University of Southern California Dissertations and Theses

Biochemical characterization and structural analysis of two hexameric helicases for eukaryotic DNA replication

(USC Thesis Other)

Biochemical characterization and structural analysis of two hexameric helicases for eukaryotic DNA replication

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

BIOCHEMICAL CHARACTERIZATION AND STRUCTURAL ANALYSIS

OF TWO HEXAMERIC HELICASES FOR EUKARYOTIC DNA REPLICATION

by

Meng Xu

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
(GENETIC, MOLECULAR AND CELLULAR BIOLOGY)

December 2012

Copyright 2012 Meng Xu

ii

Epigraph

Build one’s body, cultivate one’s mind
Forge a benevolent heart, and bring harmony to the family
Lead the nation in order, and spread fairness to the world

-The Great Learning, Book of Rites

iii

Dedication

For my grandmother, DU Xinan, and my grandfathers, XU Yao and GAO Xin.
I am so sorry for not being there for you.

iv

Acknowledgements

My gratitude to those who supported and helped me all the way along for the past several
years cannot be fully expressed in this single page. It has been a tough run and I can
barely imagine this moment of fulfillment and inner peace without help from the
following people.

Dr. Xiaojiang Chen, for being a great and, most importantly, a patient mentor;
Dr. Y. Paul Chang, for being my role model in science and a good friend;
Dr. Lin Chen and Dr. Ian Haworth, for your advice as my committee members;
Dr. Ganggang Wang, for all the conversations that brought me patience and tranquility;
Dr. Xian Jessica Yu, for being a supportive colleague;
My parents, for keeping a cozy home for me where I can always retreat to once in a while;
Yingjing, Yan and Haibin, for your encouragements whenever I feel down;
Xiaoou, for being the lighthouse when I sail in the darkness.

To those I do not name individually, please forgive my blanket thanks. Trust me, I will
do my best to show my gratefulness when you need me.

v

Table of Contents

Epigraph ii

Dedication iii

Acknowledgements iv

List of Tables vi

List of Figures vii

Abbreviations viii

Abstract ix

Chapter 1: Introduction 1

Chapter 2: Expression, Purification and Biochemical Characterization of
Schizosaccharomyces pombe Mcm4, 6 and 7 12
2.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.5 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

Chapter 3: Crystallization and Preliminary X-ray Crystallographic studies on SV40
Large Tumor Antigen Binding to Origin DNA 44
3.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
3.5 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

Bibliography 69

vi

List of Tables

2.1 Summary of biochemical properties of fragments of Mcm4, 6 and 7 . . . . . . . . . . . . .16
3.1 Diffraction data and refinement statistics of two LTag/origin DNA
crystals/structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

vii

List of Figures

1.1 Models of helicase loading and activation for genomic DNA replication . . . . . . . . . . 3

1.2 Overview of SV40 LTag and replication origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2.1 Designs of truncated fragments of Mcm4, 6 and 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.2 Oligomeric states and interactions of N-terminal fragments of Mcm4, 6 and 7 . . . . . 19

2.3 Gel filtration chromatography profiles of core and near-full-length
fragments of Mcm4, 6 and 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

2.4. Interactions and oligomeric states of co-expressed fragments of Mcm4, 6 and 7 . . . 22

2.5 Schematics of proposed models of the Mcm4/6/7 hexamer and
the Mcm2-7 double-hexamer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

S.1 Sequence alignment of MCM proteins from various organisms . . . . . . . . . . . . . . . . 40

S.2 Disordered profile plots of Mcm6 and 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

3.1 SV40 LTag domain structures and core replication origin DNA sequence . . . . . . . . 46

3.2 Diffraction image of the hexameric LTag/dsDNA crystal . . . . . . . . . . . . . . . . . . . . . 47

3.3 Initial density map of the hexameric LTag/dsDNA crystal . . . . . . . . . . . . . . . . . . . . 48

3.4 Overall structure of the LTag dimer in complex with EP-ori DNA . . . . . . . . . . . . . . 50

3.5 Detailed interactions between LTag and EP-ori DNA . . . . . . . . . . . . . . . . . . . . . . . . 53

3.6 The narrow minor groove geometry of the 5’-C
5
T
6
T
7
C
8
T
9
region of EP-ori DNA
where H513 residues on both β-hairpins bind in our crystal structure . . . . . . . . . . . . 56

3.7 Mutations introduced on the interface between adjacent Zn and AAA+ domains . . . 58

3.8 Generating a stable LTag dimer intermediate and analysis of its activities . . . . . . . . 59

viii
Abbreviations

MCM Minichromosome maintenance
pre-RC Pre-replicative complex
ORC Origin recognition complex
CDK Cyclin dependent kinase
DDK Dbf4-dependent kinase
LTag Simian virus 40 large tumor antigen
ssDNA Single strand DNA
dsDNA Double strand DNA
PEN Pentanucleotides
AT AT-rich
EP Early palindrome
S. pombe S. pombe, Schizosaccharomyces
sp S. pombe
S. cerevisiae Saccharomyces cerevisiae
sc S. cerevisiae
Mt Methanothermobacter thermautotrophicus
Sso Sulfolobus solfataricus
E. coli Escherichiai coli
PCR Polymerase chain reaction

ix
Abstract

The hetero-hexamer of the eukaryotic minichromosome maintenance (MCM) proteins
plays an essential role in replication of genomic DNA. The ring-shaped Mcm2-7
hexamers comprising one of each subunit show helicase activity in vitro, and form
double-hexamers on DNA. The Mcm4/6/7 also forms a hexameric complex with helicase
activity in vitro. We used an E. coli expression system to express various domains of
Schizosaccharomyces pombe Mcm4, 6 and 7 in order to characterize their domain
structure, oligomeric states, and possible inter-/intra-subunit interactions. We also
successfully employed a co-expression system to express Mcm4/6/7 at the same time in E.
coli, and have purified functional Mcm4/6/7 complex in a hexameric state in high yield
and purity, providing a means for generating large quantity of proteins for future
structural and biochemical studies. Based on our results and those of others, models were
proposed for the subunit arrangement and architecture of both the Mcm4/6/7 hexamer
and the Mcm2-7 double-hexamer.

To circumvent the apparent complexity in eukaryotic DNA replication, Simian virus
large tumor antigen (SV40 LTag) has been studied as in a simplified model system for
eukaryotic replication. LTag alone can recognize and assemble as double-hexamers on
the SV40 replication origin, leading to melting/unwinding of the dsDNA origin. We
obtained an LTag construct that is composed of the origin binding domain (OBD), Zn
domain and AAA+ domain co-crystalized the origin dsDNA in both hexameric and

x
dimeric forms. The structure of the dimeric LTag/dsDNA complex reveals how OBDs
recognize and bind to the origin. The interaction between the β-hairpin and the minor
groove region on the origin suggests a shape based read-out mechanism. Monte Carlo
(MC) simulations and in vitro biochemical assays indicate the structure we obtained
represents a snapshot of an early stage assembly.

1
Chapter 1
Introduction
Eukaryotic DNA replication
Precise replication of genomic DNA is a crucial step in the cell cycle and is carried out
by the replisome, a complex composed of multiple proteins. The replisome assembles and
initiates replication of parental DNA duplex at certain specific genomic DNA sites,
named replication origins. When there is only one origin in the whole circular genome,
such as in Escherichia coli, a pair of replisomes assembles at the single origin and
duplicates the genome bi-directionally. However, eukaryotic genomes are much larger
than prokaryotic genomes. For example, the human genome is 700-fold larger than the E.
coli genome, which requires eukaryotes to evolve additional levels of complexity to adapt
for repaid and controlled replication of whole genomes.

In eukaryotes, a large number of replication origins are used for a complete DNA
duplication. Given the apparent efficiency advantage of this strategy, starting of all
replisomes has to be highly coordinated in S phase of the cell cycle, resulting in neither
unreplicated nor multi-replicated genomic DNA. To ensure this, replication origins have
to be marked or “licensed” with stably binding complexes at the end of mitosis and G1
phase and maintain inactive until S phase [1]. Minichromosome maintenance (MCM)
proteins, specifically, Mcm2-7 in eukaryotes serve as both licensing factors and the
helicase in DNA replication (Figure 1.1). Unlike bacteria, no consensus sequence of

2
replication origins has been identified in metazoans [2]; the initiator complex, comprising
origin recognition complex subunit 1-6 (Orc1-6) and Cdc6, assemble on the origin to
subsequently recruit loading of Mcm2-7 (Figure 1.1A) [3, 4]. It should be noted that the
helicase loading in eukaryotes is negative controlled by CDKs. In the most recent model,
Mcm2-7 formed the heptamer with Cdt1, and the two heptamers are recruited by a single
Orc6 subunit (Figure 1.1B) [5]. In this way, Mcm2-7 hexamers are loaded around the
origin as head-to-head double-hexamers (Figure 1.1C) [6, 7]. The loading of Mcm2-7 by
ORC and Cdc6 involves ATP hydrolysis, indicating the ring-shaped Mcm2-7 hexamer
might be opened to encircle dsDNA (Figure 1.1D) [8]. The pre-replicative complex (pre-
RC) composed of ORC, Cdc6, Cdt1 and Mcm2-7 finishes assembly around the origin and
ready for replication initiation in S phase. When cells enter S phase, Mcm2-7 and other
factors are activated by CDKs and DDK [9-12]. On the other hand, the activation of pre-
RC is also negatively controlled by checkpoint pathways to prevent replication of
damaged DNA [13, 14]. Upon phosphorylation of Mcm2-7 and binding of GINS and
Cdc45, the Mcm2-7 hexamer is activated and functions as a helicase to melt/unwind
dsDNA (Figure 1.1E) [15]. Meanwhile DNA polymerase α-primase (Pol α-primase),
DNA polymerases (Pol) δ and ε join the Cdc45/Mcm2-7/GINS (CMG) complex to form
a complete replisome [16]. As replication forks progress, Mcm2-7 migrate away from the
origin. In summary, such periodic association and dissociation of Mcm2-7 on the origin
ensures each origin can be fired only once in each cell cycle.

3

Figure 1.1. Models of helicase loading and activation for genomic DNA replication
initiation in eukaryotes and Simian virus 40 (SV40). (A-E) Stages of eukaryotic DNA
licensing/replication that are described in the text. Asterisk, replication initiation in
eukaryotes still requires activation and assembly of several other factors on replication
forks, such as Mcm10 and Dpb11 [4]. (F-H) Stages of SV40 DNA replication. All DNA
polymerase and replication protein A (RPA) in the replisome are not illustrated for
conciseness.

4
Biochemical and structural characterization of MCM proteins
MCM proteins were first identified in S. cerevisiae by mutants screening, and then
identified in metazoans by immunological detection or copurification with DNA
polymerase α in mice [17], cows and humans [18]. As long as the highly conserved
MCM core domain that contains Walker A, B and R-finger motifs was initially
characterized in MCM proteins in S. cerevisiae by sequence alignments, homologs in this
family have been discovered in eukaryotes from yeast to human [19]. Homologs found
from higher eukaryotes are corresponding to the six Mcm2-7 classes, except Mcm8 and
Mcm9, which are only found in multicellular organisms [20]. The conservation among
MCM proteins from yeast to human within each Mcm2-7 classes, suggests divergence in
functions of each MCM. Interestingly, MCM proteins are also found in Archaea, which
are believed to process DNA replication mechanism similar to eukaryotes. However, only
a few MCM classes have been identified in Archaea. For example, Methanothermobacter
thermautotrophicus and Archaeoglobus fulgidus have just one MCM protein, which is
closer to Mcm4 than to the other Mcm2-7 classes in terms of sequence homology [21].

Interactions among MCM proteins have been demonstrated by different methods for
different organisms. MCM proteins can form various oligomeric complexes, including
double-hexamers [6, 7], hexamers [22, 23], tetramers [24], trimers [25], and dimers [22,
26, 27]. In addition to interactions among different MCM proteins in eukaryotes with
various combinations, self-interactions of some MCM proteins have also been identified
[22, 28]. The intact MCM hexamer is stable enough to be purified by co-expression and

5
co-pulldown with a single tagged subunit, and remains active under various conditions
for in vitro activity assays [22, 29, 30]. The biological significance of these distinct
oligomeric states still remains elusive and has invited extensive studies recently. Thus, a
high-resolution structure of any of those complexes would be very helpful to gain
insights into how they assemble and function as essential roles in eukaryotic DNA
replication.

As an alternative and enzymatically active version, the Mcm4/6/7 hexamer, composed of
two of each subunit, has been studied as a model helicase to elucidate the mechanism of
this DNA unwinding machinery [29, 31-33]. Given both Mcm2 and Mcm3/5 have been
found as negative regulatory factors for helicase activity of the Mcm4/6/7 hexamer [29,
34, 35], they might play modulating roles in the Mcm2-7 hexamer. In comparison, Mcm4,
6 and 7 directly engage in ATP binding and DNA unwinding.

Ultrastructural data of the Mcm2-7 double-hexamer, the Mcm2-7 hexamer, the
Mcm2/4/6/7 tetramer and the Mcm4/6/7 hexamer are available now [6, 28, 30, 35].
Unlike most hexameric helicases that form a planar ring-shape, S. pombe Mcm2-7
hexamer purifed from G2-phase cells has a staggered globular shape of about 27 nm in
diameter [30]. In contrast, human Mcm4/6/7 hexamer shows only a planar shape in two
sizes, one is 19.5 nm in mean diameter and another one is 27 nm [35]. Both the globular
and planar hexamers have a central cavity which resembles DNA-encircling channels
found on other hexameric helicases [36].

6

High-resolution structures of MCM proteins have been obtained from only archaeal
MCM, including Methanothermobacter thermautotrophicus MCM (MtMCM, PDB:1LTL)
and Sulfolobus solfataricus MCM (SsoMCM, PDB:2VL6 and 3F9V) [37-39]. Two of the
three structures contain only the N-terminal domain and form an N-terminus-to-N-
terminus (N-N) double-hexamer (MtMCM) and a ring-shaped hexamer (SsoMCM),
whereas the near-full-length SsoMCM structure is in monomeric state. More specifically,
MtMCM N-terminal fragments that do not contain the MCM core domain form two
homo-hexamers and then assemble head to head with their N-terminal facing inwards,
and the missing C-terminal fragments are thought to be on the outer sides of the double-
hexamer. Several features of this partial MtMCM structure are also present on eukaryotic
MCM proteins. First, a zinc-finger motif is crucial in mediating hexamer-hexamer
interaction; putative zinc-finger motifs are also found on MCM2, 4, 6 and 7. The
functional importance of this motif has not only been demonstrated in vivo by
mutagenesis studies on the Mcm2-7 hexamer [40], but also suggested by in vitro ssDNA
binding assay on the Mcm4/6/7 hexamer with a mutation on Mcm4 which destabilize the
hexamer [41]. Second, the N-termini of MCM proteins play important roles in hexamer
formation as well, which were indicated by the deletion of 204 amino acid residues at N-
terminal of yeast Mcm2 [42]. Besides, the N-terminus of an archaeal MCM is also
suspected to stimulate helicase activity of the C-terminal MCM core [43]. Third, the N-N
double-hexamer of Mcm4/6/7 has been observed on forked DNA structure and capable of
unwinding duplex DNA region of about 600 bp [32]. However, it has been reported that

7
the function of MCM subunits in eukaryotes seems to be divergent [33, 41]. Therefore,
the archaeal MCM homo-hexamers shall only serve as restricted and simplified versions
to study the eukaryotic DNA replication apparatus.

Simian virus 40 large T antigen: a much simplified model
Given the complexity of genomic DNA replication in eukaryotes, Simian virus 40 (SV40)
large T antigen (LTag) has been used as a model system to investigate the initiation of
DNA replication [44, 45]. As a member of polyomaviridae, SV40 has a circular double
strand genome of 5243 bp, which encodes three structural virion proteins (VP1, VP2 and
VP3) and two nonstructural proteins (large T antigen and small T antigen) [46]. SV40
can infect permissive cells and subsequently recruits the cellular transcription system to
express its own proteins. As a result, LTag is expressed first and directly involves in viral
DNA replication and virion assembly.

In sharp contrast to the human genome that has 30,000 to 50,000 origins activated in each
cell cycle [47], the whole SV40 genome has only one well-defined replication origin,
which is recognized by LTag and results in assembly of two LTag hexamers on the origin
(Figure 1.1F and G) [48]. LTag hexamers alone are competent to melt and unwind
dsDNA with energy provided by ATP hydrolysis (Figure 1.1H). Then DNA replication
machinery of the host cells is recruited to function with LTag, carrying out viral DNA
replication. As shown in the side-by-side comparison presented in Figure 1.1, replication
initiation for SV40 is much simplified than that for eukaryotes: one single protein, LTag,

8
fulfills the functions of Orc1-6, Cdc6, Cdt1, Mcm2-7, GINS and Cdc45, in origin
recognition, melting and unwinding [49]. In addition to the simplicity, unlike Mcm2-7,
structures of LTag and the LTag hexamers in various nucleotide binding states are
available for detailed elucidation of this multi-functional protein.

As shown in Figure 1.2A and B, NMR and X-ray structures have been obtained for all
three major domains of LTag: the DnaJ homology domain, the origin binding domain
(OBD) and the helicase domain composed of a Zn domain and an AAA+ domain [50, 51].
DnaJ homology and HR domain are not involved in in vitro helicase activity of LTag.
OBDs play essential roles in origin recognition. LTag belongs to the SF3 superfamily
helicase and possesses 3’ to 5’ helicase activity, the same as Mcm2-7 [36].

Similar to all hexameric ring-shaped helicases, the LTag hexamer has a negative charged
central channel, in which ssDNA or dsDNA is encircled [50]. The AAA+ domain directly
contact encircled DNA with a β-hairpin of residue 508-517, and six β-hairpins point
inward the central channel to form a planar “shutter” that is the narrowest part in the
central channel (Figure 1.2C). Dramatic conformational change of the AAA+ domain
relative to the Zn domain has been founded in various nucleotide binding states. As a
result, the β-hairpin shifts upward about 17 Å from nucleotide-free (Nt-free) state to
ATP-bound state. Meanwhile the central channel also shrinks from ~22 Å to ~14 Å, as
measured from peptide backbone (Figure 1.2D) [51]. The β-hairpins, especially the two

9
residues at the tip of the β-hairpins, K512 and H513, play essential roles in DNA melting
and unwinding [52]. More details are discussed in Chapter 3.

The function of 64 bp SV40 core origin DNA has been well characterized [53]. Four 5’-
GAGGC-3’ pentanucleotides (PEN1-4) are flanked by an early palindrome (EP) region
and an AT-rich (AT) region (Figure 1.2F). Each PEN is recognized by one OBD and
each half of the origin can support assembly of one LTag hexamer. Due to the inverted
orientations of PEN1/2 and PEN3/4, two LTag hexamers bind to the origin and form a
head-to-head double-hexamer with their AAA+ domains in contact with EP and AT
region, where the DNA is locally melted [54]. This helicase loading process resembles
the formation of the Mcm2-7 double-hexamer but does not involve interactions with any
other protein factors (Figure 1.1).

In Chapter 3, we determine and discuss the first crystal structure of a LTag construct
bound with the origin DNA, which starts shedding some light on the details of how
OBDs recognize and initiate assembly of the LTag on the origin, as well as how the β-
hairpins interact with the dsDNA.

10

11
Figure 1.2. Overview of SV40 LTag and replication origin. (A) Schematic of the
domains arrangement on LTag. The amino acid numbers are indicated at the bottom.
DnaJ, DnaJ homology domain; OBD, origin binding domain; Zn, Zn domain; AAA+,
AAA+ domain; HR, host range domain. (B) Known structures of domains of LTag. The
amino acid numbers are indicated. Three published X-ray and NMR structures are shown
for DnaJ (PDB:1GH6), OBD (PDB:2FUF) and the helicase (Zn and AAA+ domains)
(PDB:1N25) domains. All structures are drawn to scale and peptides connecting the
domains are represented as dashed lines. N- and C-terminal are labeled as N and C with
amino acid numbers. The C-terminal HR domain is thought unstructured and represented
by thick yellow lines. All domains are colored with the same colors as shown in Panel A.
(This panel was reproduced with permission from Reference [51]). (C) Bottom view
(viewing from C-terminal AAA+ domains) of the ring-shaped nucleotide-free structure of
the LTag 131-627 hexamer (PDB:1SVO). The six β-hairpins (colored, residue 508-517)
arrange on a circle along the central channel (dashed circle). (D) Overlap of the LTag
131-627 monomer structures of the ATP bound structure (pink, PDB:1SVM) and the
nucleotide-free structure (green) as in the respective hexamer structures (shown in Panel
E). Upon ATP binding, the whole AAA+ domain moves toward the central channel with
the Zn domain remains still. As a result, the β-hairpin moves upward (closer to the N-
terminal domains) in the central channel (represented as the dashed cylinder). All these
conformation changes result in a narrower central channel. (E) Top view of structures of
the LTag 131-627 hexamers in ATP-bound, ADP-bound (PDB:1SVL) and nucleotide-
free (Nt-free) states. The central channel is the narrowest in ATP-bound state and the
widest in Nt-free state. (Panel C-E were reproduced with permission from Reference
[50]); (F) SV40 replication origin. The base pair numbers are indicated according to the
positions on SV40 genome. Four 5’-GAGGC-3’ pentanucleotides (PEN1-4) are flanked
by the early palindrome (EP) and the AT-rich (AT) regions. Each PEN is specifically
recognized by one OBD.

12
Chapter 2
Expression, Purification and Biochemical
Characterization of Schizosaccharomyces pombe
Mcm4, 6 and 7

Reproduced with permission from Xu, M., Chang, Y.P., Chen, X.S. 2012. Expression,
Purification and Biochemical Characterization of Schizosaccharomyces pombe Mcm4, 6
and 7. (Submitted)
Author contributions: M.X. designed, expressed and purified the protein constructs;
carried out all assays for biochemical characterization. Y.P.C. provided advice for
designs of truncated proteins. X.S.C. supervised the project.

2.1 Overview
Within the MCM family, Mcm2-7 proteins are revealed as key components of the pre-
replicative complex (pre-RC). Pre-RC initiates DNA synthesis at the origin in all
eukaryotes [55-57]. Mcm2-7 are six proteins that are homologous to each other and are
conserved among Archaea and eukaryotes [21]. Mcm2-7 functions as the replicative
helicase, and can form various oligomeric complexes, including double-hexamers [6, 7],
hexamers [22, 23], tetramers [24], trimers [25], and dimers [22, 26, 27].

It has been well demonstrated that Mcm2-7 are vital in the initiation and the elongation of
genomic DNA replication as a eukaryotic replicative helicase. Purified Mcm2-7 hexamer

13
has helicase activities in vitro if glutamate is included in the reaction buffers [58]. In
addition, helicase activity has been shown in vitro for MCM sub-complex comprising
only three of the six subunits, Mcm4/6/7 hexamers (two copies of each subunit).

To further understand the subunit arrangement and architecture of the Mcm4/6/7 hexamer
assembly, we characterized individual domains and near-full-length polypeptides of each
of subunits using E. coli expression. Various truncated fragments of
Schizosaccharomyces pombe Mcm4, 6 and 7 were purified, and then their oligomeric
states and inter-subunit interactions were investigated in vitro by gel filtration and pull-
down assays. By using a co-expression system developed in E. coli, we successfully
purified in large quantity of soluble and pure S. pombe Mcm4/6/7 complex in hexameric
state.

2.2 Results
Designs of truncated fragments of Mcm4, 6 and 7
To get stably expressed and soluble constructs of Mcm4, 6 and 7 in E. coli, it is important
to make truncations around disordered regions or less conserved areas, but not in highly
conserved and well folded regions. We first predicted the disordered parts of the native
proteins using the DISOPRED server at University College London (Figure 2.1B, S.1).
The secondary structures of Mcm4, 6 and 7 were predicted using the PSIPRED server at
University College London (Figure 2.1B). In addition, we also performed structural
alignments and comparison using solved archaeal MCM structures, such as structures of

14
Methanothermobacter thermautotrophicus MCM (MtMCM, PDB:1LTL) and Sulfolobus
solfataricus MCM (SsoMCM, PDB:3F9V), to more precisely determine the appropriate
boundaries of the predicted secondary structures [37, 39]. These results form the basis
for deciding where to make truncations/deletions for protein expression.

15
Figure 2.1. Designs of truncated fragments of Mcm4, 6 and 7. (A) Schematic of fission
yeast Mcm4, 6 and 7. Locations of putative zinc finger (white boxes labeled with Z), the
MCM core region (gray boxes) was shown. Three ATPase consensus motifs in the MCM
core region were labeled with A (the Walker A motif), B (the Walker B motif) and R (the
R-finger motif). All conversed amino acid residues that define each motif were shown.
All truncation fragments reported in this paper were designed according to three domains,
N-terminal, core and C-terminal domains. This figure was generated from the sequence
alignment results shown in S.1 and each Mcm protein was aligned with the MCM box
region. (B) Disordered profile plot and predicted secondary structure of Mcm4. Only
sampled secondary structure prediction was shown and aligned with the disordered
profile. A disordered N-termini was present and aligned well with a region (1-150 aa)
that lacks any defined secondary structure, while regions with very low disorder
probability were predicted to show ordered secondary structures. The disordered profiles
were generated by DISOPRED server, and secondary structure prediction was generated
by PSIPRED server at University College London [59-61]. Disordered profile plots of
Mcm6 and 7 were shown in S.2.

We made three major MCM constructs groups in this study, N-terminal fragments, MCM
core fragments, and the nFL fragments. The summary of the constructs and the observed
biochemical properties were shown in Table 2.1.

16
Table 2.1. Summary of biochemical properties of fragments of Mcm4, 6 and 7.

(continued on next page)

17

Schematic of truncated fragments of Mcm4, 6 and 7 tested in this study. The
nomenclature for the fragments is as follows, the first numbers represent the Mcm 4, 6, or
7; the letters in the middle indicate domain locations (“N”-N terminal fragments, “C”-
core fragments, “F”-near-full-length fragments); the last numbers denotes construct
number. a, decreased expression level or plasmid instability; b, oligomeric states
depended on protein concentration; c, little equilibrium between monomeric and dimeric
states and proteins in the two states could be separated by ion-exchange chromatography;
d, a stable large complex identified with a molecular weight equal to a double-hexamer;
n/a, not available, due to lack of enough samples.

Purification and characterization of N-terminal fragments of Mcm4, 6 and 7
Because the N-terminal fragment of MtMCM and SsoMCM oligomerize into hexamers
[37, 38], we want to investigate the role of the N-terminal fragments of Mcm4/6/7 in
modulating oligomerization. Analysis of the purified proteins by gel filtration
chromatography showed that most of the N-terminal fragments behaved as monomers
(Table 1). However for some N-terminal fragments of Mcm6 and Mcm7, peaks
corresponding to a dimer formation were observed. As shown in Figure 2.2A, two out of
three Mcm6 N-terminal fragments with intact N-terminus, 6N1 and 6N2, formed single

18
peaks at the dimer position on gel filtration profiles. In contrast, the other three N-
terminal fragments with N-terminal truncation, 6N4, 6N5 and 6N6, only had peaks at
monomer position.

For Mcm7 N-terminal fragments, 7N1 and 7N2, they showed two oligomeric peaks at the
positions expected for dimers and monomers (Figure 2.2Ag and j). The fact that the two
oligomeric states could be separated by Resource Q anion-exchange chromatography
showed there was little equilibrium between the monomeric and dimeric states (Figure
2.2Ah and i).

To test whether the N-terminal fragments of Mcm4, 6 and 7 are competent to form
hetero-oligomers, several combinations of the N-terminal fragments from Mcm4/6/7
were incubated together after purified individually. A relatively low salt concentration
(50mM NaCl) was used to favor oligomerization. However, no oligomer was identified
under our tested conditions (Figure 2.2B).

19

Figure 2.2. Oligomeric states and interactions of N-terminal fragments of Mcm4, 6 and
7. (A) Gel filtration chromatography profiles of N-terminal fragments of Mcm4, 6 and 7.
Schematic of each fragment was shown in accordance with its gel filtration profile. N-
terminal fragments of Mcm6 were aligned with the zinc finger motif and a 62 amino acid
residues protruding N-termini was shown. 7N1a, separated monomeric 7N1 fragment;
7N1b, separated dimeric 7N1 fragment. Gel filtration analysis was carried out a described
under “Materials and Methods”. (B) In vitro incubation of purified N-terminal fragments
of Mcm4, 6 and 7. Interactions among the N-terminal fragments of Mcm4, 6 and 7 were
characterized by gel filtration analysis. Samples from peak fractions (pointed by arrows)
were quantitated by SDS-PAGE and mixed together in approximate equal molar ratio.
The mixture was buffer-exchanged to 50mM NaCl, 50mM Tris pH8 and 1mM DTT and
then incubated on ice for 30 minutes. For 7N1 and 7N2, only samples from peak fraction
of monomeric states were used. The incubation mixtures were subjected to gel filtration
analysis and no large complex was detected. Two groups of N-terminal fragments of
Mcm4, 6 and 7 were used, as shown in top and bottom panels.

20
Purification and characterization of core fragments of Mcm4, 6 and 7
Most of core fragment constructs of Mcm4, 6 and 7 suffered from heavy precipitation
and only soluble protein of one fragment, 4C1, could be obtained (Table 2.1). The
oligomeric states of this fragment appeared at peaks with ~250 or 500 kDa, respectively
in agreement with hexamers and 12-mers, depending on the protein concentration (Figure
2.3b and c). When a center fraction of the 12-mer peak was injected to the same gel
filtration column, a hexamer peak appeared (Figure 2.3c), indicating that the two
oligomeric states can equilibrate with each other. The protein concentration for the
hexamer peak is much lower, compared to that of the 12-mer peak. Addition of ATP and
Mg
2+
did not affect the transition between the two oligomeric forms.

Purification and characterization of nFL of Mcm4, 6 and 7
To help with purification, nFL fragments of Mcm4, 6 and 7 were tagged with GST or
8xHis and expressed in E. coli. In contrast to N-terminal fragments, these 70~90 kDa
fragments were either insoluble or degraded when expressed in E. coli. Only one nFL
fragment of Mcm7, 7F4, could be successfully expressed and purified. We also found
that the oligomeric states of this fragment changed when different N-terminal tags were
used. As shown in Figure 2.3f, His tagged 7F4 can form a very large and broad complex
peak (about 1000 kDa) and a monomer peak. The large complex peak of His tagged 7F4
was quite stable even at 1M NaCl. In comparison, the same 7F4 fragment that was
cleaved from the GST-7F4 fusion only appeared as in monomeric state (Figure 2.3e),
suggesting the N-terminal GST tag may influence the self-interaction of this fragment.

21

As for nFL Mcm6 fragments, most of them precipitated in the cell pellets. Fragment 6F9
could be purified but formed aggregates (Figure 2.3d). All nFL Mcm4 fragments had
very low expression level. For 4F5, the expressed protein seemed to be toxic to E. coli
cells as the plasmid was instable (data not shown).

Figure 2.3. Gel filtration chromatography profiles of core and near-full-length fragments
of Mcm4, 6 and 7. Schematic of each fragment was shown in accordance with its gel
filtration profile. Gel filtration analysis was carried out a described under “Materials and
Methods”. Gel filtration profile of 7N1 was chosen as a reference, and its dimer peak was
used to align with monomer peaks of 7F4. The other molecular weight shown was
determined by Bio-Rad Gel Filtration Standard (data not shown).

22

(continued on next page)

23

Figure 2.4. Interactions and oligomeric states of co-expressed fragments of Mcm4, 6 and
7. (A) Schematic of the polycistronic co-expression strategy that involves two compatible
vectors. ORF1 and ORF2 were linked by a ribosome binding site (RBS) with a spacer.
ORF3 was cloned in pXA-BN vector. Two plasmids were co-transformed into E. coli.,
followed by dual screening of ampicillin (50 µ g/ml) and chloramphenicol (17 µ g/ml). (B)
Interactions of co-expressed and copurified fragments of Mcm4, 6 and 7. E. coli. lysates
co-expressing various fragments with or without tags were passed through either
glutathione or Ni-NTA resins, then the resins were washed as described under “Materials
and Methods”. GST tags were cleaved by PreScission protease on the resin to release the
MCM proteins. His tagged proteins were eluted by imidazole. All elutions were analyzed
by SDS-PAGE. Asterisk denotes the co-lysis (instead of co-expression) of the indicated
near-full-length fragments. (C) Gel filtration chromatography profiles of purified
complexes of Mcm4, 6 and 7. Gel filtration analysis was carried out a described under
“Materials and Methods”. Asterisk: Gel filtration profile of Mcm4/6/7 hexamers
expressed and purified from insect cells in our laboratory. (D) SDS-PAGE analysis of
purified complexes of Mcm4, 6 and 7 from the gel filtration fractions shown in Panel C.
(E) Helicase assay results of the Mcm4/6/7 hexamers. No protein added in lane1 and 2. B,
boiled substrate; UB, un-boiled substrate; lane 3 and 5, 100ng protein added; lane 4 and 6,

24
200ng protein added; Asterisk, Mcm4/6/7 hexamers expressed and purified from insect
cells.

Co-expression, copurification and characterization of complexes of Mcm4, 6 and 7
Because the nFL fragments of individual Mcm4, 6, and 7 expressed in E. coli did not
behave well, we tried co-expression of all three proteins together to see if any stable
complexes of them can be obtained. A polycistronic strategy (Figure 2.4A) using two
compatible vectors was employed to co-express Mcm4, 6 and 7 in the same host cells.
Various combinations of constructs were tested and the results were summarized in
Figure 2.4B. A series of pull-down assays was also performed with either Ni-NTA or
glutathione resin. It should be noted that the GST tag had been removed by PreScission
protease in the elution, while either N-terminal or C-terminal 8xHis tag still remained. As
shown in Figure 2.4B, not all ORFs were translated, as in the cases of 6N2-His, 6N2, and
4F5. A new nFL fragment of Mcm7, 7F8, which includes an untruncated N-terminus,
was used. As for the N-terminal fragments, no interactions between 4N2 and 7N2 was
observed, given the negative reciprocal pull-down results.

For the nFL fragments, strong interactions of 4F5/6F9, 4F5/7F8 and 6F9/7F8 were
characterized by positive pull-down results. Most positive pull-down results were verified
in two directions and showed little difference no matter which fragments was tagged,
except 6F9/7F8 pair. When 7F8 was tagged and used to pull-down 6F9, only a weak
interaction was detected, indicated by a very faint band of 6F9. 1:1 molar stoichiometry
of those binding pairs was also shown by SDS-PAGE analysis. Further gel filtration
analysis clearly showed dimer peaks of 4F5/6F9 and 4F5/7F8 (Figure 2.4Ca and b),

25
whereas only aggregates were observed on gel filtration profile of 6F9/7F8 (Figure
2.4Cc). Fractions obtained from gel filtration analysis were characterized by SDS-PAGE
analysis, as shown in Figure 2.4D. Several co-expression combinations were able to
produce all three nFL fragments of Mcm4, 6 and 7, however, combinations with N-
terminal GST tagged 6F9 still suffered from poor folding, which was implied by its very
low yield and background binding with Mcm4 and 7 fragments.

We also compared co-expression results to co-lysis results. In co-lysis, GST-6F9, GST-
4F5 and 7F8-His were expressed individually, and cell pellets of their host cells were
lysed together to provide binding environment similar to intracellular condition. As
shown in Figure 2.4B Asterisk, when 7F8-His was purified by Ni-NTA resin, only a
small amount of GST-4F5 were co-pulled down, and none of GST-6F9 could be co-
pulled down. In contrast, when these fragments were co-expressed, much stronger
bindings were identified, indicating improved folding of these fragments. However, no
hexamer could be purified when each protein was expressed separately first and the cells
of each were co-lysed and incubated together, indicating that co-expression is needed for
stable complex formation.

pGEX-6F9-His-7F8/pXA-4F5 eventually produced the Mcm4/6/7 hexamer with a yield
of 10mg from 12L culture. 8xHis tag was tagged on C-terminal of the Mcm6 fragment,
and the Mcm4 and 7 fragments were not tagged. The three nFL fragments were co-
expressed and copurified with a Ni-NTA affinity column that was able to bind 8xHis tags

26
on the Mcm6 fragment. The Mcm4 and Mcm7 fragments were co-pulled down,
indicating strong bindings among the three subunits. The complexes showed a single
peak of about 500 kDa on gel filtration profiles, which is equivalent to the theoretical
molecular weight (497 kDa) of the hexamer, consisting of the nFL fragments of Mcm4, 6
and 7 (Figure 2.4Cd). The size of the peak was also verified by aligning with the hexamer
peak composed of Mcm4, 6 and 7 purified from insect cells (Figure 2.4Cf).

A 1:1:1 molar stoichiometry for three proteins was shown in Figure 2.4Dd. The central
fraction of the peak was sent to N-terminal sequencing to confirm that the complex was
composed of the Mcm4, 6 and 7 fragments. The salt resistance of the hexamer was also
tested in various concentrations of NaCl ranging from 50 to 1000 mM, and appeared as a
stably assembled oligomer that is suitable for further studies (data not shown). The final
purified Mcm4, 6 and 7 hexamer could be concentrated to as high as 50mg/ml in 50mM
NaCl, 20mM Tris pH8 and 1mM DTT, with an estimated purity over 95%. The
Mcm4/6/7 hexamer we obtained from E. coli showed helicase activity on forked dsDNA
substrate, which was made by anneal a labeled ssDNA to the circular M13 ssDNA
(Figure 2.4E). The helicase activity of this hexamer was comparable to the Mcm4/6/7 we
purified from insect cells.

As shown in Table 2.1, the nFL fragment of Mcm6 expressed for this Mcm4/6/7 hexamer
contains a highly disordered internal loop that is close to its C-terminus, which might be a
problem for future crystallographic studies. Thus a nFL fragment of Mcm6 without that

27
disordered loop was used in the co-expression and copurification. A hexamer peak still
appeared but the yield of the hexamer is much lower (Figure 2.4Ce).

2.3 Discussion
Eukaryotic MCM proteins can form various complexes including dimers, trimers,
tetramers, hexamers and double-hexamers. In addition to interactions between different
subunits, self-interactions of some MCM proteins have also been shown [22, 28]. Most of
those studies performed yeast two-hybrid assays and co-immunoprecipitation (co-IP) to
investigate and demonstrate the interactions, and there are some disagreement of MCM
protein interaction pairs in the literature [62, 63]. Gel filtrations have been used to study
interactions among Saccharomyces cerevisiae MCM proteins (scMCM) [64], in which all
full-length scMCM proteins (except scMcm5) form large aggregates, implying folding
problems of full-length MCM proteins, especially when expressed individually.

In this study, we expressed and purified a series of Mcm 4, 6 and 7 fragments as a way to
investigate domain structures, folding, and roles in oligomerization. At the same time, we
have obtained a soluble, stable and functional complex of Mcm4/6/7 from E. coli,
potentially useful for future structural and functional studies.

Oligomeric states of N-terminal fragments of Mcm4, 6 and 7
High-resolution structural data were available from the N-terminal fragments of MtMCM
and SsoMCM, which forms head-to-head double hexamers [37] or single hexamer [38].

28
In addition to sequence similarity, several features in this partial MtMCM structure are
also shown for MCM proteins in eukaryotes. First, a zinc-finger motif is crucial in
mediating hexamer-hexamer interaction. Putative zinc-finger motifs are also found on
Mcm4, 6 and 7 (Figure 2.1A), which are defined by C(X)
2
C(X)
18
C(X)
2
C. The
biochemical importance of this motif has been shown by mutagenesis studies on archaeal
and eukaryotic MCM proteins [40, 41, 65]. Second, the N-termini of MCM proteins play
important roles in hexamer formation as well, which were shown by the deletion of 204
amino acid residues at N-terminus spMcm2 [42]. Furthermore, the N-terminals of an
archaeal MCM are also shown to stimulate helicase activity of C-terminals [43].

One question to be investigated in this study is if the N-terminal domains of eukaryotic
Mcm2-7 also play the same structural role in hexamerization. According to the structural
prediction (Figure 2.1B), both Mcm4 and Mcm6 have very disordered N-termini. It was
noted previously that three yeast MCM proteins, Mcm2, 4 and 6, have extended N-
termini when compared to the other MCM proteins (Figure S.1) [19]. These extended N-
termini are rich in serine and threonine residues and was reported to play a redundant role
in initiation of DNA replication through phosphorylation [66]. Unlike the N-terminal
domains of MtMCM and SsoMCM, which can form stable hexamers, no strong inter-
subunit interactions were identified of the N-terminal domains of spMCM [37, 38].
However, we found the extended N-terminus, the first 62 amino acid residues on Mcm6,
is required for self-interaction, as deleting the 62 amino acid residues shifted the dimer to
the monomer peak (Figure 2.2Aa-f). Self-interactions of Mcm6 have been demonstrated

29
by yeast two-hybrid assays, co-IP and gel filtration, even though unclear about the
oligomeric states [28, 62, 64, 67, 68]. Even though the 62 amino acid residues were
required for dimerization of N-terminal fragments of Mcm6, an nFL Mcm6, 6F9, formed
Mcm4/6/7 hexamers (Figure 2.4Cd). This result suggests the extended N-terminus of
Mcm6 is only involved in the interactions between two N-terminal fragments.
Furthermore, unlike the extended N-termini found on Mcm2 and 4 in all eukaryotic
organisms (Figure S.1), the extended N-termini of Mcm6 only exists in S. cerevisiae and
S. pombe, suggesting that the roles associated with the extended N-termini of MCM6
may only be restricted to yeast.

For the N-terminal fragments of Mcm7, we observed two elution peaks in the gel
filtration profile that were in agreement with dimers and monomers (Figure 2.2Ag and j).
Self-interactions of Mcm7 were previously reported [28, 62, 64, 68]. Our observation that
the N-terminal fragment of Mcm7 form dimers may suggest their potential involvement
in the self-association of Mcm7. Unlike weak self-interactions of Mcm7 reported
previously, the two oligomeric states of the N-terminal fragment can be separated by ion-
exchange chromatography (Figure 2.2Ah and i), which indicated a relatively strong
interactions between the two N-terminal fragments.

In contrast to Mcm6 and 7, the extended N-terminus of spMcm4 is not likely to play a
role for intersubunit interactions. No self-interactions of N-terminal fragments of Mcm4
were identified.

30

Oligomeric states of core fragments of Mcm4, 6 and 7
The MCM core is conserved in MCM proteins from archaea to human (Figure S.1) [19,
69]. As shown in Figure 2.1A, the MCM core contains there ATPase consensus motifs,
the Walker A motif, the Walker B motif and the R-finger motif. Mutagenesis has been
done on the putative ATP binding site in this region to prove the importance of this
region in modulating oligomerization of MCM proteins [39, 42, 70]. However, possibly
due to the lack of structurally important zinc finger motifs, core fragments alone have
been reported incapable to oligomerize by themselves [43]. In our study, a core
fragment of Mcm4, 4C1, formed two oligomeric forms consistent with hexamers and 12-
mers (Figure 2.3b and c). Given the fact that the zinc finger motifs were required for
head-to-head double-hexamerization of the MtMCM [37], the 12-mers we identified here
are not likely to be the head-to-head double hexamers.

Oligomeric states of individually expressed near-full-length Mcm4, 6 and 7
Most full length eukaryotic MCM proteins have been reported to form aggregates when
expressed individually [64]. In this report, we also found that most expressed nFL
fragments of Mcm4, 6 and 7 formed aggregates or and did not behave well in solution.
Nonetheless, one nFL fragment of Mcm7 was found to be soluble and form two
oligomeric states. As shown in Figure 2.3e, a Mcm7 construct with His tag, 7F4, elute in
gel filtration chromatography as a monomer peak and a peak of about 1000 kDa
(equivalent to a 12-mer) (Figure 2.3f). The ~1000 kDa complex of 7F4 seemed to be

31
quite stable and sensitive to neither salt concentration nor protein concentration. Unlike
the core fragment of Mcm4 we reported, which showed a large complex peak on gel
filtration as well, this nFL fragment of Mcm7 contains all key elements for
oligomerization of MCM proteins, and the large complex observed here likely is a
double-hexamer.

When the same Mcm7 construct, 7F4, was fused to GST at its N-terminus, aggregates
and monomeric peaks on gel filtration were observed, which is different from the
behavior of His-7F4. This result indicates the usage of different tags fused even to the
same end can have a different effect on protein oligomerization.

Surprisingly, no dimer of the 7F4 fragment was observed, though the N-terminal
fragments of Mcm7 were capable of dimerization. One explanation might be that the
addition of the MCM core region on the nFL fragment further strengthens the protein’s
capability to oligomerize, which results in a cooperative shift from dimeric state to a
higher oligomeric state. On the other hand, if the dimer interfaces of the N-terminal
fragments are head-to-head instead of side-to-side, the interfaces may not be strong
enough to overcome the entropy increase of the much longer molecule as formed when
the fragments are long enough to include the MCM core region. This may also explain
why the longest N-terminal fragment of Mcm6, 6N3, was only found in monomeric state
(Figure 2.2Ae).

32
Co-expression of Mcm4/6/7 and purification of the soluble complex
Both Mcm2-7 hexamers and Mcm4/6/7 hexamers were co-expressed and copurified from
the same host cell cultures as reported before [28, 29]. Individually expressed MCM
proteins tend to aggregate, especially when expressed in E. coli [64]. In vivo, it has been
evaluated that MCM proteins are very abundant in cells and expression level of most
MCM proteins are very stable through the cell cycle [21, 22]. Some sub-complexes of
MCM proteins were also identified by in vivo cross-linking [62].

We used a polycistronic strategy to achieve co-expression of the fragments of Mcm4, 6
and 7 (Figure 2.4A). It should be noted that the success of the polycistronic expression is
highly dependent on the sequence around the ribosome binding sites (RBS), known as
Shine-Dalgarno sequence in E. coli [71, 72]. Which explains why we were unable to
have some ORFs expressed, such as 6N2, 6N2-His and 4F5 (Figure 2.4B).

Both Ni-NTA and glutathione resin was used to pull down the tagged fragments. If two
fragments bind each other strongly, non-tagged or otherwise tagged fragments would be
co-pulled down. However, unfolded or misfolded proteins often aggregate together on
resin, leading to false positive results. Thus, we analyzed elution instead of protein-bound
resin by SDS-PAGE to elucidate the binding pairs. Co-expression results were compared
to co-lysis results to demonstrate that some nFL fragments have to be co-expressed to
fold properly (Figure 2.4B Asterisk).

33
As shown in Figure 2.4Cd and 2.4Dd, the Mcm4/6/7 hexamer composed of the nFL
fragments was obtained from co-expression in E. coli. Helicase assay with this hexamer
was carried out and showed an activity comparable to that of the Mcm4/6/7 hexamer we
purified from insect cells before. The yield (10mg/12L culture) and purity (over 95%
purity) obtained using this E. coli co-expression provide a system for future structural and
functional studies of this MCM sub-complex.

A summary of binding pairs identified by our results was illustrated as in Figure 2.5A.
The Mcm4/Mcm6 dimer and the Mcm4/Mcm7 dimer were identified and characterized
by both gel filtration and SDS-PAGE in this study (Figure 2.4Da and b). We also showed
self-interactions of Mcm7 and Mcm6, especially in the case of 7F5, which formed a large
complex that might be a double-hexamer. These results are consistent with previous
reports [28, 62, 64, 68]. Our data support the arrangement model of the Mcm4/6/7
hexamer for the six subunits of spMCM (Figure 2.5B) that was proposed for S. cerevisiae
MCM [64], and human MCM [28, 35, 62]. The literature has some disagreement about
the interactions between Mcm6 and Mcm7. Evidence showing no direct binding [62], or
weak binding [28], or strong binding [68] of Mcm6/7 pair was reported. Our results
showed a strong binding of Mcm6 and Mcm7 by affinity pull down from co-expression
(Figure 2.4B). However, no stable Mcm6/7 dimer was present on gel filtration analysis
(Figure 2.4Dc). It should be noted even though there is no direct contact between Mcm6
and Mcm7 in the proposed planar ring-shaped hexamer structures, contact between them
might exist in a staggered globular shaped structure in which each MCM subunit has

34
direct contact with at least four subunits [30]. This staggered globular shaped Mcm2-7
hexamer is composed of two layers of trimmers and was only reported for S. pombe, but
not for human. It is likely the strong interaction between Mcm6 and 7 is unique to S.
pombe and contributes to the formation of the globular hexamer.

As for the N-terminal fragments of Mcm4, 6 and 7, we did not identify any binding pairs
between different subunits, even in co-expression (Figure 2.4B). But we observed stable
homo-dimers of N-terminal fragments of Mcm6 and 7 (Figure 2.2Aa, c, g and j). Given
the strong structural evidence for double-hexamers from MtMCM and scMCM [6, 7, 37,
73], the interfaces of these dimers are likely to be head-to-head. The 3D reconstruction
model processed from C2 point group symmetry clearly showed the two MCM hexamers
are connected by head-to-head “protein bridges” [6]. Based on the 30Å 3D EM
reconstruction model, we proposed a Mcm2-7 double-hexamer model as illustrated in
Figure 2.5C. The head-to-head interactions between two identical subunits from each
hexamer can only occur at most twice in the hexamer-hexamer interface, and all other
interactions should be between different MCM subunits. We observed homo-dimer only
for the N-terminal fragment of Mcm6 and 7. Because the 62 amino acid residues
extended N-terminus required for Mcm6 dimerization is only found in yeast, the N-N
interactions between two identical subunits should be through Mcm7. In our model, the
orientation between two hexamers is locked by the specific interaction between two
Mcm7 subunits, and double-hexamerization is further stabilized by nonspecific
interactions between zinc finger domains of the other Mcm subunits.

35

Figure 2.5. Schematics of proposed models of the Mcm4/6/7 hexamer and the Mcm2-7
double-hexamer. (A) Summary of interactions identified in this report. Double arrows:
reciprocal interactions. Single arrows: unidirectional interactions. Hallow arrows: weak
interactions. Solid lines: stable homogeneous oligomeric states, such as dimers. Dashed
lines: heterogeneous oligomeric states: such as aggregates. (B) Model of the Mcm4/6/7
hexamer. This model is based on the interactions identified in Panel A, which is
consistent with the model proposed previouslyfor S. cerevisiae MCM [64], and human
MCM [28, 35, 62]. (C-D) Model of hexamer-hexamer interactions for the Mcm2-7
double-hexamer. This model is based on Figure 5A of [6], showing a proposed Mcm7/7
interaction that locks the orientation of two hexamers. The convex and the concave on
each subunit in this figure represent the P-loop of the Walker A motif and the R-finger
motif, respectively.

36
2.4 Conclusion
Here we described a systematic characterization of the biochemical properties of different
domains of S. pombe Mcm4, 6 and 7 using E. coli expression. The oligomeric states and
inter-subunit interactions have been determined with purified protein in vitro. A co-
expression strategy was also developed to obtain large amount of soluble, stable and
functional Mcm4/6/7 hexamer complex from E. coli, which can be useful for future
structural and biochemical studies. Based on our results and the literature, we suggest an
arrangement model of S. pombe Mcm4/6/7 hexamer and the hexamer-hexamer
interactions in the Mcm2-7 double-hexamer.

2.5 Materials and Methods
Reagents
Oligonucleotides were synthesized by Integrated DNA Technologies (IDT) or Eurofins
MWG Operon. Pfu Turbo polymerase was purchased from Stratagene. Ni-NTA affinity
resin is purchased from QIAGEN. pGEX-6P-1 vector, PreScission protease, Glutathione
affinity column, Resource Q column, Superdex 200 and Superose 6 10/300 GL gel
filtration column were purchased from GE Healthcare Biosciences Amersham. The
pXA/BN-based vectors, used for protein co-expression, were engineered from the
original pAC vector described [74]. PMSF is purchased from Sigma-Aldrich.

37
MCM fragments designs and plasmid construction
To design various spMcm fragments, native disorder in proteins is determined by
DISOPRED server at University College London [59]. Secondary structure prediction
was performed on PSIPRED server at University College London [60, 61]. To determine
the precise boundaries of the fragments, conserved amino acid residues were identified
by protein sequence alignment among MCM proteins from various organisms (S.Fig.1).
Structural alignment to solved MCM structures was also conducted [37]. The multiple
sequence alignment was performed using ClustalX [75].

DNAs containing cDNA fragments encoding full length spMCM 4 (GenBank:P29458), 6
(GenBank:CAB75412) and 7 (GenBank:O75001) (generously provided by Dr. J. Hurwitz,
Memorial Sloan-Kettering Cancer Center, United States) were used as template in PCR
with Pfu Turbo polymerase to obtain amplified coding sequences of various fragments.
cDNA of N-terminal GST tagged fragments were subcloned to the NheI-AscI sites of
pGEX-6P-1 or the NgoMIV-AscI sites of pXA-BN. cDNAs of N-terminal His Tagged
fragments were subcloned to the NheI-AscI sites of pGEX-6P-1 with cDNA of GST
removed. For co-expression (Figure 2.4A), ORF1s were subcloned to the NheI-NgoMIV
sites followed by ORF2s to the NdeI-AscI sites, on pGEX-6P-1; ORF3s were subcloned
to the NgoMIV-AscI sties of pXA-BN.

38
Expression and purification of the fragments of Mcm4, 6 and 7
For the expression of various fragments of Mcm4, 6 and 7, constructs expressing each
spMcm4, 6 and 7 fragments were transformed into E. coli by electroporation. Then the
expression of proteins was induced by adding IPTG to 2mM at 18° C when the cell
density reached OD ~ 0.6. After cells were lysed by French Press, GST and His tagged
fragments were purified by glutathione and Ni-NTA affinity chromatography,
respectively. For GST tagged fragments, GST tags were subsequently removed by
PreScission protease treatment in standard lysis buffer containing 250 mM NaCl, 50mM
Tris pH8 (buffer A) and 1mM DTT. For His tagged fragments, buffer A containing 5mM
β-mercaptoethanol was used to lysate cell pellets and buffer A containing 5mM β-
mercaptoethanol and 100~150mM imidazole was used for elution. The elution was
loaded to a Superdex 200 or Superose 6 gel filtration column that is equilibrated with
buffer A containing 1mM DTT to finish the purification.

Co-expression and co-purification of near-full-length fragments of Mcm4, 6, and 7
The near-full-length (nFL hereafter) fragments of Mcm4, 6, and 7 were cloned into two
compatible vectors (pGEX-6P-1 and pXA-BN) and co-expressed in E. coli (Figure 2.4A).
Dual screening of ampicillin (50 µ g/ml) and chloramphenicol (17 µ g/ml) was used to
maintain the stable expression. Then co-purification was conducted the same as described
for individual fragments of Mcm4, 6, and 7. For the Mcm4/6/7 complex purification, cell
pellets were resuspended and lysed in buffer A containing 5mM β-mercaptoethanol.
PMSF is added to 1mM to prevent degradation. The supernatant from the lysis was

39
passed through a Ni-NTA resin column. After extensive wash (10 x column volume) of
the resin with buffer A containing 5mM β-mercaptoethanol, the Mcm4/6/7 complex
bound to the column through the C-terminal 8xHis tagged Mcm6 nFL was eluted by
imidazole (150mM). The eluted proteins were further purified using Resource Q anion-
exchange chromatography with a 50 to 1000 mM NaCl gradient elution, followed by gel
filtration chromatography with a Superdex-200 column that was pre-equilibrated with
buffer A and 1mM DTT. The proteins from the hexamer peak fractions were analyzed by
SDS-PAGE and concentrated to ~50mg/ml.

Gel filtration analysis
A portion of the purified fraction (Glutathione affinity column eluate, 100~500 μg) was
loaded to an analytical Superdex 200 or Superose 6 gel filtration column that is
equilibrated with buffer A and 1mM DTT. Fractions were collected and analyzed for
composition by SDS PAGE and then staining with Coomassie brilliant blue (R250).

Heilicase Assay
Helicase assay was performed as described [76]. To obtain the dsDNA substrate, ~10
fmol of [γ-
32
P]-ATP ssDNA (60nt) was annealed to the circular M13mp18 ssDNA. The
complementary sequence is 35nt, leaving a 25nt 5’ overhang on the substrate. Labeled
substrate DNA was incubated with 100~200ng Mcm4/6/7 hexamer in helicase buffer
containing 25mM Hepes pH7.5, 10mM magnesium acetate, 5mM ATP, 1mM DTT and

40
0.1mg/ml BSA for 45 min at 37° C. The reaction was analyzed on 12% native
polyacrylamide gel. The gels were then dried and autoradiographed.

41

42

Figure S.1. Sequence alignment of MCM proteins from various organisms. SsoMCM,
Sulfolobus solfataricus MCM. MtMCM, Methanothermobacter thermautotrophicus
MCM. This result was generated by ClustalX as described under “Materials and
Methods”.

43

Figure S.2. Disordered profile plots of Mcm6 and 7. The disordered profiles were
generated by the DISOPRED server at University College London.

44
Chapter 3

Crystallization and preliminary X-ray
crystallographic studies on SV40 Large Tumor
Antigen Binding to Origin DNA

This chapter contains contents adapted with permission from Chang, Y.P., Xu, M.,
Machado, A.C.D., Rohs, R., Chen, X.S. 2012. Crystal Structure of SV40 Large Tumor
Antigen Binding to Origin DNA: Structural Basis for Origin Recognition and Initiator
Assembly.” (Submitted)
Author contributions: Y.P.C. purified and crystallized the protein construct with DNA in
dimeric forms; collected diffraction data and solved the structure. M.X. purified and
crystallized the protein construct with DNA in hexameric form; collected diffraction data
and obtained the initial density map; designed purified mutants and carried out gel
filtration assay, ATPase assay, helicase assay and KMnO4 Reactivity Assay. A.C.D.M.
conducted analysis and prediction of the unbound DNA structure. R.R. and X.S.C.
supervised the project.

3.1 Overview
Simian virus 40 large tumor antigen (SV40 LTag) is not only required for viral DNA
replication but also acts as a dominant acting oncoprotein that can transform eukaryotic
cells [46]. Cellular replication proteins, such as primase and polymerase, are required for
SV40’s genome replication [45]. However, LTag alone functions as a highly integrated

45
machinery that is potent to recognize the origin, and melt/unwind DNA [49]. LTag is
composed of three defined domains: an origin binding domain (OBD), a Zn domain and
an AAA+ domain (4-6) (Figure 3.1A). Through its OBDs, LTag recognizes and binds
specifically to four pentanucleotides (PEN1-4, 5’-GAGGC-3’) on SV40 core origin DNA
[77, 78]. SV40 core origin DNA is comprised of an early palindrome (EP) region and an
AT-rich (AT) region, with each region contains two PENs (Figure 3.1B) [53]. LTag
assemble on each half origin into a hexamer, and a head-to-head double-hexamer on the
full origin [79, 80], then act as the active replicative helicase that melts the origin and
unwinds dsDNA on replication forks [50-52, 79-84].

Even with the presented LTag helicase domain structure and the structure of individual
OBDs [77, 85], little is known about how OBDs cooperate with the helicase domains to
recognize and assemble onto the origin. In this report, an initial density map showing
EP-half origin encircled by hexameric LTag constructs was presented, and a high-
resolution structures of the EP-half origin bound by dimeric LTag constructs was
determined.

46

Figure 3.1. SV40 LTag domain structures and core replication origin DNA sequence. (A)
Representation of the LTag domains. The construct used for co-crystallization is boxed in
blue (LTag131-627), which contains origin binding domain (OBD), Zn domain, and
AAA+ domain. (B) The core origin DNA sequence of SV40. Each of the four penta-
nucleotide GAGGC sequences (PEN1-4, labeled in red) is recognized by an OBD, and
the EP and AT-rich regions are indicated. The pseudo-PEN sequence (labeled) is
revealed here and discussed in the text, and a full PEN sequence is listed below. The EP-
half origin DNA (boxed in blue) was used for co-crystallization.

3.2 Results
Crystallization and preliminary X-ray analysis of the hexameric LTag-dsDNA
complex
The LTag 131-627 (LTag hereafter) co-crystallized with EP-half origin dsDNA. The
crystal morphology was improved from clusters of microneedles to half hexagon and
hexagon plates after multiple rounds of optimization and additive screening. The
diffraction image of the crystal and the statistics are shown in Figure 3.2 and Table 3.1,
respectively. The initial electron density map was obtained from Molecular Replacement
results, using a LTag hexamer structure (PDB:1SVO) that only has Zn and AAA+
domains (Figure 3.3). However, no OBDs could be revealed from the Molecular

47
Replacement search. Sphere-shaped density is present and likely to correspond to OBDs
(Figure 3.3b). In the central channel of the hexameric LTag complex, sausage-shaped
density is present and corresponds to the encircled EP-half origin dsDNA (Figure 3.3a).
The phasing of OBDs and the dsDNA could not be obtained/improved due to the
flexibility or multiple orientations of those components.

Figure 3.2. Diffraction image of the hexameric LTag/dsDNA crystal (1º oscillation)

48

Figure 3.3. Initial density map of the hexameric LTag/dsDNA crystal. The yellow
colored protein structure model that is enclosed by the corresponding electron density
was the result from Molecular Replacement search with a LTag hexamer structure (PDB:
1SVO). a, electron density corresponding to the EP-half origin DNA, encircled in the
central channel of the hexameric LTag complex; b, electron density probably
corresponding to OBDs; Zn, Zn binding domain; AAA+, AAA+ domain.

49
Table 3.1. Diffraction data and refinement statistics of two LTag/origin DNA
crystals/structure.

Hexameric LTag/EP origin Dimeric LTag/EP origin
Crystal dimensions (μm) 20×150×300 20× 250× 250
X-ray source 23-ID-D (APS) 23-ID-D (APS)
Wavelength (Å) 1.03324 1.03324
Space group P1 P2
1

Cell dimensions
a, b, c (Å) 116.8, 116.9, 118.1 73.63, 128.31, 166.12
α, β, γ (˚) 90.6, 89.8, 119.7 90, 89.911, 90
Mosaicity (° ) 0.6~1.2 n/a
Resolution range (Å) 50–3.50 (3.63-3.50) 50–2.8 (2.9-2.8)
No. of observations 118012 172018
No. of unique reflections 68392 65350
Mean I/σI (I) 16.8 (1.1) 12.1 (1.2)
Completeness (%) 94.6 (72.7) 93.2 (88.4)
R
sym
0.080 (0.633) 0.083 (0.592)
Refinement
Resolution (Å) n/a 50.0–2.8
No. reflections n/a 65350
R
work
/ R
free
n/a 22.69/25.49
B-factor (Averaged): Protein n/a 58.554
DNA n/a 55.396
R.m.s deviations: Bond lengths (Å) n/a 0.008655
Bond angles (˚) n/a 1.20198
Highest-resolution shell values are shown in parentheses.
n/a, not available.

50

Figure 3.4. Overall structure of the LTag dimer in complex with EP-ori DNA. (A) The
well-defined electron density map of the EP-ori DNA, showing clear density for the base-
pair sequence. (B) The map section corresponding to PEN1 in Panel A being enlarged to
show the excellent fitting of the DNA model into the density. (C) Top view of the two
OBDs on the opposite faces of the ori DNA around PEN1. (D) The overall structure of
the dimeric LTag-ori DNA complex. OBD1 binds PEN1 (highlighted), but OBD2 does
not bind to PEN2. Instead OBD2 binds to a pseudo PEN site near PEN1 but distal to
PEN2. The helicase domains of the two LTag subunits are arranged similarly as two
adjacent subunits in a LTag hexamers. (E, F) The interactions between EP-ori with
subunit 1 (E) or subunit 2 (F), showing the contacts between EP-ori with OBD, linker

51
region, Zn and AAA+ domains along the dsDNA’s consecutive major-minor grooves
(indicated by arrows) in the two subunits. Note the difference of the linker conformations
and the OBD orientations of the two subunits.

Overall architecture of the dimeric LTag-dsDNA complex
As shown in Figure 3.4, the LTag construct used in co-crystallization with EP-half origin
dsDNA was composed of three domains, OBD, Zn and AAA+ domains. SV40 origin can
be recognized and bound specifically by this LTag through bindings between OBDs and
PENs. Unlike the hexameric LTag-dsDNA co-crystal, in this crystallization condition,
this LTag co-crystallized as a dimer with the 32 bp EP-half origin dsDNA (Figure 3.1B
blue box). The space group is P2
1
as shown in Table. Thus 2-fold NCS averaging was
carried out to obtain an electron density map that is legible enough for unambiguous
determination of the DNA base pairs (Figure 3.4A and B). Therefore, the orientation and
registry of the EP-half origin in the complex was determined. The average measured
pitch/turn of the EP-half origin in the crystal is 3.4 Å/35.0º , close but not identical, to
those of standard B-form dsDNA (3.32 Å/35.4º ) [86]. The LTag interacts with the EP-
half origin DNA through OBD, Zn and AAA+ domains (Figure 3.4E and F blue arrows).

OBDs of LTag dimer recognizes PEN sequence on the origin DNA
For OBDs, instead of the expected binding of PEN1 and PEN2 by OBD1 and OBD2 of
the two subunits, only PEN1 is bound by OBD1, leaving PEN2 unbound (Figure 3.4D
and E, OBD1 in pink, PEN1 and PEN2 labeled). Surprisingly, OBD2 is found inverted
almost 180º compared to OBD1, and interacts with a 5’-GC-3’ dinucleotide (referred as

52
pseudo-PEN sequence) that is further away from PEN2 (Figure 3.4D and F). The flexible
linker (10 aa in length) connecting OBDs and Zn domains allows such distinct
orientations of the two OBDs, leaving them to bind the DNA on almost opposite faces
with inverted orientations (Figure 3.4C). Given the current position of Zn and AAA+
domain of subunit2, OBD2 seems unable to reach PEN2 even with the linker completely
extended.

As reported previously, OBD1 recognizes and binds to PEN1 in the major groove [77,
85]. As shown in Figure 3.4E and Figure 3.5A, R154, S152, and N153 are in contact with
PEN1 -G
21
A
22
G
23
G
24
. In addition, on the complementary strand of PEN1, C
24
and C
23

interact with the OBD1 peptide backbone; while G
25
interacts with R204 (Figure 3.5A).
Altogether, 7 bases of PEN1 (4 on PEN1, 3 on complementary strand of PEN1)
contribute in the highly specific sequence recognition by OBD1. The unexpected
interaction between OBD2 and the pseudo-PEN site happens in the major groove and
shows a sequence specific binding as well (Figure 3.5B, pseudo-PEN labeled in Figure
3.1B, 3.4D). G
17
of the pseudo-PEN site is in contact with N153; and G
16
and C
17
of the
complementary strand interact with R204 and peptide backbone. In contrast to OBD1,
S152 and R154 do not involve in any interaction and R154 swings away to form a
hydrogen bond with N227 (Figure 3.5B). Given that only 3 bases are directly involved in
the sequence read-out, the interactions here are not as specific as OBD1-PEN1
interactions.

53

54
Figure 3.5. Detailed interactions between LTag and EP-ori DNA. For all panels, the
upper DNA strand is in grey, lower strand in yellow, subunit 1 in pink, and subunit 2 in
green. Protein backbone oxygen and nitrogen are represented by red and blue spheres. (A)
Sequence specific interactions between OBD1 and PEN1. (B) Interactions between
OBD2 and the pseudo-PEN sequence 5’-G
17’
C
16’
and 5’- G
16
C
17
(see Figure 3.1B). (C)
The interactions of the Zn-domains of both subunits with EP-ori DNA, which is mediated
by charge-charge interactions with the DNA backbones. Water molecules are represented
by white spheres. (D) Interactions between the AAA+ domains of both subunits and the
EP region, which are mainly mediated by residues on the β-hairpin (K512/H513). Note
that H513 of both subunits insert into the minor groove to interact with both strands, the
K512s of both subunits track only one strand. (E,F) The same interactions as shown in
Panel D, except that only subunit 1 in (E), and subunit 2 in (F).

Zn domain
The Zn domain interacts with EP-half origin DNA through the dsDNA backbones by
charge-charge interactions without any identified DNA sequence specificity (Figure
3.5C). Such non-specific binding on the DNA may provide an anchoring point that
intermediates loading of the AAA+ domain onto the DNA, leading to the assembly of
LTag.

AAA+ domain interactions with DNA
The AAA+ domains of both subunit1 and 2 bind to the EP-half origin DNA on the minor
groove (Figure 3.5D and F). For subunit1, four residues that locate at a β-hairpin tip
(K512, H513, L514, and N515), and an adjacent helix (F459 on α14) interact with the
DNA via three phosphates on the DNA backbones, two sugar moieties and three bases
(C
8
, G
8
, and A
7
) (Figure 3.5E). For subunit2, similarly, K512, H513, R456, F459 and
L514 interact with the DNA backbone (Figure 3.5F).

55
It should be noted that the two residues at the β-hairpin tip, K512 and H513, have been
reported to played critical roles in origin DNA melting/unwinding [52, 83, 87]. H513 of
the two subunits interact with two adjacent bases on the minor groove edges (Figure 3.4E
and F). Exactly at the same binding spot, the minor groove region is narrowed by 1.3 Å
(4.5 Å compared to 5.8 Å for minor groove width of a standard B-form DNA). Such
conformational distortion may be induced by the binding of the β-hairpins, by which the
origin could be melted. It is still possible such distortion is inherent in the DNA itself,
and is caused by the DNA sequence. Thus, Monte Carlo (MC) simulations of the origin
DNA structure were carried out, and indicated that the DNA region (the region that was
contacted by H513 in the presence of protein binding) has an intrinsically narrowed
groove in the absence of protein binding (Figure 3.6B, green plots).

Dimer-DNA complex: an assembly intermediate without melting activity
As shown in Figure 3.6, MC simulation results indicated the narrowed minor groove
where the β-hairpins contact with the origin DNA was not the result of protein binding.
The slight distortion is a pre-existing structural feature the DNA, contributed by the
particular DNA sequence. Similar results have been reported on crystal structures of
archaeal ORC initiators-dsDNA complexes [88, 89], showing no severe distortion or
melting of DNA. However, another hexameric helicase, E1, was shown to be able to melt
DNA as in the trimeric form [90]. Since the dimeric LTag/dsDNA co-crystal was
obtained in absence of ATP, the question remained whether the intermediate dimeric
form we reported here was able to melt the origin DNA when ATP is present.

56

Figure 3.6. The narrow minor groove geometry of the 5’-C
5
T
6
T
7
C
8
T
9
region of EP-ori
DNA where H513 residues on both β-hairpins bind in our crystal structure. (A) The shape
of the molecular surface is shown with GRASP2 (concave surfaces in dark gray; convex
surfaces in green) [91]. A red mesh represents an isopotential surface at -5 kT/e
calculated with DelPhi at physiologic ionic strength [92]. The H513 residues from both
subunits intrude the minor groove in a region with enhanced negative electrostatic
potential through closing down the width of the minor groove to 4.5 Å (with an
electrostatic potential of -9.4 kT/e) from the normal width of 5.8 Å (with an electrostatic
potential of -7.2 kT/e, see panel-B). (B) Minor groove width of bound ori DNA in our
crystal structure (blue plot) and unbound ori DNA predicted in MC simulations (green
plots), and electrostatic potential in the center of the minor groove calculated with DelPhi
(red plots) illustrate that the H513 residues bind a region with intrinsically narrow minor
groove. The enhanced negative electrostatic potential in the narrower groove region
attracts H513 residues through favorable electrostatic interactions, a mechanism known
as shape readout.

57
To address this question, we designed mutations to capture a stable dimer intermediate of
LTag in solution. Two mutants (V350E/P417D and L286Dt/R567E) are designed to
introduce mutations on only one side of the interface for each mutant, so that inter-
subunit interactions are disrupted (Figure 3.7). We predicted that, when alone, these two
mutants would be monomeric. But when mixed together, each mutant contributes one
native side to form a dimer. The two outer surfaces of the dimer carry mutations to
prevent further oligomerization (Figure 3.8A). Gel filtration chromatography results
indicated that each individual mutant was indeed monomeric, and the equimolar mixture
of the two mutants was dimeric (Figure 3.8B). To distinguish the two mutant proteins on
the captured dimer, we added a 14 amino acid residues C-terminal tail on L286Dt/R567E
mutant. SDS-PAGE analysis of such a dimer revealed a 1:1 stoichiometry of the two
mutants based on gel quantification (Figure 3.8B inset).

To further validate the dimer’s interface, ATPase assay was conducted to test whether the
ATP pocket formed at the dimer interface can hydrolyze ATP. The results showed clearly
that each mutant alone had severely disrupted ATPase activity, while the mixture of the
two mutants has wt-level ATPase activity (Figure 3.8C), indicating the captured mutant
dimer intermediate that reconstitutes a wt ATP pocket at the dimer interface. Helicase
assay showed no unwinding activity for the two mutants individually or as a mixture
(data not shown), consistent with the fact that only functional hexamer of LTag can
unwind DNA.

58

Figure 3.7. Mutations introduced on the interface between adjacent Zn and AAA+
domains. (A) In wild type LTag, two Zn domains are brought together by multiple
hydrophobic interactions between two α helices, such as V350 and L286. On the interface
between two AAA+ domains, P417 and R567 are close to each other. All four sites are
mutated to E or D, thus the positive-positive repulsive force prevents oligomerization of
monomers. Both V350E/P417D and L286Dt/R567E mutants have only one intact side of
the interface thus they can only exist as monomers, individually. But when mixed
together, each mutant contributes one intact side thus forming a complete interface, while
leaving only the disrupted interfaces exposing to prevent formation of any oligomers
larger than dimers. (B and C) Close-up views of the interactions

59
With the captured dimer intermediate at hand, we assayed the origin melting activity of
the dimer intermediate in the presence or absence of ATP. We used the potassium
permanganate (KMnO
4
) reactivity assay on a 92 bp SV40 ori containing DNA substrate.
The results showed that the dimer intermediate had no detectable DNA melting activity
(Figure 3.8D). Thus the dimeric state should be an assembly intermediate that only
recognizes and binds to origin DNA, and is not capable of melting origin DNA.

Figure 3.8. Generating a stable LTag dimer intermediate and analysis of its activities. (A)
Strategy to obtain a stable dimer intermediate of LTag through two independent
mutations. Mut 1 carries V350E/P417D mutation on one side of a subunit, and mut 2 has
L286Dt/R567E mutations on the other. Both mutants are predicted to exist only as
monomers when alone, but as a stable dimer when mixed in equal molar. (B) Superdex-

60
200 column chromatography, showing mut 1 and mut 2 alone was monomeric, but when
mixed in equimolar, dimers were formed. Inset: the two mutants in the dimer peak were
detected in 1:1 ratio by SDS-PAGE, with mut 1 being slightly larger due to a C-terminal
extension. Note, WT equilibrates between monomeric and hexameric forms, no dimer or
any other intermediate oligomeric form can be detected. (C) ATPase assay result shows a
recovery of ATPase activity only when mixing mut1 and mut 2, indicating a WT ATP
pocket at the dimer interface. (D) DNA melting assay by KMnO
4
reactivity. Radiolabeled
92bp ori DNA was incubated with increasing quantities of LTag protein (lanes 2-5 and
lanes 6-9: 100, 200, 400 and 800ng for WT and mutant dimer respectively). Lane 1 has
no protein added.

3.3 Discussion
Importance of the Zn-domain and assembly of LTag at replication origin
The Zn domain plays an important role in hexamerization of the AAA+ domain [50]. In
the dimeric LTag/dsDNA structure, interactions between the two Zn domains are similar
to those of hexameric LTag structure in various nucleotide binding states [50, 51],
suggesting the relative orientation between adjacent Zn domains seems to be independent
of nucleotide binding states, DNA binding states and oligomeric states. Quantitated
analysis revealed the Zn domain and DNA interface contributes ~25% of the total
protein-DNA interface, implying its important role for initial loading and assembly of the
LTag at the origin.

Role of β-hairpin in ori DNA binding
The two residues at the tip of the β-hairpin, K512 and H513, have been reported to play
essential roles in DNA melting/unwinding activity of various helicases [52, 87, 93]. The
lysine residue (K512 of LTag) at the β-hairpin is conserved among all helicases in SF3

61
helicase family, including LTag, E1 and Rep. In contrast, the histidine (H513 of LTag) is
found only present and conserved in the helicases from organisms that have dsDNA
genome [87]. Thus it is believed that the histidine may be required for local melting of
the origin. The interactions we present in this report of these two residues may have
several implications.

First, the β-hairpins of the two subunits contact the DNA minor groove side-by-side,
following the helical path of the groove like two steps of a helical staircase. Such
arrangement is stabilized by electrostatic interactions with the DNA, and also probably a
hydrogen bond between the two H513 residues themselves. Similar helical arrangement
of the β-hairpins was also found on E1 hexamer structures [93, 94]. If β-hairpins in the
LTag hexamer all follow the minor groove and make interactions with 6 consecutive base
pairs, the dsDNA is likely to be distorted/melted since the standard A or B form dsDNA
has ~10bps per helical turn (360° ). Actually, in two structures of hexameric LTag/dsDNA
complexes (this report and unpublished data), the β-hairpins arrange almost on the same
plane, instead of following a helical path. Furthermore, dsDNA in one of the structure is
shown to be straightened thus distorted.

Second, given E1 has been reported to be able to locally melt DNA as an intermediate
trimer, instead of a complete hexamer [87], biochemical characterization of this dimeric
LTag was carried out in vitro, especially in the presence of ATP. Hexamerization of
LTag from monomeric state has been reported to be a very cooperative process, and the

62
intermediate oligomeric states, from dimeric to pentameric states, could be only captured
by crosslinking [50, 54]. For the first time, we obtained the LTag in stable dimeric form
with ATPase activity as potent as a wild type (Figure 3.8B and C), implying that the
dimeric state was stable and biochemically competent. KMnO
4
reactivity assay is able to
detect unpaired bases, and therefore was carried out to evaluate DNA melting ability of
the dimeric LTag. Our results indicated the dimeric LTag was unable to melt the EP-half
origin. Taken together with the MC simulation results, the inherent DNA distortion we
found on the EP-half origin is not likely to be further distorted upon LTag binding, and
the dimeric LTag form is just an assembly intermediate.

Third, the narrowed minor groove present in the dimeric LTag/DNA complex structure
was an intrinsic feature of the particular DNA sequence, according to the MC simulation
results (Figure 3.6). A narrowed minor groove linked to the 5’-ACTTC-3’ sequence
shown in Figure 3.6B has also been reported on X-ray structures of free B-DNA [95].
Due to the narrowed opening of the groove, the electronegativity of that region increases
by ~2kT/e, resulting in stronger attraction to the two H513 residues. It also implies that
the mechanism that the protein recognizes the DNA here, also known as read-out, is more
likely shape/electronegativity based, rather than sequence based. This shape based read-
out mechanism is supported by a previous study in which mutation of this DNA region
demolished viral DNA replication [78]. In that study, PEN1-4 were inverted from one
strand to another, even though the particular sequence remained, the actual facing
orientations of the PENs shifted and viral DNA replication was inhibited, probably due to

63
the fact that the LTag could no longer recognize the PENs properly. Such protein-DNA
interaction that is non-specific to DNA sequence has been found always to happen in the
minor groove, which was characterized from structures of 21 DNA-binding protein
families [96].

Fourth, LTag has been reported to be able to encircle and translocate along ssDNA [97,
98]. However, there is still disagreement about whether the central channel of the LTag
hexamer is wide enough to encircle dsDNA [93]. Electron microscopy results have been
reported to show LTag hexamer encircle the dsDNA origin [48, 99], whereas how the
~15 Å central channel (which is the narrowest bottleneck at the β-hairpins)
accommodates a ~20 Å dsDNA still remained unclear until our results. In the dimeric
LTag/dsDNA structure, the two H513 residues at the β-hairpin insert into the minor
groove and contact with both DNA strands. While the two K512 residues contact with
only one DNA strand and the long rigid side chains of the two residues take a
conformation that is parallel to the DNA single strand (Figure 3.5E and F). Besides, in
the initial density map of the hexameric LTag/DNA crystal, the density corresponding to
the EP-half origin DNA is clearly present and the diameter of the central channel at the β-
hairpins is ~19 Å, wide enough to accommodate dsDNA (Figure 3.3). The seemingly
snug fit of the dsDNA in the central channel might enable LTag distinguish dsDNA from
ssDNA, which is consistent with the distinct ATPase stimulation results of the LTag
incubated with dsDNA or ssDNA (unpublished data).

64
3.4 Conclusion
We have obtained an initial density map of the LTag in hexameric form, with the EP-half
origin DNA in the central channel. We also determined the crystal structure of the LTag
in dimeric form with the EP-half origin DNA, revealing how OBDs recognize and bind to
the origin. A shape based read-out mechanism was proposed for the interaction between
the β-hairpin and the minor groove region on the origin. In vitro biochemical
characterization of the dimeric LTag complex indicated the LTag dimer in the co-crystal
with the EP-half origin DNA is likely to be a snapshot of an early stage assembly,
representing how the EP-half origin recruits the LTag and initiate the assembly of the
LTag hexamer/double-hexamer.

3.5 Materials and Methods
Cloning, mutations, protein purification
LTag131–627 cDNA was cloned into pGEX-6P-1 vector with an N-terminal GST tag.
Then the construct was transformed in to E. coli by electroporation. The LTag protein
was expressed by adding IPTG to 2mM at 18° C when the cell density reached OD ~ 0.6.
After cells were lysed by French Press, the protein was purified by glutathione affinity
chromatography in a buffer containing 50 mM Tris-HCl (pH 8.0), 250 mM NaCl (buffer
A) and 1mM DTT. The N-terminal GST tag was cleaved by treatment with PreScission
protease and the eluted LTag was purified by Superdex-200 gel filtration chromatography.
To generate monomeric mutants for forming a stable dimer intermediate, V350E/P417D
and L286Dt/R567E mutants were made with the QuickChange Site-Directed

65
Mutangenesis kit (Stratagene) according to the manufacturer’s instruction. All mutations
were confirmed by sequencing the entire LTag coding sequence.

Co-crystallization and data collection
To prepare EP-half origin DNA for co-crystallization, oligonucleotides 5’-
ACTACTTCTGGAATAGCTCAGAGGCCGAGGCGT-3’ (top strand) and 5’-
CGCCTCGGCCTCTGAGCTATTCCAGAAGTAGTT-3’ (bottom strand) were purified
by ion exchange chromatography with a Mono Q column. The two purified
oligonucleotides incubated in 1:1 molar ratio and annealed by heating to 94 º C, followed
by slow cooling to room temperature. Gel filtration chromatography with a Superdex 75
column was used to separate the unannealed ssDNA thus to further purify the annealed
EP-half origin dsDNA. The purified EP-half origin dsDNA and LTag131–627 (10mg/ml)
were mixed in a 1.25:1 molar ratio and incubated at 10º C for 1 hour. The Dimer crystals
were grown at 4 º C in the sitting drop vapor diffusion trays within a solution containing
100 mM bis-Tris (pH 6.75) and 20 % (v/v) PEG3350. The hexamer crystals were grown
at 18 º C in the sitting drop vapor diffusion trays within a solution containing 60mM
Hepes (pH 7.25) and 140~160mM Na/K Tartrate and 30~50mM MgCl
2
as the additive.
The crystals grew to full sizes after 7~14 days and transferred to the crystallization buffer
containing 20~30% glycerol. Then the crystals were flash frozen and stored in liquid
nitrogen. Diffraction data were collected at the synchrotron beamline (APS) and datasets
were processed using HKL2000. The diffraction statistics are summarized in Table 3.1.

66
Structure determination and refinement
For the hexamer crystal, only initial phase restricted to the Zn and AAA+ domains were
obtained by Molecular Replacement (PHASER) with one copy of the LTag hexamer
(PDB:1SVO). Due to the fuzzy density of OBD domains and the fact that the EP-half
origin dsDNA has several orientations in the crystal, no further phasing or refinement
could be done that would yield credible statistics.

For the dimer crystal forms, the initial phase were solved by Molecular Replacement
(PHASER) with two copies of each of the LTag helicase domain (PDB:1SVM) and the
OBD domain (PDB:2ITL). Then 2-fold NCS averaging was performed to improve the
density to the extent that the registry of the DNA base pairs in the complex could be
determined based on the shape of the electron density. The phase was further improved
by iterative model-rebuilding with O or Coot, and refinement with CNS. The refinement
statistics are shown in Table 3.1.

Analysis and prediction of origin DNA structure
CURVES was used to analyze the dimeric LTag/DNA structure [100]. The electrostatic
potential was calculated as described (20, 29)[101, 102], with the reference points at the
center of the minor groove. AMBER force field, an implicit aqueous solvent model, and
explicit sodium counter ions were used in the all-atom Monte Carlo (MC) simulations to
predict the conformation of DNA unbound with protein [103].

67
ATPase Assay and Helicase Assay
The ATPase assay for LTag WT or mutants was performed by detecting the phosphate
generated by ATP hydrolysis as described previously [104]. For helicase DNA substrate,
Y fork-shaped DNA with 44nt ssDNA tails and a 44nt duplex was made by annealing
two oligonucleotides of 5’-(dT)
44
GCTCGTGCAGACGTCGAGGTGAGGACGAGCT
CCTCGTGACCACG-3’ and 5’-CGTGGTCACGAGGAGCTCGTCCTCACCTCGACG
TCTGCACGAGC(dT)
44
-3’. Helicase assay was performed as described [104]. Briefly,
~10 fmol of [ -32P]-ATP labeled substrate DNA was incubated with 300ng LTag in
helicase buffer containing 20mM Tris-Cl pH7.5, 10mM MgCl
2
, 5mM ATP, 1mM DTT
and 0.1mg/ml BSA for 45 min at 37° C. The reaction was analyzed on 12% native
polyacrylamide gel, and the amount of radioactively labeled oligonucleotide was
determined by autoradiography.

KMnO
4
Reactivity Assay
KMnO
4
reactivity assay to test DNA melting activity was performed as follows. A 92 bp
dsDNA substrate that contains EP-half origin was generated by annealing two ssDNA.
The top strand sequence is 5’-GCGTGTCATTGGGGGCTTATACAGGCGTAGACTA
CTTCTGGAATAGCTCAGAGGCCGAGGCGACTACAATGGGCCCAACTCAATCA
CAGCTC-3’, and the bottom strand sequence is 5’-GAGCTGTGATTGAGTTGGGCC
CATTGTAGTCGCCTCGGCCTCTGAGCTATTCCAGAAGTAGTCTACGCCTGTAT
AAGCCCCCAATGACACGC-3’. Only the bottom strand was 5’end-labeled with [ -32P]
ATP. 10~15 fmol DNA was incubated with 100~800ng LTag protein in buffer containing

68
20mM Tris-Cl pH7.5, 10mM MgCl
2
, 5mM ATP, 1mM DTT and 0.1mg/ml BSA for 30
min at 37° C. Then KMnO
4
was added to a final concentration of 6mM and reactions
were incubated at 37° C for 2 min. To stop the oxidation, reactions were reduced by
adding β-mercaptoethanol to 160mM, SDS to 0.3% and EDTA to 10mM. Modified DNA
substrates were deproteinized by first digestion with proteinase K (20 μg/ml, 60min at
37° C) then bound to Qiagen QIAEX II beads. Piperidine was added to 20% (v/v) on
dried beads to cleave DNA substrates at modified nucleotides (30 min at 90° C).

69

Bibliography

1. Blow JJ, Dutta A: Preventing re-replication of chromosomal DNA. Nat Rev
Mol Cell Biol 2005, 6(6):476-486.
2. Mechali M: Eukaryotic DNA replication origins: many choices for
appropriate answers. Nat Rev Mol Cell Biol 2010, 11(20861881):728-738.
3. Speck C, Chen Z, Li H, Stillman B: ATPase-dependent cooperative binding of
ORC and Cdc6 to origin DNA. Nat Struct Mol Biol 2005, 12(11):965-971.
4. Remus D, Diffley JF: Eukaryotic DNA replication control: lock and load, then
fire. Curr Opin Cell Biol 2009, 21(6):771-777.
5. Takara TJ, Bell SP: Multiple Cdt1 molecules act at each origin to load
replication-competent Mcm2-7 helicases. Embo J 2011, 30(24):4885-4896.
6. Remus D, Beuron F, Tolun G, Griffith JD, Morris EP, Diffley JF: Concerted
loading of Mcm2-7 double hexamers around DNA during DNA replication
origin licensing. Cell 2009, 139(4):719-730.
7. Evrin C, Clarke P, Zech J, Lurz R, Sun J, Uhle S, Li H, Stillman B, Speck C: A
double-hexameric MCM2-7 complex is loaded onto origin DNA during
licensing of eukaryotic DNA replication. Proc Natl Acad Sci U S A 2009,
106(48):20240-20245.
8. Randell JC, Bowers JL, Rodriguez HK, Bell SP: Sequential ATP hydrolysis by
Cdc6 and ORC directs loading of the Mcm2-7 helicase. Mol Cell 2006,
21(1):29-39.
9. Tanaka S, Tak YS, Araki H: The role of CDK in the initiation step of DNA
replication in eukaryotes. Cell division 2007, 2:16.
10. Tanaka S, Umemori T, Hirai K, Muramatsu S, Kamimura Y, Araki H: CDK-
dependent phosphorylation of Sld2 and Sld3 initiates DNA replication in
budding yeast. Nature 2007, 445(7125):328-332.
11. Zegerman P, Diffley JF: Phosphorylation of Sld2 and Sld3 by cyclin-
dependent kinases promotes DNA replication in budding yeast. Nature 2007,
445(7125):281-285.
12. Sheu YJ, Stillman B: Cdc7-Dbf4 phosphorylates MCM proteins via a docking
site-mediated mechanism to promote S phase progression. Mol Cell 2006,
24(1):101-113.
13. Zegerman P, Diffley JF: DNA replication as a target of the DNA damage
checkpoint. DNA repair 2009, 8(9):1077-1088.
14. Diffley JF: Regulation of early events in chromosome replication. Curr Biol
2004, 14(18):R778-786.
15. Moyer SE, Lewis PW, Botchan MR: Isolation of the Cdc45/Mcm2-7/GINS
(CMG) complex, a candidate for the eukaryotic DNA replication fork
helicase. Proc Natl Acad Sci U S A 2006, 103(27):10236-10241.

70
16. Labib K, Gambus A: A key role for the GINS complex at DNA replication
forks. Trends Cell Biol 2007, 17(6):271-278.
17. Hennessy KM, Lee A, Chen E, Botstein D: A group of interacting yeast DNA
replication genes. Genes Dev 1991, 5(6):958-969.
18. Thommes P, Fett R, Schray B, Burkhart R, Barnes M, Kennedy C, Brown NC,
Knippers R: Properties of the nuclear P1 protein, a mammalian homologue of
the yeast Mcm3 replication protein. Nucleic Acids Res 1992, 20(5):1069-1074.
19. Tye BK: MCM proteins in DNA replication. Annu Rev Biochem 1999, 68:649-
686.
20. Maiorano D, Lutzmann M, Mechali M: MCM proteins and DNA replication.
Curr Opin Cell Biol 2006, 18(2):130-136.
21. Kearsey SE, Labib K: MCM proteins: evolution, properties, and role in DNA
replication. Biochim Biophys Acta 1998, 1398(2):113-136.
22. Lei M, Kawasaki Y, Tye BK: Physical interactions among Mcm proteins and
effects of Mcm dosage on DNA replication in Saccharomyces cerevisiae. Mol
Cell Biol 1996, 16(9):5081-5090.
23. Richter A, Knippers R: High-molecular-mass complexes of human
minichromosome-maintenance proteins in mitotic cells. Eur J Biochem 1997,
247(1):136-141.
24. Ishimi Y, Ichinose S, Omori A, Sato K, Kimura H: Binding of human
minichromosome maintenance proteins with histone H3. J Biol Chem 1996,
271(39):24115-24122.
25. Musahl C, Schulte D, Burkhart R, Knippers R: A human homologue of the yeast
replication protein Cdc21. Interactions with other Mcm proteins. Eur J
Biochem 1995, 230(3):1096-1101.
26. Burkhart R, Schulte D, Hu D, Musahl C, Gohring F, Knippers R: Interactions of
human nuclear proteins P1Mcm3 and P1Cdc46. Eur J Biochem 1995,
228(2):431-438.
27. Sherman DA, Forsburg SL: Schizosaccharomyces pombe Mcm3p, an essential
nuclear protein, associates tightly with Nda4p (Mcm5p). Nucleic Acids Res
1998, 26(17):3955-3960.
28. Yabuta N, Kajimura N, Mayanagi K, Sato M, Gotow T, Uchiyama Y, Ishimi Y,
Nojima H: Mammalian Mcm2/4/6/7 complex forms a toroidal structure.
Genes Cells 2003, 8(5):413-421.
29. Lee JK, Hurwitz J: Isolation and characterization of various complexes of the
minichromosome maintenance proteins of Schizosaccharomyces pombe. J
Biol Chem 2000, 275(25):18871-18878.
30. Adachi Y, Usukura J, Yanagida M: A globular complex formation by Nda1 and
the other five members of the MCM protein family in fission yeast. Genes
Cells 1997, 2(7):467-479.
31. Ishimi Y: A DNA helicase activity is associated with an MCM4, -6, and -7
protein complex. J Biol Chem 1997, 272(39):24508-24513.

71
32. Lee JK, Hurwitz J: Processive DNA helicase activity of the minichromosome
maintenance proteins 4, 6, and 7 complex requires forked DNA structures.
Proc Natl Acad Sci U S A 2001, 98(1):54-59.
33. You Z, Komamura Y, Ishimi Y: Biochemical analysis of the intrinsic Mcm4-
Mcm6-mcm7 DNA helicase activity. Mol Cell Biol 1999, 19(12):8003-8015.
34. Ishimi Y, Komamura Y, You Z, Kimura H: Biochemical function of mouse
minichromosome maintenance 2 protein. J Biol Chem 1998, 273(14):8369-
8375.
35. Sato M, Gotow T, You Z, Komamura-Kohno Y, Uchiyama Y, Yabuta N, Nojima
H, Ishimi Y: Electron microscopic observation and single-stranded DNA
binding activity of the Mcm4,6,7 complex. J Mol Biol 2000, 300(3):421-431.
36. Patel SS, Picha KM: Structure and function of hexameric helicases. Annu Rev
Biochem 2000, 69:651-697.
37. Fletcher RJ, Bishop BE, Leon RP, Sclafani RA, Ogata CM, Chen XS: The
structure and function of MCM from archaeal M. Thermoautotrophicum.
Nat Struct Biol 2003, 10(3):160-167.
38. Liu W, Pucci B, Rossi M, Pisani FM, Ladenstein R: Structural analysis of the
Sulfolobus solfataricus MCM protein N-terminal domain. Nucleic Acids Res
2008, 36(10):3235-3243.
39. Brewster AS, Wang G, Yu X, Greenleaf WB, Carazo JM, Tjajadi M, Klein MG,
Chen XS: Crystal structure of a near-full-length archaeal MCM: functional
insights for an AAA+ hexameric helicase. Proc Natl Acad Sci U S A 2008,
105(51):20191-20196.
40. Yan H, Gibson S, Tye BK: Mcm2 and Mcm3, two proteins important for ARS
activity, are related in structure and function. Genes Dev 1991, 5(6):944-957.
41. You Z, Ishimi Y, Masai H, Hanaoka F: Roles of Mcm7 and Mcm4 subunits in
the DNA helicase activity of the mouse Mcm4/6/7 complex. J Biol Chem 2002,
277(45):42471-42479.
42. Sherman DA, Pasion SG, Forsburg SL: Multiple domains of fission yeast
Cdc19p (MCM2) are required for its association with the core MCM
complex. Mol Biol Cell 1998, 9(7):1833-1845.
43. Barry ER, McGeoch AT, Kelman Z, Bell SD: Archaeal MCM has separable
processivity, substrate choice and helicase domains. Nucleic Acids Res 2007,
35(3):988-998.
44. Bullock PA: The initiation of simian virus 40 DNA replication in vitro.
Critical reviews in biochemistry and molecular biology 1997, 32(6):503-568.
45. Fanning E, Zhao K: SV40 DNA replication: from the A gene to a nanomachine.
Virology 2009, 384(19101707):352-359.
46. Ahuja D, Saenz-Robles MT, Pipas JM: SV40 large T antigen targets multiple
cellular pathways to elicit cellular transformation. Oncogene 2005,
24(52):7729-7745.
47. Huberman JA, Riggs AD: Autoradiography of chromosomal DNA fibers from
Chinese hamster cells. Proc Natl Acad Sci U S A 1966, 55(3):599-606.

72
48. Valle M, Gruss C, Halmer L, Carazo JM, Donate LE: Large T-antigen double
hexamers imaged at the simian virus 40 origin of replication. Mol Cell Biol
2000, 20(1):34-41.
49. Simmons DT: SV40 large T antigen functions in DNA replication and
transformation. Adv Virus Res 2000, 55(11050941):75-7134.
50. Li D, Zhao R, Lilyestrom W, Gai D, Zhang R, DeCaprio JA, Fanning E,
Jochimiak A, Szakonyi G, Chen XS: Structure of the replicative helicase of the
oncoprotein SV40 large tumour antigen. Nature 2003, 423(6939):512-518.
51. Gai D, Zhao R, Li D, Finkielstein CV, Chen XS: Mechanisms of conformational
change for a replicative hexameric helicase of SV40 large tumor antigen. Cell
2004, 119(1):47-60.
52. Shen J, Gai D, Patrick A, Greenleaf WB, Chen XS: The roles of the residues on
the channel beta-hairpin and loop structures of simian virus 40 hexameric
helicase. Proc Natl Acad Sci U S A 2005, 102(32):11248-11253.
53. Deb S, DeLucia AL, Baur CP, Koff A, Tegtmeyer P: Domain structure of the
simian virus 40 core origin of replication. Mol Cell Biol 1986,
6(3023900):1663-1670.
54. Dean FB, Borowiec JA, Eki T, Hurwitz J: The simian virus 40 T antigen double
hexamer assembles around the DNA at the replication origin. J Biol Chem
1992, 267(20):14129-14137.
55. Tye BK: The MCM2-3-5 proteins: are they replication licensing factors?
Trends Cell Biol 1994, 4(5):160-166.
56. Chong JP, Mahbubani HM, Khoo CY, Blow JJ: Purification of an MCM-
containing complex as a component of the DNA replication licensing system.
Nature 1995, 375(6530):418-421.
57. Todorov IT, Attaran A, Kearsey SE: BM28, a human member of the MCM2-3-
5 family, is displaced from chromatin during DNA replication. J Cell Biol
1995, 129(6):1433-1445.
58. Bochman ML, Schwacha A: The Mcm2-7 complex has in vitro helicase activity.
Mol Cell 2008, 31(2):287-293.
59. Ward JJ, Sodhi JS, McGuffin LJ, Buxton BF, Jones DT: Prediction and
functional analysis of native disorder in proteins from the three kingdoms of
life. J Mol Biol 2004, 337(3):635-645.
60. Buchan DW, Ward SM, Lobley AE, Nugent TC, Bryson K, Jones DT: Protein
annotation and modelling servers at University College London. Nucleic
Acids Res 2010, 38(Web Server issue):W563-568.
61. Jones DT: Protein secondary structure prediction based on position-specific
scoring matrices. J Mol Biol 1999, 292(2):195-202.
62. Yu Z, Feng D, Liang C: Pairwise interactions of the six human MCM protein
subunits. J Mol Biol 2004, 340(5):1197-1206.
63. Kneissl M, Putter V, Szalay AA, Grummt F: Interaction and assembly of
murine pre-replicative complex proteins in yeast and mouse cells. J Mol Biol
2003, 327(1):111-128.

73
64. Davey MJ, Indiani C, O'Donnell M: Reconstitution of the Mcm2-7p
heterohexamer, subunit arrangement, and ATP site architecture. J Biol Chem
2003, 278(7):4491-4499.
65. Poplawski A, Grabowski B, Long SE, Kelman Z: The zinc finger domain of the
archaeal minichromosome maintenance protein is required for helicase
activity. J Biol Chem 2001, 276(52):49371-49377.
66. Masai H, Taniyama C, Ogino K, Matsui E, Kakusho N, Matsumoto S, Kim JM,
Ishii A, Tanaka T, Kobayashi T et al: Phosphorylation of MCM4 by Cdc7
kinase facilitates its interaction with Cdc45 on the chromatin. J Biol Chem
2006, 281(51):39249-39261.
67. Tran NQ, Dang HQ, Tuteja R, Tuteja N: A single subunit MCM6 from pea
forms homohexamer and functions as DNA helicase. Plant molecular biology
2010, 74(4-5):327-336.
68. Liang DT, Forsburg SL: Characterization of Schizosaccharomyces pombe
mcm7(+) and cdc23(+) (MCM10) and interactions with replication
checkpoints. Genetics 2001, 159(2):471-486.
69. Forsburg SL: Eukaryotic MCM proteins: beyond replication initiation.
Microbiol Mol Biol Rev 2004, 68(1):109-131.
70. Schwacha A, Bell SP: Interactions between two catalytically distinct MCM
subgroups are essential for coordinated ATP hydrolysis and DNA replication.
Mol Cell 2001, 8(5):1093-1104.
71. Hall MN, Gabay J, Debarbouille M, Schwartz M: A role for mRNA secondary
structure in the control of translation initiation. Nature 1982, 295(5850):616-
618.
72. Chen H, Bjerknes M, Kumar R, Jay E: Determination of the optimal aligned
spacing between the Shine-Dalgarno sequence and the translation initiation
codon of Escherichia coli mRNAs. Nucleic Acids Res 1994, 22(23):4953-4957.
73. Chong JP, Hayashi MK, Simon MN, Xu RM, Stillman B: A double-hexamer
archaeal minichromosome maintenance protein is an ATP-dependent DNA
helicase. Proc Natl Acad Sci U S A 2000, 97(4):1530-1535.
74. Chen XS, Stehle T, Harrison SC: Interaction of polyomavirus internal protein
VP2 with the major capsid protein VP1 and implications for participation of
VP2 in viral entry. Embo J 1998, 17(12):3233-3240.
75. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG: The
CLUSTAL_X windows interface: flexible strategies for multiple sequence
alignment aided by quality analysis tools. Nucleic Acids Res 1997,
25(24):4876-4882.
76. Greenleaf WB, Shen J, Gai D, Chen XS: Systematic study of the functions for
the residues around the nucleotide pocket in simian virus 40 AAA+
hexameric helicase. J Virol 2008, 82(18400864):6017-6023.
77. Bochkareva E, Martynowski D, Seitova A, Bochkarev A: Structure of the
origin-binding domain of simian virus 40 large T antigen bound to DNA.
EMBO J 2006, 25(17139255):5961-5969.

74
78. Deb SP, Tegtmeyer P: ATP enhances the binding of simian virus 40 large T
antigen to the origin of replication. J Virol 1987, 61(2824805):3649-3654.
79. Valle M, Chen XS, Donate LE, Fanning E, Carazo JM: Structural basis for the
cooperative assembly of large T antigen on the origin of replication. J Mol
Biol 2006, 357(16481006):1295-1305.
80. Mastrangelo IA, Hough PV, Wall JS, Dodson M, Dean FB, Hurwitz J: ATP-
dependent assembly of double hexamers of SV40 T antigen at the viral origin
of DNA replication. Nature 1989, 338(6217):658-662.
81. Joo WS, Kim HY, Purviance JD, Sreekumar KR, Bullock PA: Assembly of T-
antigen double hexamers on the simian virus 40 core origin requires only a
subset of the available binding sites. Mol Cell Biol 1998, 18(5):2677-2687.
82. Sreekumar KR, Prack AE, Winters DR, Barbaro BA, Bullock PA: The simian
virus 40 core origin contains two separate sequence modules that support T-
antigen double-hexamer assembly. J Virol 2000, 74(18):8589-8600.
83. Borowiec JA, Hurwitz J: Localized melting and structural changes in the SV40
origin of replication induced by T-antigen. EMBO J 1988, 7(10):3149-3158.
84. Borowiec JA, Dean FB, Bullock PA, Hurwitz J: Binding and unwinding--how T
antigen engages the SV40 origin of DNA replication. Cell 1990, 60(2):181-184.
85. Meinke G, Phelan P, Moine S, Bochkareva E, Bochkarev A, Bullock PA, Bohm
A: The crystal structure of the SV40 T-antigen origin binding domain in
complex with DNA. PLoS Biol 2007, 5(2):e23.
86. Olson WK, Gorin AA, Lu XJ, Hock LM, Zhurkin VB: DNA sequence-
dependent deformability deduced from protein-DNA crystal complexes. Proc
Natl Acad Sci U S A 1998, 95(9736707):11163-11168.
87. Liu X, Schuck S, Stenlund A: Adjacent residues in the E1 initiator beta-
hairpin define different roles of the beta-hairpin in Ori melting, helicase
loading, and helicase activity. Mol Cell 2007, 25(6):825-837.
88. Dueber ELC, Corn JE, Bell SD, Berger JM: Replication origin recognition and
deformation by a heterodimeric archaeal Orc1 complex. Science 2007,
317(17761879):1210-1213.
89. Gaudier M, Schuwirth BS, Westcott SL, Wigley DB: Structural basis of DNA
replication origin recognition by an ORC protein. Science 2007,
317(17761880):1213-1216.
90. Schuck S, Stenlund A: Surface mutagenesis of the bovine papillomavirus E1
DNA binding domain reveals residues required for multiple functions related
to DNA replication. J Virol 2006, 80(15):7491-7499.
91. Petrey D, Honig B: GRASP2: visualization, surface properties, and
electrostatics of macromolecular structures and sequences. Methods Enzymol
2003, 374:492-509.
92. Rocchia W, Sridharan S, Nicholls A, Alexov E, Chiabrera A, Honig B: Rapid
grid-based construction of the molecular surface and the use of induced
surface charge to calculate reaction field energies: applications to the
molecular systems and geometric objects. JComputChem 2002, 23(1):128-137.

75
93. Enemark EJ, Joshua-Tor L: Mechanism of DNA translocation in a replicative
hexameric helicase. Nature 2006, 442(16855583):270-275.
94. Sanders CM, Kovalevskiy OV, Sizov D, Lebedev AA, Isupov MN, Antson AA:
Papillomavirus E1 helicase assembly maintains an asymmetric state in the
absence of DNA and nucleotide cofactors. Nucleic Acids Res 2007,
35(17881379):6451-6457.
95. Oguey C, Foloppe N, Hartmann B: Understanding the sequence-dependence of
DNA groove dimensions: implications for DNA interactions. PloS one 2010,
5(12):e15931.
96. Luscombe NM, Thornton JM: Protein-DNA interactions: amino acid
conservation and the effects of mutations on binding specificity. J Mol Biol
2002, 320(5):991-1009.
97. Goetz GS, Dean FB, Hurwitz J, Matson SW: The unwinding of duplex regions
in DNA by the simian virus 40 large tumor antigen-associated DNA helicase
activity. J Biol Chem 1988, 263(1):383-392.
98. Morris PD, Byrd AK, Tackett AJ, Cameron CE, Tanega P, Ott R, Fanning E,
Raney KD: Hepatitis C virus NS3 and simian virus 40 T antigen helicases
displace streptavidin from 5'-biotinylated oligonucleotides but not from 3'-
biotinylated oligonucleotides: evidence for directional bias in translocation
on single-stranded DNA. Biochemistry 2002, 41(11841230):2372-2378.
99. Gomez-Lorenzo MG, Valle M, Frank J, Gruss C, Sorzano CO, Chen XS, Donate
LE, Carazo JM: Large T antigen on the simian virus 40 origin of replication: a
3D snapshot prior to DNA replication. Embo J 2003, 22(23):6205-6213.
100. Lavery R, Sklenar H: Defining the structure of irregular nucleic acids:
conventions and principles. J Biomol Struct Dyn 1989, 6(4):655-667.
101. Rohs R, West SM, Sosinsky A, Liu P, Mann RS, Honig B: The role of DNA
shape in protein-DNA recognition. Nature 2009, 461(19865164):1248-1253.
102. Honig B, Nicholls A: Classical electrostatics in biology and chemistry. Science
1995, 268(7761829):1144-1149.
103. Rohs R, Sklenar H, Shakked Z: Structural and energetic origins of sequence-
specific DNA bending: Monte Carlo simulations of papillomavirus E2-DNA
binding sites. Structure 2005, 13(16216581):1499-1509.
104. Greenleaf WB, Shen J, Gai D, Chen XS: Systematic study of the functions for
the residues around the nucleotide pocket in simian virus 40 AAA+
hexameric helicase. Journal of virology 2008, 82(12):6017-6023.

Abstract (if available)

Abstract The hetero-hexamer of the eukaryotic minichromosome maintenance (MCM) proteins plays an essential role in replication of genomic DNA. The ring-shaped Mcm2-7 hexamers comprising one of each subunit show helicase activity in vitro, and form double-hexamers on DNA. The Mcm4/6/7 also forms a hexameric complex with helicase activity in vitro. We used an E. coli expression system to express various domains of Schizosaccharomyces pombe Mcm4, 6 and 7 in order to characterize their domain structure, oligomeric states, and possible inter-/intra-subunit interactions. We also successfully employed a co-expression system to express Mcm4/6/7 at the same time in E. coli, and have purified functional Mcm4/6/7 complex in a hexameric state in high yield and purity, providing a means for generating large quantity of proteins for future structural and biochemical studies. Based on our results and those of others, models were proposed for the subunit arrangement and architecture of both the Mcm4/6/7 hexamer and the Mcm2-7 double-hexamer. ❧ To circumvent the apparent complexity in eukaryotic DNA replication, Simian virus large tumor antigen (SV40 LTag) has been studied as in a simplified model system for eukaryotic replication. LTag alone can recognize and assemble as double-hexamers on the SV40 replication origin, leading to melting/unwinding of the dsDNA origin. We obtained an LTag construct that is composed of the origin binding domain (OBD), Zn domain and AAA+ domain co-crystalized the origin dsDNA in both hexameric and dimeric forms. The structure of the dimeric LTag/dsDNA complex reveals how OBDs recognize and bind to the origin. The interaction between the β-hairpin and the minor groove region on the origin suggests a shape based read-out mechanism. Monte Carlo (MC) simulations and in vitro biochemical assays indicate the structure we obtained represents a snapshot of an early stage assembly.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

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

PDF

Structure and function of archaeal McM helicase

PDF

Simulating the helicase motor of SV40 large tumor antigen

PDF

Mechanism study of SV40 large tumor antigen atpase and helicase functions in viral DNA replication

PDF

Structural and biochemical studies of two DNA transaction enzymes

PDF

Structural and biochemical studies of DNA helicase complexes: conformational diversity of archaeal MCM

PDF

Structural studies of two key factors for DNA replication in eukaryotic cells

PDF

Structural and biochemical determinants of APOBEC1 substrate recognition and enzymatic function

PDF

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

PDF

Mechanism of human nonhomologous DNA end joining

PDF

The function of Rpd3 in balancing the replicaton initiation of different genomic regions

PDF

Characterization of three novel variants of the MAVS adaptor

PDF

Exploring three-dimensional organization of the genome by mapping chromatin contacts and population modeling

PDF

Biochemical mechanism of TopBP1 recruitment to sites of DNA damage

PDF

Structural and biochemical analyses on substrate specificity and HIV-1 Vif mediated inhibition of human APOBEC3 cytidine deaminases

PDF

Identification and characterization of PR-Set7 and histone H4 lysine 20 methylation-associated proteins

PDF

The calcium-sensing receptor in the specification of normal and malignant hematopoietic cell localization in the bone marrow microenvironment

PDF

Mechanistic basis for chromosomal translocations at the E2A gene

Asset Metadata

Creator Xu, Meng (author)

Core Title Biochemical characterization and structural analysis of two hexameric helicases for eukaryotic DNA replication

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Genetic, Molecular and Cellular Biology

Publication Date 11/21/2012

Defense Date 10/16/2012

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

Tag cell cycle proteins,DNA-binding proteins,Escherichiai coli,helicase,OAI-PMH Harvest,protein binding,protein oligomerization,recombinant proteins,schizosaccharomyces pombe,simian virus 40

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Chen, Xiaojiang S. (committee chair), Chen, Lin (committee member), Haworth, Ian S. (committee member)

Creator Email mengsimonxu@gmail.com,mengxu@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-112770

Unique identifier UC11289469

Identifier usctheses-c3-112770 (legacy record id)

Legacy Identifier etd-XuMeng-1326.pdf

Dmrecord 112770

Document Type Dissertation

Rights Xu, Meng

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