and
overexpression of UmuC can be toxic to cells, so transformation efficiency was very low,
and we were not able to obtain any mutations despite multiple trials at optimization of
reaction conditions. We instead had a piece of the UmuC gene sequence containing the
desired mutation synthesized by Genscript and performed appropriate restriction digests
and ligations with this synthetic piece of DNA and the pHUC25 plasmid. In this way, we
were able to generate two substitutions on UmuC: Phe1 and Phe18. The plasmid with the
amber codon was co-transformed into our competent Pol V expression strain, RW644, in
addition to another plasmid containing an arabinose inducible umuD′ gene required for
expression of Pol V, and the appropriate pEVOL vector. The transformed strain displayed
zero to limited growth at 37ºC and 30ºC, so protein expression was carried out at 16ºC
and growth time extended. When the optical density at 600nm (OD
600
) was
approximately 0.2-0.3, the un-natural amino acid was added. When the OD
600
reached
0.4-0.5, the vectors were induced with 4g/L arabinose. The earlier addition of the
unnatural amino acid was found to increase protein yield.
Pol V purification was performed using slight deviations from the standard
protocol (Karata, Vaisman, Goodman, & Woodgate, 2012). In brief, the cells were lysed
using a French Press and centrifuged to clear cell debris. The clarified lysate was
incubated with Qiagen’s Nickel-NTA agarose resin for 20 minutes at 4ºC. The Ni-NTA
resin was then poured into a column and washed with several column volumes of buffer.
48
Bound protein was eluted with imidazole containing buffer (50mM HEPES pH6.8, 1M
NaCl, 20% glycerol, 500mM imidazole). The eluted protein is loaded onto a pre-
equilibrated gel filtration column, Superdex 200. The Pol V-containing peak fractions are
applied to HTP hydroxyapatite resin. After washing the resin with several column
volumes of buffer, bound protein is eluted in a phosphate buffer (100mM NaPO
4
pH 6.5,
20% glycerol, 1M NaCl, 2mM DTT) in 500μl fractions. Pol V containing fractions are
aliquoted and stored at −80 °C.
2.3.2 Fluorescent Labeling
For preliminary FRET studies, UmuC
F1AzF
and RecA
F21AzF
were used. These
substitutions do not interfere with protein folding or activity. The fluorophores contain an
alkyne group that reacts with the azide. Some were synthesized and others were
purchased from Invitrogen. Currently we have Cy3, Cy5, AlexaFluor488 (a fluorescein
analog), and TAMRA (a rhodamine analog). Initially, the dyes were synthesized such
that the click chemistry labeling reaction could be performed with a copper catalyst.
However, the metal catalyst caused major aggregation and precipitation of the proteins.
Forward reactions testing for activity also revealed that copper killed the enzymatic
activity of what little labeled protein remained. After this, dyes were synthesized with
dibenzocyclooctyne (DIBO) groups which enable click chemistry to occur in the absence
of a copper catalyst (Ning et al., 2008). Using the DIBO dyes, RecA with the pAzF
substitution at positions Phe21, Lys106, and Lys317 was labeled with AlexaFluor488 and
Cy5. For preliminary studies in this dissertation, the RecA
F21AzF
was used because
49
purification of this protein yielded the largest amount of protein and this substitution was
also closest in enzymatic activity to wt RecA.
In the case of UmuC
AzF
, the protein is incubated with dye at 4ºC overnight after it
is eluted from the Ni-NTA resin and before it is loaded onto the Superdex 200 column.
Labeling of the protein at this step of the purification eliminates having to purify away
free dye at a later step and greatly increases the final yield of labeled protein. We’ve
labeled UmuC
F1AzF
with Cy3 and AlexaFluor 488 and both retain a wildtype level of
activity.
2.3.3 Pol V Mut
Pol V Mut for the crosslinking studies was made as previously described (Jiang et
al., 2009). Briefly, a biotinylated 30mer oligonucleotide (100 μM) was incubated with
strepavidin-coated agarose resin (Stratagene) for 20 min at room temperature. Unbound
oligomers were removed by washing with reaction buffer three times. The concentration
of bound biotinylated 30-mer was about 20 μM. RecA
pBF
or RecA
Fl-AzF
was immobilized
on resin-bound ssDNA by incubating the DNA oligomers with saturated amount of RecA
(1 RecA per 3 nt) for 10 min at 37 °C in the presence of 0.5mM ATPγS. Unbound RecA
and ATPγS were removed by washing with reaction buffer three times. The resin-RecA*
was resuspended in reaction buffer. His-tagged Pol V (5 μM) was added to immobilized
resin-RecA* and incubated for 10 min at 37°C before removing RecA* by centrifugation.
Pol V Mut remained in the supernatant.
For FRET studies, a 30nt oligomer with a 6 carbon amino linker attached at the
5′-end was covalently bound to Sigma’s cyanogen-bromide-activated Sepharose 4B
50
beads according to manufacturer’s instructions. The coupling efficiency was calculated to
be 60%. Resin was made such that the final concentration of DNA-resin would be 1
nmole DNA/30μL resin. RecA* was bound to the resin as described above. Pol V Mut
was obtained as above.
DNA oligonucleotides were synthesized using Applied Biosystems 3400 DNA
synthesizer. The sequence of the biotinylated oligo is 5′-GbioA AAT TCA GAG ACT
GCG CTT TAA GAA CTT. The sequence of the oligomer used for cyanogen bromide
resin is 5′-aminoGT AAA TTC AGA GAC TGC GCT TTA AGA ACT T.
2.3.4 Photocrosslinking
Photo-crosslinking reactions were carried out with purified proteins. A 365nm
hand held, UV lamp was used to activate the p-Benzoyl-phenylalanine. Reactions were
all performed in a round bottom, 96-well plate on ice in a dark room. The reactions were
carried out for a maximum of one hour with aliquots of the samples taken at the times
indicated in the gels. The reactions were quenched with equal parts of enzymatic stop
solution (95% formamide, 50mM EDTA) and protein loading dye (50mM Tris-HCl
pH6.8, 2% SDS, 10%glycerol, 1% β-mercaptoethanol, 12.5mM EDTA, 0.02%
bromophenol blue). They were loaded on a SDS-PAGE and then transferred to PVDF
membrane for immunoblotting. Immunoblotting was performed using Pierce’s
SuperSignal West Femto Chemiluminescent kit. The Westerns were imaged with Bio-
Rad’s Chemidoc XRS system.

51
2.3.5 FRET
Steady-state bulk studies were carried out using PTI’s fluorescence system.
Emission curves were acquired with an excitation wavelength of 532nm for protein with
Cy3 donor and Cy5 acceptor, Cy3 donor alone, Cy5 acceptor alone, and unlabeled
protein. Background subtractions were made to correct for protein internal fluorescence,
direct excitation of acceptor by the donor wavelength, and crosstalk between donor and
acceptor. Fӧrster distance was calculated by acquiring donor emission spectra and
acceptor excitation spectra. The orientation factor was assumed to be 2/3, the refractive
index was 1.33, the door quantum yield used was 0.15 and the acceptor extinction
coefficient used was 250,000/M*cm.
Steady state FRET parameters were calculated using PTI’s Felix32 software.
Normalized emission curves for the enzyme with donor alone or donor in the presence of
acceptor were used to calculate FRET. Background subtractions were made using
unlabeled enzyme, to correct for internal protein fluorescence and for acceptor only
labeled enzyme, to correct for direct excitation of acceptor and for donor-acceptor
crosstalk. The calculation is based on donor fluorescence quenching in the presence of
acceptor.
52

CHAPTER 3
SINGLE MOLECULE IN VIVO VISUALIZATION OF
REPLICATING CELLS WITH DNA DAMAGE

53
3.0 Introduction
Advances in microscopy have made it possible to visualize molecular processes in
live cells at the single molecule level. These methods allow for an examination of cellular
and biological processes not afforded by bulk or in vitro measurements alone and are able
to directly sample the heterogeneity present at the molecular level in vivo. They have
great potential to answer outstanding questions whether it be the existence and function
of multiple oligomeric states of proteins (Leake et al., 2008), determining the
stoichiometry and turnover for membrane proteins involved in the bacterial flagellar
rotary motor (Reid et al., 2006), or determining replisome composition. A recent lot of
studies have successfully endeavored to break down the components of the E. coli
replisome and image replication in live cells (Reyes-Lamothe, Sherratt, & Leake, 2010)
(Lia, Michel, & Allemand, 2012). These studies have already answered important
longstanding questions about replication. Reyes-Lamothe et al used stepwise photo-
bleaching to reveal that the active replisome in vivo contained three polymerases and not
two, thus solving an important question in the field. Further single molecule studies have
gone on to demonstrate that the third polymerase is most likely involved in lagging strand
DNA synthesis (Georgescu, Kurth, & O'Donnell, 2012; Lia et al., 2012).
Although there is some information known about the interplay between
replicative and specialized polymerases at the replication fork (Indiani, Langston,
Yurieva, Goodman, & O'Donnell, 2009; Lehmann et al., 2007), polymerase switching is
certainly an area that can be illuminated by in vivo single molecule microscopy. To that
end, we are in collaboration with Antoine van Oijen of the Single Molecule Biophysics
54
Group in the Netherlands and Micheal M. Cox at the University of Wisconsin, Madison
to investigate the interplay between polymerases V and III at stalled replication forks. We
have made fluorescently labeled chromosomal replacements of Pol V and various
subunits of Pol III, as well as Pol IV. Currently we have chromosomally tagged umuC
(catalytic subunit of Pol V), dnaQ (ε subunit of Pol III), dnaX (τ subunit of clamp loader),
dnaE (α catalytic subunit of Pol III) and dinB (Pol IV) with fluorescent proteins Ypet,
mKate2, and mTFP1. We also have dual color strains in which two different proteins are
tagged allowing for direct visualization of co-localization. We’ve also expressed some of
the tagged proteins in different recA and umuD backgrounds relevant to SOS
mutagenesis: recA730 (E38K) and recA1730 (S117F) and umuD-K97A and umuD′-K97A.
Our preliminary studies have yielded two very important findings. We’ve found
that Pol III’s ε proofreading subunit is rapidly degraded upon induction of UmuD,
dependent on the presence of the protease motif in UmuD needed for self-cleavage to
form UmuD′. The unexpected destruction of ε suggests a possible mechanism for
replacement of Pol III core by any of the three SOS polymerases, as opposed to a
mechanism that operates via mass action, as is currently thought to be the case (Fujii,
Gasser, & Fuchs, 2004; Uchida et al., 2008). The second finding is that Pol V is imaged
at the cell membrane. This is reminiscent of Evelyn Witkin’s early finding that in SOS
induced cells, RecA protein becomes associated with the cell membrane (Garvey, St
John, & Witkin, 1985). It also raises the possibility of TLS as a membrane associated
phenomenon.

55
3.1 Results
3.1.1 Disappearance of the ε Subunit of Polymerase III upon Induction of
Polymerase V
When we started imaging the dual color strains, a very exciting early discovery
was that upon UV Induction of Pol V, we saw a complete disappearance of the ε subunit
of Pol III (Figure 3.1A). We labeled the ε subunit in this strain to get co-localization and
potentially, polymerase switching information between Pol V and Pol III. But much to
our surprise, we saw that ε completely disappeared ~30-35 minutes after UV induction of
Pol V (Figure 3.1, A). What was more visually striking was that as the ε signal
disappeared (blue color in Figure 3.1, A), the Pol V signal (fluorescently tagged UmuC,
red color in Figure 3.1, A) increased. We then looked at ε signal alone in umuDC+ or
ΔumuDC strains. The results were very clear. In the umuDC+ strain, one hour post-UV
irradiation, there was almost no fluorescence attributed to ε, whereas in the ΔumuDC the
cells stay lit up with ε signal for one hour after irradiation with UV (Figure 3.1, B).

56

Figure 3.1. Polymerase III Epsilon Disappearance Upon UV Induction of Pol V. A. Dual-color time-lapse images showing Pol V
(red) and ε subunit (blue) in E. coli cells with chromosomal umuC-mKate2 and dnaQ-YPET at 0, 35 min, 1 h and 2 h following SOS
induction by UV-light. The appearance of Pol V (red) in filamentous SOS-induced cells coincides with the disappearance of ε (blue)
in the same cells. B. A significant decrease of fluorescent ε signal, 1 h post UV irradiation is observed in wt umuDC+ cells, but not in
ΔumuDC cells.
To investigate the effects of the UmuD proteins on ε activity biochemically, we
performed DNA exonuclease digestion assays with purified Pol III core (α, ε, and θ
subunits). We used a P32 labeled, 28 nucleotide primer annealed to an 80 nucleotide
template which contained biotin on both ends (Figure 3.2, A). The primer/template was
loaded with streptavidin to allow for the β processivity clamp to stay associated (Bertram,
Bloom, O'Donnell, & Goodman, 2004). We observed that the exonuclease activity of Pol
III core is inhibited by UmuD
2
and UmuD
2
′ (Figure 3.2, B).
57

Figure 3.2. Inhibition of Exonuclease Activity of Pol III Core By UmuD 2 and UmuD′ 2. A. An illustration of the primer/template
DNA used. Streptavidin on both ends of the template prevents the β processivity clamp from dissociating. B. Exonuclease inhibition
of Pol III core is observed as an increase in the amount of primer. Lane 1: 20nM Pol III core with 200nM β clamp and 50nM γ clamp-
loading complex (β/γ). Lane 2: PolIII core, β/γ, 500nM UmuD 2. Lane 3: PolIII core, β/γ, 1μM UmuD 2. Lane 4: PolIII core, β/γ,
500nM UmuD′ 2. Lane 5: PolIII core, β/γ, 1μM UmuD′ 2. Lanes 6-8: β/γ, 1μM UmuD 2, 1μM UmuD′ 2 to rule out contaminating
exonuclease activity in protein preparations. Lane 9: Primer alone.
An under-appreciated fact about the UmuD protein is that it is a peptidase. It
undergoes a self-cleavage reaction, with the assistance of RecA*, to form a truncated
version of itself, UmuD′, by cleaving between residues 24 and 25 (Burckhardt et al.,
1988; Nohmi et al., 1988; Shinagawa et al., 1988). The residues important for this
cleavage reaction are Lys97 and Ser60, residues which are both retained in UmuD′. To
test whether the disappearance of ε we observed in live cells could somehow be related to
the peptidyl activity of UmuD
2
/D
2
′, we made strains in which we replaced umuD on the
chromosome with umuD-K97A and umuD′-K97A. The results clearly implicated this
58
catalytic residue in the disappearance of the ε subunit (Figure 3.3). In both backgrounds,
there is very little observable loss of ε fluorescence post UV induction relative to the wt
strain.

Figure 3.3. Epsilon Degradation is Dependent on Lys97 in UmuD/D′. Single color imaging of Ypet-ε cells. The top panel is wt
umuD. There is a decrease in Ypet-ε fluorescence for over an hour, post UV induction. The center and lower panels are umuD-K97A
and umuD′-K97A cells, respectively. The Ypet fluorescence signal in these cells stays stable for over an hour post UV irradiation.

To test whether the inhibition we observed was due to cleavage of ε by either
UmuD
2
or UmuD
2
′, we incubated the purified proteins together in various combinations
and ran the reactions on SDS-PAGE gels. We tested cleavage of ε both in the context of
Pol III core and purified ε. So far, we have not been able to demonstrate in vitro cleavage
of ε.

59
3.1.2 Pol V Location in Live Cells
UmuC-mKate2 is visualized in the cell as tight fluorescent foci, appearing ~ 30 -
35 min after exposure to UV, a timescale that agrees with data for the temporal UV-
induction of Pol V (Woodgate & Ennis, 1991). The tagged UmuC is mutagenically
active. A rifampicin resistance assay showed that fluorescently tagged Pol V was
mutagenic in a dose-dependent manner, at levels of 40% compared to WT Pol V (Figure
3.4, open circles compared to closed circles).

Figure 3.4. UV Mutagenesis of Strains. Rifampicin resistance assays were performed to score for UV mutagenesis.

Initially, the sm-imaging data showed Pol V foci were mostly along the E. coli
membrane, colocalizing with the membrane protein LacY (Figure 3.5). This immediately
brought to mind an early SOS phenomenon observed by Evelyn Witkin in which she
60
observed that both SOS-inducing agents and strains in which SOS is constitutive showed
enhanced levels of the RecA protein in membrane fractions relative to non-induced cells
(Garvey et al., 1985). It raised the possibility that perhaps Pol V activation or TLS were
membrane associated events.

Figure 3.5. UV-induced Pol V foci localize at the cell membrane. Dual color imaging of an E. coli strain, containing chromosomal
umuC-mKate2 allele and lacY-eYFP on a plasmid. The strain was irradiated with UV light (10 J/m
2
). Individual molecules of Pol V
(red spots) and LacY (blue spots) are localized at the cell membrane.

To test this theory, we constructed several strains with different genetic
backgrounds relevant to SOS mutagenesis. In particular, we have fluorescent tagged Pol
V in the following backgrounds: lexA(def), recA730, recA730 lexA(def), recA730 ΔpolB
ΔdinB, and recA1730. As you’ll recall, LexA is the transcriptional repressor that under
non-damage conditions, binds to the operator region of >40 SOS genes. RecA730 (E38K)
is a RecA mutant that is constitutively active for mutagenesis. RecA1730 (S117F) is dead
for mutagenesis. polB and dinB are the genes for Polymerases II and IV, respectively,
both of which, along with Pol V, are induced as part of the SOS response.
61
Figure 3.6. umuC-mKate2 Localization in Different Genetic Backgrounds. UmuC-mKate2 fluorescence (yellow spots) was
examined in several different genetic backgrounds. In almost all of them, with the exception of recA1730, there are fluorescent foci in
the cells with or without UV.

We looked at the location of fluorescent UmuC-mKate2 spots in these strains
±UV. What we observed is that all of these strains (with the exception of the recA1730
background), had most of their fluorescent spots located in the cytosol (Figure 3.6).
These strains express UmuC without UV induction, and are mutagenic in the absence of
it (Figure 3.4, Figure 3.6), When we applied a UV dose to these strains, there was an
increase in the number of fluorescent spots overall, both cytosolic and membrane
associated, but in every case, there was a definite increase in membrane associated spots
(Figure 3.7). The recA1730 strain +UV was the only strain that showed membrane
associated spots and almost no cytosolic spots. This strain is also not mutagenic at all.

62

Figure 3.7. umuC-mKate2 Fluorescent Spots Characterized by Location in Cell. Individual cells were analyzed for the number of
fluorescent spots they had in the cytosol (top row) or on the membrane (bottom row). In the absence of UV, both the recA730 strain
and recA730 ΔpolB ΔdinB strains have more cells with a large number of spots in the cytosol. Upon induction of UV in these strains,
the fluorescent spots increase. The mutagenically dead recA1730 strain has very fewe cytosolic spots even upon induction of UV.

63
3.2 Discussion and Future Directions
The disappearance of ε raises an intriguing possibility about the regulation of
polymerases and their access to a replication fork. As cells respond to SOS inducing
signals, levels of UmuD increase. We have shown that upon UV induction, Pol V signal
increases and ε signal decreases. This result is dependent on UmuD, and not on UmuC,
since we observe it in a ΔumuDC strain in which we express only UmuD on a plasmid.
Specifically, the disappearance is dependent on the Lys97 residue of UmuD or UmuD′.
Since this is the residue responsible for self-cleavage of UmuD, it is tempting to
speculate that the disappearance of ε we observe is due to a cleavage reaction mediated
by UmuD/D′. Perhaps UmuD/D′ nicks ε so that it is targeted for degradation by a
protease. To that end, we have also tested ε disappearance in a clpP- strain. ClpP is the
protease implicated in the degradation of ε (Bressanin et al., 2009). We observed that the
disappearance of ε is also dependent on the ClpP protease. We are cautious to take this
leap, however, because we have not yet detected any cleavage of ε in vitro either with
purified proteins or with cell extracts in different genetic backgrounds. There may yet be
some un-discovered mechanism that destroys ε upon UV induction, but it needs to be
further investigated. Whatever the mechanism is, dual color imaging of these strains with
umuC and another Polymerase III subunit, τ, will shed light on whether epsilon
disappearance translates to easier access of replication forks by Pol V.
The observation that UmuC is found on the cell membrane in induced cells,
combined with Evelyn Witkin’s earlier observation that RecA is also on the cell
membrane in SOS induced cells, raises the issue of whether TLS or Pol V activation are
64
membrane related phenomena. The data is not un-ambiguously clear yet. Since there are
so many UmuC spots in the cytosol in backgrounds where SOS is constitutive without
any UV induction, I would venture to say that TLS is occurring in the cytosol. The
number of membrane spots increase with UV induction, but the number of spots on the
whole increases in response to UV. Perhaps when excess UmuC is present, the enzyme is
deactivated rapidly and its fate is the membrane as a form of regulation. There is also the
possibility that Pol V gets activated on the membrane. There is membrane associated
DNA present in damaged cells (Garvey et al., 1985). If this DNA is ssDNA and attracts
RecA, that is a situation right for Pol V activation. A careful time course of these strains
should be able to address some of these issues. Dual color strains will also shed light on
where Pol III replication is occurring in damaged cells and how Pol V gains access to
replication forks.

65
Chapter 3.3 Experimental Procedures
3.3.1 Strain Construction and Single Molecule Imaging
Strain construction was performed in Michael Cox’s lab at UW, Madison. All
proteins were chromosomally tagged at their C-terminal ends with fluorescent protein by
P1 transduction. The wildtype E. coli strain MG1655 was used as the background strain
for all the constructs. This strain behaves more like wildtype E. coli upon UV induction
then other “wildtype” E. coli strains, such as AB1157. There is an 11 amino acid linker
from the C-terminal end of the tagged protein and start of the fluorescent protein: Ser Ala
Gly Ser Ala Ala Gly Ser Gly Glu Phe. The Ypet fluorescent protein gene was purchased
from Addgene and the mKate2 from Evrogen. The constructs are verified, tested for
mutagenesis and then sent to the van Oijen lab for single molecule imaging.
To visualize live cells, they use a commercial microscope with a high NA (1.49)
objective. Coupled into the microscope are home-built optics, with excitation light from
laser sources at wavelengths of 405 nm, 458 nm, 514 nm, 568 nm and 637 nm.
Computer-controlled shutters and filters are used for automated control over the
generation of laser pulses with a large flexibility in intensity, exposure duration,
alternation of colors, and exposure frequency. Using an electron-multiplying CCD
camera, the fluorescence is imaged with high sensitivity and low background enabling
the detection and counting of individual molecules. The bacteria are placed in
temperature-controlled flow cells that allow the rapid exchange of medium and nutrients.
They have developed a protocol that allows the bacteria to be immobilized to the flow-
cell surface by derivatizing the glass with positively charged amine groups. Under these
66
conditions, they can observe doubling rates of the bacteria similar to those obtained in
solution, indicating that surface immobilization does not affect growth.
3.3.2 In Vitro Exonuclease Protection Assays
DNA oligonucleotides were synthesized on ABI’s 3400 DNA Synthesizer. The 28
nt primer was 5′end labeled with γP32-ATP. Free ATP was purified away using Bio-
Rad’s P6 BioSPin columns. The primer was annealed to an 80mer template containing a
biotin on both ends. The primer/template DNA (20nM) was loaded with 100nM
streptavidin. β processivity clamp (200nM) and γ (50nM) loading complex were
incubated with the DNA for 3 minutes with ATP (1mM). The loaded DNA was incubated
with 20nM Pol III core (a gift from Mike O’Donnell) and 0.5μM or 1.0μM of UmuD
2
or
UmuD′
2
. An aliquot of the reactions was taken at 7.5 minutes and 20 minutes and
quenched with STOP solution (95% formamide, 50mM EDTA). The radiolabeled primer
was resolved on a 20% Urea-PAGE gel and imaged using Bio-Rad’s Storm
PhosphorImager.
3.3.3 Rifampicin Mutagenesis
Strains to be examined were streaked out on Luria-Bertani(LB)-agar media and
grown overnight at 37ºC. At least seven individual colonies were picked for each strain
and overnight cultures were made in liquid LB. Liquid overnight cultures for each of the
strains were diluted 1/100 in fresh LB broth, grown to an optical density of 600 nm of 0.2
to 0.3, centrifuged, and then re-suspended in half the volume of ice-cold 0.9% NaCl.
These suspensions were UV irradiated with a hand held 254nm lamp in 6-ml aliquots in a
petri dish. All procedures were performed in subdued light. One mL of irradiated cells
67
was added to 5mL of fresh LB and allowed to recover overnight at 37ºC with gentle
shaking. 100μL of overnight cells were plated onto an LB-agar petri dish containing
50μg/mL Rifampicin. 10μL of cells were diluted 10
6
fold in ice cold 0.9% NaCl and
100μL of the dilution was plated on LB agar plates to determine Rif mutation
frequency/10
8
cells.

68

CHAPTER 4
TOWARDS A POLYMERASE V CRYSTAL STRUCTURE

69
4.0 Introduction
Protein structural information is vital to understanding the interactions proteins
make with their substrates, other proteins, and their environment. There are crystal
structures available for the RecA-nucleoprotein filament (Chen et al., 2008) and for
UmuD
2
(Peat et al., 1996) and an NMR structure for UmuD′
2
(Ferentz, Walker, &
Wagner, 2001). This gives us some idea of what an active Pol V Mut complex may be
composed of, but it would be ideal to know exactly what the structure of an active Pol V
Mut complex looks like.
In the past, obtaining structural information for Pol V in any form would’ve been
a pipe dream because of how difficult Pol V was to purify and the insoluble nature of Pol
V’s catalytic subunit UmuC. Once it was realized that UmuC was purifiable as a co-
complex with UmuD′
2
(Bruck et al., 1996), more protein was obtainable but the
purification process was still complex and laborious. The purification process yielded
only a few milligrams of pure protein from hundreds of liters of induced starting culture
(Schlacher, Jiang, Woodgate, & Goodman, 2006).
Roger Woodgate’s group came up with an expression and purification scheme for
Pol V in which the N-terminal of UmuC was tagged with a hexahistidine tag (Karata et
al., 2012). This construct is on a low copy plasmid upstream of the umuD′ gene and not
behind any inducible promoter. When this plasmid along with another one expressing an
arabinose inducible umuD′ was co-transformed into an expression strain, they were able
to retrieve 1mg protein from 4 liters of starting culture, a great improvement over ~15mg
of protein from 150 liters. The purification time was also cut down from three weeks in
70
the traditional method, to two days with the new construct. What once seemed like a
distant hope was now a very real possibility. We decided to pursue crystallization of Pol
V.

71
4.1 Results/Methods
4.1.1 Improvements on Purification to Increase Yield
Although Karata’s protocol was a great improvement over the traditional
purification and would yield a large amount of protein for most biochemical purposes, it
still was not enough to undertake the task of crystallization. Karata’s scheme also calls
for purification through a hydroxyapatite column (HAP) in the final step. This step is not
suitable for protein that is to be used for crystallography since it would require elution
with phosphate buffers. Phosphate is known to form crystals and it is generally
recommended to avoid the use of phosphate at any point in the purification scheme when
obtaining protein for use in crystallization.
Instead, we changed the purification scheme, so that yield was greatly increased
and the protein was suitable for downstream crystallization. Below, I will briefly describe
some of these changes. Instead of starting with 2L of cultured cells, we used 6L. The salt
concentration was increased in the lysis buffer to 1M NaCl instead of 300mM NaCl. The
method of lysis was also changed from sonication to French Press. This was because the
amount of time required to lyse a larger re-suspension of cells was having an adverse
effect on protein yield.
The first step of the purification requires batch purification with Ni-NTA resin.
Instead of using one Ni-NTA column, the lysate is applied to three columns, each ~1mL.
The columns are washed thoroughly with four washes of buffer (1M NaCl, 50mM Tris-
HCl pH 7.5, 10% glycerol, 2mM DTT, 35mM Imidazole), resuspending the Ni-NTA
resin each time. Bound protein is eluted in two elution steps from each column. In the
72
next step, the pooled elute from the Ni-NTA columns is passed through a gel filtration
column, Pharmacia’s Superdex 200. We include a small amount of Triton-X 100 (0.1%
final concentration) in the buffer to break down larger protein complexes that have
formed and eliminate contaminating proteins from the preparation.
It is after this step that the purification protocol for crystallization changes
dramatically. In the regular purification scheme, the peak fractions from the gel filtration
column containing Pol V are applied to a HAP resin, but for reasons mentioned
previously, we forgo this step to avoid phosphate in the final protein preparation. We
instead dialyse the protein into a more suitable buffer for crystallization. After the
Superdex column, the protein is in HEPES buffer containing high amounts of salt and
glycerol (50mM HEPES pH6.8, 1M NaCl, 20% glycerol, 0.1% Triton-X 100, 2mM
DTT). Since these buffer conditions are not ideal for crystal formation, we stepwise
dialyze the protein into buffer containing lower amounts of salt and glycerol (50mM
HEPES pH6.8, 300mM NaCl, 10% glycerol, 2mM DTT). We do this in steps because
immediate dialysis of the protein into the lower salt and glycerol buffer causes protein
precipitation.
The protein after the gel filtration column and dialysis is very dilute, typically
<0.4mg/mL in ~30mL. The protein must be concentrated before setting up crystal trays.
We are unable to concentrate the protein using membrane filtration because it causes
precipitation on the filters and reduces protein recovery. Instead, we employ the use of
PEG dialysis to concentrate the protein. A dialysis buffer containing 15% PEG 20kDa is
used to increase the concentration of the protein (50mM HEPES pH6.8, 300mM NaCl,
73
10% glycerol, 2mM DTT, 15% PEG 20kDa). With this method we can increase the
concentration of the protein from <0.4mg/mL to ~1.5mg/mL with 85% recovery. The
total protein yield using this method is on average 6mg of protein from 6L of starting
culture.
In an attempt to increase protein concentration, we dialysed the protein in the
presence of primer/template DNA substrate or with RecA. We tried different
combinations of lengths and sequences (Table 4.1). With RecA we were able to increase
the concentration of protein to 2mg/mL with 75% recovery before we observed
precipitation. With the best combination of primer/template we were able to purify the
protein to a concentration of 2.7mg/mL with 55% recovery. We were now ready to set-up
crystal screens with Pol V alone or co-crystals with Pol V + RecA or Pol V + DNA.

Table 4.1. Co-Concentration Conditions For Pol V. Pol V was concentrated with the indicated protein or DNA. Two different DNA
sequences were tried. Sequence A had better recovery. Different variations of length on Sequence A were also tried. Co-crystals were
set-up with the original sequence A.

74
4.1.2 Crystal Screens
Crystal screens were set up in 96-well sitting drop trays at 18ºC by vapor
diffusion with assistance from the Xiaojiang Chen lab. Some of the screens tested
included Qiagen’s Classics Lite, pH Clear, Cations, Am
2
SO
4
and JCSG+. Several drops
showed aggregation almost immediately. Some drops showed phase separation or small
crystals and were chosen for optimization. The conditions which produced crystals were
optimized first.
One promising condition we chose to pursue was Pol V alone with a reservoir
solution of 0.1M HEPES pH7.5, 1.0M MgAcetate. The original crystal was square
shaped and appeared 3-dimensional (Figure 4.1, A). Optimizations were carried out in
hanging-drop vapor diffusion trays. Almost all of these drops became dehydrated within
three weeks, leaving behind metallic looking crystals in some of the conditions. These
crystals were difficult to handle. Preliminary X-ray diffraction on a home-source revealed
no diffraction pattern (Figure 4.1, B) and these crystals could not be verified by running
them on a SDS-PAGE gel. They were difficult to dissolve in solution.
Another promising condition we pursued was Pol V in a complex with a high
affinity RecA mutant, RecA730/ΔC17. The co-crystal was originally rectangular in shape
and the reservoir well condition was 0.2M Ammonium Acetate, 0.1M Tris pH8.5. When
optimizing around this condition, we got several positive hits and the shapes were either
cross- or star-shaped (Figure 4.1, C). These crystals also appeared metallic and were
difficult to dissolve. They could not be verified on a SDS-PAGE gel and X-ray
75
diffraction on a home source revealed water rings and salt spots for some of them (Figure
4.1, D).

Figure 4.1. Crystals and Diffraction Patterns. A. Crystal produced with Pol V Alone. B. Diffraction pattern for crystal in A. C. Co-
crystal of Pol V + RecA. D. Diffraction pattern for crystal in C

The Pol V + DNA conditions had some promising visible phase separation in the
drops. We optimized around the following three conditions: (1) 0.1 mM Tris pH 8.5,
1.1M CaCl
2
; (2) 1.6M Ammonium Sulfate, 0.1M Citric Acid pH 5.0; (3) 0.1M Na-
Acetate pH 4.6, 1.6M Zinc sulfate. The first condition had lots of precipitation in the
drops and there were salt crystals present in the reservoir buffer indicative of the
diffusion buffer producing salt crystals. There was lots of phase separation in the
optimization drops for the second condition, but never any crystals. The third condition
also had great phase separation in most of the optimization drops, but no crystals formed.

76
4.2 Discussion and Future Directions
Although we attempted to crystallize Pol V in many different forms, with various
binding partners, we still have not attempted to crystallize active Pol V Mut. Long term
stability of Pol V Mut may preclude the possibility of trying to “capture the enzyme
alive”. Deactivation of Pol V Mut at lower temperatures is something that needs to be
characterized to answer this question. If the complex is only active once made for a
matter of hours, then crystallizing in this form may prove to be difficult. If on the other
hand, the protein can remain on ice for days or even weeks, without losing activity, then
concentration of the active complex and then crystallization would be a real possibility.
Although it would be best to try to crystallize active Pol V Mut, any form of Pol
V we can crystallize would be a giant step forward in understanding this enzyme. It is my
belief that conditions with Pol V alone or Pol V with RecA and/or DNA still hold
promise. It may be possible to increase the concentration of Pol V alone to more than
2.0mg/mL, but that may not be entirely necessary to obtain crystals. There is lots of
activity in crystal drops at even lower concentrations of Pol V, down to 1mg/mL. There is
precipitation, but also phase separation, and small crystals that form at these lower
concentrations. Some of these conditions are worth pursuing further.

77

BIBLIOGRAPHY
78

Bertram, J. G., Bloom, L. B., O'Donnell, M., & Goodman, M. F. (2004). Increased dNTP
binding affinity reveals a nonprocessive role for Escherichia coli beta clamp with
DNA polymerase IV. [Research Support, U.S. Gov't, P.H.S.]. J Biol Chem,
279(32), 33047-33050.
Boudsocq, F., Campbell, M., Devoret, R., & Bailone, A. (1997). Quantitation of the
inhibition of Hfr x F- recombination by the mutagenesis complex UmuD'C. J Mol
Biol, 270(2), 201-211.
Brenner, S. L., Zlotnick, A., & Griffith, J. D. (1988). RecA protein self-assembly.
Multiple discrete aggregation states. [Research Support, U.S. Gov't, P.H.S.]. J
Mol Biol, 204(4), 959-972.
Bressanin, D., Stefan, A., Piaz, F. D., Cianchetta, S., Reggiani, L., & Hochkoeppler, A.
(2009). Proteolysis of the proofreading subunit controls the assembly of
Escherichia coli DNA polymerase III catalytic core. Biochim Biophys Acta,
1794(11), 1606-1615.
Bruck, I., Woodgate, R., McEntee, K., & Goodman, M. F. (1996). Purification of a
soluble UmuD'C complex from Escherichia coli. Cooperative binding of UmuD'C
to single-stranded DNA. J Biol Chem, 271(18), 10767-10774.
Burckhardt, S. E., Woodgate, R., Scheuermann, R. H., & Echols, H. (1988). UmuD
mutagenesis protein of Escherichia coli: overproduction, purification, and
cleavage by RecA. Proc Natl Acad Sci U S A, 85(6), 1811-1815.
Chen, Z., Yang, H., & Pavletich, N. P. (2008). Mechanism of homologous recombination
from the RecA-ssDNA/dsDNA structures. Nature, 453(7194), 489-484.
Chin, J. W., Martin, A. B., King, D. S., Wang, L., & Schultz, P. G. (2002). Addition of a
photocrosslinking amino acid to the genetic code of Escherichia coli. Proc Natl
Acad Sci U S A, 99(17), 11020-11024.
Chin, J. W., Santoro, S. W., Martin, A. B., King, D. S., Wang, L., & Schultz, P. G.
(2002). Addition of p-azido-L-phenylalanine to the genetic code of Escherichia
coli. J Am Chem Soc, 124(31), 9026-9027.
Dorman, G., & Prestwich, G. D. (1994). Benzophenone photophores in biochemistry.
Biochemistry, 33(19), 5661-5673.
Dutreix, M., Burnett, B., Bailone, A., Radding, C. M., & Devoret, R. (1992). A partially
deficient mutant, recA1730, that fails to form normal nucleoprotein filaments.
Mol Gen Genet, 232(3), 489-497.
Dutreix, M., Moreau, P. L., Bailone, A., Galibert, F., Battista, J. R., Walker, G. C., &
Devoret, R. (1989). 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, 171(5), 2415-2423.
Egelman, E. H., & Stasiak, A. (1986). Structure of helical RecA-DNA complexes.
Complexes formed in the presence of ATP-γ-S or ATP. J Mol Biol, 191(4), 677-
697.
79
Ferentz, A. E., Walker, G. C., & Wagner, G. (2001). Converting a DNA damage
checkpoint effector (UmuD2C) into a lesion bypass polymerase (UmuD'2C).
EMBO J, 20(15), 4287-4298.
Fernandez de Henestrosa, A. R., Ogi, T., Aoyagi, S., Chafin, D., Hayes, J. J., Ohmori, H.,
& Woodgate, R. (2000). Identification of additional genes belonging to the LexA-
regulon in Escherichia coli. Molecular Microbiology, 35, 1560-1572.
Frank, E. G., Cheng, N., Do, C. C., Cerritelli, M. E., Bruck, I., Goodman, M. F., Steven,
A. C. (2000). Visualization of two binding sites for the Escherichia coli
UmuD'(2)C complex (DNA pol V) on RecA-ssDNA filaments. J Mol Biol,
297(3), 585-597.
Friedberg, E. C., Walker, G. C., Siede, W., Wood, R. D., Schultz, R. A., & Ellenberger,
T. (2006). DNA Repair and Mutagenesis (2nd ed. ed.). Washington, D. C.: ASM
Press.
Fujii, S., Gasser, V., & Fuchs, R. P. (2004). The biochemical requirements of DNA
polymerase V-mediated translesion synthesis revisited. J Mol Biol, 341(2), 405-
417.
Galardy, R. E., Craig, L. C., & Printz, M. P. (1973). Benzophenone triplet: a new
photochemical probe of biological ligand-receptor interactions. Nat New Biol,
242(117), 127-128.
Garvey, N., St John, A. C., & Witkin, E. M. (1985). Evidence for RecA protein
association with the cell membrane and for changes in the levels of major outer
membrane proteins in SOS-induced Escherichia coli cells. J Bacteriol, 163(3),
870-876.
George, J., Devoret, R., & Radman, M. (1974). Indirect ultraviolet-reactivation of phage
lambda. Proc Natl Acad Sci U S A, 71(1), 144-147.
Georgescu, R. E., Kurth, I., & O'Donnell, M. E. (2012). Single-molecule studies reveal
the function of a third polymerase in the replisome. Nat Struct Mol Biol, 19(1),
113-116.
Heller, R. C., & Marians, K. J. (2006). Replisome assembly and the direct restart of
stalled replication forks. Nat Rev Mol Cell Biol, 7(12), 932-943.
Heuser, J., & Griffith, J. (1989). Visualization of RecA protein and its complexes with
DNA by quick-freeze/deep-etch electron microscopy. J Mol Biol, 210(3), 473-
484.
Indiani, C., Langston, L. D., Yurieva, O., Goodman, M. F., & O'Donnell, M. (2009).
Translesion DNA polymerases remodel the replisome and alter the speed of the
replicative helicase. Proc Natl Acad Sci U S A, 106(15), 6031-6038.
Jewett, J. C., & Bertozzi, C. R. (2010). Cu-free click cycloaddition reactions in chemical
biology. Chem Soc Rev, 39(4), 1272-1279.
Jiang, Q., Karata, K., Woodgate, R., Cox, M. M., & Goodman, M. F. (2009). The active
form of DNA polymerase V is UmuD'(
2
)C-RecA-ATP. Nature, 460(7253), 359-
363.
Karata, K., Vaisman, A., Goodman, M. F., & Woodgate, R. (2012). Simple and efficient
purification of Escherichia coli DNA polymerase V: cofactor requirements for
optimal activity and processivity in vitro. DNA Repair (Amst), 11(4), 431-440.
80
Kato, T., & Shinoura, Y. (1977). Isolation and characterization of mutants of Escherichia
coli deficient in induction of mutations by ultraviolet light. Mol Gen Genet,
156(2), 121-131.
Kauer, J. C., Erickson-Viitanen, S., Wolfe, H. R., Jr., & DeGrado, W. F. (1986). p-
Benzoyl-L-phenylalanine, a new photoreactive amino acid. Photolabeling of
calmodulin with a synthetic calmodulin-binding peptide. J Biol Chem, 261(23),
10695-10700.
Kurumizaka, H., & Shibata, T. (1996). Homologous recognition by RecA protein using
non-equivalent three DNA-strand-binding sites. J Biochem, 119(2), 216-223.
Lata, S., Gavutis, M., Tampe, R., & Piehler, J. (2006). Specific and stable fluorescence
labeling of histidine-tagged proteins for dissecting multi-protein complex
formation. J Am Chem Soc, 128(7), 2365-2372.
Leake, M. C., Greene, N. P., Godun, R. M., Granjon, T., Buchanan, G., Chen, S., Berks,
B. C. (2008). Variable stoichiometry of the TatA component of the twin-arginine
protein transport system observed by in vivo single-molecule imaging. Proc Natl
Acad Sci U S A, 105(40), 15376-15381.
LeClerc, J. E., Borden, A., & Lawrence, C. W. (1991). The thymine-thymine pyrimidine-
pyrimidone(6-4) ultraviolet light photoproduct is highly mutagenic and
specifically induces 3' thymine-to-cytosine transitions in Escherichia coli. Proc
Natl Acad Sci U S A, 88(21), 9685-9689.
Lehmann, A. R., Niimi, A., Ogi, T., Brown, S., Sabbioneda, S., Wing, J. F., . . . Green, C.
M. (2007). Translesion synthesis: Y-family polymerases and the polymerase
switch. DNA Repair (Amst), 6(7), 891-899.
Lia, G., Michel, B., & Allemand, J. F. (2012). Polymerase exchange during Okazaki
fragment synthesis observed in living cells. Science, 335(6066), 328-331.
Little, J. W., Edmiston, S. H., Pacelli, L. Z., & Mount, D. W. (1980). Cleavage of the
Escherichia coli LexA protein by the RecA protease. Proc Natl Acad Sci U S A,
77(6), 3225-3229.
Marians, K. J. (2008). Understanding how the replisome works. Nat Struct Mol Biol,
15(2), 125-127.
McGrew, D. A., & Knight, K. L. (2003). Molecular design and functional organization of
the RecA protein. Crit Rev Biochem Mol Biol, 38(5), 385-432.
Ning, X., Guo, J., Wolfert, M. A., & Boons, G. J. (2008). Visualizing metabolically
labeled glycoconjugates of living cells by copper-free and fast huisgen
cycloadditions. Angew Chem Int Ed Engl, 47(12), 2253-2255.
Nohmi, T., Battista, J. R., Dodson, L. A., & Walker, G. C. (1988). RecA-mediated
cleavage activates UmuD for mutagenesis: mechanistic relationship between
transcriptional derepression and posttranslational activation. Proc Natl Acad Sci U
S A, 85(6), 1816-1820.
Patel, M., Jiang, Q., Woodgate, R., Cox, M. M., & Goodman, M. F. (2010). A new model
for SOS-induced mutagenesis: how RecA protein activates DNA polymerase V.
Crit Rev Biochem Mol Biol, 45(3), 171-184.
81
Peat, T. S., Frank, E. G., McDonald, J. P., Levine, A. S., Woodgate, R., & Hendrickson,
W. A. (1996). Structure of the UmuD' protein and its regulation in response to
DNA damage. Nature, 380(6576), 727-730.
Pham, P., Bertram, J. G., O'Donnell, M., Woodgate, R., & Goodman, M. F. (2001). A
model for SOS-lesion-targeted mutations in Escherichia coli. Nature, 409(6818),
366-370.
Reid, S. W., Leake, M. C., Chandler, J. H., Lo, C. J., Armitage, J. P., & Berry, R. M.
(2006). The maximum number of torque-generating units in the flagellar motor of
Escherichia coli is at least 11. Proc Natl Acad Sci U S A, 103(21), 8066-8071.
Reuven, N. B., Arad, G., Maor-Shoshani, A., & 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(45), 31763-31766.
Reyes-Lamothe, R., Sherratt, D. J., & Leake, M. C. (2010). Stoichiometry and
architecture of active DNA replication machinery in Escherichia coli. Science,
328(5977), 498-501.
Sakmar, T. P. (2011). Receptors: clicking class B GPCR ligands. Nat Chem Biol, 7(8),
500-501.
Schlacher, K., Cox, M. M., Woodgate, R., & Goodman, M. F. (2006). RecA acts in trans
to allow replication of damaged DNA by DNA polymerase V. Nature, 442(7105),
883-887.
Schlacher, K., Jiang, Q., Woodgate, R., & Goodman, M. F. (2006). Purification and
characterization of Escherichia coli DNA polymerase V. Methods Enzymol, 408,
378-390.
Schlacher, K., Leslie, K., Wyman, C., Woodgate, R., Cox, M. M., & Goodman, M. F.
(2005). DNA polymerase V and RecA protein, a minimal mutasome. Mol Cell,
17(4), 561-572.
Scrima, A., Konickova, R., Czyzewski, B. K., Kawasaki, Y., Jeffrey, P. D., Groisman, R.,
Thoma, N. H. (2008). Structural basis of UV DNA-damage recognition by the
DDB1-DDB2 complex. Cell, 135(7), 1213-1223.
Shinagawa, H., Iwasaki, H., Kato, T., & Nakata, A. (1988). RecA protein-dependent
cleavage of UmuD protein and SOS mutagenesis. Proc Natl Acad Sci U S A,
85(6), 1806-1810.
Sommer, S., Boudsocq, F., Devoret, R., & Bailone, A. (1998). Specific RecA amino acid
changes affect RecA-UmuD'C interaction. Mol Microbiol, 28(2), 281-291.
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(1), 87-93.
Story, R. M., Weber, I. T., & Steitz, T. A. (1992). The structure of the E. coli recA
protein monomer and polymer. Nature, 355(6358), 318-325.
Sweasy, J. B., Witkin, E. M., Sinha, N., & Roegner-Maniscalco, V. (1990). RecA protein
of Escherichia coli has a third essential role in SOS mutator activity. J Bacteriol,
172(6), 3030-3036.
Szekeres, E. S., Jr., Woodgate, R., & Lawrence, C. W. (1996). Substitution of mucAB or
rumAB for umuDC alters the relative frequencies of the two classes of mutations
82
induced by a site-specific T-T cyclobutane dimer and the efficiency of translesion
DNA synthesis. J Bacteriol, 178(9), 2559-2563.
Tang, M., Pham, P., Shen, X., Taylor, J. S., O'Donnell, M., Woodgate, R., & Goodman,
M. F. (2000). Roles of E. coli DNA polymerases IV and V in lesion-targeted and
untargeted SOS mutagenesis. Nature, 404(6781), 1014-1018. doi:
10.1038/35010020
Tang, M., Shen, X., Frank, E. G., O'Donnell, M., Woodgate, R., & Goodman, M. F.
(1999). UmuD'(
2
)C is an error-prone DNA polymerase, Escherichia coli pol V.
Proc Natl Acad Sci U S A, 96(16), 8919-8924.
Uchida, K., Furukohri, A., Shinozaki, Y., Mori, T., Ogawara, D., Kanaya, S., Akiyama,
M. (2008). Overproduction of Escherichia coli DNA polymerase DinB (Pol IV)
inhibits replication fork progression and is lethal. Mol Microbiol, 70(3), 608-622.
Wang, Q., Chan, T. R., Hilgraf, R., Fokin, V. V., Sharpless, K. B., & Finn, M. G. (2003).
Bioconjugation by copper(I)-catalyzed azide-alkyne [3 + 2] cycloaddition. J Am
Chem Soc, 125(11), 3192-3193.
Weigle, J. J. (1953). Induction of mutation in a bacterial virus. Proceedings of the
National Academy of Sciences USA, 39, 628-636.
Witkin, E. M. (1967). The radiation sensitivity of Escherichia coli B: a hypothesis
relating filament formation and prophage induction. Proc Natl Acad Sci U S A,
57(5), 1275-1279.
Woodgate, R., & Ennis, D. G. (1991). Levels of chromosomally encoded Umu proteins
and requirements for in vivo UmuD cleavage. Mol Gen Genet, 229(1), 10-16.
Woodgate, R., Rajagopalan, M., Lu, C., & Echols, H. (1989). UmuC mutagenesis protein
of Escherichia coli: purification and interaction with UmuD and UmuD'. Proc
Natl Acad Sci U S A, 86(19), 7301-7305.
Xie, J., & Schultz, P. G. (2005). An expanding genetic code. Methods, 36(3), 227-238.
Young, T. S., & Schultz, P. G. (2010). Beyond the canonical 20 amino acids: expanding
the genetic lexicon. J Biol Chem, 285(15), 11039-11044.
Yu, X., & Egelman, E. H. (1993). The LexA repressor binds within the deep helical
groove of the activated RecA filament. J Mol Biol, 231(1), 29-40.

Abstract (if available)

Abstract In Escherichia Coli, cell survival and genomic stability after UV radiation or other DNA damaging agents depends on repair mechanisms induced as part of the SOS response. The SOS response involves upregulation of over 40 genes, which are normally repressed by the LexA protein under non-damage conditions. The early phase of the SOS response is centered around error-free DNA repair mechanisms. These include base excision repair (BER), nucleotide excision repair (NER), and recombinational repair. If these mechanisms are insufficient to surmount the replicational impedance brought about by DNA damage, the bacterial cell advances into the later part of the SOS response. This later phase of the response helps ensure cell survival albeit at the cost of faithful DNA repair. Known as SOS mutagenesis, this phase is characterized by elevated mutation levels, which are a direct result of error-prone DNA replication. SOS mutagenesis is the result of the low fidelity DNA polymerase V. DNA polymerase V (Pol V) is a heterotrimeric protein composed of the catalytically active UmuC and a dimer of UmuD'. The heterotrimer is essentially inactive, however, and must strip a RecA monomer and an ATP cofactor from the 3' tip of a RecA-DNA nucleoprotein filament to form an active Pol V mutasome composed of UmuC:UmuD'₂:RecA:ATP. This dissertation focuses on investigating the Pol V mutasome including deducing protein-protein interactions in the active, inactive, and de-activated states of the mutasome, detecting a conformational change in the different activity states of the mutasome, visualizing the SOS response in live cells, and attempting to ascertain the crystal structure of Pol V.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Decrypting Escherichia coli DNA polymerase V mutasome ATP regulation

PDF

Physiological roles and evolutionary implications of alternative DNA polymerases in Escherichia coli

PDF

Activation mechanism of damaged-induced DNA polymerase V in Escherichia coli

PDF

The role of ATP in the regulation of Escherichia coli DNA polymerase V activity

PDF

Investigating highly mutagenic polymerases in pathogenic bacteria

PDF

The mechanism of damage-induced mutations in Escherichia coli

PDF

Structural and biochemical studies of large T antigen: the SV40 replicative helicase

PDF

Genetic diversity and bacterial death in the context of adaptive evolution

PDF

Expansion of deoxynucleotide analog probes for studying DNA polymerase mechanism: synthesis of novel beta, gamma-CXY deoxycytidine series

PDF

Generation of monoclonal antibodies via phage display and in vitro affinity maturation using activation induced deoxycytidine deaminase and DNA polymerase eta

PDF

Synthetic studies of chemical probes for i) DNA, ii) RNA polymerases and iii) tropomyosin receptor kinase

PDF

Mechanisms of long-term survival in Escherichia coli

PDF

The contribution of chemical steps to DNA polymerase beta catalysis and fidelity

PDF

Unraveling the role of iron-sulfur clusters in DNA repair enzymes

PDF

Mechanism of human nonhomologous DNA end joining

PDF

Dps contributes to typical growth, survival, and genome organization in E. coli

PDF

X-ray structural studies on DNA-dependent protein kinase catalytic subunit:DNA co-crystals

PDF

Nutritional competence in Escherichia coli: the use of double-stranded DNA as a nutrient

PDF

Simulating the helicase motor of SV40 large tumor antigen

PDF

The intra-S phase checkpoint and its effect on replication fork dynamics in saccharomyces cerevisiae

Asset Metadata

Creator Patel, Meghna (author)

Core Title Investigating the E. Coli DNA polymerase V mutasome

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Molecular Biology

Publication Date 01/29/2013

Defense Date 10/05/2012

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

Tag DNA Polymerase V,error-prone DNA repair,OAI-PMH Harvest,translesion DNA synthesis,UV mutagenesis

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Goodman, Myron F. (committee chair), Pinaud, Fabien (committee member), Reisler, Hanna (committee member)

Creator Email meghnapa@usc.edu,meghnayadav@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-129615

Unique identifier UC11290526

Identifier usctheses-c3-129615 (legacy record id)

Legacy Identifier etd-PatelMeghn-1414.pdf

Dmrecord 129615

Document Type Dissertation

Rights Patel, Meghna

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