Biophysical study of DNA at single molecule level using solid … · 2017-02-18 · Commercial No...

Biophysical study of DNA at single molecule

level using solid-state nanopores

By

Azadeh Bahrami

A dissertation submitted to

The Department of Chemistry, Faculty of Natural Sciences

The Imperial College of Science, Technology and Medicine

In partial fulfillment of the requirements for the degree of

Doctor of Philosophy

January 2014

London, United Kingdom

i

Abstract Since the discovery of deoxyribonucleic acid (DNA) over 140 years ago, this biomolecule

still remains one the most studied macromolecules in nature. The modifications and

interactions associated with this duplex biopolymer have been shown to play fundamental

roles in cellular machinery. This research project exploited sets of single-molecule detection techniques in parallel with

conventional molecular biology methodologies to study i) chemical modification

(methylation) of the DNA and its functional role and ii) electrostatic interaction between two

homologous DNA duplexes. In this project solid-state nanopores were utilised as a novel

approach to probe structural and conformational changes of linear and circular DNA.

To begin with, the effect of DNA methylation level in breast cancer cell-lines was

investigated. Using solid-state nanopore sensors and a methyl specific antibody (5’-mc), the

methylated and unmethylated regions of FOXA1 (a gene associated with breast cancer

development) promoter were differentiated. Simultaneously, the methylation level of this

gene was evaluated in various breast cancer cell-lines and confirmed the impact of DNA

methylation in gene silencing. In addition, using atomic force microscopy analysis, the

binding affinity of the antibody to the methylated DNA was determined

Furthermore, by employing the same methodologies, the presence of an electrostatic

recognition step in homologous segments of a bacteria plasmid within the framework of

Kornyshev-Leikin theory was investigated. However to this end, the verification of this

model was inconclusive. Nevertheless it was serendipitously found that the plasmid with

homologous regions was dimerised and then formed a single loop. This finding would be the

motivation behind further experiments to gain a better understanding of the possible sequence

dependence of the DNA topology and configuration during cloning and amplification

procedures. Furthermore, using a combination of various techniques, the biophysical

properties of the monomeric and dimeric plasmids were characterised.

Overall, the combined findings of the mentioned projects provided remarkable insights on the

molecular biophysics of DNA-DNA and DNA-protein interactions within the framework of

the central dogma of molecular biology.

ii

Declarations

I, the author, hereby declare that the work presented in following thesis is original and

performed by myself; otherwise it is clearly cited and acknowledged.

This work has not been previously submitted in any form to satisfy any degree requirement at

this or any other university.

Azadeh Bahrami

29.01.2014

Copyright Notice

The copyright of this thesis rests with the author and is made available under a Creative Commons Attribution Non-

Commercial No Derivatives licence. Researchers are free to copy, distribute or transmit the thesis on the condition that they

attribute it, that they do not use it for commercial purposes and that they do not alter, transform or build upon it. For any

reuse or redistribution, researchers must make clear to others the licence terms of this work.

iii

Acknowledgments

Throughout my years at Imperial College, I have had the great pleasure to meet and work

with the most wonderful and remarkable people who made this work possible and enjoyable.

First and foremost, I would like to express my sincere gratitude and appreciation to my PhD

supervisor Dr. Tim Albrecht for giving me the opportunity to undertake a PhD at the most

crucial time! I am always thankful for your continual support, encouragement and guidance

throughout my academic and non-academic life. In addition to scientific skills, the time

management and multi-tasking are two of the best lessons that I have learnt from you.

My special gratitude also goes to my second supervisor, Professor Eric Lam for his kind

supports and advice as well as providing me with an opportunity to experience a life as a

molecular oncologist.

I have been privileged to work with other academics during my PhD study. I would like to

extend my appreciation to Professor Alexei Kornyshev and his co-workers, Ruggero and

Dominic to whom I owe my theoretical backgrounds on DNA physics. I am also grateful to

Professor Tony Cass for allowing me to use his laboratory facilities at any time I needed.

I have had a wonderful time to work with current and former members of the Albrecht

research group who made my lab’s life to be an incredible experience! Together we shared

many laughs, excitements and of course to some extent frustrations. In particular, I would to

like acknowledge the invaluable advice and trainings I have received from Alex, Agnieszka,

DJ, Fatma, Pippa, Billy and Tom. Thanks for your friendship and encouraging me to keep up

whenever I was down! I wish you all the best in your future careers.

Herewith, I would also like to acknowledge the financial support that I received from my

PhD scholarship sponsor, E F Kernahan Fund by facilitating my path towards this research

degree.

I believe that is also fair to give the British weather a credit over the course of past three year

for motivating me to stay in the lab and work harder during all the brutal cold rainy days!

iv

I also wish to thank all my family members and friends for being there for me. Sorry for

neglecting you and being unsociable in past couple of years.

As always, my brother, Hossein was my main accompanier throughout my time in London.

You are the best flat-mate that anyone can ask for! Good luck with your PhD studies too!

My highest appreciation goes to my fiancé, Ashkan, who was the most amazing listener and

supporter of my work! I cannot thank you enough for always believing in me. You patiently

listened to me whenever I was unstoppably going on and on about every single detail of my

experiments, no matter how boring they might have sounded to you! You stood by me

through the good times and bad! Thanks for being there to cheer me up. I cannot wait to start

the next chapter of our lives together.

Lastly and most importantly, my deepest gratitude belongs to my mum and dad who were my

best friends and the greatest role models throughout my life. I could not have achieved

anything without your endless love and support. You always helped me to become a better

person as I went along. There are no words to describe my appreciation to you, so simply put,

THANK YOU for everything! To this end, I am dedicating this thesis to both of you who

made it possible.

v

To my parents …

vi

Table of Contents

Abbreviations ............................................................................................................................ ix

List of Figures ........................................................................................................................... xi

List of Tables .......................................................................................................................... xix

Thesis Outline .......................................................................................................................... xx

Chapter 1: Introduction .......................................................................................................... 1

1.1 The Discovery of DNA.......................................................................................................... 2

1.2 Biophysics of DNA ............................................................................................................... 3

1.2.1 Helical Geometries and Alternate Structures of DNA ............................................................... 3

1.2.2 Thermodynamics of DNA .......................................................................................................... 5

1.2.3 Flexibility and Elasticity of DNA ............................................................................................ 10

1.2.4 Polyelectrolyte properties of DNA ........................................................................................... 13

1.3 DNA Topology .................................................................................................................... 18

1.3.1 Global Confirmations of DNA ................................................................................................. 19

1.3.2 Topological domains ................................................................................................................ 19

1.3.3 DNA Supercoiling.................................................................................................................... 20

1.3.4 Nicked DNA ............................................................................................................................ 23

1.3.5 Knots ........................................................................................................................................ 24

1.3.6 Catenanes ................................................................................................................................. 25

1.3.7 Cruciforms ............................................................................................................................... 25

1.3.8 Biological Role of DNA Topology .......................................................................................... 26

1.4 DNA-DNA Interaction ........................................................................................................ 27

1.4.1 DNA Condensation .................................................................................................................. 28

1.4.2 Multimolecular Aggregates ...................................................................................................... 29

1.4.3 Liquid Crystalline Phases ......................................................................................................... 30

1.4.4 Homologous Pairing ................................................................................................................ 31

1.5 DNA-Protein Interaction ..................................................................................................... 32

1.5.1 DNA Binding Proteins ............................................................................................................. 33

1.5.2 DNA Modifying Enzymes ....................................................................................................... 34

1.5.3 DNA Binding Antibodies ......................................................................................................... 36

1.6 Summary .............................................................................................................................. 37

1.7 References ........................................................................................................................... 38

Chapter 2: Materials and Methods ...................................................................................... 44

2.1 Molecular Biology Laboratory ............................................................................................ 45

2.1.1 Cell culture ............................................................................................................................... 45

2.1.2 Gel Electrophoresis .................................................................................................................. 45

2.1.3 Western blotting and antibodies ............................................................................................... 45

2.1.4 Quantitative real-time polymerase chain reaction (qRT-PCR) ................................................ 46

2.1.5 Methylated DNA Immunoprecipitation assay (MeDIP) .......................................................... 46

2.1.6 Long range PCR ....................................................................................................................... 47

vii

2.1.7 In-vitro methylation ................................................................................................................. 47

2.1.8 Binding assay of DNA-Antibody complex .............................................................................. 47

2.1.9 Electrophoretic mobility shift assay (EMSA) .......................................................................... 48

2.1.10 Construction of “parallel” and “antiparallel” supercoiled plasmids ......................................... 48

2.1.11 Relaxation and linearisation of supercoiled DNA .................................................................... 50

2.1.12 S1 Nuclease digestion .............................................................................................................. 50

2.1.13 Ethanol precipitation of DNA .................................................................................................. 50

2.2 Physical Chemistry Laboratory ........................................................................................... 51

2.2.1 Ultraviolet- Visible (UV-Vis) spectroscopy measurement of DNA ........................................ 51

2.2.2 Nanopore fabrication ................................................................................................................ 51

2.2.3 Silver/Silver chloride (Ag/AgCl) electrode preparation........................................................... 52

2.2.4 Nanopore membrane preparation and device assembly ........................................................... 52

2.2.5 Electrochemical measurements ................................................................................................ 52

2.2.6 DNA translocation and data acquisition ................................................................................... 53

2.2.1 Statistical analysis of translocation experiments ...................................................................... 53

2.2.2 Atomic force microscopy (AFM) ............................................................................................. 54

2.2.3 Dynamic light scattering (DLS) of plasmid DNA .................................................................... 54

2.3 References ........................................................................................................................... 56

Chapter 3: Solid-State Nanopore Based Detection of Sonicated DNA ............................. 57

3.1 Background .......................................................................................................................... 58

3.1.1 Biological nanopores................................................................................................................ 59

3.1.2 Solid-state nanopores ............................................................................................................... 60

3.1.3 Electrophoresis in Nanopores .................................................................................................. 62

3.1.4 Surface Charge Effect .............................................................................................................. 63

3.1.5 Electroosmosis in Nanopores ................................................................................................... 64

3.1.6 Entropic Effect ......................................................................................................................... 64

3.1.7 Applications of Nanopores ....................................................................................................... 66

3.2 Experimental Objectives...................................................................................................... 67

3.3 Results and Discussions....................................................................................................... 67

3.3.1 Fabrication of Single Nanopore by Focused Ion Beam Milling ............................................... 67

3.3.2 Device Platform ....................................................................................................................... 70

3.3.3 Ionic Conductance of Cylindrical Solid-State Nanopores ........................................................ 71

3.3.4 DNA Extraction, Purification and Sonication .......................................................................... 74

3.3.5 Stochastic sensing of DNA at single molecule level ................................................................ 75

3.3.6 Translocation Dynamics .......................................................................................................... 76

3.4 Conclusion ........................................................................................................................... 83

3.5 References ........................................................................................................................... 84

Chapter 4: Probing DNA Methylation in Breast Cancer Cell Lines................................. 90

4.1 Background .......................................................................................................................... 91

4.1.1 Forkhead Box A1(FOXA1)...................................................................................................... 91

4.1.2 Epigenetic Modifications ......................................................................................................... 92

Current technologies in DNA methylation analysis: ............................................................................... 95

4.2 Experimental Objectives...................................................................................................... 96

4.3 Results and Discussion ........................................................................................................ 97

viii

4.3.1 Effect of Cytosine Methylation on FOXA1 Expression........................................................... 97

4.3.2 In- vitro Methylation of FOXA1 Promoter .............................................................................. 99

4.3.3 Formation of DNA-Antibody Complex ................................................................................. 101

4.3.4 Probing DNA Methylation Using Solid-State Nanopores...................................................... 106

4.4 Conclusion ......................................................................................................................... 114

4.5 References ......................................................................................................................... 116

Chapter 5: Characterisation of Homologous Pairing in Closed Circular DNA ............ 121

5.1 Background ........................................................................................................................ 122

5.1.1 Homology Recognition in DNA Duplexes ............................................................................ 123

5.1.2 Reported Studies on Homologous DNA Segments Interaction ............................................. 123

5.2 Experimental Objectives.................................................................................................... 124

5.3 Construction of the DNA Plasmids ................................................................................... 124

5.4 Results and Discussion ...................................................................................................... 125

5.4.1 Atomic Force Microscopy (AFM) ......................................................................................... 125

5.4.2 Gel Electrophoresis ................................................................................................................ 128

5.4.3 Dynamic Light Scattering (DLS) ........................................................................................... 138

5.4.4 Nanopore Translocation ......................................................................................................... 139

5.5 Conclusion ......................................................................................................................... 144

5.6 References ......................................................................................................................... 147

Chapter 6: Conclusion ......................................................................................................... 153

Appendices ............................................................................................................................ 158

Appendix I: Sequencing Data ...................................................................................................... 159

Appendix II: Nanopore Data ........................................................................................................ 161

Appendix III: AFM Data .............................................................................................................. 162

Appendix IV: Copyright Permissions .......................................................................................... 164

ix

Abbreviations

5’-mc 5’-methylcytosine

A Adenosine

ACF Auto correlation function

AdDA Anti-dsDNA antibody

AFM Atomic force microscopy

Ag/AgCl Silver/silver chloride electrode

ap-DNA Anti-parallel DNA

apT-DNA TOPO(I) treated anti-parallel DNA

ATP Adenosine triphosphate

bp Base pair (0.34 nm)

BRCA1 Breast cancer associated gene 1

C Cytosine

cc-DNA Closed circular DNA

cDNA Complementary DNA

CE Counter electrode

CH3-DNA (In-vitro) methylated DNA

ChIP Chromatin Immunoprecipitation

CMOS complementary metal-oxide semiconductor

CpG Cytosine-(phosphate)-Guanine dinucleotide

CV Cyclic Voltammetry

DBD DNA binding domain

DBP DNA binding prtoein

DH Debye-Hückel (theory)

DLS Dynamic light scattering

DMEM Dulbecco Modified Eagle Medium

DNA Deoxyribonucleic acid

DNTM DNA methyltransferase

dNTP deoxynucleoside 5’-triphosphates

dpore Nanopore diameter

ds-DNA Double-stranded DNA

e Elementary charge

E. coli Escherichia coli

ECD Event charge deficit (C)

EDTA Ethylenediaminetetraacetic acid

EMSA Electrophoretic mobility shift assay

EO Electroosmosis

EP Electrophoresis

ER Estrogen Receptor

ERα Estrogen Receptor alpha

F Forward primer

FCS Fetal calf serum

FIB Focused ion beam

FOX Forkhead box

FOXA1 Forkhead box A1 (Gene)

FOXA1 Forkhead box A1 (protein)

G Guanine

GC Gouy-Chapman (model)

Gpore Nanopore (ionic) conductance

H Histone

H. sapiens Homo sapiens

HDAC Histone deacetylase

HMG High-mobility group (protein)

Ig Immunoglobulin

IP Immunoprecipitated

I-t Current-time (curve)

I-V Current-Voltage (curve)

Kan Kanamycin

KL Kornyshev-Leikin (theory)

L Ligand

LDNA DNA contour length

Lk Linking number

lKuhn Kuhn length

lp Persistence length

lpore Nanopore membrane length

Lys Lysine (amino acid)

MBD Methyl binding domain

MCF-7 Michigan Cancer Foundation – 7 (breast carcinoma cell-line)

MeDIP Methyl DNA Immunoprecipitation

miRNA MicroRNA

MLET-2 Endocrine resistant MCF-7clone (cell-line)

mRNA Messenger RNA

MspA Mycobacterium smegmatis porin A

x

n Number of trials

NN Nearest-neighbour (model)

NTP Nucleoside triphosphates

OmpG Outer membrane protein G

ORF Open reading frame

P Polymer

PB Poisson-Boltzmann (theory)

PBS Phosphate buffered saline

PCR Polymerase chain reaction

PDMS Poly(dimethylsiloxane)

p-DNA Parallel DNA

pI Isoelectronic point

pT-DNA TOPO(I) treated parallel DNA

PTFE Polytetrafluoroethylene

qRT-PCR Quantitative real time PCR

R Reverse primer

RE Reference electrode

Rg Radius of gyration

RIE Reactive ion etching

RNA Ribonucleic acid

SAM S-adenosyle methionine

SEM Scanning electronic microscopy

Ser Serine (amino acid)

siRNA Small interference RNA

SLE Systemic Lupus Erythematosus

SNR Signal-to-noise ratio

ss-DNA Single-stranded DNA

T Thymine

TAE (1××××) 40 mM Tris, 20mM acetic acid, and 1mM EDTA (buffer)

TBP TATA-binding protein

TE Tris HCl-EDTA

Tm DNA melting temperature (K)

TOPO Topoisomerase (enzyme)

Tyr Tyrosine (amino acid)

U Uracil

Vbias Bias potential (V)

w/v Weight per volume

WE Working electrode

WLC Worm-like chain (model)

α-HL alpha-haemolysin

β-tub Beta tubulin (protein)

∆I Current blockade (A)

ε Permittivity (dielectric constant)

κ -1 Debye length

λ-DNA Lambda DNA (48.502 kbp)

σcond. Specific conductivity

σpore Nanopore surface charge density

τd Dwell time (s)

xi

List of Figures

Chapter 1: Introduction

Figure 1.1: Double helix structure of DNA. (a) Franklin's X-ray picture of DNA. The regularity of this pattern

indicated that DNA is a helix. This image is reprinted from ref. 3 (copyright licence number: 3310781191367).

(b) Watson and Crick’s first schematic of the double helix strands of DNA. The two ribbons represent sugar-

phosphate chains. The arrows indicate the two strands are antiparallel. This image is reprinted from ref. 4

(copyright licence number: 3310780241604). ........................................................................................................ 2

Figure 1.2: Computer graphic models for (a) right-handed A-DNA, (b) right-handed B-DNA, c) left-handed Z-

DNA. The base pairs are represented with blue and sugar-phosphate backbones with red balls. This image is

reprinted from ref. 1 (copyright permission from Mc-Graw Hill requested on 16/01/2014). ................................ 4

Figure 1.3: Schematic of a DNA melting curve, measured by an increase in absorbance at 260nm. The melting

temperature (Tm) is indicated at 358 K which is determined at the point where the melting curve is half

completed. .............................................................................................................................................................. 7

Figure 1.4: Schematic of a DNA molecule confirmation in free solution at two length scales. (a) A connection

of three rigid Kuhn segments of length of 300 bp at smallest scale (b) A coiled DNA molecule with

radius of gyration at the largest scale. This figure is adapted from ref. 37. .................................................... 12

Figure 1.5: Schematic of a negatively charged spherical polyelectrolyte of radius R> upon application of

an electric field of E in an ionic solution. This figure is adopted from ref. 40. ................................................... 16

Figure 1.6: schematics of topological domains. (a) cc-DNA, (b) Linear DNA loops attached to a nuclear matrix,

(c) Linear DNA affixed to a membrane, (d) Linear DNA wrapped around proteins aggregates. This figure is

adopted from ref. 47. ............................................................................................................................................ 19

Figure 1.7: Schematic of ds-DNA supercoiling configurations: (a) Supercoiling of a relaxed cc-DNA to

plactonemic coils in prokaryotes. (b) Supercoiling of a linear DNA to solenoidal coils in eukaryotes. .............. 20

Figure 1.8: Electron micrographs cc-DNA supercoiling transitions from relaxed to tightly supercoiled plasmid

DNAs. The molecule on the left is the most relaxed configuration. The degree of supercoiling increases from

left to right. This figure is reprinted from ref. 24 (copyright permission from W. H. Freeman on 16/01/14). ..... 23

Figure 1.9: Schematic of nicking of a double stranded supercoiled DNA using a nickase enzyme. The

supercoiled cc-DNA lost the topological features and became a relaxed and open circular nicked DNA. .......... 23

Figure 1.10: Schematic of formation of a right-handed elementary knot from a double stranded cc-DNA. ....... 24

Figure 1.11: Schematic of right-handed catenation of two double stranded cc-DNA molecules. ....................... 25

Figure 1.12: Schematic of formation of cruciform from a negatively supercoiled B-form ds-DNA. The blue and

red segments represent complementary halves of an inverted repeat. The flanking DNA is shown in black. ..... 25

Figure 1.13: Phase transition in DNA by increasing DNA concentration. This figure is reprinted from ref. 71

(copyright licence number 3310820233895). ....................................................................................................... 30

xii

Figure 1.14: Schematic of electrostatic homology recognition in DNA duplexes. A zipper-alignment of

phosphate strands with positively charged counterions in the grooves of the opposing molecule. s is the

curvilinear coordinate along each DNA molecule. On the right, the sequence-dependent variation in the twist

Ω(s) and axial rise h(s) per bp is depicted. This figure is reprinted from ref. 74 (copyright licence number:

3310820770863). .................................................................................................................................................. 32

Chapter 2: Materials and Methods

Figure 2.1: Vector map of pET-24a (+) plasmid. This figure is reprinted from ref. 2. ........................................ 49

Chapter 3: Solid-State Nanopore Based Detection of Sonicated DNA

Figure 3.1: (a) The heptameric α-hemolysin (α-HL), The cross-sectional view on the right displays the inner

cavity (green), inner constriction (red), and β-barrel (blue). This image is reprinted from ref. 5 (copyright

licence number: 331082116402). (b) Schematic of translocation of ss-DNA through α-HL. This figure is

reprinted from ref. 15 (copyright licence number: 3310830225419)................................................................... 60

Figure 3.2: Illustration of a solid-state nanopore device. (a) Schematic of threading and translocation of a single

DNA molecule through a solid-state nanopore in KCl solution. (b) Scanning electron microscopy (SEM) image

of a 40 nm nanopore fabricated on a Si3N4 membrane. (c) Schematic of current- time trace, before and after

addition of DNA during translocation process. .................................................................................................... 61

Figure 3.3: Schematic of current-time traces in three systems (a) No DNA is added to KCl solution. A steady-

state ionic current (open pore current) upon the application of Vbias is generated due to flux of K+ and Cl- ions

(not in scale) across the pore (b) Translocation of DNA in Ogston regime: when Rg < dpore, there is a very

small entropic effect and no stretching of DNA is required during the translocation through the pore, hence a

very fast current blockade events are resulted. (c) Translocation of DNA in entropic trapping regime: when Rg ≥

dpore, a very large entopic effect is associated with translocation of DNA through the pore; linear DNA has to

stretch to travel across the pore, hence, the resulting blockade events are slower. ............................................... 65

Figure 3.4: Schematic of Si3N4 membrane fabrication: Deposition of a freestanding membrane, followed by

photolithography and RIE. Then a KOH wet etching was applied to create a 50 µm × 50 µm Si3N4 window.

Lastly, FIB milling can subjected to fabricate a nanopore. This scheme is adopted from ref. 52. ....................... 68

Figure 3.5: SEM image of a Si3N4 membrane before fabrication of a nanopore by FIB milling. (a) top view-

472 µm × 472 µm Si3N4 membrane window patterned by semiconductor lithography (b) bottom view- 50 µm ×

50 µm membrane opened by RIE technique. ........................................................................................................ 68

Figure 3.6: Carl Zeiss XB1540 FIB/SEM instrument for nanopore milling. ....................................................... 69

Figure 3.7: SEM image of a fabricated nanopore on 50 µm × 50 µm Si3N4 membrane. (a) ~ 88 nm pore (Mag =

83.89 kX). (b) after 10s exposure to SEM, the pore is shrunk to ~67 nm (Mag = 102.21 kX). (c) followed by

another 10s SEM exposure, pore shrunk to ~34 nm (Mag = 86.54 kX). ............................................................. 70

xiii

Figure 3.8: Schematic of device platform: Two 1 ml chambers, two Ag/AgCl electrodes, two PDMS (1cm outer

and 0.35 cm inner diameters) and a Si3N4 nanopore chip (blue). This Figure is adopted from ref. 55 ................ 70

Figure 3.9: Ionic conductance measurements of two single nanopores fabricated on Si3N4 membranes

(thickness L = ~100nm). Cyclic Voltammetry performed at 50 mV/s scan rate in 1 M KCl, with ~1 cm2

Ag/AgCl electrodes. Bias of -500 to 500 mV is applied. The pore conductance can be determined by the IV

curve slope Red: G = 874 nS, dpore= ~ 89 nm. Black: G=34 nS, dpore= ~18 nm. .................................................. 72

Figure 3.10: Agarose gel electrophoresis, (1% agarose, 5V/cm, 1hr). Lane 1: 1 kbp DNA Ladder (New England

Biolabs), 0.5, 1, 2, 3 kbp bands are indicated. Lane 2: MCF-7 sonicated DNA-500-3000 bp. ........................... 75

Figure 3.11: Current-time (I-t) curve of a ~18 nm pore with Vbias of 200 mV, 1 M KCl-Tris HCl (pH 8.5) during

translocation of sonicated DNA. (a) before (control) and (b) after addition of sonicated MCF-7 DNA (800 pM)

(c) Magnified image of the indicated translocation events, which shows the pattern and shape of 4 individual

blocked events. ..................................................................................................................................................... 76

Figure 3.12: Schematic of a translocation process, where td is translocation (dwell) time, Io (pA) is open pore

current, Ib (pA) is the blocked pore current, ∆I (pA) is the current blockade amplitude and ECD (fC) is the

integrated event area. ............................................................................................................................................ 76

Figure 3.13: Histogram analysis of (a) τd and (b) ∆I (c) cluster plot (∆I vs. τd) of translocation of sonicated

MCF-7 DNA through a ~18 nm pore, in 1M KCl-10mM Tris-HCl pH 8.5, at 200 mV applied potential and

room temperature. The (stretched) Gaussian fits are indicated with red curves in graph (a) and (b). .................. 77

Figure 3.14: Schematic of translocation of (a) linear, (b) folded, (c) semi-folded ds-DNA through the pore and

its effect on current-time trace. This scheme is adopted from ref. 62. .................................................................. 78

Figure 3.15: Histogram analysis of ECD upon translocation of sonicated MCF-7 DNA through ~18 nm pore, at

1M KCl-10mM Tris-HCl pH 8.5, 200mV applied potential and room temperature. The (stretched) Gaussian fit

is indicated with red curve. ................................................................................................................................... 80

Figure 3.16: Semi log plot of the effect of applied potential (Vbias) on frequency of events per second, upon

translocation of sonicated MCF-7 DNA through ~18 nm pore, at 1M KCl-10mM Tris-HCl (pH 8.5) and room

temperature. The linear fit is indicated with dashed-red lines. ............................................................................. 81

Chapter 4: Probing DNA Methylation in Breast Cancer Cell Lines

Figure 4.1: Schematic illustration of the effect of a promoter’s CpG islands hypermethylation in gene

expression. (a) Essential level of cytosine methylation on the promoter segment is required for functionality and

expression of the gene (b) Hyper-cytosine-methylation of promoter results in repression of the gene. ............... 94

Figure 4.2: Hypothetical illustration of a current-time trace upon an electrokinetically driven of (translocation)

of (a) methylated (CH3) DNA with 3 methylated CpG regions and (b) CH3-DNA-5’-mc antibody-complex

through a Si3N4 nanopore. The assigned sub-peaks represent the sites where an antibody is bound to methylated

CpGs. .................................................................................................................................................................... 96

xiv

Figure 4.3: Schematic of a MeDIP assay key stages: Following the cell lysis and DNA extraction, the genomic

DNA is sonicated to 100-600 bp fragments and then denatured at 95 ˚C to generate ss-DNA. Subsequently the

ss-DNA is incubated with the 5'-mc antibody which is already bound to specific magnetic beads. The enriched

DNA is precipitated and isolated by a magnet. At the end, the DNA is purified and prepared for PCR analysis.97

Figure 4.4: (a) MeDIP assay of FOXA1 in MCF-7 and MLET-2 cells, followed by ethidium bromide detection

on 1% agarose gel, GAPDH is a negative control gene. (b) qRT-PCR of mRNA of FOAX1 gene in MCF-7 and

MLET-2 cells. The Y-axis values are arbitrary and normalised to the L19 housekeeping gene. (c) Western-blot

analysis of FOXA1 protein in MCF-7 and MLET-2 cells. β-tubulin is a positive control protein. ...................... 98

Figure 4.5: Scheme of 5’ cytosine (in-vitro) methylation reaction using SAM as a methyl donor and M.SssI

enzyme as a catalyst. ............................................................................................................................................ 99

Figure 4.6: Restriction site of methyl sensitive HpaII enzyme. ........................................................................ 100

Figure 4.7: 1% agarose gel electrophoresis of HpaII digested in-vitro methylatated FOXA1 promoter: Lane1:1

kbp DNA ladder (New England BioLabs), 1 kbp and 3 kbp fragments are indicated. Lane 2: unmethylated

FOXA1 promoter (long range PCR product). Lane 3: unmethylated FOXA1 promoter + HpaII. Lane 4:

methylated FOXA1 promoter Lane 5: methylated FOXA1 promoter + HpaII. ................................................... 100

Figure 4.8: Electrophoretic mobility shift assay (EMSA). (a) The schematic of EMSA with CH3-DNA

fragments and 5’mc antibodies. The first (left) lane: the electrophoresis of CH3-DNA (black) without antibody.

The second (right) lane: the electrophoresis of the CH3 after incubation with 5’mc antibody. The unbound DNA

fragments (grey) migrates at the same speed as the first lane and CH3 DNA-antibody complexes (blue) exhibit

lower mobility. Here 5 configurations of bindings are shown. (b) 0.4% agarose gel electrophoresis at 2V/cm for

4-5 hr on ice. Post-stained with 3x GelRed for 30 min. All samples incubated in 100mM KCl-Tris (pH 8.5), for

2 hr at 37 ˚C. Lane 1: unmethylated FOXA1 promoter (3.4 kbp; long range PCR product). Lane 2: mixture of

unmethylated FOXA1 promoter +5’mc antibody. Lane 3: 5’mc antibody in (negative control). Lane 4: CH3-

FOXA1 promoter (3.4 kbp). Lane 5: mixture of CH3-FOXA1 promoter + 5’mc antibody. Lower band represents

the fraction of DNA that is not bound to antibody. The upper band (indicated with an arrow) represents the

fraction of DNA that formed a complex with the antibody. ............................................................................... 102

Figure 4.9: AFM topography images (flatten-3-order) of (a) CH3 FOXA1 promoter, (b) 5’-mc antibody

(appeared as small dots) (c) unmethylated FOXA1 promoter + 5’-mc antibody, (d)-(f) CH3-FOXA1 promoter +

5’-mc antibody (white arrows indicate the sites where an antibody is bound). The corresponding height (z) scale

bar is shown underneath of each image. ............................................................................................................. 103

Figure 4.10: AFM analysis: (a) Histogram analysis of the antibody height (z direction; n = 87), (b) Histogram

analysis of the antibody (5’-mc) diameter (x-y direction; n = 151) (c) Histogram analysis of DNA (CH3-FOXA1

promoter) contour length (n = 55), (d) Column bar of the number of bound antibodies per DNA molecule (n =

46). Histograms in (a-c) are fitted with Gaussian distribution indicated with blue curves. ................................ 104

Figure 4.11: The conductance measurement (IV curve) of a ~ 40 nm pore fabricated on a Si3N4 membrane

(Lpore= ~60 nm) at two KCl concentration of 1 M (black) and 0.1 M (red). (Inset) The SEM image of the same

nanopore that was used in translocation experiments. The I-V measurement was performed at 50 mV/s scan rate

xv

and bias of -500 -500 mV. The IV curve slope yields Gpore ≈ 338 nS (black; 1M KCl) and Gpore ≈ 47 nS (red; 0.1

M KCl)................................................................................................................................................................ 107

Figure 4.12: Detection of CH3-DNA, CH3-DNA-Antibody complex and 5’-mc antibody using a solid-state

nanopore. The figure displays the representative ionic current traces and the typical individual translocation

events observed during translocation of each analyte. I-t traces were recorded at 0.1 M KCl-Tris-HCl (pH 8.5),

at room temperature, sampled at 200 kHz and low-pass (Bessel) filtered at 10 kHz. (a) CH3-DNA molecules

were detected at 500 mV and current blockades observed. (b) CH3-DNA-Antibody complex molecules were

detected at -500 mV and current blockades observed. No translocation events were detected at 500 mV. (c)

Translocation of 5’-mc antibody molecules was only observed at -1000 mV with current enhancement

characteristics. No translocation events were detected at lower Vbias, including ± 500 mV. ............................... 108

Figure 4.13: The histogram analysis of 5’-mc antibody translocations with a ~40 nm pore, at -1000 mV in 100

mM KCl-TrisHCl (pH 8.5). (a) Dwell time (τd) and (b) current-enhancement (∆I) distributions (n = 1001). The

(stretched) Gaussian fits are indicated with navy curves. ................................................................................... 110

Figure 4.14: Frequency of events (s.M)-1 analysis of 1000 events of each individual analyte. Translocation of

CH3-DNA performed at 500 mV, 5’-mc antibody at -1000 mV and the CH3-DNA-Antibody complex at -500

mV in 100 mM KCl-TrisHCl (pH 8.5). .............................................................................................................. 110

Figure 4.15: The nanopore translocation data (n = 1001) of CH3 DNA and the complex in 100 mM KCl-

TrisHCl (pH 8.5). Event number density plots (2-D histogram of ∆I vs. τd ) of (a) CH3 DNA at 500 mV and (b)

the complex at -500 mV. 2-D histograms are normalised to 1 and the point densities are colour coded from blue

(low) to red (high). Comparison of (c) τd and (d) ∆I histograms of CH3-DNA (black) and the complex (blue).111

Figure 4.16: Hypothetical illustration of electrophoretic (EP) and electreoosmotic (EO) effects in nanopore

translocations in 100 mM KCl (pH 8.5). The Si3N4 pore walls are negatively charged. (a) EP governed

translocation of CH3-DNA at 500 mV. (b) EO governed translocation of CH3-DNA –Antibody complex at -500

mV. (c) EO governed translocation of 5’-mc antibody at -1000 mV. The nanopore and analyte sizes, as well as

the magnitude of the electokinetic forces are not to scale. ................................................................................. 113

Chapter 5: Characterisation of Homologous Pairing in Closed Circular DNA

Figure 5.1: Schematic of the preparation of the parallel and anti-parallel DNA plasmids from the native pET-

24-a(+) plasmid. In the presence of homology recognition, the "parallel" sample is expected to have shape and

physical properties, compared to the control. ..................................................................................................... 124

Figure 5.2: AFM imaging data for relaxed (TOPO treated) plasmids in air. (a) apT-DNA on Mg2+-modified

mica (image size: 2.5 µm by 2.5 µm), (b) pT-DNA on Mg2+-modified mica (image size: 5 µm by 5 µm), (c)

apT-DNA on APTES +-modified mica (image size: 1.5 µm by 1.5 µm), (d) pT-DNA on APTES +-modified mica

(image size: 1.0 µm by 1.0 µm). ......................................................................................................................... 126

Figure 5.3: Histogram analysis of the contour length (LDNA) of apT-DNA (blue; n=62) and pT-DNA (red; n=41)

using ImageJ software. The Gaussian fits are indicated with black lines. .......................................................... 127

xvi

Figure 5.4: Hypothetical schematic of ap- and p- DNA structures after cloning and amplification. According to

AFM analysis, ap-DNA remained as a 6.3 kbp circular DNA, while p-DNA dimerised after cloning and

amplification and formed a 12.6 kbp single loop circular DNA. ........................................................................ 128

Figure 5.5: 0.8% agarose gel electrophoresis (1× TAE, 5 V/cm, 1 hr) of ap- and p-DNA in supercoiled,

linearised (EcoRI) and relaxed (TOPO treated) forms. Lanes 1-7: 1) ap-DNA, 2) p-DNA, 3) linear ap-DNA, 4)

linear p-DNA, 5) apT-DNA, 6) pT-DNA and 7) 1kb DNA ladder (New Englan BioLabs; 3 kb band is indicated).

The majority species of circular ap-DNA always moved faster than p-DNA. As expected, the most relaxed

topomer before topoisomerase treatment moved at the same speed as majority species after topoisomerase

treatment. ............................................................................................................................................................ 129

Figure 5.6: A typical image of supercoiled, linear and relaxed ap- and p-DNA samples electrophoresis at

various agarose gel percentages. (a) 0.5% (w/v), (b) 0.8% (w/v), (c) 1.0% (w/v), (d) 2.0% (w/v), (e) 3.0% (w/v).

The lanes1-7 are the same in each gel: 1) 1 kb DNA ladder, 2) ap-DNA (supercoiled) 3) p-DNA (supercoiled) 4)

linear ap-DNA (see above) 5) linear p-DNA 6) apT-DNA 7) pT-DNA. Electrophoresis was conducted in 1×

TAE buffer at 23 ˚C with an applied field of 5 V/cm for 1.5 hr, except (b) 2% gel which carried out for 2.5 hr.

Gels were post stained with 3× GelRed DNA stain for 30 min before taking the images. ................................ 130

Figure 5.7: Ferguson plot of DNA mobility as a function of gel percentage (gel pore size). As the gel

percentage decreases, the mobility increases; extrapolation towards 0% yielded the mobility values for the gel-

free case. The slope determines the retardation coefficient. Linearised DNA (black starsc), ap- DNA (red, open

circles); p-DNA (blue squares); apT- DNA (purple triangles) and pT-DNA (green crosses). ........................... 130

Figure 5.8: S1 digestion.0.8% agarose gel electrophoresis (1× TAE, 5 V/m, 1 hr). Lanes 1-11: 1 kb DNA

ladder (0.5 kb band is indicated); 2) ss-DNA M13mp18 (control); 3) ssDNA M13mp18 + S1 endonuclease; 4)

ap-DNA; 5) ap-DNA + S1; 6) p-DNA; 7) p-DNA + S1; 8) apT-DNA; 9) apT-DNA + S1; 10) pT-DNA; 11) pT-

DNA + S1. .......................................................................................................................................................... 132

Figure 5.9: Effect of metal chlorides on mobility. 0.8% agarose gel electrophoresis (1× TAE, 5 V/m, 1 hr) of

ap- and p-DNA in presence of MgCl2 and CaCl2 (inset). Lanes 1,8 and 15 are the 1 kb DNA ladders (the 1 kb

bands are indicated). Lanes 2-7 are the negative control where no MgCl2 was added during incubation, same as

Figure 5.5: 2) ap-DNA, 3) p-DNA, 4) linear ap-DNA, 5) linear p-DNA, 6) apT-DNA, 7) pT-DNA. Samples in

lanes 9-14 were incubated with 40 mM MgCl2 overnight at 37˚C (lanes13+14 were relaxed by TOPO in

presence of this metal chloride): 9) ap-DNA+ Mg2+, 10) p-DNA+ Mg2, 11) linear ap-DNA + Mg2+, 12) linear p-

DNA + Mg2+, 13) ap-T DNA + Mg2+, 14) pT-DNA + Mg2+. The inset represents a typical 0.8% gel of

supercoiled and relaxed samples in presence of 40 mM CaCl2 in the same condition as above. Lanes i-iv: i) ap-

DNA + Ca2+, ii) p-DNA + Ca2+, iii) apT-DNA + Ca2+, iv) pT-DNA + Ca2+. apT + ion2+ topoisomers (smears) are

indicated by yellow (dashed) ellipses. ................................................................................................................ 134

Figure 5.10: Increasing ionic strength of the running buffer and the gel matrix during electrophoresis. 0.8%

agarose gel electrophoresis (2.5 V/m, 1.5 hr) in (a) 1× TAE (pH 8.3), (b) 1× TAE + 0.1M KCl (pH 8.3), (c) 1×

TAE + 0.5M KCl (pH 8.3). Lanes 1-7 are the same in all gels: 1) ap-DNA, 2) p-DNA, 3) linear ap-DNA, 4)

linear p-DNA, 5) apT- DNA, 6) pT-DNA, 7) 1 kp DNA ladder (3 kb band is indicated). ................................. 135

Figure 5.11: Increasing ionic strength of the DNA samples buffers. 0.8% agarose gel electrophoresis (1× TAE,

5 V/m, 1 hr). DNAs were initially eluted in 10 mM Tris-HCl (pH 8.5) before 30 min incubation with KCl. (a)

xvii

Increasing KCl concentration of supercoiled DNA buffers. Linearised ap and p-DNAs are used as the controls

(reference bands). (b) Increasing KCl concentration of relaxed (TOPO treated) DNA buffers. Supercoiled ap

and p-DNAs are used as the controls (reference bands). Lane 1-15 in (a): 1) 1 kb DNA ladder ( 2 kb band

indicated), 2) ap-DNA with no KCl, 3) ap-DNA + 38 mM KCl, 4) ap-DNA + 75 mM KCl, 5) ap-DNA + 113

mM KCl, 6) ap-DNA + 150 mM KCl, 7) ap-DNA + 300 mM KCl, 8) p-DNA with no KCl, 9) p-DNA + 38 mM

KCl, 10) p-DNA + 75 mM KCl, 11) p-DNA + 113 mM KCl, 12) p-DNA + 150 mM KCl, 13) p-DNA + 300

mM KCl, 14) linear ap-DNA (no KCl), 15) linear p-DNA (no KCl). Lanes 16-30 in (b): 16) 1 kb DNA ladder (

2 kb band indicated), 17) supercoiled ap-DNA (no KCl), 18) apT-DNA with no KCl, 19) apT-DNA + 38 mM

KCl, 20) apT-DNA + 75 mM KCl, 21) apT-DNA + 113 mM KCl, 22) apT-DNA + 150 mM KCl, 23) apT-

DNA + 300 mM KCl, 24) supercoiled p-DNA (no KCl), 25) pT-DNA with no KCl, 26) pT-DNA + 38 mM KCl,

27) pT-DNA + 75 mM KCl, 28) pT-DNA + 113 mM KCl, 29) pT-DNA + 150 mM KCl, 30) apT-DNA + 300

mM KCl. ............................................................................................................................................................. 137

Figure 5.12: DLS study of ap- and p-DNA ionic strength dependence (n = 3). Ratio of translational diffusion

coefficients of p- and ap-DNA as a function of ionic strength on a semi-log graph. Open squares: supercoiled

DNA, filled triangles: relaxed (TOPO treated) DNAs the error bars denote three independent measurements with

three repeats each. (Source: courtesy of W. Pitchford). ..................................................................................... 138

Figure 5.13. Effect of 1 M KCl on supercoiled plasmids (buffer) during incubation for 2 hr, 0.8% agarose gel

electrophoresis (1× TAE, 5 V/m, 1 hr). All DNAs were initially eluted in 10 mM Tris-HCl (pH 8.5) before

incubation. Lanes 1-5: 1) DNA ladder (3 kb band is indicated), 2) ap-DNA+1 M KCl , 3) p-DNA + 1 M KCl, 4)

ap-DNA with no KCl 5) p-DNA with no KCl. ................................................................................................... 139

Figure 5.14: Nanopore translocation data at 1 M KCl-Tris-HCl (pH 8.5) (a) Schematic of the nanopore setup

(cross-sectional view). (b) Ion current/voltage trace for the pore used (conductance = 305.4 nS, solution

conductivity σs = 10.98 Ω-1m-1; pore channel length Lpore = 70 nm; estimated pore diameter dpore = 44 nm

assuming cylindrical geometry); inset: SEM image of the pore utilised. (c) Examples of DNA translocation

events for p- and ap-DNA at 150 mV bias. ΔI vs. τd event number density plots of (d) ap-DNA and (e) p-DNA

at Vbias of (i) 150 mV, n = 748 (ii) 200 mV, n = 1254 and (iii) 300 mV, n = 995. The histograms are normalised

to 1, colour code in panel e.iii. ............................................................................................................................ 141

Figure 5.15: A semi-logarithmic plot of the most probable translocation time τd max for ap-DNA (open

squares) and p-DNA (closed circles) vs. applied bias voltage (Vbias). The error bars were estimated from the

fitting procedures. The linear fit for ap-DNA is indicated with a dashed line. ................................................... 143

Appendices

Figure App. 1: (a) Bloackage current (∆I) and (b) dwellt time (d ) histogram anaalysis of p- and ap-DNA

translocation through a ~ 44 nm pore at voltage bias of (i) 150 mV, (ii) 200 mV and (iii) 300 mV. The ap-DNA

is blue column bars and p-DNA is black columns bars. The histograms are fitted with (skewed) Guaasian

distrubutions and colour coded with blue (ap-DNA) and black. p-(DNA) curves. The peak value of each curve

is presneted as the mot probale value in Table 5.1, section 5.4.3. ...................................................................... 161

xviii

Figure App. 2: AFM data in air on Mg2+ modified mica. (a) Supercoiled ap-DNA, 2.5 µm × 2.5 µm scan (b)

supercoiled p-DNA, 2.5 µm × 2.5 µm scan, (c) relaxed apT-DNA, 5.0 µm × 5.0 µm scan, (d) relaxed pT-DNA,

10 µm × 10 µm scan. .......................................................................................................................................... 162

Figure App. 3: AFM data in air on silinised (APTES) modified mica. (a) Supercoiled ap-DNA, 1.0 µm × 1.0

µm scan (b) supercoiled p-DNA, 2.65 µm × 2.65 µm scan, (c) relaxed apT-DNA, 1.5 µm × 1.5 µm scan, (d)

relaxed pT-DNA, 1.0 µm × 1.0 µm scan. ........................................................................................................... 163

xix

List of Tables

Chapter 1: Introduction

Table 1.1: Geometry parameters of A, B and Z-DNA.1, 19-21. ................................................................................. 5

Table 1.2: Nearest-neighbour thermodynamic parameters for 10 different Watson-Crick pairwise interactions at

1 M NaCl , 37˚C, pH 7.0 (1 cal = 4.184 J.). The slash (/) indicates the sequence at complementary strand

(antiparallel orientation). The values are taken from ref. 29................................................................................... 9

Chapter 4: Probing DNA Methylation in Breast Cancer Cell Lines

Table 4.1: Comparison of the main methodologies and principles in DNA methylation analysis. 40-42 ............... 95

Table 4.2: Summary of blockade events parameters from the histogram analysis (n = 1001). The errors denote

the standard deviation resulting from the fitting procedure. ............................................................................... 112

Chapter 5: Characterisation of Homologous Pairing in Closed Circular DNA

Table 5.1 Summary of nanopore data at three applied potential. The most probable (max) ΔI and τd values

obtained from the dwell time-histograms fitted with the (skewed) Gaussian distribution. (see Appendix II ). The

error associated with each data point, denotes the standard deviation resulted from the fitting procedure. ....... 142

Appendices

Table App. 1: Summary of permissions for third party copyright works. ......................................................... 164

xx

Thesis Outline

This thesis is divided into 6 chapters which represent the main works carried out during this

PhD study. The main research focussed on the biological, chemical and physical properties of

DNA at single molecule level, in prokaryotes and eukaryotes. In particular, it attempted to

address two current challenges in the field, including:

i) Ultrafast sensing of DNA methylation with implications in cancer diagnosis.

ii) Investigation of the homology recognition in closed circular DNA.

In this project, solid-state nanopore sensors were utilised as novel biosensors and an

alternative method to study the above modifications and features at the nanoscale.

Chapter 1 reviews the key biophysical properties of a DNA molecule, as well as its

biological function within the cell machinery.

The materials and methods used throughout the experiments presented in the thesis are

covered in Chapter 2.

A detailed operational set-up and the fabrication process of nanopore sensors are discussed in

Chapter 3. Furthermore, efficacy and efficiency of the nanopore chips were tested and

characterised using translocation of sonicated genomic DNA.

Chapter 4 outlines the significance of DNA methylation sensing in breast cancer cell lines,

exemplified by a study of methylation levels in the breast cancer-associated FOXA1 gene. It

was shown that by employing the properties of DNA-protein (antibody) interactions, the

detection of methylated regions of DNA with solid-state nanopores can be enhanced.

Chapter 5 presents the experimental studies conducted in an attempt to evaluate the effect of

electrostatic forces in facilitating the homologous pairing during the recombination process in

a protein free solution. However, it was found that homologous DNA was dimerised and

then formed a single loop DNA, resulting in the modulation of physical and structural

properties of the homologous DNA.

The last chapter, Chapter 6, concludes this thesis and underlines once more the key impacts

and discoveries of this project. Moreover, possible future experimental pursuits are outlined.

1

Chapter 1

Introduction 1.1 The Discovery of DNA ............................................................................................................................. 2

1.2 Biophysics of DNA ................................................................................................................................... 3

1.3 DNA Topology ....................................................................................................................................... 18

1.4 DNA-DNA Interaction ............................................................................................................................ 27

1.5 DNA-Protein Interaction ......................................................................................................................... 32

1.6 Summary ................................................................................................................................................. 37

1.7 References ............................................................................................................................................... 38

Synopsis: Deoxyribonucleic acid (DNA) is the carrier of the genetic information in nearly all living

organisms. This chapter aims to briefly highlight the key concepts and properties of this biomolecule which are

directly relevant to these experimental studies, as well as providing a general introduction to DNA biophysics

and its significance in mechanisms of cellular machinery.

Chapter 1 Introduction

2

1.1 The Discovery of DNA

In 1869, Fredrich Miescher discovered that the cell nucleus consists of a compound called

nuclein and the major component of nuclein is DNA. By the end of nineteenth century, it

was established that the building blocks of this long chain polymer are nucleotides and were

composed of a sugar, a phosphate group and a base.1 In 1953, based on the X-ray images

taken by Franklin and Gosling, Figure 1.1.a, James Watson and Francis Crick rejected the

triple helix model of Pauling and Corey2 and proposed that DNA is a double helix: two

strands wound around each other in such a way that the sugar-phosphate backbones are on

the outside of the ladder and the bases of each strand are on the inside of the helix (see Figure

1.1.b).3

Figure 1.1: Double helix structure of DNA. (a) Franklin's X-ray picture of DNA. The regularity of this pattern

indicated that DNA is a helix. This image is reprinted from ref. 3 (copyright licence number: 3310781191367).

(b) Watson and Crick’s first schematic of the double helix strands of DNA. The two ribbons represent sugar-

phosphate chains. The arrows indicate the two strands are antiparallel. This image is reprinted from ref.4

(copyright licence number: 3310780241604).

DNA, the carrier of genetic information, consists of four different bases: adenine (A),

thymine (T), cytosine (C) and guanine (G).

Earlier in 1952, Erwin Chargaff revealed the base composition of DNA from various sources

have roughly equal amounts of purines and pyrimidines. Further studies showed A and T

present equally in DNA, as were the amounts of G and C.5 These findings known as

Chargaff’s rules, provided crucial information for Watson and Crick’s model, where in their


3

classic paper in Nature, they outlined that uniformity of the double helix can be only kept if

two strands are made of complementary base pair sequences: a purine in one stand is always

paired with pyrimidine in the other. Based on their postulation, A and C from one strand

form a double and triple hydrogen bond, respectively, with T and G of the other strand.4

Franklin’s X-ray suggested that the spacing between base pairs (bp) is 3.32 Å and the overall

helix repeat distance is about 33.2 Å, i.e. 10 bp per turn of helix.3

Moreover, Watson and Crick proposed that, as a result of complementary strands, DNA

undergoes a semiconservative replication. This mechanism ensures that the two daughter

DNA duplexes will be exactly the same. In 1958, Matthew Meselson and Franklin Stah

confirmed this theory by autoradiography visualisation of old and new strands within

replicated chromosome of Escherichia Coli (E.coli) Bacteria.6

1.2 Biophysics of DNA

Different studies on naturally occurring DNA sequences and synthetic polynucleotides have

shown that the DNA molecules could have structural polymorphism which plays an

important role on its biological function. It is established that the global conformation and

structure of DNA molecules can be adapted to its environment by twisting, turning and

stretching.7 DNA conformation can be dependent on its base-composition, chemical

modification, direction and degree of supercoiling, hydration level and presence of

counterions and polyamines in solution.8-10

1.2.1 Helical Geometries and Alternate Structures of DNA

In spite of the great diversity of living organisms, majority of DNA molecules follow the

complementary principle of Watson and Crick’s model. However, their proposed structure

represents the sodium salt of DNA in a fibre produced at very high relative humidity (92%).

This is called the B-form of DNA (Figure 1.2.b). Under physiological conditions (200mM

NaCl, pH 7.4 and 37˚C), this is the most probable form within the cell. In B-DNA the helix is

right-handed. X-ray crystallography of DNA showed that the surface of this double helix is

indeed not cylindrical and consists of two grooves: a) major grove- 22 Å wide and b) minor

groove-12Å wide;11 The larger groove width of the major groove implies greater accessibly

of edge bases in this group, hence provides specific sites for DNA binding proteins (see

section 1.5.1).12 In B-DNA, the two strands run in opposite directions and base pairs are


4

planar and perpendicular to the axis of double helix.4 The crystal structure of B-DNA

indicated that the B helix is packed in 10 bp per turn. However, later studies by J.C. Wang

(1979) using gel electrophoresis analysis, demonstrated that the helical repeat of B-DNA in

solution is 10.5 units per turn.13

If the relative humidity of surrounding DNA fibre is reduced to 75%, the resulting structure

of the sodium salt of DNA called the A-form (Figure 1.2.a).14 Similarly to B-DNA, the two

complementary strands are antiparallel and form a right handed helix. However, the A-form

differs from the B-DNA in several respects. In A-DNA, there are 10.7 bp per helical turn

instead of the 10.5 found in B–DNA crystal structure. The bases are planar but their plane is

no longer perpendicular to helix axis and tilts 20 degrees away from the horizontal plane.

Hence, the base pairs shift from the centre of duplex, forming an empty channel in the

middle. Each turn of A-DNA occurs in 24.6 instead of 33.2 Å as in the B-form.1, 15

Figure 1.2: Computer graphic models for (a) right-handed A-DNA, (b) right-handed B-DNA, c) left-handed Z-

DNA. The base pairs are represented with blue and sugar-phosphate backbones with red balls. This image is

reprinted from ref.1 (copyright permission from Mc-Graw Hill requested on 16/01/2014).

In 1979, Alexander Rich and co-workers discovered the most striking example of DNA that

can be deviated from the B-form. They presented a left-handed antiparallel duplex called Z-

form (Figure 1.2.c) with alternating purines (A or G) and pyrimidines (T or C) (e.g., poly

[dG-dC] ∙ poly [dG-dC]).16 The Z-DNA has six dinucleotides per turn and its DNA backbone

exhibits “zig-zag” characteristics, hence the “Z” form name. Formation of Z-DNA is


5

generally unfavourable under physiological conditions and requires alternating purine–

pyrimidine sequences, negative supercoiling (see section 1.3.3) or high salt concentration.1, 7,

15 The biological significance of Z-DNA is still under debate. Interestingly, in 2001, Keji

Zhao and colleagues discovered that activation of a gene (CSF1) requires a regulatory

sequence switch to Z-DNA form.17 They showed that one of the primary functions of Z-DNA

formation is to facilitate transcriptional initiation. Thus, Z-forms are mainly found at

promoter regions ofgenes.18

The helical parameters and geometry of the three described forms are summarised in Table

1.1. It should be noted that DNA conformations are not limited to the above forms. Other

structures such as B’, C, E, G, H, L, M, N, O, P, R, S, T, W, and X-DNA have also been

discovered, although most of these forms are generated synthetically and do not occur

naturally in biological cells.7

Table 1.1: Geometry parameters of A, B and Z-DNA.1, 19-21.

Geometry attribute A-DNA B-DNA Z-DNA

Helix sense right-handed right-handed left-handed

Repeating unit 1 bp 1 bp 2 bp

Helical repeat (bp/turn) 10.7 10.5 12

Inclination of bp from horizontal +19° −1.2° −9°

Rise/bp along axis 2.3 Å 3.32 Å 3.8 Å

Pitch of helix 24.6 Å 33.2 Å 45.6 Å

Diameter 23 Å 20 Å 18 Å

1.2.2 Thermodynamics of DNA

Duplex formation of nucleic acids is a fundamental process during replication of parental

DNA, recognition of codon versus anticodon in translation, as well as various laboratory

procedures such as hybridisation of probes in sequencing and Southern blotting, cDNA

expression profiling and most importantly polymerase chain reaction (PCR). Therefore, an

understanding of the thermostability and energetics of DNA pairing is vital in order to gain a

better insight on the molecular details of many biological processes in the cell.22

The structure of the DNA double helix is stabilised by hydrogen bonds and base-stacking

interactions. As mentioned above, three hydrogen bonds are formed between bases C and G


6

and two between A and T. Thus, it is speculated that the stability of double stranded (ds)-

DNA relative to single stranded (ss)-DNA is dependent on the percentage of C-G pairs, as it

has the largest influence on the average number of hydrogen bonds within the double helix.

Furthermore, structural analysis revealed that the van der Waals interactions between

adjacent bases of the same strand also contribute to the overall stability of ds-DNA.23

A. DNA melting:

When a solution of ds-DNA is sufficiently heated (above melting temperature), the non-

covalent forces that hold the two complementary strands weaken and eventually the two

strands of DNA molecules unwind and dissociate into single strands. This process is known

as DNA melting or DNA denaturation. In addition to heat, DNA melting can be promoted by

lowering of the ionic strength and increasing of the pH of the solution as well as addition of

dimethyl sulfoxide and formamide which disrupt the hydrogen bonding of duplex DNA.1

Denaturation of DNA occurs over a narrow temperature range. The temperature at which

half of the DNA strands are in a denatured and random coil state is called the melting

temperature (). The amount of separated (denatured) strands in solution can be determined

by ultra-violet-visible (UV-Vis) spectroscopy at a wavelength of 260 nm. DNA is the best

known example of the Hypochromicity effect due to the close proximity of bases. When a

DNA molecule is in a duplex-state, a low UV absorbance results due to the base pair

interactions. When these interactions are removed and the DNA is in a single-strand-state,

the absorbance rises by 30-40%. Figure 1.3 shows a typical profile of UV absorbance against

temperature, known as a DNA melting curve. The midpoint of this transition is the where

50% of duplex DNA is denatured.1


7

Figure 1.3: Schematic of a DNA melting curve, measured by an increase in absorbance at 260nm. The melting

temperature (Tm) is indicated at 358 K which is determined at the point where the melting curve is half

completed.

The DNA melting process is generally reversible. The intact double helix structure can be

restored by incubating the ss-DNA solution at a temperature below the , in a process

known as reannealing and generally involves reassociation and reformation of the original

helix. The efficiency and efficacy of this reversal process can be studied by a decrease in UV

absorbance or insusceptibility to single-strand specific nucleases.24

B. Two-State model:

is influenced by DNA length, sequence, salt concentration, pH and is significantly

affected by the GC content of the double helix. Previous studies on long DNA molecules

from various species, such as Yeast, Bacteriophage T4, E.coli and Calf thymus showed that

there is a linear relationship between and GC percentage,1,25

! " #!$!% " & (1. 1)

For short oligo-nucleotides (~12 bp), the two-state model can be used to approximate . In

this model there is no intermediate state, therefore in equilibrium,

'()*+ , (()*+- " (()*+. (1. 2)

The equilibrium constant for this reaction is


8

/(()*+-0/(()*+.0/'()*+0 (1. 3)

According to Van’t Hoff equation, the standard Gibbs free energy1 2%˚, is given by,

2%˚ !34! (1. 4)

where, is the ideal gas constant (1.987 cal K-1mol-1) and is the absolute temperature. At

the midpoint of the melting curve, . During dissociation of ds-DNA and in the

absence of additional nucleic acids, /(()*+-0=/(()*+.0=/'()*+0. This is equal to half of

the initial concentration of ds-DNA, /'()*+05657, hence

2%˚!34 8 /'()*+565709 (1. 5)

The standard thermodynamic relationship defines Gibbs free energy as,

2%˚ 2:˚ 2;˚ (1. 6)

The terms 2:˚ and 2;˚ are the standard enthalpy and entropy of the duplex melting,

respectively. Thereby,

2:˚2;˚ !34 8 /'()*+565709 (1. 7)

In accordance with Eq. (1. 7), 2:˚ and 2;˚ can be determined from a plot of

34 8<= /'()*+565709 vs. ! <>? (van’t Hoff plot). However it should be noted that this equation is

based on the assumption that only two states are involved in denaturation of ds-DNA: duplex-

state and single stranded-state. This statement may not be prevalent for long nucleic acids

where denaturing occurs via several transition steps. Therefore, a statistical mechanical

model is required for accurate predictions.23, 26

C. Nearest Neighbour model:

Measuring the thermodynamic parameters may not be practical for every single sequence. In

1970, D.M. Gray and I. Tinoco proposed the Nearest-Neighbour (NN) model to approximate

these parameters and consequently predict the melting temperature.27 The NN model

postulated that the stability of the DNA double helix is dependent on the identity and


9

interaction of neighbouring base pairs. i.e. the thermostabilty of the DNA is dictated by the

base sequence rather than the base composition. Considering Watson-Crick base pairing,

there are ten possible NN interactions. Table 1.2 presents the thermodynamic parameters of

these interactions in 1 M NaCl at 37 ˚C and pH 7.0.28, 29

Table 1.2: Nearest-neighbour thermodynamic parameters for 10 different Watson-Crick pairwise interactions at

1 M NaCl, 37 ˚C, pH 7.0 (1 cal = 4.184 J). The slash (/) indicates the sequence at complementary strand

(antiparallel orientation). The values are taken from ref. 29

Sequence

(5’-3’/3’5’)

@A˚ (kcal mol

-1)

@B˚ (cal mol

-1K

-1)

@CDE˚

(kcal mol-1

)

AA/TT -7.6 -21.3 -1.00

AT/TA -7.2 -20.4 -0.88

TA/AT -7.2 -21.3 -0.58

CG/GC -10.6 -27.2 -2.17

CA/GT -8.5 -22.7 -1.45

CT/GA -7.8 -21.0 -1.28

GA/CT -8.2 -22.2 -1.30

GC/CG -9.8 -24.4 -2.24

GT/CA -8.4 -22.4 -1.44

GG/CC -8.0 -19.9 -1.84

In addition to 10 NN dimers, other factors including initiation of duplex formation, entropic

penalty to maintain the C2 symmetry (sym.) of the self-complementary duplex, as well as

counterion condensation (see section 1.4.1) need to be taken into account. In general, the free

energy of forming a nucleic acids duplex is expressed as,

2%°!FFG3 H852%°I9<J5K<

" 2%°LMN " 2%°I4IF O " 2%°I4IF P (1. 8)

Where 2%°I is the standard energy of NN dimers, 5 is the number of occurrence of each

type of NN,!Q and 2%°LMN equals 0.43 kcal mol-1 for a self-complementary sequence and

zero for two complementary sequences. Two initiation parameters were introduced to account

for the difference between the AT and GC terminals. The corresponding values for

2%°I4IF O and 2%°I4IF P are 0.98 and 1.03 kcal mol-1 respectively.29


10

The parameters in Table 1.2 are only applicable to a DNA solution of 1 M NaCl, pH 7.0. As

mentioned earlier, the relative stability of the DNA duplex is also dependent on the ionic

strength of the DNA buffer. The ∆G˚ and Tm rise with increasing ionic strength. A high ionic

strength results in the electrostatic shielding of negative phosphate groups by positive counter

ions, hence the strands repulsion of a duplex DNA is weakened.23, 30 The salt factor correction

of the NN model is beyond the scope of this thesis, for further details see SantaLucia,

J.A.,1998.29

Overall, considering above discussions, it becomes evident that in addition to hydrogen

bonding, the stacking interaction as well as the ionic strength play significant roles in the

energetics of denaturation and formation of double helices in nucleic acids. Nevertheless,

one should note that in all of the above models the double helix is assumed to be in the B-

form structure.

1.2.3 Flexibility and Elasticity of DNA

The ease with which a DNA molecule can be deformed into a compact structure is an

important issue in discussions of packaging of the DNA into chromatin and nucleoid in

eukaryotic and prokaryotic cells.31 Conformational studies have showed that the bond angles

that characterise the DNA biomolecular chain span a limited range; hence, there is some

rigidity in the structure. On the other hand, Coulomb repulsion between the negatively

charged phosphate groups eventually results in reduction of the rigidity of this double helix.

Thus, the DNA is modelled as a semi-flexible polymer or a worm-like chain (WLC). This

biopolymer consists of *.R monomers (base pairs) of length '.R!(~0.34 nm). Therefore, the

contour length of DNA can be described as,

S *.R'.R (1. 9)

When some degree of stiffness is involved in any chain like DNA, the tangent to the two

segments of the chain contour will tend to be pointed in the same direction, provided that the

segments are sufficiently close to each other; i.e. the local contour persists in a given

direction. In polymer science, the persistence length!T, is defined as the length over which

correlations of the direction of the tangent are lost.!!T is a basic mechanical property to

quantify the stiffness of a polymer. 15, 31


11

In addition to elastic deformation in response to external stress, DNA molecules undergo

“worm-like” thermal motions, i.e. they experience bending and torsional fluctuations.

Indeed, fluctuations are responsible for adjustment of the DNA to various key proteins such

as repressors and activators during genetic regulations. DNA damage by radiation and

chemical agents is also possible due to fluctuations where it makes the buried reaction groups

within the double helix accessible to external stress. These thermal fluctuations can be

described within the simplest model of WLC- Kratky-Porod theory (1949).10, 15 For instance,

considering a polymer that behaves like an elastic beam, the force (F) per unit length (L)

required to bend a beam through a curvature <U is

VR W= (1. 10)

where, B is the bending module. T counts how short a segment can bend (e.g. in a circle) by a

fluctuation order of thermal energy, kBT (kB, Boltzmann constant). Form Eq. (1. 10) one can

conclude that a fluctuation V~ kBT will bend a length T into a circle (~T) for,

RX! WYZ (1. 11)

This simple expression shows the correlation between R and the elastic parameter B. In

polymer physics, a polymer is stiff when S [ R and flexible when S \ R.32 In DNA, the

persistence length is much larger than the size of one monomer (0.34 nm), thus capturing the

stiffness and resulting in the semi-flexible nature of the molecule. A typical value for the

persistence length of DNA is about 50 nm (~150 bp). The persistence length can be varied by

the ionic strength of a solution and any electrostatic forces inside the molecule chain.33-35

When describing DNA dynamics, Kuhn length, lKuhn is often used and defined as

]^_6 R (1. 12)

The persistence length of a chain is directly proportional to the chain’s intrinsic elastic

constant. In the simplest elastic model of DNA, each Kuhn segment is modeled as freely

joined with the next segment (see Figure 1.4.a).

*]^_6 S]^_6 (1. 13)


12

When the contour length is much smaller than the Kuhn length,!S [ ]^_6( and thus also

[ R), the molecule behaves as a flexible elastic rod and if it is the opposite case where

!S \ ]^_6 (thus for \ R) the molecules can be described by a random walk of Kuhn

segments (see Figure 1.4.a).36

Figure 1.4: Schematic of a DNA molecule confirmation in free solution at two length scales. (a) A connection

of three rigid Kuhn segments of length of ]^_6 300 bp at smallest scale (b) A coiled DNA molecule with

radius of gyration ` at the largest scale. This figure is adapted from ref. 37.

For a random walk of Kuhn segments without excluded-volume, the average dimension of a

molecules radius of gyration (see Figure 1.4.b) is given by,

` ]^_6!a*]^_6 (1. 14)

In other words, a free DNA molecule in solution adopts a fluctuating random coil of typical

size . For very long polymer chains, the excluded-volume interactions introduce an extra

repulsion which expands the chains. However, short DNA molecules exhibit a high bending

rigidity due to the high ratio of their persistence length to their diameter, hence a very small

excluded volume effect under ambient conditions would result.15

As the DNA polymer consists of two strands, the common mechanism of polymer flexibility,

due to rotation around single bonds, is not applicable. DNA flexibility is due to accumulation

of small changes of angles between adjacent base pairs; therefore the elastic rod model (also

referred to as WLC) can be used to model the DNA double helix. In this model, the sequence

dependence of the DNA bending and torsional rigidity is neglected and the DNA is treated as

a homogenouse and isotropic elastic rod. However, this simple model can still be a practical


13

description of both linear and circular ds-DNA. Not only does it capture the qualitative

physics but can also provide a quantitative description of the elastic strains and stresses in

biological helices.

The total elastic energy of a rod-like macromolecule is described by

b c dW<e< e<J= " W=!e= e=J= " &7 f'g'( Jh= " &i f '('(J h=j '(kJ (1. 15)

where s is the coordinate along its centerline and the first two terms of the integrand give the

bending energy: e< and e=!are the two principle curvatures, W< and W=!are the corresponding

bending rigidities, and e<J and e=J are intrinsic curvatures in an underformed state. The third

term is the torsional energy, g is the twist angle of the rod, lmli is the twist per unit length

(torsional strain), &7 is the torsional rigidity and J is the intrinsic twist. The last term of

integrand expresses the energy associated with axial strain (lilin caused by displacement

of material from the centreline position (J to ( upon stretching; &i is the corresponding

stretching elasticity.

In summary, within the framework of the elastic rod model, bending, twist and stretching are

coupled to each other to give the total potential energy.

b b.o6l " b7p5i7 " bi7qor_ (1. 16)

This model offers a good first estimation in describing the global macromolecular properties

of DNA chain.10

1.2.4 Polyelectrolyte properties of DNA

The DNA polymer is an acid where H+ dissociates from each phosphate group in aqueous

solution. Thus, each base pair carries two elementary negative charges which provide the

electrostatic repulsion. The electrostatic properties of DNA and the electric field generated

near its surface is crucial in various biological functions, such as the activation of

chromosomes for genetic transcription which is controlled through enzymatic modification of

histone tail charges that neutralise DNA wound around nucleosomes.10, 38, 39


14

For the simplicity, the DNA is treated as a highly charged rod with uniformly distributed

linear charge density1 stuv.This model is the simplified version of cylindrical model where

DNA is characterised as a uniformly charged cylinder of diameter '` (geometrical diameter),

whereas in the former molecule '` w #.15

A. DNA at rest in electrolyte solution:

When a charged molecule like DNA is immersed in a fluid containing mobile ions, the

molecule perturbs the distribution of ions. Generally, the ions equilibrium is determined by

the balance between electrostatic and Brownian forces.40

The conformation of the molecules in free solution depends on the ionic strength and the type

of counterions in the medium. The negatively charged DNA attracts the cations and repulses

the anions present in solution. In fluid proximate to the duplex chain, the counterions

concentration decays exponentially.

In mean-field theory, the electrostatic potential in equilibrium may be found using the

Poisson-Boltzmann (PB) equation,

where e is the elementary charge , x. is the dielectric constant of the medium and xJ the

permittivity of a vacuum. y5 is the valence and e5 is the concentration of species Q in the

electrolyte solution. The PB theory states that the equilibrium concentrations of the ions are

related to electric field via the Boltzmann factor. In the PB equation, one assumes:

i) The chemical potential of each of the ionic species is homogeneous in the absence of

fixed charges.

ii) There is no correlation between ions, which is only applicable to dilute solutions.

iii) Ions are point-like charges.

Despite the above limitations, the PB theory is still a rational model to study the electrostatic

interaction of polyelectrolyte molecules like DNA.41

x.xJz=R |Hy55e5 f|y5RYZ h (1. 17)


15

The general solution of the PB equation is called the Gouy-Chapman (GC) solution. In the

GC the charge distribution of ions as a function of distance from the metal surface is

modelled. The simpler model of GC is the Debye-Hückel (DH) theory where the PB equation

is linearised. The DH theory is only applicable to sufficiently dilute electrolyte solutions

(≤100mM).42 According to the DH equation, external fields are screened in an electrolyte

with the potential R deceasing exponentially from its value on boundary i, R i (1. 18)

The Debye length (~< is the characterising length scale for the decay of the charge layer

around the DNA or any charged object in salt media,

where !is the ionic strength of solution.

Hy5=5e5 (1. 20)

Furthermore, as a first approximation, the total charge of DNA would be! |*.R. However

according to Manning condensation theory (for further details see section 1.4.1), the mobile

counterions of solution are hypothesised to “condense” onto the chain of where the

effective charge spacing on the backbone is equal to the Bjerrum length (Z) scale. Z

characterises the length scale over which the electrostatic interaction between the charges

along the backbone is equal to the thermal energy.

Z |=x.xJYZ (1. 21)

In water at room temperature, Z is ~7Å . According to Manning’s theory the effective charge

of ds-DNA in the presence of condensed counterions (multivalent cations) would be

considerably less than! |*.R!.35, 36, 38, 43-45 In addition, the electrostatic repulsion of negatively

charged phosphate groups leads to an increase in the persistence length of DNA, as a result

the apparent persistence length1 R′ is dependent on two factors,

~< xJx.YZ |= (1. 19)


16

R′ R + (1. 22)

where is R is the intrinsic persistence length which is discussed in section 1.2.3 and is

the Odijk-Skolnick-Fixman (OSF) length which characterises the additional contribution

arising from electrostatic repulsion,

Z=S= (1. 23)

As a result of the electrostatic interaction, at least to some extent, the thickness of the Debye

layer (Eq.(1. 19)) and hence the persistence length of DNA (Eq.(1. 22)) can be tuned in the

experiments by changing the ionic strength of fluids. At ionic strengths of 10 mM and 100

µM, ~< is ~1 nm and ~10 nm respectively.33, 38, 40

B. DNA in an external electric field:

Let us consider an applied external electric field with a uniform current in solution. This

electrical field acts on the polyelectrolyte as well as its neighbouring ions. In particular the

excess counterions in the electrical double layer (thickness of ~<) are dragged in the

opposite direction to the applied field. This results in a hydrodynamic interaction of the

surrounding ion cloud with the charged object (see Figure 1.5).

Figure 1.5: Schematic of a negatively charged spherical polyelectrolyte of radius R>~< upon application of an

electric field of E in an ionic solution. This figure is adopted from ref. 40.


17

The motion of charged particles such as negatively charged DNA molecules in a fluid under

influence of a uniform electric field is known as an electrophoresis process. At equilibrium

there is no net force and the electrical force (Voor) is opposed by a frictional (viscous drag)

force (Vq5r). Voor Vq5r (1. 24)

The Voor can be expressed as

Voor stuvb (1. 25)

and the Vq5r is given by

Vq5r !r (1. 26)

where is the velocity and r is the friction coefficient of the DNA chain. Balancing these

two forces results in

VoorVq5r J!b (1. 27)

where # is the free-solution electrophoretic mobility of DNA,

J stuvr ! (1. 28)

In order to evaluate #, one needs to work out r, which is dependent on the viscosity of

solution (). For a cylinder length of S and diameter of '`,

r S34 f S'`h

! (1. 29)

In case of a duplex DNA, S is the ds-DNA contour length and the diameter is equal to '`.

The electrophoretic mobility of DNA can be described in two limited regimes using relative

magnitudes of persistence length, the Debye length and the radius of gyration of the chain:


18

i) For a thick Debye layer (`<!~<) , the counterions are far from the duplex chain and

do not interact hydrodynamically with the DNA polymer. Hence, for a charge density

per Bjerrum length (stuvZ) of Z, the electric force and viscosity drag can be

determined independently, leading to

J Z!34 S'`! (1. 30)

In this regime the velocity and hence the mobility is dependent on the size and shape of the

charged object. This is a common case when the molecules are small (`is small) and/or at a

very low salt concentration (~< is large).

ii) For a thin Debye layer (`>!~<), the hydrodynamic interactions between the

counterions and the chain must be taken into account. These interactions are screened

over the Debye layer and the mobility of the chain is equivalent to,

J o !34 ~<r ! (1. 31)

where the o is the effective charge density of DNA according to Manning condensation. In

this regime the DNA acts as a freely-draining polyelectrolyte and its charge to friction ratio is

not a function of the length of the molecule (Smoluchowski regime).36, 46

This section only described the free solution electrophoresis of DNA. Once the DNA is

placed in a matrix such as gel or a micro/ nanochannel, other parameters such as degree of

confinement, ζ potential of the surface and hence the electro-osmotic effect, introduce an

additional contribution in electrophoretic mobility. For further details see chapter 3, section

3.1.5.

1.3 DNA Topology

DNA chains exist at various sizes; from below one persistence length up to millions of

persistence length.15 However, sizes, sequences and helical geometries (see section 1.2.1)

collectively known as primary and secondary structures, are not the only features in which

DNA molecules vary from each other. They can also adopt various configurations and

tertiary structures. In this section, the conformational and topological properties of ds-DNA

Chapter 1

and their significance in genome functioning

are largely based on an article written by Sergei M. Mirkin from

(2001).47

1.3.1 Global Confirmations of DNA

The global confirmations of DNA can be classified into two major forms: a)

closed circular (cc-) DNA.

The DNA axis is often linear as in the case of human genomic DNA and

eukaryotic species. In the simplest form, DNA has free ends

other hand in 1963, it was unexpectedly found that ds

in polyoma virus.48, 49 Currently

the typical configuration found in bacteria (also known as

cytoplasm of eukaryotes (e.g. mitochondria and plastid).

1.3.2 Topological domains

A decade after the discovery

strands can be highly intertwined and coiled. This finding initiated

topological aspects of the DNA structur

the free rotation of its ends is restricted,

Figure 1.6: schematics of topological domains.

(c) Linear DNA affixed to a membrane,

adopted from ref. 47.

The most well-known example of a topological domain is the cc

free ends at all. Even in the case of linear DNA

n genome functioning are briefly surveyed. The following sub

are largely based on an article written by Sergei M. Mirkin from University of Illinois

Global Confirmations of DNA

The global confirmations of DNA can be classified into two major forms: a)

The DNA axis is often linear as in the case of human genomic DNA and

In the simplest form, DNA has free ends which can rotate freely.

other hand in 1963, it was unexpectedly found that ds-DNA existed as a closed circular form

Currently, it is acknowledged that in addition to viruses, cc

the typical configuration found in bacteria (also known as plasmid DNA), archaea and

cytoplasm of eukaryotes (e.g. mitochondria and plastid).15

Topological domains

A decade after the discovery of the double helix, it was demonstrated that the two DNA

strands can be highly intertwined and coiled. This finding initiated researches into

DNA structure. When a DNA segment is constrained in a way that

restricted, it is called a topological domain (see

schematics of topological domains. (a) cc-DNA, (b) Linear DNA loops attached to a nuclear matrix,

membrane, (d) Linear DNA wrapped around proteins aggregates. This figure

known example of a topological domain is the cc-DNA,

ll. Even in the case of linear DNA in vivo, the free rotation

Introduction

19

The following sub-sections

niversity of Illinois

The global confirmations of DNA can be classified into two major forms: a) linear and b)

The DNA axis is often linear as in the case of human genomic DNA and the majority of

can rotate freely. On the

DNA existed as a closed circular form

it is acknowledged that in addition to viruses, cc-DNA is

DNA), archaea and

uble helix, it was demonstrated that the two DNA

researches into the

e. When a DNA segment is constrained in a way that

(see Figure 1.6).

Linear DNA loops attached to a nuclear matrix,

d proteins aggregates. This figure is

where there are no

the free rotation of its ends is

Chapter 1

repressed as i) it consists of large loops firmly attached to the nuclear matrix (e.g.

chromosomal DNA) or ii) its ends are affixed to a membrane (e.g. Phage DNA) or iii) is

wrapped around the proteins (e.g. chromatin), which make

a closed circular form.47 Therefore for

topological characteristics of a cc

be applied to all of the above cases.

1.3.3 DNA Supercoiling

Supercoiling is one of the major features of any biologically active

or unwinding of a DNA strand. Its role becomes significant in compaction and packaging of

DNA in both prokaryotic and eukaryotic cell lines.

configurations of supercoiled DNA: a)

prokaryotic DNA and b) solenoidal

eukaryotes.

Figure 1.7: Schematic of ds-DNA supercoiling configurations:

plactonemic coils in prokaryotes. (b)

A covalently closed double helical DNA such as cc

twists. Supercoiling is characterised by two types of coils:

i) Twist (Tw), is the number of helical turns of closed DNA under given conditions. T

a large positive number for any natural DNA.

consists of large loops firmly attached to the nuclear matrix (e.g.


wrapped around the proteins (e.g. chromatin), which makes all these topologies equivalent to

Therefore for simplicity, in upcoming sections, mainly the

topological characteristics of a cc-DNA will be discussed. However, the same principles can

cases.

DNA Supercoiling

Supercoiling is one of the major features of any biologically active DNA and refers to over

winding of a DNA strand. Its role becomes significant in compaction and packaging of

DNA in both prokaryotic and eukaryotic cell lines. Figure 1.7 shows the two principle

configurations of supercoiled DNA: a) plactonemic, the typical interwound coiling in

solenoidal, where DNA is wrapped around nucleosomal particles in

DNA supercoiling configurations: (a) Supercoiling of a relaxed cc

(b) Supercoiling of a linear DNA to solenoidal coils in eukaryotes.

A covalently closed double helical DNA such as cc-DNA has a certain numbers of coils or

twists. Supercoiling is characterised by two types of coils:

), is the number of helical turns of closed DNA under given conditions. T

a large positive number for any natural DNA.

Introduction

20

consists of large loops firmly attached to the nuclear matrix (e.g.


all these topologies equivalent to

in upcoming sections, mainly the

the same principles can

DNA and refers to over-

winding of a DNA strand. Its role becomes significant in compaction and packaging of

ws the two principle

, the typical interwound coiling in

, where DNA is wrapped around nucleosomal particles in

Supercoiling of a relaxed cc-DNA to

Supercoiling of a linear DNA to solenoidal coils in eukaryotes.

numbers of coils or

), is the number of helical turns of closed DNA under given conditions. Tw is


21

ii) Writhe (Wr), is the number of turns that the duplex axis makes around the superhelix

axis, i.e. the number of superhelical winding. Wr can be in any sign, and generally its

absolute value is much smaller than that of Tw.

The algebraic sum of above intersection parameters results in the third topological component

of a cc-DNA, called the linking number (Lk),

SY p "q ! (1. 32)

Lk is always an integer, even though neither Tw nor Wr should be such. Also, if there is no

break introduced into one or both strands of DNA, Lk value cannot be changed by any

deformation of double helical DNA (topologically invariant), despite the fact that values of

both Tw and Wr can be easily altered by changes of the ambient condition.

The topological state of a cc-DNA is often described by the specific linking difference or

super helical density (σ) which is normalized to the molecule’s length,

SY SYSY 2SYSY (1. 33)

where SY is the linking number of a relaxed DNA; in other words, it is the number Watson-

Crick turns/twists (!p) found in a DNA molecule. The SY of *.R-long cc-DNA

corresponds to

SY p *.R (1. 34)

The number of bp per turn is designated as γ; For instance, in a B-form DNA the γ is

determined as 10.5 bp per turn (see section 1.2.1). This parameter can vary with the changing

of the ambient conditions, such as, ionic strength of solution, temperature, etc.

From equation (1. 32), one can rearrange equation (1. 33) to obtain

p "qSY 2p "qSY (1. 35)

The above equation shows that when a cc-DNA is under a topological stress, as a result of a

linking difference, the twist deviates from its optimal value and writhe is introduced. In

addition, the sign of σ can be a good description of supercoiling nature. A positive value of σ


22

means that the molecule is positively supercoiled or overwound and the negative σ refers to a

negatively supercoiled or unwound DNA. It is found that chromosomal DNA extracted from

E.coli is negatively supercoiled and ##.

Moreover, the Gibbs free energy associated with supercoiling under physiological conditions

can be expressed as a quadratic function of σ,

2% #*.R= (1. 36)

This equation indicates that 2% is proportional to the square of σ, hence, small changes in

superhelical density of DNA results in significant modulation in 2%. Since supercoiling

induces energetically unfavorable torsional and bending deformation into DNA, local DNA

changes leading to supercoil relaxation of DNA become favorable. e.g. in 1050 bp long

negatively supercoiled DNA with linking difference (2SY) of -4, it is sufficient to unwind a

42 bp long segment of DNA ( four helical turns, = 4) to relax the molecule. This simple

example also suggests that for a negatively supercoiled DNA, unwinding and for a positively

supercoiled DNA, overwinding is thermodynamically more favorable.23, 32, 47

Additionally, when closed loop DNA molecules such as plasmids are extracted from bacteria,

the supercoiled confirmation is the most dominant topology. However, within the same

sample solution, various degrees of supercoiling can be observed. The probability of each

supercoiling state can vary, depending on physiological conditions and species strain. For

instance, since temperature affects the winding of the DNA double helix, by decreasing the

temperature, the twist angle in DNA increases. As a result, value is reduced and SY is

increased. Consequently, σ is affected and leads to an increase in the level of negative

supercoiling. Another parameter can be the ionic strength of the solution. Due to the

screening of DNA negative charges by the addition of monovalent or divalent cations, the

DNA double helix winds less tightly and the twist angle increases. Hence, as a result, SY

and the level of negative supercoiling also increases.19 Figure 1.8 shows classic electron

micrographs of a cc-DNA supercoiling with different ranges of helical densities.


23

Figure 1.8: Electron micrographs cc-DNA supercoiling transitions from relaxed to tightly supercoiled plasmid

DNAs. The molecule on the left is the most relaxed configuration. The degree of supercoiling increases from

left to right. This figure is reprinted from ref. 24 (copyright permission from W. H. Freeman on 16/01/14).

The removal of supercoiling features (twists and writhes) of a DNA duplex is an initial and

crucial step for the majority of genetic processes, even though untangling of the double

strands is topologically impossible, unless one of the strands is broken. Therefore, cell

machinery utilises additional mechanisms, such as employing the topoisomerase enzyme

family to address this problem.

1.3.4 Nicked DNA

The presence of a single-strand break (nick) in a double stranded supercoiled cc-DNA

removes the topological constraint and effectively allows the DNA to be in its relaxed (open

circular) configuration (see Figure 1.9).

Figure 1.9: Schematic of nicking of a double stranded supercoiled DNA using a nickase enzyme. The

supercoiled cc-DNA lost the topological features and became a relaxed and open circular nicked DNA.

When a phosphodiester bond of two adjacent nucleotides breaks, the strands can rotate freely

and adopt any twist and writhe with no coupling between them and any torsional stress within


24

the strands can be released. It is found that the nicked conformations play a key role at the

beginning of the DNA replication process and can be generated through enzymatic reactions

in vivo and in vitro.

1.3.5 Knots

The linking number is not the only topological characteristics of a cc-DNA. When a long

linear DNA (>10 kbp) cyclise into a circular DNA, knots of different types and complexity

can be formed (see Figure 1.10) Knotting is topologically invariant and cannot be changed by

any conformational changes without any strand breakage in DNA. Knotted molecules can be

found in living cells and it is speculated that they are the side products of various genetic

processes such as recombination.15, 50

Figure 1.10: Schematic of formation of a right-handed elementary knot from a double stranded cc-DNA.

Knotted DNA was first detected in 1976 by Lie and Davis, using the type I topoisomerase

enzyme (TOPO) during preparation of single-stranded circular DNA,51 However, it was still

unclear how knots could be generated in double stranded DNA. Later on it was found that the

second class of topoisomerase enzymes –type II (e.g. DNA gyrase) is capable of untying and

tying of knots in double stranded cc-DNA.52 TOPO I relaxes the supercoiled DNA by nicking

one strand of DNA, rotating the strand about the other one and rejoining it again without

input of energy. On the contrary, TOPO II changes the linking number in cc-DNA (2SY ) by breaking both strands simultaneously and rotating one strand pass the other one,

followed by religating them via hydrolysis of adenosine triphosphate (ATP). Ultimately, both

types of topoisomerases establish a complete equilibrium distribution of topological states in

cc-DNA where there is an interplay of supercoiling and knotting.53


25

1.3.6 Catenanes

In addition to knots, it is possible that two or more DNA molecules interlink like chain links

in a process of cyclisation, leading to an invariant topology known as catenanes (see Figure

1.11). This topology usually occurs during the late stages of DNA replication, where the two

single strands are catenated. Catenanes can still be replicated but cannot be separated into the

two daughter strands.54 Since TOPO II can break double stranded DNA, two DNA molecules

of catenanes can be separated by this category of enzymes.24, 50

Figure 1.11: Schematic of right-handed catenation of two double stranded cc-DNA molecules.

1.3.7 Cruciforms

The Cruciform is another topology which forms readily under negative supercoiling which

requires palindromic regions of sequence elements called inverted repeats.15 Inverted repeats

are a sequence of nucleotides where they are equidistant from symmetry center in a DNA

strand and are reverse complements of each other (see Figure 1.12).

Figure 1.12: Schematic of formation of cruciform from a negatively supercoiled B-form ds-DNA. The blue and

red segments represent complementary halves of an inverted repeat. The flanking DNA is shown in black.

Unpairing of two complementary strands, followed by self-pairing of each strands results in

the formation of a cruciform. Extrusion of inverted repeats as a cruciform allows the removal


26

of negative supercoiling by reducing the number of helical twists (2p). Topologically, this

process is equivalent to a total unwinding of the inverted repeats.

Energetically, formations of cruciforms are very costly (20 kcal mol-1) due to several

energetic barriers, including: i) unpairing of a ds-DNA, ii) formation of a four way junction

of DNA duplexes and iii) presence of single stranded bases at the central loop.

Consequently, the cruciform formation is not feasible in linear DNA. Since, topologically,

cruciforms are equivalent to unwound DNA, the necessary energy can be provided by

relaxing the torsional tensions in a negatively supercoiled DNA.

Furthermore, the length of the inverted repeats is another parameter which affects the

energetics of cruciforms. Longer inverted segments will relax more supercoils upon

formation of cruciforms making the process more probable and thermodynamically more

favourable in comparison with short inverted repeats.

1.3.8 Biological Role of DNA Topology

As mentioned above, all genomic DNA molecules, whether they are linear or circular,

contain topological domains. At first glance, it seems to be ambiguous that how a cell

benefits from these constraints features. Among different classes of DNA topology, the role

of DNA supercoiling during various genomic processes including gene expression is

relatively well documented. Eq.(1. 32) showed that any changes in local secondary structure

including helical turn (Tw) will influence global shape (Wr) of the DNA molecule. For

instance, upon binding of a protein to a DNA segment, the local unwinding is promoted and

ultimately reflected by a change of supercoiling of the whole DNA molecule. Sensing the

link between the local and global changes is speculated to assist the cell machinery in two

ways: i) leads the cell to assess the integrity of DNA, which is a crucial perception before

initiation of DNA replication. As an example, the presence of a single strand nick results in

removal of supercoiled topologies and relaxation of DNA. Therefore if a DNA molecule is

supercoiled, that implies that there are no breaks in DNA. ii) allows the distant

communication among topologically constrained DNA segments. Simply put, changing the

topology of one DNA segment can be instantly detected at a remote segment. This

characteristic is particularly important during initiation of transcription and genetic

recombination, when two or more separated DNA segments interact with each other.50 The


27

thermodynamic behaviour of cruciform formation is another evidence for “distant

communication”. As previously stated, a specific free energy is required to form cruciforms.

Interestingly, it has been found that the rate and extent of cruciform formation decreases with

increasing temperature. This is in contrast to predictions as one anticipates that an increase in

thermal energy would speed up reactions. One explanation for this behaviour is that non-

palindromic regions also unwind as the temperature decreases, consequently resulting in the

relaxation of negative supercoiling, hence the reduction in free energy available to activate

the transition of cruciforms.19

Moreover, the majority of genomic DNA in both prokaryotes and eukaryotes are found to be

negatively supercoiled. In spite of the high energy requirements to supercoil the DNA,

unwinding of the negatively supercoiled DNA is energetically favourable, hence it is the

dominant DNA topology in variety living organisms. Besides in majority of the genetic

processes, unwinding of DNA, at least transiently, is essential for initiation steps.50

Lastly, even though there are few definitive biological roles for cruciforms, the significance

of inverted repeats in replication and transcription processes is determined. For instance,

inverted repeats are frequently associated with the origins of DNA replication. It has also

been found that they provide the recognition and binding site for specific proteins. It is

speculated that the presence of inverted repeats in origin regions facilitates the formation of

cruciforms which then perhaps enhance the binding of proteins and thus promote the

replication process.19 Additionally, one of the best known examples of DNA cruciforms is the

Holliday Junction. This intermediate configuration was first proposed by Robin Holliday in

1964 to address a mechanism for exchange of genomic information during homologous

recombination.55 In this process, two homologous chromosomes align and strands are

exchanged between the two DNA duplexes. Following branch migration of exchanged

regions, the four-stranded Holliday Junction is formed. These junctions can be isomerised

into four different configurations and ultimately can be cut by specific enzymes called

resolvase to create two hetro-duplex DNA molecules.19

1.4 DNA-DNA Interaction

In sections 1.2 & 1.3, we outlined the mechanical properties of DNA such as persistence

length, polyelectrolyte property and torsional rigidity which are involved in the packaging of


28

DNA within the cell were outlined. In addition to mentioned parameters, it has been shown

that a DNA molecule interacts strongly with itself and other DNA molecules due to the high

bare charge density of this long biopolymer. In this section, different types of DNA-DNA

interaction and electrostatic origin theories on this topic are outlined.

1.4.1 DNA Condensation

DNA condensation is the collapse of long DNA chain into compact and usually highly

ordered toroidal structures. This phenomenon is commonly described by Manning’s theory

(1969) which states that counterions condense onto the polyelectrolyte molecule until the

charge densities of adjacent monomers along the polyelectrolyte chain are reduced below a

critical value.43 In 1971, sedimentation analysis of DNA in presence of polymer and salt

showed the first evidence of DNA condensation.56 In aqueous solutions, condensation is

normally triggered by the addition of condensing agents-trivalent or higher valence cations.

The most common condensing agents are polyamine spermidine3+ and spermine4+, inorganic

cobalt-amine cations such as Co(NH3)63+

and basic proteins such as histones H1 and H5.57

Initially it was found that divalent ions do not provoke DNA condensation in water at room

temperature. However, Ma and Bloomfield (1994) showed that Mn2+ ions are effective

enough to condense the supercoiled circular but not the linear DNA in aqueous solutions at

room temperature.58 Shortly after, Kornyshev and Leikin (1999) also reported that it is not

only the valence of counterions that influences the condensation process but that the type of

condensing agent is also important. For instance, in aqueous solution at room temperature,

alkali metal ions such as Mg2+ and Ca2+ do not induce condensation or aggregation as

opposed to transition metals like Mn2+ and Cd2+ ions.59

Several parameters can affect the condensation of DNA including the length of the polymer,

dielectric constant of the solvent, type of counterions or condensing gents, valence of the

counterions, ionic strength and temperature.60, 61

Multivalent cations primarily promote DNA condensation through an electrostatic

mechanism. Using Manning’s theory, Wilson and Bloomfield (1979) showed that when the

negative charges of a DNA molecule are screened and neutralised (80-90%) by the presence

of multivalent cations, the inter-helix repulsion is reduced.62 This phenomenon results in the

curvature and bending of the polymer which plays a critical role in facilitating electrostatic


29

attractive potential energy. As mentioned in section 1.2.3, the total electrostatic potential is

dependent on twisting, stretching and bending. Theoretical models predicted that the total

electrostatic potential is minimised with the bending of a straight DNA into a circular form in

the presence of counterions. This finding reveals that i) why it is energetically favourable for

a negatively charged DNA to condense in the first place and ii) why the toroidal structures

are the most common morphology.60

1.4.2 Multimolecular Aggregates

Multimolecular aggregation in DNA is closely-related to DNA condensation. Aggregation is

an intermolecular interaction where multiple DNA duplex chains attract each other and in a

variety of complex structures. This is in contrast to condensation which is an intramolecular

interaction and involves the collapse of a single DNA molecule into a compact structure.

Observation of the existence of an attractive force between the negatively-charged strands

was intriguing and as a result, many theories were developed to account for it. Post and Zimm

(1979) proposed a general phase-transition theory to evaluate and compare the tendency of

condensation and aggregation in polymers, using the change in free energy and an interaction

parameter (χ). The thermodynamic descriptions and the modelling parameters are beyond the

scope of this thesis, for further details refer to Post & Zimm, 1982b.63 It has been reported

and confirmed experimentally using light-scattering studies 64 that in sufficiently dilute DNA

solutions (< 1 µg ml-1), the collapse of a single polymer (condensation) is the dominant

process, whereas in more concentrated DNA solution, aggregation is more frequent.63 As in

the case of DNA condensation, multivalent counterions play a key role in aggregation by

screening the electrostatic repulsion between DNA chains. As the molecular weight of DNA

increases, the aggregates become the more favourable features, unless the concentration is

very dilute.

Furthermore, various experimental studies reported that in the presence of monovalent and

multivalent counterions the aggregation and precipitation occurs above a certain threshold

concentration of multivalent ions. Burak et al. theoretically addressed the dependence of this

threshold on the concentration of DNA itself. They demonstrated that when the DNA

concentration is smaller than the monovalent salt concentration, the threshold of multivalent

counterions such as spermidine3+ and spermine4+ depends linearly on DNA concentration.

Three key findings can be deduced from their overall analysis:65

Chapter 1

i) The number of condensing multivalent ions required to initiate the aggregation

decreases with addition of monovalent salt.

ii) DNA is not overcharged

aggregation.

iii) Provided high monovalent salt concent

in the vicinity of DNA is smaller than the PB

1.4.3 Liquid Crystalline Phases

Liquid crystalline (LC) is a state of matter that has properties between the conven

and solid crystal, i.e. LC may flow like a liquid, despite its molecules

crystal-like way. In 1961, K. Robinson reported the first evidence of the LC

phase behaviour study of a calf thymus DNA.

studies using electron microscopy, X

found that the DNA LC phase can be classified

its nature is dependent on DNA length and stiffness, salt type and concentration,

osmotic stress and activity of cond

constraints align the DNA molecules in parallel

induce complete crystallinity, thus,

axis.67-69

Later studies revealed that by changing concentration of DNA,

series of phase transitions.70 Figure 1.

transforms the isotropic solution of DNA

Figure 1.13: Phase transition in DNA by

(copyright licence number 3310820233895)

The number of condensing multivalent ions required to initiate the aggregation

decreases with addition of monovalent salt.

overcharged by multivalent ions (such as spermine) at the onset of

Provided high monovalent salt concentration, the number of multivalent counterions

inity of DNA is smaller than the PB prediction.

Liquid Crystalline Phases

Liquid crystalline (LC) is a state of matter that has properties between the conven

may flow like a liquid, despite its molecules being

In 1961, K. Robinson reported the first evidence of the LC

phase behaviour study of a calf thymus DNA.66 Later, following various characterising

ing electron microscopy, X-ray diffraction and magnetic resonance methods, it was

DNA LC phase can be classified as a cholesteric (chiral nematic

its nature is dependent on DNA length and stiffness, salt type and concentration,

osmotic stress and activity of condensing agents. In the cholesteric phase, entropic packaging

constraints align the DNA molecules in parallel. However, they are not strong enough to

induce complete crystallinity, thus, the molecules rotate continuously along the cholesteric

Later studies revealed that by changing concentration of DNA, the LC phase undergoes a

Figure 1.13 illustrates how an increase in DNA concentration,

transforms the isotropic solution of DNA.

Phase transition in DNA by increasing DNA concentration. This figure is reprinted from ref.

(copyright licence number 3310820233895).

Introduction

30

The number of condensing multivalent ions required to initiate the aggregation

by multivalent ions (such as spermine) at the onset of

ration, the number of multivalent counterions

Liquid crystalline (LC) is a state of matter that has properties between the conventional liquid

being orientated in a

In 1961, K. Robinson reported the first evidence of the LC-DNA during

, following various characterising

ray diffraction and magnetic resonance methods, it was

chiral nematic) type, where

its nature is dependent on DNA length and stiffness, salt type and concentration, temperature,

In the cholesteric phase, entropic packaging

they are not strong enough to

ntinuously along the cholesteric

LC phase undergoes a

illustrates how an increase in DNA concentration,

is reprinted from ref. 71


31

At relatively low concentration, due to minimisation of chiral energy, the cholesteric phase is

dominant. However, at higher concentration, minimisation of excluded volume is favoured by

a hexagonal phase. In dilute solutions (< 1 mg ml-1), DNA exists as random coils and the

solution is an isotropic liquid. When the DNA concentration slowly increases (100-200 mg

ml-1), the intermediate blue or precholesteric phase occurs. This dynamic and transient phase

consists of microscopic textures with regular three-dimensional (3D) cubic structures with

lattice periods of several hundred nanometers. At higher concentration (200-300 mg ml1), the

most common LC phase- cholesteric structures start forming as described above. Additional

increase of DNA concentration (> 350 mg ml-1) results in the formation of a two-dimensional

(2D) columnar hexagonal phase. This phase exhibits a long-range positional order with

translational symmetries where DNA molecules are undirectionally and longitudinally

aligned in a lateral hexagonal order. However, these are not true crystals. It should be noted

that in most of experimental studies, the ionic strength ranged from 1 mM to 3 M, indicating

the role of DNA shielding and electrostatic interaction by counter ions in LC phases. For

theoretical modelling and prediction of LC-DNA phase transitions and processes see the

review by A.D. Rey (2010).71

The presence of chlolesteric torque is believed to have a major role in packaging of DNA,

hence in determining the shape of DNA in chromosome, bacterial nucleotide, sperm heads

and virus. This hypothesis is in accordance with Monte Carlo simulations by Marenduzzo et

al., in which a transition from a classical isotropic phase to a cholesteic phase inside

bacteriophage capsids is demonstrated.72

1.4.4 Homologous Pairing

In 1999, A. Kornyshev and S. Leikin proposed the electrostatic zipper motif theory to model

the attraction of two homologous DNA segments at low DNA concentration, in a protein-free

electrolyte solution.59, 73 In the Kornyshev-Leikin (KL) theory, the DNA-DNA interaction

between two homologous sequences is governed by two properties of DNA molecules: i)

high charge density rods and ii) the double helix structure. According to their predictions, the

electrostatic repulsion between negatively charged strands is reduced by a zipper-alignment

of phosphate strands with positively charged counterions in the grooves of the opposing

molecule (see Figure 1.14). This reduction of electrostatic energy favours the pairing of

homologous double helices. As the helical pitch parameters (twist and axial rise per bp) are


32

sequence-dependent, the pairing of non-homologous segments requires costly elastic

deformation.

Figure 1.14: Schematic of electrostatic homology recognition in DNA duplexes. A zipper-alignment of

phosphate strands with positively charged counterions in the grooves of the opposing molecule. s is the

curvilinear coordinate along each DNA molecule. On the right, the sequence-dependent variation in the twist

Ω(s) and axial rise h(s) per bp is depicted. This figure is reprinted from ref. 74 (copyright licence number:

3310820770863).

In addition, the electrostatic interaction of homologous DNA strands explains a number of

observed features of DNA-DNA interactions, including multimolecular DNA aggregates as

discussed in section 1.4.2.59, 73-75 Since the KL theory of homologous pairing is elaborated in

some of the experimental findings of this thesis, a more detailed discussion is proffered in

chapter 5.

1.5 DNA-Protein Interaction

Almost all functions of DNA are dependent on proteins. There are many proteins that bind

specifically or non-specifically to ss- and ds-DNA molecules. Proteins can form a complex

with DNA helices via hydrogen bonds, salt bridges, hydrophobic and electrostatic effects,

covalent and van der Waals interactions, etc. Interactions of proteins with DNA may affect

the structure and topological features of DNA and lead to determining the role of DNA in

genomics.

In this section, the major classes of proteins that are involved in the formation of complexes

with DNA duplex are enumerated.


33

1.5.1 DNA Binding Proteins

DNA binding proteins (DBPs) are the key players in various cellular activities including

replication, transcription, packaging, rearrangement, repairs and regulation and expression of

genes. DBPs contain DNA binding domains that allow affinity to DNA. DNA binding

domains are protein folds that consist of at least one motif that recognises DNA strands

generally or sequence specifically. The structural element that is frequently employed in

DNA binding is α-helix such helix-turn-helix motif and Lucien Zippers. However some cases

of β-sheets, β-ribbons, loops or mix combination two or three elements are found, such as

helix-loop-helix and zinc fingers. A zinc finger domain generally consists of one α-helix and

2 β-sheets which is stabilised by coordinating Zinc ions.76, 77 For a more detailed review on

DNA binding domains and their role in DNA-protein interactions see Luscombe et al.,

2000.76

Non-specific or general affinity DBPs can bind to DNA strands regardless of their sequences

via the interaction of functional groups of protein and sugar phosphate backbone of DNA.

For instance, histones (proteins) bind non-specifically to DNA helices through ionic bonds to

organise the DNA into a compact structures known as chromatin. As a result, histones form a

disc-shape complex with DNA called a nucleosome.

Sequence-specific or recognition-specific proteins mostly interact with the major groove of

B-DNA to identify the base pair as more functional groups exposed. As it was discussed

above, the orientation of DNA bases which can be altered by adopting various topologies

such as supercoiling can play a crucial role in binding affinity. The most common

recognition-specific proteins are transcription factors which regulate gene expression by

binding to a specific sequence of DNA at various stages of the transcription process.

Nevertheless, in some cases, recognition is mediated by binding to minor grooves, such as

DNA complexes of TATA-binding protein (TBP). A TBP binds to the TATA box (5'-

TATAAA-3') of DNA by partially unwinding the helix and introducing the double kinks. In

addition to recognition-specific bindings, some DBPs can also interact with DNA non

specifically, such as high-mobility group (HMG) proteins which involve in dynamic

organisation and structure of chromatin.77


34

1.5.2 DNA Modifying Enzymes

Enzymes are globular proteins, ranging from 62 to over 2500 amino acids residues and act as

highly selective catalysers in eukaryotic and prokaryotic cells process, such as DNA damage

and miss-match repair, genetic recombinations and various stages of RNA and proteins

synthesis, etc. The most common and major families of DNA modifying enzymes are

surveyed below.

A. Nucleases and Ligases:

Nucleases cleave the ss- and ds- enzymes sequence specifically or non- specifically by

catalysing the hydrolysis of the phosphodiester bonds. The sequence-specific nucleases are

referred as restriction enzymes which cut the DNA at restriction sites.

DNA ligases can reverse the process by rejoining the cut or broken DNA strands. Restriction

enzymes (e.g. EcoRI) and DNA ligases (e.g. T4 Ligase) require divalent cations to function.

This activates water molecules for nucleophilic attack and stabilise the negatively charged

strands at the transition step. Restriction enzymes have been shown to have a key role in

protection of bacteria from phage infection by cleaving the phage DNA as part of the

restriction modification system. In addition to their role in vivo, they are utilised in molecular

cloning and DNA fingerprinting technologies. Furthermore, DNA ligases are particularly

important in the replication process and are used to join the short segments of DNA that are

generated as part of the replication process into a complete copy of the DNA template.77

B. Topoisomerases and Helicases:

All organisms have evolved enzymes to adjust DNA topology by relaxing superhelical

constraints generated at various stages of replication, transcription and recombination or by

introducing supercoiling through inputting energy. Regulating and controlling DNA topology

is vital for genomic packaging and performance. In section 1.3.5, some cases were briefly

mentioned, where the two classes of topoisomerase (TOPO) enzymes, type I and type II are

employed to introduce various topological features in circular and linear DNA. Strand

cleavage by TOPO enzymes involves a nucleophilic attack of the phosphodiester backbone

by tyrosine-hydroxyl group. This leads to a covalent bond between the enzyme and the 5’

end of the strand and rotation of 3’ end, subsequently followed by a re-ligation.


35

As discussed earlier, type I relaxes the supercoiled DNA by nicking one strand of DNA,

rotating the strand around the other one and rejoining it again without input of energy,

whereas type II breaks both strands, passes one region of DNA through the gap, resulting in

two-unit changes in DNA linking number (2SY ). Ultimately, TOPO II re-anneals the

strands by hydrolysis of ATP molecule- the universal energy donor.77

Helicases are another class of enzymes that contribute to topological adjustment of nucleic

acids and participate in nearly all cellular processes that are involved with the genome. They

act by unpackaging the gene through the unwinding of DNA duplex. These motor proteins

move directionally along a nucleic acid phosphodiester backbone to separate the nucleic acid

strands by disrupting the hydrogen bonds between the bases through ATP hydrolysis.

Mechanically, these enzymes can be categorized into two classes. One class moves in the

5’— 3’ direction, e.g. the Bacteriophage T7 gene 4 helicase. The other class translocates in

the opposite direction, from 3’— 5’, e.g. PcrA helicase from Bacillis stearothermophilus.77

C. Polymerases:

Polymerases take part in the synthesis of polynucleotide chains from nucleoside triphosphates

(NTP), in addition to their proof-reading activity during DNA replication. There are two

classes of polymerases: DNA-directed DNA polymerase which makes a copy of DNA from

the template in the replication process and the second type, DNA directed-RNA polymerase,

which specialises in copying the sequence of a DNA into an RNA strand during the

transcription process. Crystal structures of polymerases reveal a two-metal ion mechanism for

enzyme activity. The first metal ion in the active site of the enzyme activates the

deprotonation of the 3’-hydroxyl group at the terminus of the growing chains. As a result the

α-phosphate of the incoming dNTP is attacked. The second metal ion is involved in the

reorientation of α-phosphate, stabilisation of the transition state and departure of the

pyrophosphate from the catalytic site. DNA polymerases are not site- or sequence-specific,

but are required to be specific for correct Watson-Crick base pairing. RNA polymerases

cannot recognize the DNA segment directly, instead auxiliary proteins such as transcription

factors are required to target and recruit the enzyme where it binds to a DNA promoter

sequence.77


36

D. Recombinases

As mentioned earlier, the recombinase family is a group of enzymes that catalyse the

chromosomal cross-over and genetic recombination at various stages to control gene

expression. The recombinase family is split into two fundamental groups based on the active

amino acid within the catalytic domain: the tyrosine (Tyr) and serine (Ser) recombinases.78

Following up the recognition and alignment of two homologous segments of DNA, the

strand-exchanges proceed with cleavage and rejoining of the DNA strands by

transesterification reactions. In this process, the DNA is cut at fixed points within the

crossover regions (30-40 bp) resulting in the release of deoxyribose hydroxyl group.

Concurrently, the recombinase enzymes bind to the DNA backbone via formation of transient

covalent phosphodiester bonds between the hydroxyle group of the nucleophilic Tyr or Ser

residues of enzyme and phosphate groups of the DNA backbone. These reactions, catalysed

by different types of recombinase enzymes such as RecA, Cre, Tre, FLP, etc., enable

excision, insertion, inversion, translocation and cassette exchange during the recombination

process.79, 80

1.5.3 DNA Binding Antibodies

The antibody or immunoglobulin (Ig) is a large Y-shaped protein that is produced by immune

system cells. Based on site- and conformational-specificity, the antibody recognises a specific

foreign target, called an antigen. An antibody-antigen interaction can be stabilised by

hydrogen bonds, van der Waals interactions, hydrophobic effects and electrostatic forces. The

affinity of antibodies is highly dependent on temperature, salt concentration and pH of a

solution. 81

The first discovered anti-nuclear antibody was anti-dsDNA antibody, a member of the super

family of autoantibodies. Its target antigen is double stranded DNA and is mostly developed

in the autoimmune disease, Systemic Lupus Erythematosus (SLE).1 The exact generation

mechanism of anti-dsDNA antibody in cell is still unclear. However, it is speculated that the


37

immune response against extracellular ds-DNA resulted from dead or dying cells or that the

malfunction of the apoptosis process* leads to production of anti-dsDNA antibody.

Synthetically, nucleic acids can be made immunogenic so as to become detectable by

antibodies with a broad range of specificities and affinities. These antibodies can interact with

ss-DNA and/or ds-DNA, Z-DNA, tRNA, oligonuleotides, etc. It is also possible to prepare

antibodies specifically against purines and pyrimidines, chemically modified bases such as

cytosine methylation and hydroxymethylation, as well as photoproducts and cyclobutane

pyrimidine dimers generated from UV-induced DNA damage.82, 83

In general, nucleic acids-binding antibodies are mostly IgG† isoforms and proven to be

valuable tools in clinical medicine, cell biology and molecular biology, as they can be

employed as biomarkers to detect certain autoimmune diseases and DNA damages, as well as

acting as labels in the detection of DNA modifications.

1.6 Summary

In this chapter, the main features and biophysical properties of a ds-DNA molecule were

reviewed, including structural geometries, thermodynamics, elasticity, polyelectrolyte

properties and the topological domains. In addition, the interactions of this biopolymer with

itself and other macromolecules including proteins, enzymes and antibodies are described,

where their significance can be applied to various stages of the “central of dogma of

molecular biology” including replication, transcription translation.

* Apoptosis is the highly organised process of programmed cell death in multicellular organisms. In this process, the cell degrades the nuclear DNA and signals for phagocytosis.

† Immunoglobulin G (IgG) is an antibody isotype, which is composed of four peptide chains, two identical heavy chains and two identical light chains arranged in a Y-shape. Each IgG monomer has two antigen binding sites.


38

1.7 References

1. Weaver, R.F. Molecular Biology, Edn. 4th. (McGraw-Hill, New York; 2008).

2. Pauling, L. & Corey, R.B. A Proposed Structure For The Nucleic Acids. Proc. Natl.

Acad. Sci. USA 39, 84-97 (1953).

3. Franklin, R. & Gosling, R.G. Molecular Configuration of Sodium Thymonucleate.

Nature 171, 740-741 (1953).

4. Watson, J.D. & Crick, F.H.C. A Structure for Deoxyribose Nucleic Acid. Nature 171,

737-738 (1953).

5. Zamenhof, S., Brawerman, G. & Chargaff, E. On the desoxypentose nucleic acids

from several microorganisms. Biochim Biophys Acta 9, 402-405 (1952).

6. Meselson, M. & Stahl, F.W. THE REPLICATION OF DNA IN ESCHERICHIA

COLI. Proc. Natl. Acad. Sci. USA 44, 671-682 (1958).

7. Ghosh, A. & Bansal, M. A glossary of DNA structures from A to Z. Acta Crystallogr

D Biol Crystallogr 59, 620-626 (2003).

8. Basu, H., Feuerstein, B., Zarling, D., Shafer, R. & Marton, L. Recognition of Z-RNA

and Z-DNA determinants by polyamines in solution - experimental and theoretical-

studies. Journal of Biomolecular Structure & Dynamics 6, 299-309 (1988).

9. Thomas, T. & Thomas, T. Conformational transitions of polynucleotides in the

presence of rhodium complexes. Journal of Biomolecular Structure & Dynamics 7,

1221-1235 (1990).

10. Kornyshev, A.A., Lee, D.J., Leikin, S. & Wynveen, A. Structure and interactions of

biological helices. Reviews of Modern Physics 79, 943-996 (2007).

11. Wing, R. et al. Crystal-structure analysis of a complete turn of B-DNA. Nature 287,

755-758 (1980).

12. Pabo, C. & Sauer, R. Protein-DNA Recognition. Annual Review of Biochemistry 53,

293-321 (1984).

13. Wang, J.C. Helical repeat of DNA in solution. Proc Natl Acad Sci U S A 76, 200-203

(1979).

14. Fuller, W., Wilkins, W.H.F., Wilson, H.R. & Hamilton, L.D. The molecular

configuration of deoxyribonucleic acid: IV. X-ray diffraction study of the A form.

Journal of Molecular Biology 12, 60-IN69 (1965).

15. Frank-Kamenetskii, M.D. Biophysics of the DNA Molecule: A New Trend.

Molecular Biology 36, 232-235 (2002).


39

16. Wang, A. et al. Molecular-structure of a left-handed double helical DNA fragment at

atomic resolution. Nature 282, 680-686 (1979).

17. Liu, R. et al. Regulation of CSF1 promoter by the SWI/SNF-like BAF complex. Cell

106, 309-318 (2001).

18. Mulholland, N., Xu, Y., Sugiyama, H. & Zhao, K. SWI/SNF-mediated chromatin

remodeling induces Z-DNA formation on a nucleosome. Cell Biosci 2, 3 (2012).

19. Sinden, R.R. DNA structure and function. (Academic Press, San Diego ; London;

1994).

20. Rich, A., Nordheim, A. & Wang, A.H. The chemistry and biology of left-handed Z-

DNA. Annu Rev Biochem 53, 791-846 (1984).

21. Kladde, M.P., Kohwi, Y., Kohwi-Shigematsu, T. & Gorski, J. The non-B-DNA

structure of d(CA/TG)n differs from that of Z-DNA. Proc Natl Acad Sci U S A 91,

1898-1902 (1994).

22. SantaLucia, J. & Hicks, D. The thermodynamics of DNA structural motifs. Annu Rev

Biophys Biomol Struct 33, 415-440 (2004).

23. Haynie, D.T. Biological thermodynamics. (Cambridge University Press, Cambridge;

2001).

24. Kornberg, A. & Baker, T. DNA replication, Edn. 2nd ed /. (W.H. Freeman, New

York; 1991).

25. Marmur, J. & Doty, P. Thermal renaturation of deoxyribonucleic acids. J Mol Biol 3,

585-594 (1961).

26. Owczarzy, R. et al. Predicting sequence-dependent melting stability of short duplex

DNA oligomers. Biopolymers 44, 217-239 (1997).

27. Gray, D. & Tinoco, I. A new approach to study of sequence-dependent properties of

polynucleotides. Biopolymers 9, 223-& (1970).

28. Breslauer, K.J., Frank, R., Blöcker, H. & Marky, L.A. Predicting DNA duplex

stability from the base sequence. Proc Natl Acad Sci U S A 83, 3746-3750 (1986).

29. SantaLucia, J. A unified view of polymer, dumbbell, and oligonucleotide DNA

nearest-neighbor thermodynamics. Proc Natl Acad Sci U S A 95, 1460-1465 (1998).

30. Manning, G.S. The molecular theory of polyelectrolyte solutions with applications to

the electrostatic properties of polynucleotides. Q Rev Biophys 11, 179-246 (1978).

31. Hagerman, P. Flexibility of DNA. Annual Review of Biophysics and Biophysical

Chemistry 17, 265-286 (1988).


40

32. Sneppen, K. & Zocchi, G. Physics in molecular biology. (Cambridge University

Press, Cambridge; 2005).

33. Salieb-Beugelaar, G.B., Dorfman, K.D., van den Berg, A. & Eijkel, J.C.

Electrophoretic separation of DNA in gels and nanostructures. Lab Chip 9, 2508-2523

(2009).

34. Skolnick, J. & Fixman, M. Electrostatic persistence length of a wormlike

polyelectrolyte. Macromolecules 10, 944-948 (1977).

35. Manning, G.S. The persistence length of DNA is reached from the persistence length

of its null isomer through an internal electrostatic stretching force. Biophys J 91,

3607-3616 (2006).

36. Salieb-Beugelaar, G.B., Dorfman, K.D., van den Berg, A. & Eijkel, J.C.T.

Electrophoretic separation of DNA in gels and nanostructures. Lab on a Chip 9, 2508-

2523 (2009).

37. Salieb-Beugelaar, G.B. et al. Field-dependent DNA mobility in 20 nm high nanoslits.

Nano Letters 8, 1785-1790 (2008).

38. Dorfman, K. DNA electrophoresis in microfabricated devices. Reviews of Modern

Physics 82, 2903-2947 (2010).

39. Alberts, B. Molecular biology of the cell, Edn. 5th. (Garland Science, New York;

2008).

40. Viovy, J.L. Electrophoresis of DNA and other polyelectrolytes: Physical mechanisms.

Reviews of Modern Physics 72, 813-872 (2000).

41. Cortini, R. (Department of Chemistry, Imperial College London, 2013.

42. Russel, W.B., Saville, D.A. & Schowalter, W.R. Colloidal dispersions. (Cambridge

University Press, Cambridge; 1989).

43. Manning, G.S. Limiting laws and counterion condensation in polyelectrolyte solutions

I. Colligative properties. J. Chem. Phys. 51, 924-933 (1969).

44. Manning, G.S. Counterion binding in polyelectrolyte theory. Accounts of Chemical

Research 12, 443-449 (1979).

45. Manning, G.S. Electrostatic free energies of spheres, cylinders, and planes in

counterion condensation theory with some applications. Macromolecules 40, 8071-

8081 (2007).


41

46. Desruisseaux, C., Long, D., Drouin, G. & Slater, G.W. Electrophoresis of composite

molecular objects. 1. Relation between friction, charge, and ionic strength in free

solution. Macromolecules 34, 44-52 (2001).

47. Mirkin, S.M. in Encyclopedia of Life Sciences (John Wiley & Sons, Ltd, 2001).

48. Dulbecco, R. & Vogt, M. Evidence for a ring structure of polyoma virus DNA. Proc

Natl Acad Sci U S A 50, 236-243 (1963).

49. Weil, R. & Vinograd, J. The cyclic helix and cyclic coil forms of polyoma viral DNA.

Proc Natl Acad Sci U S A 50, 730-738 (1963).

50. Mirkin, S.M. in eLS (John Wiley & Sons, Ltd, 2001).

51. Liu, L., Depew, R. & Wang, J. Knotted single-stranded-DNA rings - novel

topological isomer of circular single-stranded-DNA formed by treatment with

Escherichia-coli omega protein. Journal of Molecular Biology 106, 439-452 (1976).

52. Liu, L., Liu, C. & Alberts, B. Type II DNA topoisomerases: enzymes that can unknot

a topologically knotted DNA molecule via a reversible double-strand break. Cell 19,

697-707 (1980).

53. Podtelezhnikov, A.A., Cozzarelli, N.R. & Vologodskii, A.V. Equilibrium

distributions of topological states in circular DNA: interplay of supercoiling and

knotting. Proc Natl Acad Sci U S A 96, 12974-12979 (1999).

54. Sundin, O. & Varshavsky, A. Terminal stages of SV40 DNA replication proceed via

multiply intertwined catenated dimers. Cell 21, 103-114 (1980).

55. Holliday, R. The Induction of Mitotic Recombination by Mitomycin C In Ustilago

and Saccharomyces. Genetics 50, 323-335 (1964).

56. Lerman, L.S. & Frisch, H.L. Why does the electrophoretic mobility of DNA in gels

vary with the length of the molecule? Biopolymers 21, 995-997 (1982).

57. Bloomfield, V.A. DNA condensation by multivalent cations. Biopolymers 44, 269-

282 (1997).

58. Ma, C. & Bloomfield, V.A. Condensation of supercoiled DNA induced by MnCl2.

Biophys J 67, 1678-1681 (1994).

59. Kornyshev, A. & Leikin, S. Electrostatic zipper motif for DNA aggregation. Physical

Review Letters 82, 4138-4141 (1999).

60. Mukherjee, A.K. Electrostatic contribution to DNA condensation--application of

'energy minimization' in a simple model in the strong Coulomb coupling regime. J

Phys Condens Matter 23, 325102 (2011).


42

61. Podgornik, R., Rau, D.C. & Parsegian, V.A. Parametrization of direct and soft steric-

undulatory forces between DNA double helical polyelectrolytes in solutions of several

different anions and cations. Biophys J 66, 962-971 (1994).

62. Wilson, R.W. & Bloomfield, V.A. Counterion-induced condesation of

deoxyribonucleic acid. a light-scattering study. Biochemistry 18, 2192-2196 (1979).

63. Post, C. & Zimm, B. Theory of DNA condensation - collapse versus aggregation.

Biopolymers 21, 2123-2137 (1982).

64. Post, C.B. & Zimm, B.H. Light-scattering study of DNA condensation: competition

between collapse and aggregation. Biopolymers 21, 2139-2160 (1982).

65. Burak, Y., Ariel, G. & Andelman, D. Onset of DNA aggregation in presence of

monovalent and multivalent counterions. Biophys J 85, 2100-2110 (2003).

66. Robinson, C. Liquid-crystalline structures in polypeptide solutions. Tetrahedron 13,

219-234 (1961).

67. Robinson, C. The Cholesteric Phase in Polypeptide Solutions and Biological

Structures. Molecular Crystals 1, 467-494 (1966).

68. Rill, R., Livolant, F., Aldrich, H. & Davidson, M. Electron-microscopy of liquid-

crystalline DNA - direct evidence for cholesteric-like organization of DNA in

dinoflagellate chromosomes. Chromosoma 98, 280-286 (1989).

69. Strey, H., Podgornik, R., Rau, D. & Parsegian, V. DNA-DNA interactions. Current

Opinion in Structural Biology 8, 309-313 (1998).

70. Livolant, F. Ordered phases of DNA in vivo and in vitro. Physica A: Statistical

Mechanics and its Applications 176, 117-137 (1991).

71. Rey, A.D. Liquid crystal models of biological materials and processes. Soft Matter 6,

3402-3429 (2010).

72. Marenduzzo, D. et al. DNA–DNA interactions in bacteriophage capsids are

responsible for the observed DNA knotting. Proceedings of the National Academy of

Sciences 106, 22269-22274 (2009).

73. Kornyshev, A. & Leikin, S. Theory of interaction between helical biomolecules.

Biophysical Journal 72, WAMK8-WAMK8 (1997).

74. Cortini, R., Kornyshev, A., Lee, D. & Leikin, S. Electrostatic Braiding and

Homologous Pairing of DNA Double Helices. Biophysical Journal 101, 875-884

(2011).


43

75. Kornyshev, A. & Leikin, S. Sequence recognition in the pairing of DNA duplexes.

Physical Review Letters 86, 3666-3669 (2001).

76. Luscombe, N.M., Austin, S.E., Berman, H.M. & Thornton, J.M. An overview of the

structures of protein-DNA complexes. Genome Biol 1, REVIEWS001 (2000).

77. Blackburn, G.M. & Royal Society of Chemistry (Great Britain) Nucleic acids in

chemistry and biology, Edn. 3rd. (RSC Pub., Cambridge; 2006).

78. Wang, Y., Yau, Y.Y., Perkins-Balding, D. & Thomson, J.G. Recombinase

technology: applications and possibilities. Plant Cell Rep 30, 267-285 (2011).

79. Pan, G., Luetke, K. & Sadowski, P.D. Mechanism of cleavage and ligation by FLP

recombinase: classification of mutations in FLP protein by in vitro complementation

analysis. Mol Cell Biol 13, 3167-3175 (1993).

80. Turan, S. & Bode, J. Site-specific recombinases: from tag-and-target- to tag-and-

exchange-based genomic modifications. FASEB J 25, 4088-4107 (2011).

81. Janeway, C., Travers, P., Walport, M. & Murphy, K.P. Janeway's immunobiology,

Edn. 7th. (Garland Science, New York; 2008).

82. Lee, J.S. Antibodies to nucleic acids. Biochemical Education 12, 98-101 (1984).

83. Jacob, T.M. & Srikumar, C. Nucleic acid reactive antibodies— specificities and

applications. Journal of Biosciences 7, 61-73 (1985).

44

Chapter 2

Materials and Methods

2.1 Molecular Biology Laboratory ............................................................................................................... 45

2.2 Physical Chemistry Laboratory .............................................................................................................. 51

2.3 References .............................................................................................................................................. 56

Synopsis: The materials and methods used throughout the experiments presented in the thesis are outlined in

this chapter. The methodologies have been divided into the subgroups of molecular biology and physical

chemistry techniques. These methods were used to study the single molecule sensing of sonicated (linear) DNA,

methylation level of linear DNA and homology recognition of cc-DNA.

Chapter 2 Material and Methods

45

2.1 Molecular Biology Laboratory

2.1.1 Cell culture

The human breast carcinoma MCF-7, and HBL-100 cell lines used in this study originated

from the American Type Culture Collection and were acquired from the Cell Culture Service,

Cancer Research UK (London, UK). They were maintained in DMEM supplemented with

10% FCS, 2 mM glutamine, and 100 U*/ml penicillin/streptomycin. The cells were incubated

in 10% CO2 humidified atmosphere at 37 °C.

2.1.2 Gel Electrophoresis

Approximately 30 ng of each DNA sample was loaded into varying concentrations of 7 cm ×

10 cm agarose gel (Sigma-Aldrich) as indicated. Normally, electrophoresis was conducted in

40 mM Tirs-HCl (pH 8.0), 20 mM Acetic acid, 1 mM EDTA (pH 8.0) at 23 ˚C, with an

applied field of 5 V/cm, unless stated otherwise. The gel was stained in 50 ml of 3× Gel Red

(Biotium) for 30 min followed by 10 min destaining in distilled H2O. Photographs were taken

by GelDoc XR+ system (BioRad) and analysed by ImageJ software.

2.1.3 Western blotting and antibodies

Western blotting was performed on whole cell extracts. Cells were washed twice with ice-

cold Phosphate-buffered saline (PBS), scraped and centrifuged at 2000 rpm for 2 min at 4 °C.

A lysis buffer (20 mM Tris-HCl (pH 7.4), 20 mM dithiothreitol, 2 mM EDTA, 1% Triton X-

100, 1% NP40, 1% sodium deoxycholate, 1 mM sodium pyrophosphate, 1 mM sodium

orthovanadate and 1 mM phenylmethylsulfonyl-fluoride) was added directly to the cell

pellets. Cells were re-suspended in the lysis buffer and centrifuged at 13000 rpm for 10 min

at 4 °C. Proteins were in the resulting supernatant and were determined by a Bradford assay.

Once extracted, they were boiled at 100 °C for 5 min in 2x SDS buffer. Primary antibodies of

FOXA1 (Rabbit, Abcam) and β-tubilin (rabbit, Santa Cruz Biotechnology) were detected

using horseradish peroxidase linked anti-rabbit and anti-mouse conjugates and subsequently

visualized using the Enhanced Chemiluminescence (ECL) detection system.

* U: unit


46

2.1.4 Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was extracted using the RNeasy Mini kit (Qiagen), and cDNA was prepared using

SuperScript III reverse transcriptase and random primers (Invitrogen). For qRT-PCR, 100 ng

of cDNA were added to SYBER-Green Master Mix (Applied BioSystems) and run in 7900-

HT Fast Real-time PCR System (Applied BioSystems). The cycling program was performed

at 95 °C for 20 min, followed by 40 cycles of 95 °C for 15s and 60 °C for 1 min. Each sample

was assayed in triplicate and the results normalized to the level of ribosomal protein, L19

RNA. The following forward (F) and reverse (R) primers (Invitrogen) at final concentration

of 1 µM were used:

FOXA1 (H.sapien†; F) 5’-GCTGGACTTCAAGGCATACGA-3’

FOXA1 (H.sapien; R) 5’-GGCAACGTAGAGCCGTAAGG-3’

L19 (H.sapien; F) 5’-GCGGAAGGGTACAGCCAAT-3’

L19 (H.sapien; R) 5’-GCAGCCGGCGCAAA-3’

2.1.5 Methylated DNA Immunoprecipitation assay (MeDIP)

Genomic DNA from cultured cells (106) was extracted by overnight proteinase-K treatment,

phenol/chloroform/isoamylalcohol extraction, ethanol precipitation and RNase-A digestion.

Purified DNA (100 ng/µl) were sonicated (Bioruptor, Diagenode) at low power for 10

minutes (15s ON, 15s OFF) to obtain fragments of 100-600 bp on average. For the MeDIP

assay, a Diagenode MagMeDIP kit was used with 1 µg of fragmented DNA being used for a

standard MeDIP assay. DNA was denatured for 10 min at 95 ˚C and immunoprecipitated (IP)

for 2 hr at 4 ˚C with 2 µl of monoclonal antibody against 5’-methylcytosine (Diagenode) in

final volume of 500 µl IP buffer (10 mM sodium phosphate (pH 7.0), 140 mM NaCl, 0.05%

Triton ×-100). The mixture was incubated in 30 µl of Dynabeads with M-280 sheep antibody

to mouse IgG overnight at 4 ˚C and washed three times with 700 µl of IP buffer. Later, the

beads were treated with proteinase-K for 3 hr at 50 ˚C and the methylated DNA recovered

using the Diagenode DNA purification kit.1

†: Homo sapiens (cell-liens)


47

2.1.6 Long range PCR

A full length of 3.4 kbp FOXA promoter was amplified according to the manufacturer’s

instructions using a Qiagen LongRange PCR kit. The DNA template was then extracted from

the MCF7-cell lines using a Qiagen DNeasy kit. The cycling program was performed in the

following order: 3 min inactivation at 95 °C, followed by 35 cycles of 15s denaturation at 93

°C, 30s annealing at 62 °C, and 4 min extension (1min/ kbp) at 68 °C. The following primers

(Invitrogen) at a final concentration of 1 µM were used:

FOXA1 (H.sapien)-Promoter (F) 5’-CTTTGTGTGAAGCGTGCATT-3’

FOXA1 (H.sapien)-Promoter (R) 5’-GGGACATCTCCCATAACACG-3’

The PCR products were purified using a Qiagen PCR purification kit visualised by 0.8%

agarose gel (5 V/cm, 1 hr) and confirmed by Sanger sequencing (Source BioScience

LifeSciences). The sequencing data is presented in Appendix I. The BLAST program‡ was

used for performing sequence alignments and similarity searches.

2.1.7 In-vitro methylation

4 µg of 3.4 kbp FOXA promoter (PCR product of section 2.1.6) was methylated for 5 hr at 37

˚C with 32 U of M.SssI methylase (New England Biolabs), 128 µM of S-adenosylmethionine,

in 50 mM NaCl, 10 mM Tris-HCl, 10 mM MgCl2, 1 mM DTT (pH 7.9), in total volume of

100 µl. The DNA was purified using a Qiagen MinElute (enzymatic) Reaction Cleanup Kit.

Complete methylation was subsequently confirmed by a restriction protection assay. 1µg of

methylated and unmethylated DNA was incubated for 1 hr with 20 U HpaII restriction

endonuclease (New England Biolabs). The reaction products were separated in a 0.8%

agarose gel, at 5 V/cm and 1 hr electrophoresis.

2.1.8 Binding assay of DNA-Antibody complex

In-vitro methylated DNA samples (see section 2.1.7 ) and 5’-methyl cytosine (5’.mc)

antibody (Mouse IgG1, Clone # 33D, Aviva System Biology) were incubated in molar ratios

of 1 to 9, respectively, in a total volume of 20 µl of 100 mM KCl-10 mM Tris-Cl (pH 8.5) at

‡ http://blast.ncbi.nlm.nih.gov/Blast.cgi


48

37 ˚C for 2<= hr in dry bath. In parallel, two negative controls were carried out: i) the same

ratio of unmethylated DNA (purified PCR product, see section 2.1.6) was incubated with the

antibody in same condition as above, ii) methylated DNA incubated in the same condition as

above in the presence of no antibody. Instead, the antibody’s phosphate buffer (10mM

phosphate buffer; 150 mM NaCl; pH 7.4) is used to make up the final 20 µl volume. The

binding efficacy was tested by gel shift assay (see section 2.1.9) and atomic force microscopy

(see section 2.2.8). Ultimately, the DNA-antibody complex were translocated with nanopore

chips (see section 2.2.6).

2.1.9 Electrophoretic mobility shift assay (EMSA)

The DNA binding assay in section 2.1.8 was performed on approximately 15 ng of

unmethylated DNA and 40 ng methylated DNA in total. The total volume of mixture + 3 µl

50% glycerol (no dye was added) was loaded into wells of a 0.4% agarose gel. The

electrophoresis performed at 2 V/cm, for 4-5 hr on ice, followed by staining with 3× Gel Red

as described earlier (see section 2.1.2).

2.1.10 Construction of “parallel” and “antiparallel” supercoiled plasmids§

Full length of 1000 bp Kanamycin (Kan) gene was amplified from 5310 bp pET-24a-d (+)

Plasmid (Novagen, see Figure 2.1 ) using HotStar HiFidelity Polymerase Kit (Qiagen). To

generate parallel and antiparallel specific sequences, the PCR reaction performed at 95 °C for

20 min, followed by 40 cycles of 95 °C for 15s and 60 °C for 1 min. Following primers

(Sigma-Aldrich) at final concentration of 1 µM are used in PCR reaction.

Parallel-EcoRI (F) 5’-ATGCGATG.GAATTC.CACCGCTGGTAGCGGTGGTTTTT-3’

Parallel EcoRI (R) 5’-TGATGACT.GGATCC.GAAAAACTCATCGAGCATCAAAT-3’

Antiparallel-EcoRI (F) 5’-TGATGACT.GGATCC.GAAAAACTCATCGAGCATCAAAT-3’

Antiparallel-BamHI(R) 5’-ATGCGATG.GAATTC.CACCGCTGGTAGCGGTGGTTTTT-3’

§ The design, engineering and initial cloning of the plasmids were carried out by Dr. Deanpen Japrung, the former research associate in the Albrecht group. The follow-up cloning, bacterial culturing, amplifications and maxi preps, etc were performed by the author.


49

Figure 2.1: Vector map of pET-24a (+) plasmid. This figure is reprinted from ref. 2.

Subsequently, 50 ng of PCR product was ligated into 50 ng of pET-24-a (+) vector

(Novagen) where it was digested by EcoRI and BamHI restriction enzymes. The ligation

reaction was performed in 50 mM Tris-HCl, 10 mM MgCl2,1 mM ATP,10 mM DTT(pH 7.5)

and 400 U T4 DNA ligase enzyme (New England Biolabs) in a total reaction volume of 20 µl

at 16˚C for 2 hr. Then 5 µl of ligation mixture was transformed by heat shock into XL10-

Gold ultra competent cells (Agilent Technologies) and plated in an LB-Kan agar dish.

Following 16 hr incubation at 37 ˚C oven, individual colonies were picked and grown in a

rotary shaker bath to an OD 600 of 0.8 at 37 ˚C in LB medium with 50 µg/ml Kan antibiotic.

The bacteria culture cells were harvested at 6000 × g in a centrifuge (RCB5 Sorvall, Thermo

Scientific) at 4 ˚C for 15 min. Finally, DNA plasmids were extracted using commercial

Maxiprep kits (Qiagen), eluted in 10 mM Tris-HCl (pH 8.5) and visualised on a 0.8% agarose

gel and confirmed by Sanger sequencing (Source BioScience LifeSciences). The insert


50

sequencing data is presented in Appendix I. The BLAST program** was used for performing

sequence alignments and similarity searches.

2.1.11 Relaxation and linearisation of supercoiled DNA

The supercoiled Plasmid DNA (300 ng) was relaxed with wheatgerm Topoisomerase I

(Promega) at 37 ˚C overnight in 50 mM Tris-HCl (pH 7.5), 0.1 mM EDTA (pH 8.0), 1 mM

DTT, 48 U enzyme and 40 mM of various types metal ions as indicated in section 5.4.1.C.i,

in a total volume of 100 µl. The reaction was stopped by incubating the samples at 65˚C for

20 min. Subsequently, the DNA was purified using commercial DNA clean up kits

(QIAQucik, Qiagen) and visualised in a 0.8% agarose gel, at 5 V/cm and 1 hr electrophoresis.

The supercoiled Plasmid DNA (300 ng) was linearised with EcoRI restriction enzyme (New

England BioLabs) at 37 ˚C for 2 hr in 100 mM Tris-HCl, 50 mM NaCl, 10 mM MgCl2,

0.025% Triton® X-100 (pH 7.5), 100 µg/ml BSA and 10 U enzyme in a total reaction

volume of 50 µl. Later, the DNA was purified and visualised using the same procedure as

above.

2.1.12 S1 Nuclease digestion

100 ng of supercoiled and relaxed plasmids as well as M13mp18 single stranded DNA

(positive control; New England BioLabs), were incubated at 23˚C for 30 min in 40 mM

sodium-acetate (pH 4.5), 300mM NaCl, 10 mM ZnSO4 and 1U S1 enzyme

(Thermoscientific) in a total reaction volume of 50 µl. The reaction was stopped by addition

of 3.5 µl of 0.5 M EDTA (pH 8.0) and heating at 70 ˚C for 10 min. Subsequently, the

samples were run in a 0.8% agarose gel, at 5 V/cm for 1 hr electrophoresis.

2.1.13 Ethanol precipitation of DNA

If it was required to increase the concentration of the DNA samples at any stage, the ethanol

perception was performed in the following order: the DNA sample volume was measured, <<J

volume of 3 M sodium acetate (pH 5.2; final concentration of 0.3 M) was added and mixed

well with the DNA solution. Next, 2 volumes of cold 100% ethanol was added and mixed.

** http://blast.ncbi.nlm.nih.gov/Blast.cgi


51

The mixture was placed at -20 ˚C in a freezer, overnight. The following day, the mixture was

centrifuged at 14000 rpm for 15 min and the supernatant decanted. Then, 1 ml of 70%

ethanol was added and centrifuged briefly at 14000 rpm. The supernatant decanted and the

pellet air-dried under the fume hood for 5 min. At the end, the pellet was suspended and

mixed in desired volume of 10 mM Tris-HCl (pH 8.5).

2.2 Physical Chemistry Laboratory

2.2.1 Ultraviolet- Visible (UV-Vis) spectroscopy measurement of DNA

To determine the DNA concentration and its purity, the UV-Vis spectroscopy performed at

260 nm wavelength on all DNA samples before each experiment using NanoDrop™ 2000

Spectrophotometer (Themo Scientific) according to the manufacturer’s manual.3 Briefly, 2 µl

of the sample’s buffer was loaded on the pedestal and the blank measurement was recorded.

Afterwards the pedestal was gently wiped and dried using a lint-free laboratory wipe and then

2 µl of the desired DNA sample was applied. The UV-Vis absorbance was recorded and

hence the concentration was measured. To evaluate samples purity 260/280 and 260/230

ratios were determined. Only DNA samples with ratio of 260/280 ≈ 1.8 and 260/230 ≈ 2.0-

2.2 were used for future experiment, indicating that DNA was free of protein, phenol, ethanol

and EDTA contaminations.

2.2.2 Nanopore fabrication

A 50-100 nm Si3N4 membrane was deposited on both sides of a 300 µm silicon wafer by low

pressure chemical vapour deposition. Then, a standard photolithography-reactive ion etching,

followed by an anisotropic KOH wet etching, was applied to fabricate a 50 µm × 50 µm

Si3N4 window.†† Next, a CrossBeam® FIB/SEM system (Carl Zeiss XB1540) was used for

focused ion beam (FIB) nanopore milling by bombarding Ga+ ions to 5 mm × 5 mm Si3N4

chip, followed by direct imaging of the nanopore fabrication process by scanning electron

†† The above Si3N4 membrane fabrication process was carried out by Thomas Gibb and Fatma Dogan, PhD candidates in the Albrecht group.


52

microscopy (SEM). The nanopore fabrication was performed at an acceleration potential of

30 kV and 1 pA beam current and exposure time of 1-5s.

2.2.3 Silver/Silver chloride (Ag/AgCl) electrode preparation

A 10 cm coiled silver (Ag) wire was immersed into 3 M HNO3 for 2 min and washed with

distilled water afterwards. Potentiometry (Gamry Reference 600, Warminster, USA) was

used to electroplate the Ag wire in 2 M HCl, 500 µA input current at bias of 6 V for 15 min.

Another Ag wire was used as the reference electrode. The redox reactions during this process

are shown in Eq. (2.1, 2). The anodized electrodes (Ag/AgCl) were stored in 1M KCl at room

temperature.

Anode: Ag(s) + Cl- (aq) ↔ AgCl (s) +e- (2. 1)

Cathode: 2H+ (aq) + 2e- ↔ H2 (g) (2. 2)

2.2.4 Nanopore membrane preparation and device assembly

Prior to any experiment, the Si3N4 membranes were cleaned with piranha‡‡ solution (1

H2SO4: 3 H2O2). The membranes were flushed with 70% of ethanol, isopropanol, rinsed with

distilled water and subjected to O2 plasma cleaning on both sides for 3 min where any

residual organics on the surface is removed. After the cleaning procedure, the membrane is

sealed between two PDMS gaskets and mounted between two chambers of PTFE cell. Two

Ag/AgCl electrodes are then immersed in each chamber filled with 1M KCl-10mM Tris-HCl

(pH8.5).4 Nanopore membranes were stored in 50% ethanol to avoid accumulation of dusts or

other contaminants.

2.2.5 Electrochemical measurements

Conductance measurements were performed with Gamry Reference 600 potentiostats

(Gamry, Warminster, USA). In the cyclic voltammetry experiments, two newly prepared

Ag/AgCl electrodes were connected to the instrument, one as a working (WE) electrode and

‡‡ Hazard warning: piranha solution reacts strongly with the organic compounds and should be handled with extreme caution. Do not store the solution in closed containers and dispose following hazardous waste disposal procedure of the institute.


53

the other one as a reference (RE)/ counter (CE) electrode. A bias of -0.5 to +0.5 V between

WE and CE/RE electrodes was applied, where a steady-state ionic current established.

2.2.6 DNA translocation and data acquisition

Translocation experiments and data acquisitions were carried out with an Axopatch 200B

patch clamp instrument (Molecular Devices, Sunnyvale, USA). The same cell and electrode

setup, as in the cyclic voltammetry (see section 2.2.5), was used. The RE was connected to

the ground and immersed in the (cis) chamber into which 0.4-1 nM analyte (plasmid, linear

DNA, Antibody, DNA-antibody complex) was added to a KCl solution. The WE electrode

was connected to the headstage of the amplifier. A bias of 0.05 V to 1 V was applied to

establish an electric field which was the driving force of the translocation experiment. The

current–time traces were sampled at 200 kHz and filtered at 10 kHz with a low pass Bessel

filter. In all experiments, 1M KCl-10 mM Tris-HCl (pH 8.5) was used as electrolyte, except

for the translocation of methylated DNA and its complex with antibody, which was

performed in 0.1M KCl-10 mM Tris-HCl (pH 8.5) solution.4

2.2.7 Statistical analysis of translocation experiments

The threshold for event detection was defined with respect to the signal distribution obtained

from the peak amplitude histogram. A Gaussian distribution was used as a fitting function to

determine a cut-off value at µ ± 3σ which excluded > 99 % of the baseline in order to identify

distinctive translocation events from noise - µ is the mean and σ is the standard deviation.

The baseline was subtracted manually. The translocation events were detected with pCLAMP

10 software. Event statistics and fittings (linear, Gaussian and stretched-Gaussian) were

obtained by OriginPro 8.5. The stretched (asymmetric) Gaussian equation,

¡= (2. 3)

was obtained from a model reported by Talaga and Li.5 The expression implemented into

OriginPro 8.5 and the fitting parameters ¡< and ¡= were computed by the software with

maximum iteration number of 100. The fit coverage was assessed by the reduce chi-square

statistics. In all cases the R2 values for the fits were greater than 0.97.


54

2.2.8 Atomic force microscopy (AFM)

All AFM images were obtained in air at 23˚C with an Agilent 5500 AFM/SPM microscope

operating in tapping mode (Agilent Technologies), using commercial “Super Sharp Silicon”

AFM probes (Windsor Scientific) with following parameters: 1024 × 1024 or 2048 × 2048

pixels, scan area of 2.5 × 2.5, 5.0 × 5.0 and 10.0 × 10.0 µm2, speed of 0.6 to 0.9 lines/s.

Images were processed with third-order “flatten filter” (PicoView 1.10, Agilent

Technologies).

Sample preparation:

1.5 ng/µL of each biological molecule or complex was freshly prepared in 1.5 mM EDTA

(pH 8.0), 10 mM HEPES (pH 7.6), 4 mM MgCl2 in total volume of 20 µL. 4 µL of this

mixture was deposited on freshly cleaved 9.9 mm diameter mica (Agar Scientific) and left to

adsorb to the surface for 5-10 min. To remove the excess buffer salt, the substrate was rinsed

with 1-2 ml of nuclease free H2O and dried with a flow of dry N2. All imaging conditions

were reproduced independently at least three times.

Silanisation of mica:

Two plastic caps of Eppendorf tubes were cut and placed in the bottom of a 2 L desiccator.

The desiccator was evacuated with a vacuum pump and filled with dry N2. Mica sheets were

freshly cleaved as thin as 0.05-0.1 mm. 30 µl of 3-aminopropyltriethoxy silane (APTES;

Sigma) was added into one plastic cap and 10 µl of N, N-diisopropylethylamine (Sigma) was

added into the other cap. Mica strips were mounted at the top of the desiccator and left for the

reaction to proceed for 1-2 hours. The caps were then removed and purged dry with N2 for 2

minutes. The mica sheets were left to cure for 24 hr to be ready for the sample deposition

without requirement of divalent ions. 6-8

2.2.9 Dynamic light scattering (DLS) of plasmid DNA

All DLS experiments and auto correlation function (ACF) analyses described below were

carried out by William Pitchford, a PhD candidate in the Albrecht group.

Scattered light intensity ACF were acquired with a Beckman Coulter, Inc. Delsa™Nano C

instrument at room temperature. This instrument was equipped with a 658 nm laser,


55

containing dual 30 mW laser diodes and detected scattered light at 165° (q2 = 6.34x1014 m-2

where q is the scattering vector).

A sample run consisted of 100 accumulations of the intensity autocorrelation function (G2(τ))

where the scattering intensity was quantified from the number of photons per sampling time

(1 µs) and the correlation function calculated over a 1s period. Correlation functions from

each accumulation were summed to reduce noise. A minimum of three runs were conducted

per sample. The Delsa Nano 2.31 operating software contained a ‘dust’ filter which rejected

accumulations where the intensity of scattered light was above a specified threshold. This

upper threshold was set at 25% above the mean scattering intensity. Accumulations were

rarely rejected due to the careful filtering of all buffer solutions prior to the experiment, using

filter paper with 0.1 µm mean pore diameter (Whatman).

Supercoiled DNA samples were studied at ionic strengths of 0.010 M, 0.048 M, 0.085 M,

0.124 M, 0.160 M and 1.01 M. Each solution contained the appropriate quantity of KCl and

was buffered at pH 8.5 using 0.01 M Tris·HCl. A fresh DNA batch (~15 ng/µl) was used for

each condition.


56

2.3 References

1. Mohn, F., Weber, M., Schübeler, D. & Roloff, T.-C. in DNA Methylation, Vol. 507.

(ed. J. Tost) 55-64 (Humana Press, 2009).

2. http://www.helmholtz-muenchen.de/fileadmin/PEPF/pET_vectors/pET-24a-

d_map.pdf (accessed on 29.01.2014).

3. http://openwetware.org/images/f/f8/Quick_Guide_Nanodrop.pdf (accessed on

29.01.2014).

4. Tabard-Cossa, V. in Engineered Nanopores for Bioanalytical Applications. (eds. J.B.

Edel & T. Albrecht) 59-88 (Elsevier, Oxford, UK; 2013).

5. Talaga, D.S. & Li, J. Single-molecule protein unfolding in solid state nanopores. J Am

Chem Soc 131, 9287-9297 (2009).

6. Lyubchenko, Y. Preparation of DNA and nucleoprotein samples for AFM imaging.

Micron 42, 196-206 (2011).

7. Lyubchenko, Y.L. et al. Atomic force microscopy imaging of double stranded DNA

and RNA. J Biomol Struct Dyn 10, 589-606 (1992).

8. Lyubchenko, Y., Shlyakhtenko, L., Harrington, R., Oden, P. & Lindsay, S. Atomic

force microscopy of long DNA: imaging in air and under water. Proc Natl Acad Sci U

S A 90, 2137-2140 (1993).

57

Chapter 3

Solid-State Nanopore Based Detection of Sonicated Genomic-DNA

3.1 Background ................................................................................................................................................. 58

3.2 Experimental Objectives ............................................................................................................................. 66

3.3 Results and Discussions .............................................................................................................................. 67

3.4 Conclusion .................................................................................................................................................. 83

3.5 References ................................................................................................................................................... 84

Synopsis: This chapter outlines the fabrication process and the operational set-up of the silicon nitride

nanopore sensors used throughout this thesis. In addition, fundamental theories and backgrounds, such as the

electrophoresis and electroosmotic process, noise effect, the pore conductance, etc are outlined. Furthermore,

functionality and sensitivity of the nanopore chips were evaluated by translocating a sonicated genomic DNA

extracted from a human breast cancer cell line. Sonication parameters are optimised in such way as to create

sub-3 kbp DNA fragments, in order to be consistent with the size-range of the DNA of under investigation in

later studies. Successful DNA translocation events are observed and the statistical analysis are performed on

>1000 events to obtain the conductance changes and the translocation speed, as well as the flux rate of DNA

translocation process through a sub-20nm pore.

Chapter 3 Solid-State Nanopore Based Detection of Sonicated DNA

58

3.1 Background

The biological cell membranes contain of various types of nanoscale (1-100nm)

transmembrane pores that actively or passively control the trafficking of ions and molecules

across the cell membrane and membranes of intracellular organelles. Examples include: the

nuclear membrane that controls passage of RNA, the cell membrane pores that allow viral

genome transfer, nano-sized potassium and sodium ion channels that keep the rhythm of

muscle contraction cells, the secretion of proteins across pores in cellular organelles

membrane and many other examples which are all essential for the functioning and

maintenance of the cell.1-3

In 1976, Neher and Sakmann demonstrated that individual ion channels in the cell can be

probed electrically using the patch-clamp technique.4 Over a decade ago, this phenomenon

inspired many researchers to use protein or solid-state nanopores as an inexpensive and

ultrafast biosensor for detection of various single biomolecules. Being label free and not

requiring the immobilisation of the analytes on a surface, this approach is of great benefit as a

suitable research tool.5 The basic principle behind this novel biosensor is electrophoretically

driven passage of individual molecules through a pore. Consequently, detectable changes in

ionic pore current are observed. For instance, DNA, a highly negatively-charged molecule,

translocates (tranverses through the pore) in a linear fashion, driven by the electric field. As

soon as the DNA blocks the pore, there is a significant reduction in ionic current as the part

of the liquid volume that carries the ionic current is occupied by the DNA.1, 6

Current blockade events are characterised by amplitude and dwell time; hence, when they are

mapped into two-dimensional space (Blockade amplitude vs. translocation time), clusters

correspond to distinct translocation events which can be correlated to predicted structure.3 In

addition, frequency and patterns (shapes) of current blockage events provide us with valuable

insight on properties of the analytes.1, 7

Bezrukov and co-workers (1994) were the first group to demonstrate that a single ion channel

incorporated into lipid bilayer can be used to count polymers based on the same principle as


59

the Coulter counter*.8 Shortly after, in 1996, John Kasianowicz and co-workers reported on

the use of the biological pore of Staphylococcal α-haemolysin (α-HL) protein for

translocation and characterisation of length distribution of individual polynucleotides.3, 9

This study raised the prospect of rapid and inexpensive genome-sequencing and led scientists

to investigate the physics of DNA translocation through pores5. The translocation process has

been studied as a function of transmembrane potential, pore diameter and electrostatics.10, 11

The first step in this process is the threading of DNA. The frequency of threading can be

calculated from the number of blockade currents per time. The frequency of blockades

increases with increasing voltage, which is consistent with the electrophoresis process. The

translocation event duration depends on the applied potential and length of the strand.5

Furthermore, the ionic strength of the electrolyte, biomolecules conformation, orientation and

interaction, pH and temperature have significant influences on translocation process.

In the following sub-sections, the main types and properties of nanopore sensors are

described, followed-up by a short review of the fundamental physics of ion transport and the

translocation process in nanopore devices.

3.1.1 Biological nanopores

In the 1970s, it became apparent that biological membranes of cells incorporate nanoscopic

channels composed of proteins that are embedded in lipid bilayers 12. The bacterial pore of α-

HL (see Figure 3.1) is currently the most widely used biological channel for DNA analysis.

This protein is formed by self-assembly of seven identical polypeptides and secreted as a

toxin by Staphylococcus and spontaneously forms a nanopore when inserted into a lipid

membrane. This protein consists of a 14-stranded transmembrane β-barrel and a bigger cap

outside of the membrane.13 The external dimensions of the pore are about 10 nm x 10 nm.

The pore has width of 1.4 nm at the narrowest point of the transmembrane channel and

allows the passage of ions at high ionic conductance. This protein has a robust structure and

lacks any moving part which makes it suitable for nanopore detection.5, 14

* A Coulter counter is an apparatus for counting and sizing particles suspended in electrolytes. It is used for cells, bacteria, prokaryotic cells and virus particles.


60

Figure 3.1: (a) The heptameric α-hemolysin (α-HL), The cross-sectional view on the right displays the inner

cavity (green), inner constriction (red), and β-barrel (blue). This image is reprinted from ref. 5 (copyright

licence number: 331082116402). (b) Schematic of translocation of ss-DNA through α-HL. This figure is

reprinted from ref. 15 (copyright licence number: 3310830225419).

The small pore size of α-HL only allows translocation of ss-DNA but not ds-DNA as 2.2 nm

width of ds-DNA is too wide for the pore. In 1996, Kasianowicz et al. demonstrated the

electrophoretic transport of individual ss-DNA and RNA molecules through α-HL9; then

revealed the ability of α-HL to distinguish between freely translocating RNA homopolymers

of adenylic and cytidylic acid15. This work was followed by a report from Meller and co-

workers on differentiating between polydeoxyadenylic acid and polydeoxycytidylic acid

strands of ss-DNA16.

α-HL is not the only biological pore used in nanopore sensing. Protein structures and their

self-assembly properties result in high reproducibility of pore dimensions, geometry and

ability of mass-production of biological pores, which attracts the attention of many

researchers in this field.7, 17 Other examples include outer membrane protein G (OmpG) porin

from the outer membrane of Gram-negative bacteria, Myobacterium smegmatis protein A

(MspA) porin, peptide antibiotics gramicidin and alamethicin channels.18-21

3.1.2 Solid-state nanopores

While initial experiments were solely performed with natural protein pores, engineered

nanopores in organic polymers or inorganic materials such as silicon, silicate, silicon nitride,

silicon oxide, graphene as well as glass nanopipettes (nanocapillaries)22 are now used on a

routine basis, as they overcome problems such as fixed size and limited stability due to

properties of proteins.5, 23


61

The first solid-state nanopore used for sensing single molecules was fabricated into silicon

nitride using ion beam sculpturing.24 Solid-state nanopores use the same operating principle

as biological pore: the nanopore membrane splits a fluidic system into two compartments.

These reservoirs are filled with an electrolyte solution and the only connection between these

compartments is the nanopore. An electrochemical potential difference is established and a

steady-state current flows when a constant potential is applied between two non-polarisable

electrodes, which are already immersed in the electrolyte solution. As the pore resistance is

considerably larger than the resistance of the bulk solution, most of the potential changes

occur at or in the vicinity of the nanopore. The resulting potential gradient is the driving force

of DNA translocation or any other charged molecules. The transient blockage of the pore by

the analyte causes significant changes in pore conductance and consequently the ionic

current. As soon the translocation of a single molecule is completed, the original steady-state

ionic current is restored.17 Figure 3.2 illustrates the schematic of the threading and

translocation of a DNA molecule in KCl electrolyte solution through a silicon nitride (Si3N4)

nanopore.

Figure 3.2: Illustration of a solid-state nanopore device. (a) Schematic of threading and translocation of a single

DNA molecule through a solid-state nanopore in KCl solution. (b) Scanning electron microscopy (SEM) image

of a 40 nm nanopore fabricated on a Si3N4 membrane. (c) Schematic of current- time trace, before and after

addition of DNA during translocation process.

(a) (c)

(b)

I

I

t

t

- DNA

+ DNA


62

One of the limitations of solid-state nanopores is the fast translocation of analyte through the

pore. Controlling DNA motion and translocation speed is the key challenge to DNA

sequencing by nanopores. The proposed approach to overcome this problem is utilising

optical tweezers which can pull DNA through a nanopore at arbitrarily slow speeds1.

Alternatively Mayer and co–workers introduced coating of solid-state nanopore with fluid

lipid bilayer, which slowed down the translocation process as well as preventing pore

clogging.25

In addition to difficult process of making well-defined (small) pores reproducibly, the high

background noise of solid-state nanopores, which limit the ultimate sensitivity and

throughput is one of their drawbacks compared to biological sensors. In 2012, Rosenstein et

al. addressed this problem by introducing an integrated nanopore with CMOS† technology to

gain a sub-microsecond temporal resolution.26

3.1.3 Electrophoresis in Nanopores

Generally, when an electric field is applied at a given point in the cell, the current-induced

field and local electrostatics is introduced which leads to the transport of ions, solvent and

analyte molecules. In most cases, the driving force of translocation is current-induced;

nevertheless, the electrostatic effects become important in the vicinity of the charged walls of

the nanopore (see section 3.1.4 for further details).

At a constant bias potential (¢£¤¥) and in the steady-state condition, the potential drop across

the cell is equal to ¢£¤¥ and the induced steady-state ionic current of I, throughout the cell is

the same. The total potential drop depends on three local “resistors” in the cells: i) the

electrode/solution interfaces, ii) the electrolyte solution and iii) the nanopore.

In a nanopore experiments, (ideal) non-polarisable and Faradaic Ag/AgCl electrodes are

used. Hence, the potential drop at the electrode/solution interface is very small. On the other

hand, the pore resistance is generally at MΩ-GΩ range, therefore relatively, the potential

drop across the electrolyte of 0.1-1 M KCl‡ solution is negligible, i.e. the “access resistance”

† CMOS: Complementary Metal-Oxide Semiconductor.

‡ The typical electrolyte used in translocation experiments is KCl solution, with the ionic strength of 0.1 to 1M.


63

is effectively zero. As a result, the potential gradient of .5-i is nearly equal to the potential

drop at the nanopore. However, this assumption is not valid for very short and very thin

membranes, such as graphene, where the pore resistance is small compared to solution

resistance.

By assuming that most of the potential drop occurs across the pore, one can conclude that the

local electric field inside the pore is the driving force of the electrophoresis process of the

charged analyte through the pore, which can be given by,

b .5-iSRqo (3. 1)

where SRqo is the length of the nanopore channel. This equation implies that in most of the

bulk solution, the electric field is not the dominant force for the dynamics of the analyte.

Instead, the diffusion governs the transport of the molecules toward the nanopore. Once the

analyte is in close proximity to the pore, the “capture” occurs by the electric field-created at

the pore entrance.

3.1.4 Surface Charge Effect

Almost all surfaces immersed in a polar solution, carry a charge due to the stabilising effect

of the solvent on ions. If the surface walls are charged, the ionic distribution is different from

the bulk solution. Generally, counterions of the bulk solution accumulate in proximity of the

walls due to electrostatic interactions. This type of characteristic becomes more important in

the case of silicon oxide or silicon nitride membranes nanopores. When these membranes are

immersed in KCl solution, the hydroxyl groups or oxide groups on the surface ionise and lead

to a negative surface charge density. As a result, due to local excess of K+ ions and the

depletion of Cl- ions, an electric double layer at interface is formed. This charged double

layer will respond to the external applied electric charge and contribute to the ion flux across

the pore.27

The Debye length (thickness of the double layer; κ-1) varies from a fraction of a nanometre to

tens of a nanometre in concentrated and diluted solutions respectively. Therefore, the ionic

strength of the electrolyte used for translocation experiments with solid-state nanopore

sensors becomes a key player on the behaviour of the conductance changes during the


64

translocation of the analyte through the pore. For instance, in the case of the translocation of a

linear DNA such as lambda phage (λ-)-DNA (48.5 kbp), provided sufficiently low salt

concentration (<150 mM KCl), the current-enhancement instead of the current-blockage, was

observed. This behaviour is rationalised by the salt dependency of the conductance change

through small pores in charged membranes.28 For further discussion, refer to section 3.3.3.

3.1.5 Electroosmosis in Nanopores

Electroosmosis (EO) is the motion of an electrically neutral liquid adjacent to a charged

surface, when an electric field is applied in parallel to the interface.

In a charged solid-state nanopore, the EO effect becomes particularly important, as an

additional electrokinetically driven flow is introduced. For instance, in case of aqueous KCl

solution, the water molecules coordinate more towards the K+ than Cl- ions , resulting in a

drag force exerted by K+ ions on the liquid, hence, when an electric field is applied the liquid

is dragged in the same direction of K+ ions movements. If a negatively charged

polyelectrolyte such as DNA is then added to the solution, this EO flow acts in opposite

direction to the electrophoretic force. Therefore, this viscous drag slows down the

translocation speed of DNA. The magnitude of EO effect depends on electrolyte solution, the

analyte, the pore surface material and length of the nanopore channel. For example, the EO

effect is less prominent on DNA due to its very high (fixed) charge density, compared to

proteins where it exhibits a lower effective charge. In addition, the protein’s charge density is

significantly affected by the solution properties, such as, ionic strength, type of ions, pH. As a

result, the speed and direction of translocation in protein sensing are dictated by an interplay

of electrophoretic and electroosmotic forces.27, 29

3.1.6 Entropic Effect

In vivo, the translocation of the (bio) molecules through the nuclear pores or ion channels are

facilitated by special proteins and/or interaction with cellular membranes. In in vitro

experiments, such as our nanopore sensing study, an external field is required to overcome

the activation barrier to transport the molecules through the pore. The activation barrier in

nanopores is entropic in origin. The degree of freedom of a biopolymer is significantly

reduced when it enters a nanostructure, hence, it is entropically unfavourable for the polymer


65

to pass through such a confined geometry only by means of thermal fluctuations. Since the

majority of the analytes used in nanopore sensing are charged molecules, the entopic barrier

can be compensated by the applied transmembrane potential. Given that the entropic barrier

faced by translocation polymer into the pore is an equilibrium property, normally the

translocation dynamics can be outlined in three distinct stages:30

i) Approach of the polymer in close proximity to the nanopore, followed by repeated

threading and unthreading one of its ends into the pore.

ii) “Capture” of the polymer, which results from the final threading when the polymer

enters the nanopore.

iii) Translocation of the polymer through the pore, resulting in transient current

blockage.

Figure 3.3: Schematic of current-time traces in three systems (a) No DNA is added to KCl solution. A steady-

state ionic current (open pore current) upon the application of Vbias is generated due to flux of K+ and Cl- ions

(not in scale) across the pore (b) Translocation of DNA in Ogston regime: when Rg < <= dpore, there is a very

small entropic effect and no stretching of DNA is required during the translocation through the pore, hence a

very fast current blockade events are resulted. (c) Translocation of DNA in entropic trapping regime: when Rg ≥

<= dpore, a very large entopic effect is associated with translocation of DNA through the pore; linear DNA has to

stretch to travel across the pore, hence, the resulting blockade events are slower.

(c)(b)(a)

Si 3

N4

- +

K+

Cl-

- +

Si 3

N4

- +S

i 3N

4

I

t


66

The scale of stretching or threading of a polymer during the capture process by the nanopore

is strongly dependent on the radius of gyration (Rg) of the polymer in respect to the nanopore

diameter (dpore). The schematic illustration in Figure 3.3 compares the characteristic of the

blocked events at two scales of DNA-lengths relative to the pore size. In general the Rg

dependence of the electrophoretic process, though a nanostructure can be described by two

major regimes: When the DNA molecule is smaller than dpore (Rg < dpore) the Ogston regime

is dominant. Under these conditions, the entopic barrier and the hydrodynamic interaction

between polymer and the pore walls are lower, thereby full threading of the polymer is not

required and fast translocation events are observed (see Figure 3.3.b). On the other hand,

when DNA structure is larger than dpore or has a comparable size (Rg ≥ dpore) the entropic

trapping occurs. In this regime, the polymer undergoes a series of threading and elongation in

order to traverse through the nanopore. As a result, long translocation times are observed (see

Figure 3.3.c).31

3.1.7 Applications of Nanopores

The diversity of analytes that can be sensed with nanopores spans a broad range of

nanoparticles, organic polymers, peptides, oligonucleotides, proteins, enzymes and large

biomolecular complexes including protein-protein and protein-DNA complexes.5

The main prospect of nanopore-sensing commenced with ultra-fast and inexpensive genomic

sequencing. Initially nanopore technology was used for nucleic acids analysis only. Studies in

this field allowed discrimination of different nucleic acids e.g. ss-DNA from ds-DNA15, 32,

identification of continuous bases (PolyA,C,T,G)33, characterization of hybridisation of

individual DNA strands34, distinguishing hairpins and trapped duplexed DNA and sensing

point–mutation35, as well as identifying orientation of 5’ and 3’-threaded strands36,

suggesting the emerging potential of nanopore-sensing as a next-generation DNA sequencing

tool37.

Moreover, wide variety of analytes that can be detected using this sensor extended the scope

of nanopore applications and led the researchers to engineer and modify both biological and

solid-state nanopore to make this sensor more effective for detection of each specific analyte.


67

Protein analysis is one of the most significant applications of nanopore sensing, which can be

categorised into structural analysis, bio-sensing and binding characterisation.3 Protein-sensing

by nanopore showed probing of different types of proteins and antibodies38-41, as well as

sensing antibody-protein (IgG-BSA)42, antibody-microparticles43, DNA-protein complexes

(dsDNA-recA)44and enzyme-ligand interaction (sulfonamide-carbonic anhydrase II)20. In

addition, further studies showed detection of small molecules such as antibiotics and

unfolded peptides.45-47

Another field that nanopore technology was introduced in is ultra-fine molecular sieving.

Striemer et al. (2006) reported on the fabrication of ordered arrays of nanopore in ultra-thin

membrane to develop an ultra-filter for separation of molecules.3, 48

3.2 Experimental Objectives

In the following sections, the experimental works conducted on a silicon nitride (Si3N4)

membrane to mill a single nanopore are described. Next, electrochemical conductance

measurements were carried out to characterise the pore conductance and diameter, followed

by translocation of a cocktail of short ds-DNA fragments. This translocation experiment

allows us to obtain a better understanding of resolution, robustness, capture rate and

interference of noise in solid-state nanopore chips as well as optimising methodologies and

instrumentation, including DNA-preparation protocol and data-acquisition parameters.

3.3 Results and Discussions

3.3.1 Fabrication of Single Nanopore by Focused Ion Beam Milling

The fabrication of a nanopore on silicon nitride membranes and other solid-state membranes

has always been a challenge for researchers and various techniques have been proposed to

improve stability, control of diameter and channel length; including KOH etching, electron

beam lithography utilising transmission electron microscopy (TEM), ion beam sculpting and

focussed ion beam (FIB) milling.1, 24, 49-51 The latter is the method employed in this project.

The Si3N4 membrane devices were fabricated by Thomas Gibb and Fatma Dogan (PhD

candidates in the Albrecht group). In summary, as Figure 3.4 illustrates, a freestanding 50-


68

100 nm Si3N4 membrane is deposited on both sides of a 300 µm silicon wafer by low pressure

chemical vapour deposition. Subsequently, a standard photolithography reactive ion etching

(RIE) followed by an anisotropic KOH wet etching were applied to fabricate 50 µm × 50 µm

Si3N4 window. At the final stage, the membrane was subjected to FIB to fabricate a nanopore.

Figure 3.4: Schematic of Si3N4 membrane fabrication: Deposition of a freestanding membrane, followed by

photolithography and RIE. Then a KOH wet etching was applied to create a 50 µm × 50 µm Si3N4 window.

Lastly, FIB milling can subjected to fabricate a nanopore. This scheme is adopted from ref. 52.

Figure 3.5 shows an SEM image of Si3N4 membrane before ion beam milling, the arrow

indicates where ion beam is focussed to drill the pore on 50 µm × 50 µm membrane.

Figure 3.5: SEM image of a Si3N4 membrane before fabrication of a nanopore by FIB milling. (a) top view-

472 µm × 472 µm Si3N4 membrane window patterned by semiconductor lithography (b) bottom view- 50 µm ×

50 µm membrane opened by RIE technique.

For nanopore milling, a Carl Zeiss FIB/SEM instrument was used (Figure 3.6). With this

technology, an ion beam is combined with an electron beam. With FIB milling, an ion beam

is used to drill a small hole in the membrane rather than sputtering the whole surface. The

ion beam consists of gallium ions (Ga+) from a liquid source with maximum energy of 30

100 µm 10 µm

(a) (b)

Chapter 3 Solid

keV. When the Ga+ ion beam is focuse

small as 20 nm can be fabricated in

Figure 3.6: Carl Zeiss XB1540 FIB/SEM instrument for nanopore milling

Moreover, a focused electron-

lateral transport of membrane material.

already milled by FIB. The critical parameter that dictate

current, milling duration, FIB probe accelerating voltage, number of layers, magnification

and exposure time to ion or electron beam. The built

is used to monitor surface changes before and after fabrication.

Initially, in order to achieve a small pore, the nanopore is milled for 5s a

voltage, 1 pA milling current and 84 kX magnification. Subsequently a 10 sec SEM

20 kV, followed by another 10 sec

(see Figure 3.7). In their study, Zhang et al. reported that pore shrinking in Si

by SEM electron beam could be due to a combination of Joule heating and electron

induced migration.53 However later on, Chansin et al. demonstrated that in addition to pore

shrinking, carbon deposition in the vicinity of the pore could be also important and affect the

surface-material properties of a

(unless otherwise stated), an attempt was made

SEM-shrinking by optimising the milling parameters. Further studies showed

lowering the milling time to 1

3 Solid-State Nanopore Based Detection of Sonicated DNA

ion beam is focused on a spot in the middle of the membrane,

small as 20 nm can be fabricated in a few seconds.

Carl Zeiss XB1540 FIB/SEM instrument for nanopore milling

-beam on a relatively large area of the membrane can initiate the

lateral transport of membrane material. Therefore, it can be utilised to shrink the large pores

The critical parameter that dictates the nanopore size are beam


ion or electron beam. The built-in field-emission SEM of the equipment

is used to monitor surface changes before and after fabrication.

to achieve a small pore, the nanopore is milled for 5s at 30 kV accelerating

current and 84 kX magnification. Subsequently a 10 sec SEM

, followed by another 10 sec was applied to shrink the pore size from 88 nm to 34 nm

In their study, Zhang et al. reported that pore shrinking in Si

by SEM electron beam could be due to a combination of Joule heating and electron

However later on, Chansin et al. demonstrated that in addition to pore


al properties of a Si3N4 nanopore.54 Hence, in nearly all future experiments

an attempt was made to mill a sufficiently small nanopore without

shrinking by optimising the milling parameters. Further studies showed

lowering the milling time to 1-2s, nanopores as small as 35-40 nm can be fabricated (

State Nanopore Based Detection of Sonicated DNA

69

e membrane, pores as

Carl Zeiss XB1540 FIB/SEM instrument for nanopore milling.

relatively large area of the membrane can initiate the

Therefore, it can be utilised to shrink the large pores

the nanopore size are beam


emission SEM of the equipment

t 30 kV accelerating

current and 84 kX magnification. Subsequently a 10 sec SEM scan at

applied to shrink the pore size from 88 nm to 34 nm

In their study, Zhang et al. reported that pore shrinking in Si3N4 membranes

by SEM electron beam could be due to a combination of Joule heating and electron–beam

However later on, Chansin et al. demonstrated that in addition to pore


Hence, in nearly all future experiments

to mill a sufficiently small nanopore without

shrinking by optimising the milling parameters. Further studies showed that by simply

40 nm can be fabricated ( see

Chapter 3 Solid

Figure 3.2.b), which are reasonably small enough for detection of the analytes of under

investigation in future works (refer to Chapter 4 & 5).

Figure 3.7: SEM image of a fabricated nanopore on

83.89 kX). (b) after 10s exposure to SEM, the pore is shrunk to ~67 nm (Mag

another 10s SEM exposure, pore shrunk to

3.3.2 Device Platform

Figure 3.8 represents a custom

chambers (1 ml volume). After fabrication and mi

two circular PDMS gaskets to ensure a mega/giga

cells. Non-polarisable Ag/AgCl electrodes with fast kinetics (

charging) connected to a setup and

the same electrolyte solution.

solution such as KCl and the cell

electromagnetic fields, unless noted otherwise.

Figure 3.8: Schematic of device platform: Two 1 ml chambers, two Ag/AgCl electrodes, two PDMS (1cm outer

and 0.35 cm inner diameters) and a Si

(a)

Mag=102.21 KX

10 s

200 nm Mag=83.89 KX

88 nm

3 Solid-State Nanopore Based Detection of Sonicated DNA

which are reasonably small enough for detection of the analytes of under

investigation in future works (refer to Chapter 4 & 5).

abricated nanopore on 50 µm × 50 µm Si3N4 membrane. (a) ~

after 10s exposure to SEM, the pore is shrunk to ~67 nm (Mag = 102.21 kX).

another 10s SEM exposure, pore shrunk to ~34 nm (Mag = 86.54 kX).

Platform

m-built polytetrafluoreothylene (PTFE) cell that consists of two

chambers (1 ml volume). After fabrication and milling, the Si3N4 chip is sandwiched between

two circular PDMS gaskets to ensure a mega/giga-ohm-range seal between two chambers of

polarisable Ag/AgCl electrodes with fast kinetics (thus, negligible capacitive

connected to a setup and immersed in reservoirs. Each compartment is filled with

the same electrolyte solution. Electrochemical measurements were carried out in 1

solution such as KCl and the cell-setup is enclosed in a Faraday cage to exclude p

ds, unless noted otherwise.

Schematic of device platform: Two 1 ml chambers, two Ag/AgCl electrodes, two PDMS (1cm outer

) and a Si3N4 nanopore chip (blue). This Figure is adopted from ref.

(c)(b)

10 s 10 s

67 nm

Mag=102.21 KX200 nm 200 nm

State Nanopore Based Detection of Sonicated DNA

70

which are reasonably small enough for detection of the analytes of under

(a) ~ 88 nm pore (Mag =

102.21 kX). (c) followed by

built polytetrafluoreothylene (PTFE) cell that consists of two

chip is sandwiched between

range seal between two chambers of

thus, negligible capacitive

immersed in reservoirs. Each compartment is filled with

carried out in 1 M salt

setup is enclosed in a Faraday cage to exclude parasitic

Schematic of device platform: Two 1 ml chambers, two Ag/AgCl electrodes, two PDMS (1cm outer

This Figure is adopted from ref. 55

Mag=86.54 KX

34 nm


71

3.3.3 Ionic Conductance of Cylindrical Solid-State Nanopores

Measuring the ionic conductance of nanopore is the first step of any translocation experiment.

This electrical paremeter is proportional to pore diameter and length. Thus, it allows us to

examine the blockage and capacitance of the pore, as well as providing valuable information

on efficiency and quality of the cell set-up, electrodes and nanopore membrane.

Cyclic Voltametry (CV) is the most widely used technique for acquiring quantitative

information about redox reactions and ions transport. In the cell set-up descibed in section

3.3.2, an oxidative electrochemical reaction occurs at the anode (+)

O¦L !"!P3~ !!w !O¦P3L !"!~ (3. 2)

The above reaction, results in the capture of Cl- ion from the aqueous KCl solution at the

electrode, as well as migration of the electron (~) through the circuit, which generates the

current. The resulting charge imbalance at the elctrode leads to migration of K+ ion towards

the nanopore membrane.

The reverse reaction, the reduction process, occurs at the cathode (−)

!O¦P3L !" ~ !w O¦L!"!P3~!! (3. 3)

where the released Cl- ion migrates towards the nanopore and the electron is used up. If the

applied bias is within the range of ±1 (V), the above redox reaction would be the sole

electrochemically active process in the cell , the resulting current–voltage (I-V) response for a

nanopore is Ohmic. At larger Vbias, water molecules and other species might become

electrochemically active, resulting in non-ohmic electrochemical processes and significant pH

instability in weak buffers. In additon, at very high value of Vbias , nanopore membranes are

more likely to become instable and fragile. Therefore , nanopore experiments are frequently

performed under biases lower than 1 V.56

In this work, the CV is used to assess the behaviour of the ionic flux through the pore.

According to Ohm’s law, the ionic current flow though the nanopore is proportional to the

applied potential difference across the pore (V=IR), hence the nanopore conductance of §¨©ª

(S) can be given by


72

%§¨©ª §¨©ª ¢£¤¥ (3. 4)

where is the steady-state ionic current (A), and Rpore is the (dominant) resistance (Ω) in the

cell. In simple terms, §¨©ª is the reciprocal Rpore which can be found from the slope of an I-

V curve.

In the current study, an Ohmic (linear) behaviour of the I-V curve is expected to be observed,

otherwise it provides evidenceof pore clogging or imperfections in the operating system.

Figure 3.9: Ionic conductance measurements of two single nanopores fabricated on Si3N4 membranes

(thickness L = ~100nm). Cyclic Voltammetry performed at 50 mV/s scan rate in 1 M KCl, with ~1 cm2

Ag/AgCl electrodes. Bias of -500 to 500 mV is applied. The pore conductance can be determined by the IV

curve slope Red: §¨©ª = 874 nS, dpore= ~ 89 nm. Black: G=34 nS, dpore= ~18 nm.

Figure 3.9 shows the I-V curves of two single nanopores fabricated on Si3N4 membranes

(thickness of ~100nm), using FIB milling for 5s. Later, one of them (black) is shrunk by a

25s exposure to the electron beam. During CV scans, Vbias of ± 500 mV is applied at the scan

rate of 50 mV/s. The measurements were carried out in 1 M KCl, with ~1 cm2 Ag/AgCl

electrodes. The §¨©ª of 875 nS (red, no SEM shrinking) and 34 nS (black, 25s SEM

shrinking) are obtained from the slope of the I-V curves.

In a typical nanopore sensing experiment, assuming the pore is cylindrical, the pore

conductance can be expressed as a function of dpore,27

-600 -400 -200 0 200 400 600-500

-400

-300

-200

-100

0

100

200

300

400

500

~ 18 nm

~ 89 nm

I (n

A)

Vbias (mV)


73

%§¨©ª '§¨©ª=¥!S§¨©ª (3. 5)

where ¥ is the specific conductivity of electrolyte solution and Lpore is the channel length

(the membrane thickness).The equation (3. 5), is a very simplified model, but still a good

description for conductance change with respect to the pore diameter. However, this model

does not take the surface charge effect into account. At low salt concentrations (< 0.1M KCl),

the salt-dependent surface charge of the nanopore contributes significantly to the ionic

current, as the electroneutrality is maintained by introducing excess movement of the

counterions that screen the walls of the membrane.28, 57 Hence, the ionic conductance of a

nanopore is not only governed by the bulk solution conductvity but also the surface

conductivity as for charged nanopores, equivalent amounts of counterions are required to

compensate for the surface charge. These counterions respond to applied electric fields and

add to total pore conductance. Thereby, the sum of bulk and surface charge conductance for a

cylindrical pore of high aspect ratio (Lpore >> dpore) can be described as

%§¨©ª '§¨©ª=S§¨©ª «¬ " ®¯ «®e | " '§¨©ªS§¨©ª ¥°©± ! «¬ (3. 6)

where «¬ and ®¯ are the electrophoretic mobilties of K+ and Cl- ions, with values of 7.616

× 10-8 m2/Vs and 7.909 × 10-8 m2/Vs respectively.23 The number density of potassium or

chloride ions «®e is a function of the KCl concentration (e; M) and is calculated by

«®e = e × 6.02×10−20 M m−3. e is the elementary charge (1.60217657 × 10-19 C) and ¥°©± is the surface charge density of the pore. The first term of the equation represents the bulk

conductance, which is equivalent to equation.(3. 5) and the second term describes the

contribution of surface charge to conductance.

If !«®e !\ =²³´µ¶l·¸µ¹!o , the first term of the equation (bulk conductance) is the dominant

conductance of the pore, therefore the hydrodynamic diameter of the pore can be estimated

simply by equation (3. 5).

The conductance measurements in Figure 3.9 were carried out in 1 M KCl solution, therefore

the surface charge contribution is negligible. At room temperature, 1 M KCl exhibits the

conductivity (L) of 10.98 Ω-1m-1, 58 hence according to Equation.(3. 5), the dpore of ~ 89 nm


74

(red, no SEM shrinking) and ~18 nm (black, after SEM shrinking) were obtained,

respectively for §¨©ª of 875 nS and 34 nS.

Furthermore, in addition to ionic strength , the surface contribution at a given surface charge

density is also large for small pores. For large pores (dpore > κ1), which is the case in this

study, the double layer is unperturbed compared to a single flat surface and the electrostatic

potential drops to zero at a distance sufficiently far from the pore wall. However if the pore is

very small (dpore ≤ κ1) the electric double layer is affected by the charge and the curvature

effects of the small pore dimensions, as well as the hydrodynamic coupling between opposing

sides of the pore. Overall, the total conductance of nanopore depends on bulk ionic strength

and pore geometry, including size, length and shape.17, 28

3.3.4 DNA Extraction, Purification and Sonication

To assess robustness and resolution of the nanopore chip for detection of relatively short

fragments of DNA (≤ 3 kbp), translocation of Sonicated DNA (H. sapien) was carried out.

As human genomic DNA size is too large for studying with current nanopore devices, it is

randomly sheered to smaller fragments (≤ 3 kbp) by sonication. DNA extraction and

purification from MCF-7 cells was carried out using commercial kits (Qiagen) and yielded 50

µl of 100 ng/µl of DNA. DNA sheering was performed by Bioruptor sonicator at low power,

15 sec ON and OFF for 2 minutes. Figure 3.10 shows the agarose gel image of MCF-7-

soniacted DNA, which confirms that DNA is sheered into 0.5 to 3 kbp fragments. This size

range was the ideal DNA contour length for future antibody assays, as the smaller contour

length makes the DNA more accessible to antibodies (see chapter 4).


75

Figure 3.10: Agarose gel electrophoresis, (1% agarose, 5V/cm, 1hr). Lane 1: 1 kbp DNA Ladder (New England

Biolabs), 0.5, 1, 2, 3 kbp bands are indicated. Lane 2: MCF-7 sonicated DNA-500-3000 bp.

3.3.5 Stochastic sensing of DNA at single molecule level

In nanopore sensing, a patch clamp amplifier is required for detection of the low current

(order of 10-12) and fast transient changes. In this set-up, one of the Ag/AgCl electrodes acts

as a working electrode (WE) which is connected to the headstage of the amplifier, whilst the

other Ag/AgCl is the reference electrode (RE) which is connected to the ground. The work

shown here was performed on a ~18 nm pore, Figure 3.9 and freshly prepared 1 M KCl-10

mM Tris-HCl (pH 8.5). A 200 mV bias was applied across the membrane to create an electric

field of ~ 2 MV/m. At this bias, a steady-state ionic current of ~5.8 nA flows (see Figure

3.11).

After addition of 1 µg of sonicated MCF-7 DNA to the chamber where the RE is immersed

(cis chamber), DNA molecules begin electrophoretically traversing from cathode’s (RE) to

anode’s (WE) reservoir (trans chamber) via the nanopore in a stochastic process. This

migration of DNA molecules results in a significant drop of ionic current, as each DNA

molecule occupies the pore volume during the translocation process. Figure 3.11 compares

the current-time (I-t) trace before and after addition of the DNA, where each downward-spike

represents an individual translocation event.

1 k

bp

Lad

der

So

nic

ate

dD

NA

1 kbp

2 kbp

3 kbp

0.5 kbp

1 2


76

Figure 3.11: Current-time (I-t) curve of a ~18 nm pore with Vbias of 200 mV, 1 M KCl-Tris HCl (pH 8.5) during

translocation of sonicated DNA. (a) before (control) and (b) after addition of sonicated MCF-7 DNA (800 pM)

(c) Magnified image of the indicated translocation events, which shows the pattern and shape of 4 individual

blocked events.

3.3.6 Translocation Dynamics

Figure 3.12 is an illustration of the main components of a translocation event where the

translocation time/ dwell time (τd), open pore current (Io), blocked pore current (Ib), current

blockade amplitude (∆I= Io-Ib), and event charge deficit (ECD; integral of obstructed ionic

current) are indicated in the I-t trace schematic. A discussion of the characteristics and

statistical analysis of the mentioned parameters follows.

Figure 3.12: Schematic of a translocation process, where td is translocation (dwell) time, Io (pA) is open pore

current, Ib (pA) is the blocked pore current, ∆I (pA) is the current blockade amplitude and ECD (fC) is the

integrated event area.

300 301 302 303 304 3055200

5400

5600

5800

6000

6200

I (p

A)

t (s)

0 1 2 3 4 55200

5400

5600

5800

6000

6200

I (p

A)

t (ms)

10

0 p

A

10 ms 200 p

A

40 ms

(a) (b)

(c)


77

Histogram and 2D scatter analysis of τd and ∆I of 1183 events are plotted in Figure 3.13.

Figure 3.13: Histogram analysis of (a) τd and (b) ∆I (c) cluster plot (∆I vs. τd) of translocation of sonicated

MCF-7 DNA through a ~18 nm pore, in 1M KCl-10mM Tris-HCl pH 8.5, at 200 mV applied potential and

room temperature. The (stretched) Gaussian fits are indicated with red curves in graph (a) and (b).

A. Dwell time:

Generally, τd can provide us with valuable information on the length of DNA. In this

experiment τd spans in range of 1-25 ms. The most probable translocation time of 3.5 ± 0.9

ms and the effective velocity of ~500 bp/ms were obtained from the histograms analysis. The

error denotes the standard deviation resulting from the fitting procedure.

Using a simple equation of force balance between the electric field in the nanopore and the

viscous drag over DNA, τd can be written as 59

l Stuvº.5-i (3. 7)

where is the viscosity of solution, Stuv is the contour length of DNA, λ is the linear charge

density and K is a constant of proportionality.

Experimental measurements with a solid-state nanopore showed that l Stuvα where α is

1.4 for shorter DNA strands (15-3500bp) and 1.28 for longer polymers.5 This observation is

in contrast to the results obtained from the biological pore, αHL, where α! ! . This

difference can be explained by considering the pore diameter with respect to DNA molecules.

Solid-state nanopores are wider compared to α-HL protein channels, therefore the whole

hydrodynamic drag on the whole DNA molecule has to be taken into account, which leads to

(a) (b) (c)

0 5 10 15 20 25

600

500

400

300

200

100

0

∆∆ ∆∆I

(pA

)

ττττd (ms)

0 5 10 15 20 250

20

40

60

80

100

ττττd (ms)

Nu

mb

er o

f E

ven

ts

0 100 200 300 400 500 6000

20

40

60

80

100

N

um

ber

of

Ev

ents

∆∆∆∆I (pA)


78

the power-law scaling of τd. In contrast, the narrow diameter of α-HL pore causes very strong

interactions between the DNA and the pore walls, therefore in α-HL, the translocation speed

is dictated by the traversing DNA segment through the pore.5, 60

B. Current modulation:

∆I analysis showed that blockade current amplitudes vary from 230 to 530 pA , with the

most probable ∆I = 260 ± 24 pA and mean value of 300 ± 60 was found for all events. The ∆I

histogram did not follow a well defined Gaussian distribution, compared to what we observed

in a size-comparable study in chapter 4 (see section 4.3.4, Figure 4.15). This deviation is

another manifestation of the structural diversity in the solution. In addition, the presence of

various hydrodymanic and folding foms of DNA insde the pore may also affect ∆I

distibution.61, 62

Figure 3.14: Schematic of translocation of (a) linear, (b) folded, (c) semi-folded ds-DNA through the pore and

its effect on current-time trace. This scheme is adopted from ref. 62.

The Rg of ~ 42-122 nm is estimated based on the DNA lengths of the sample under study.63

As the smallest Rg in the DNA sample is larger than dpore (~18 nm), at least partial threading

and elongation of the polymer is expected (entopic trapping regime; see section 3.1.6 ). In

(c)(b)(a)

I

t


79

2005, Storm et al. reported that ds-DNA translocation can take palce in a folded manner

through SiO2 pore (see Figure 3.14). They showed that during translocation process instead

of a complete threading and linear passage of molecule, part of molecule can be folded,

therefore, more counterions are excluded as the folded DNA molecule occupies a larger

volume of the pore. Consequently, this results in faster translocation but larger modulation of

current.

C. Conductance modulation:

Assuming, the conductance modulation resulted from the analyte translocation is simply

equivalent to ΔI/V,64 ΔG of ~-1.3 nS is estimated from the most probable ∆I analysis. In

theory, during the translocation of a DNA molecule, the magnitude of the 2 is dependent on

i) the exclusion of counterions from the pore because of the volume occupied by DNA and ii)

the entrance of additional cations (K+) that is facilitated by DNA translocation due to

negative charges of the phosphate group,28

2% S§¨©ª

'= !]¬ " ¼¯ " ½]¾ s1¾ (3. 8)

where dDNA is the diameter of ds-DNA (2.2 nm), ½«¾ is the effective electrophoretic mobility

of potassium ions moving along the DNA and ¿®1¾ is the effective charge on the DNA per

unit length. Here, by assuming that ½«¾ equals to the bulk ionic mobility (½«), and the mean

length of the sonicated DNA is 1750 bp, ¿®1¾ of ~0.6 electron per bp is approximated. This

estimation is remarkably close to Manning’s prediction65 and other experimental

measurements reported in literature, including a study by Keyser et. al (2006) where they

determined ¿®1¾ of 0.5 ±!0.05 electron per bp (75% reduction of the DNA bare charge) by

combing optical tweezers with solid-state nanopore sensing.23, 66

D. Event charge deficit:

The nanopore current is a measure of the net transport charged species through the pore per

unit time. Thus, by addition of a polyelectrolyte like DNA, the resulted current modulation

also indcates the amount of charged transfer through pore. As previously illustrated in Figure


80

3.12, this “charge” component is referred to as ECD (event charge deficit; C) which is the

current modulation integrated over the duration of an event,

b&) c 2'À7Á7

(3. 9)

where, t1 and t2 (τd= t2-t1) are the times at the beginning and the end of a translocation event,

respectively.

Since the charge of the DNA molecules with the same contour lengths is constant, therefore

the ECD is expected to be unchanged regardless of the conformation and event type.

Figure 3.15: Histogram analysis of ECD upon translocation of sonicated MCF-7 DNA through ~18 nm pore, at

1M KCl-10mM Tris-HCl pH 8.5, 200mV applied potential and room temperature. The (stretched) Gaussian fit

is indicated with red curve.

Figure 3.15 illustrates the ECD distribution of all detected translocation events. The

histogram analysis shows that the majority of events exhibits 0.16 ± 0.03 pC, equivalent to

~998 ke charge transfer during the translocation process.

Using a 12 nm SiO2 pore, Fologea et al. (2007) demonstrated that b&) Â Stuv<Ã which was

in agreement with an earlier report by Storm et al.62, 67 However, in this specific experiment,

ECD analysis cannot determine the DNA sizes, as the sonication process yielded a variety of

DNA lengths with differences as small as 1 bp. If the DNA sample of study contained a

mixture of fixed-size of DNA fragments, e.g. 0.5, 1, 2, 3 kbp fragment, assuming single

molecule translocation through the pore, one would expect 4 distributions and clusters that

correspond to each specific length.62

0.0 0.5 1.0 1.5 2.0 2.5 3.00

20

40

60

80

100

Nu

mb

er o

f E

ven

ts

ECD (pC)


81

E. Translocation frequency:

Further studies investigated the effect of applied potential on translocation frequency (f) at

voltage range of 50-250 mV (see Figure 3.16). Below the threshold voltage of 50 mV, no

translocation events were observed; perhaps, the driving electrophoretic force was not high

enough to overcome the entropic barrier of squeezing the DNA molecules into a narrow pore.

Figure 3.16: Semi log plot of the effect of applied potential (Vbias) on frequency of events per second, upon

translocation of sonicated MCF-7 DNA through ~18 nm pore, at 1M KCl-10mM Tris-HCl (pH 8.5) and room

temperature. The linear fit is indicated with dashed-red lines.

The translocation rate (number of events per unit time) has attracted much attention among

researchers, as it provides an intriguing perception of how a long polymer find its way into a

nanopore and interacts with the pore during the passage. Besides, the quantitative

information obtained from the frequency of events can be used in evaluating the efficiency of

the devices or the handling procedures.56

The capture of DNA molecules is strongly dependent on the pore geometry, DNA stiffness

and the applied potential. Generally, at low DNA concentration limit, where there is no

memory effect, the event occurring frequency (f) can be described by Van’t Hoff Arrhenius

law and transition-state theory,68-70

Ä ÄJ ! !Å¢£¤¥Å JÆ ! (3. 10)

0 50 100 150 200 250-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

ln [

f (e

ven

ts.

s-1)]

Vbias (mV)


82

where, Å¢£¤¥Å J y|YZÆ ! is a barrier reduction factor due to ¢£¤¥, acting on y|, the

effective charge of DNA. ÄJ is the frequency of events in absence of electric field, which is

governed by bÇ, the activation energy,70

ÄJ È! b- YZÆ ! (3. 11)

is a probability factor, È is the frequency factor and YZ is the thermal factor. b- is often

of entropic origin, however in our experiment, the electrostatic origins of surface charges

and/or dielectric effects of the negatively charged DNA molecule can contribute to activation

energy.

If the capture rate is only limited by the time required for the DNA to arrive at the nanopore,

and not by the final threading process, the DNA translocation occurs in a diffusion-limited

regime. In this regime, the capture rate is only governed by DNA’s overall charge and

diffusion coefficient, as well its concentration. Therefore, the translocation frequency scales

linearly with applied potential. However, the semi-log relationship presented in Figure 3.16.c

suggests that the blockade rate is not a diffusion-limited process and indeed is dictated by an

energy barrier. i.e. a fraction of the DNA molecules that approach to nanopore entrance, is

rejected by the pore, at voltage range of 50-250 mV.56, 69

To obtain a value for the activation energy, we assume = 1 and the È can be estimated by,70

È &)+§¨©ªS§¨©ª ! (3. 12)

where & is the bulk concentration of DNA, ) is DNA diffusion coefficient and +§¨©ª is the

cross-section area of the nanopore. Given +§¨©ª≈ 3×10-16 m2, SRqo 10-7 m, & ≈ 0.5-1 nM,

) ≈ 2×10-12 m2s-1 for 500 bp and 7×10-12 m2s-1 for 3000 bp DNA,71

then È 3×107 s-1 and

9×107 s-1 for the shortest and longest fragments, respectively. ÄJ= 0.44 s-1 obtained from the

intercept of the voltage relationship; hence b- of ~5 YZ is approximated from Eq. (3. 11),

which is relatively comparable to previous reports.70 These estimates are uncertain, because

and S§¨©ª and & are uncertain, also ) of free solution (ignoring geometry confinement) is

taken into account.


83

Finally, the translocation rate results confirmed that the dominant force in this translocation

study is the electrophoretic force. Indeed, this behaviour was expected in this system, as the

ratio between the Debye length (~0.3 nm at 1M KCl) and the pore channel’s length (~100

nm) is very small. If EO effect was not negligible in our system, one would expect the events

rate to decrease when increasing the applied voltage, as the electroosmotic velocity is also

proportional to the applied electric felid.72 Besides, if EO was the dominant driving force, the

translocation direction would have been reversed, as the force was acting in the opposite

direction.

3.4 Conclusion

For the first time, the translocation of genomic DNA of MCF-7 (breast cancer) cell-line

through a Si3N4 nanopore was demonstrated. However, due to the complicated and large

structure of the genome, the DNA was sheered into smaller fragments prior to the start of the

experiment.

The characteristic of blockade events and capture rate analysis confirmed that the

translocation process was facilitated by the electrophoresis and entropic factors are important

in the capture and translocation process. τd and ∆I data analysis somehow allowed the

investigation of the structural and conformational properties of the sonicated DNA in

electrolyte solution, as well as its dynamics through a narrow pore. One can conclude that

the mobility of the DNA inside the pore is dependent on pore geometry, DNA effective

charge, interactions between adjacent DNA segments, interactions with the pore walls and the

hydrodynamic effects.73 Further research is required to obtain a better understanding of

patterns and characteristics of sonicated DNA translocation; such as translocation of fixed–

length DNA fragments ( i.e. 0.5, 1, 2, 3 kbp) individually and in mixture, in order to

determine the corresponding sub-cluster and sub-peaks in cluster plots and histogram

analyses, respectively.

Lastly, using this experiment as the basis of a single molecule sensing with nanopore devices,

the operational set-ups and data acquisition parameters are optimised, for future studies.


84

3.5 References

1. Dekker, C. Solid-state nanopores. Nat Nano 2, 209-215 (2007).

2. Sakmann, B. & Neher, E. Single Channel Recording, Edn. 2nd. (Nature Publishing

Group, 2005).

3. Mulero, R., Prabhu, A.S., Freedman, K.J. & Kim, M.J. Nanopore-Based Devices for

Bioanalytical Applications. 15, 243-252 (2010).

4. Neher, E. & Sakmann, B. Single-channel currents recorded from membrane of

denervated frog muscle fibres. Nature 260, 799-802 (1976).

5. Howorka, S. & Siwy, Z. Nanopore analytics: sensing of single molecules. Chemical

Society Reviews 38, 2360-2384 (2009).

6. Bahrami, A., Doğan, F., Japrung, D. & Albrecht, T. Solid-state nanopores for

biosensing with submolecular resolution. Biochem Soc Trans 40, 624-628 (2012).

7. Bayley, H. & Jayasinghe, L. Functional engineered channels and pores (Review). Mol

Membr Biol 21, 209-220 (2004).

8. Bezrukov, S.M., Vodyanoy, I. & Parsegian, V.A. Counting polymers moving through

a single ion channel. Nature 370, 279-281 (1994).

9. Kasianowicz, J. Characterization of individual polynucleotide molecules using a

membrane channel. Proc. Natl Acad. Sci. USA 93, 13770-13773 (1996).

10. Henrickson, S.E. Driven DNA transport into an asymmetric nanometer-scale pore.

Phys. Rev. Lett. 85, 3057-3060 (2000).

11. Maglia, G., Restrepo, M.R., Mikhailova, E. & Bayley, H. Enhanced translocation of

single DNA molecules through alpha-hemolysin nanopores by manipulation of

internal charge. Proc Natl Acad Sci U S A 105, 19720-19725 (2008).

12. Deamer, D.W. & Branton, D. Characterization of Nucleic Acids by Nanopore

Analysis. Accounts of Chemical Research 35, 817-825 (2002).

13. Korchev, Y.E., Bashford, C.L., Alder, G.M., Kasianowicz, J.J. & Pasternak, C.A.

Low conductance states of a single ion channel are not 'closed'. J Membr Biol 147,

233-239 (1995).

14. Song, L. Structure of staphylococcal alpha-hemolysin, a heptameric transmembrane

pore. Science 274, 1859-1866 (1996).


85

15. Akeson, M., Branton, D., Kasianowicz, J.J., Brandin, E. & Deamer, D.W.

Microsecond time-scale discrimination among polycytidylic acid, polyadenylic acid,

and polyuridylic acid as homopolymers or as segments within single RNA molecules.

Biophys. J. 77, 3227-3233 (1999).

16. Meller, A. & Branton, D. Single molecule measurements of DNA transport through a

nanopore. Electrophoresis 23, 2583-2591 (2002).

17. Ayub, M. et al. Nanopore/electrode structures for single-molecule biosensing.

Electrochimica Acta 55, 8237-8243 (2010).

18. Chen, M., Khalid, S., Sansom, M.S. & Bayley, H. Outer membrane protein G:

Engineering a quiet pore for biosensing. Proc Natl Acad Sci U S A 105, 6272-6277

(2008).

19. Butler, T.Z., Pavlenok, M., Derrington, I.M., Niederweis, M. & Gundlach, J.H.

Single-molecule DNA detection with an engineered MspA protein nanopore. Proc

Natl Acad Sci U S A 105, 20647-20652 (2008).

20. Mayer, M., Semetey, V., Gitlin, I., Yang, J. & Whitesides, G.M. Using Ion Channel-

Forming Peptides to Quantify Protein−Ligand Interactions. Journal of the American

Chemical Society 130, 1453-1465 (2008).

21. Menestrina, G. & Sera, M.D. Pore-forming Peptides and Protein Toxins. (Taylor &

Francis Group, London and New York; 2003).

22. Schneider DNA Translocation through Graphene Nanopores. Nano letters 10, 3163-

3167 (2010).

23. Smeets, R.M.M. et al. Salt dependence of ion transport and DNA translocation

through solid-state nanopores. Nano Letters 6, 89-95 (2006).

24. Li, J. et al. Ion-beam sculpting at nanometre length scales. Nature 412, 166-169

(2001).

25. Yusko, E.C. et al. Controlling protein translocation through nanopores with bio-

inspired fluid walls. Nat Nanotechnol 6, 253-260 (2011).

26. Rosenstein, J.K., Wanunu, M., Merchant, C.A., Drndic, M. & Shepard, K.L.

Integrated nanopore sensing platform with sub-microsecond temporal resolution. Nat

Methods 9, 487-492 (2012).


86

27. Albrecht, T., Gibb, T. & Nuttal, P. Engineered Nanopores for Bioanalytical

Applications. (Elsevier, Oxford, UK; 2013).

28. Smeets, R.M.M. Salt-dependence of ion transport and DNA translocation through

solid-state nanopores. Nano Lett. 6, 89-95 (2006).

29. Firnkes, M., Pedone, D., Knezevic, J., DoÉblinger, M. & Rant, U. Electrically

Facilitated Translocations of Proteins through Silicon Nitride Nanopores: Conjoint

and Competitive Action of Diffusion, Electrophoresis, and Electroosmosis. Nano

Letters 10, 2162-2167 (2010).

30. Panja, D., Barkema, G.T. & Kolomeisky, A.B. Through the eye of the needle: recent

advances in understanding biopolymer translocation. J Phys Condens Matter 25,

413101 (2013).

31. Salieb-Beugelaar, G.B., Dorfman, K.D., van den Berg, A. & Eijkel, J.C.

Electrophoretic separation of DNA in gels and nanostructures. Lab Chip 9, 2508-2523

(2009).

32. Meller, A., Nivon, L., Brandin, E., Golovchenko, J. & Branton, D. Rapid nanopore

discrimination between single oligonucleotide molecules. Proc. Natl Acad. Sci. USA

97, 1079-1084 (2000).

33. Clarke, J. et al. Continuous base identification for single-molecule nanopore DNA

sequencing. Nat Nano 4, 265-270 (2009).

34. Howorka, S., Cheley, S. & Bayley, H. Sequence-specific detection of individual DNA

strands using engineered nanopores. Nature Biotechnol. 19, 636-639 (2001).

35. Vercoutere, W. Rapid discrimination among individual DNA hairpin molecules at

single-nucleotide resolution using an ion channel. Nature Biotechnol. 19, 248-252

(2001).

36. Wanunu, M., Chakrabarti, B., Mathé, J., Nelson, D.R. & Meller, A. Orientation-

dependent interactions of DNA with an alpha-hemolysin channel. Phys Rev E Stat

Nonlin Soft Matter Phys 77, 031904 (2008).

37. Venkatesan, B.M. & Bashir, R. Nanopore sensors for nucleic acid analysis. Nat Nano

6, 615-624 (2011).


87

38. Movileanu, L. Detecting protein analytes that modulate transmembrane movement of

a polymer chain within a single protein pore. Nature Biotechnol. 18, 1091-1095

(2000).

39. Fologea, D., Ledden, B., McNabb, D.S. & Li, J. Electrical characterization of protein

molecules by a solid-state nanopore. Appl Phys Lett 91, 539011-539013 (2007).

40. Kasianowicz, J.J., Henrickson, S.E., Weetall, H.H. & Robertson, B. Simultaneous

Multianalyte Detection with a Nanometer-Scale Pore. Analytical Chemistry 73, 2268-

2272 (2001).

41. Han, A. Sensing protein molecules using nanofabricated pores. Appl. Phys. Lett. 88,

093901 (2006).

42. Sexton, L.T. et al. Resistive-Pulse Studies of Proteins and Protein/Antibody

Complexes Using a Conical Nanotube Sensor. Journal of the American Chemical

Society 129, 13144-13152 (2007).

43. Saleh, O.A. & Sohn, L.L. Direct detection of antibody-antigen binding using an on-

chip artificial pore. Proc Natl Acad Sci U S A 100, 820-824 (2003).

44. Smeets, R.M.M., Kowalczyk, S.W., Hall, A.R., Dekker, N.H. & Dekker, C.

Translocation of RecA-Coated Double-Stranded DNA through Solid-State

Nanopores. Nano Letters 9, 3089-3095 (2008).

45. Nestorovich, E.M., Danelon, C., Winterhalter, M. & Bezrukov, S.M. Designed to

penetrate: Time-resolved interaction of single antibiotic molecules with bacterial

pores. Proceedings of the National Academy of Sciences of the United States of

America 99, 9789-9794 (2002).

46. Astier, Y., Braha, O. & Bayley, H. Toward Single Molecule DNA Sequencing: Direct

Identification of Ribonucleoside and Deoxyribonucleoside 5‘-Monophosphates by

Using an Engineered Protein Nanopore Equipped with a Molecular Adapter. Journal

of the American Chemical Society 128, 1705-1710 (2006).

47. Oukhaled, G. et al. Unfolding of proteins and long transient conformations detected

by single nanopore recording. Physical Review Letters 98 (2007).

48. Striemer, C.C., Gaborski, T.R., McGrath, J.L. & Fauchet, P.M. Charge- and size-

based separation of macromolecules using ultrathin silicon membranes. Nature 445,

749-753 (2007).


88

49. Storm, A.J. Fabrication of solid-state nanopores with single-nanometre precision.

Nature Mater. 2, 537-540 (2003).

50. Schmidt, C., Mayer, M. & Vogel, H. A chip-based biosensor for the functional

analysis of single ion channels. Angew. Chem. Int. Edn 39, 3137-3140 (2000).

51. Jen Lo, C. Fabrication of symmetric sub-5 nm nanopores using focused ion and

electron beams. Nanotechnology 17, 3264 (2006).

52. Wei, R., Pedone, D., Zürner, A., Döblinger, M. & Rant, U. Fabrication of metallized

nanopores in silicon nitride membranes for single-molecule sensing. Small 6, 1406-

1414 (2010).

53. Zhang, W. et al. Controllable shrinking and shaping of silicon nitride nanopores under

electron irradiation. Applied Physics Letters 90 (2007).

54. Chansin, G. et al. Resizing Metal-Coated Nanopores Using a Scanning Electron

Microscope. Small 7, 2736-2741 (2011).

55. Tabard-Cossa, V., Trivedi, D., Wiggin, M., Jetha, N.N. & Marziali, A. Noise analysis

and reduction in solid-state nanopores. Nanotechnology 18 (2007).

56. Wanunu, M. Nanopores: A journey towards DNA sequencing. Phys Life Rev 9, 125-

158 (2012).

57. Schoch, R.B. & Renaud, P. Ion transport through nanoslits dominated by the effective

surface charge. Applied Physics Letters 86, 253111-253113 (2005).

58. Horne, R.A. & Young, R.P. Electrical conductivity of aqueous 0.03 to 4.0M

potassium chloride solutions under hydrostatic pressure. The Journal of Physical

Chemistry 71, 3824-3832 (1967).

59. Fologea, D. Slowing DNA translocation in a solid-state nanopore. Nano Lett. 5, 1734-

1737 (2005).

60. Storm, A.J. Fast DNA translocation through a solid-state nanopore. Nano Lett. 5,

1193-1197 (2005).

61. Chen, P. et al. Probing Single DNA Molecule Transport Using Fabricated Nanopores.

Nano Letters 4, 2293-2298 (2004).

62. Storm, A.J., Chen, J.H., Zandbergen, H.W. & Dekker, C. Translocation of double-

strand DNA through a silicon oxide nanopore. Physical Review E 71, 051903 (2005).


89

63. Manning, G. A procedure for extracting persistence lengths from light-scattering data

on intermediate molecular weight DNA. Biopolymers 20, 1751-1755 (1981).

64. Steinbock, L.J., Lucas, A., Otto, O. & Keyser, U.F. Voltage-driven transport of ions

and DNA through nanocapillaries. Electrophoresis 33, 3480-3487 (2012).

65. Manning, G.S. The molecular theory of polyelectrolyte solutions with applications to

the electrostatic properties of polynucleotides. Q Rev Biophys 11, 179-246 (1978).

66. Keyser, U. et al. Direct force measurements on DNA in a solid-state nanopore. Nature

Physics 2, 473-477 (2006).

67. Fologea, D., Brandin, E., Uplinger, J., Branton, D. & Li, J. DNA conformation and

base number simultaneously determined in a nanopore. Electrophoresis 28, 3186-

3192 (2007).

68. Brun, L. et al. Dynamics of polyelectrolyte transport through a protein channel as a

function of applied voltage. Phys Rev Lett 100, 158302 (2008).

69. Wanunu, M., Morrison, W., Rabin, Y., Grosberg, A.Y. & Meller, A. Electrostatic

focusing of unlabelled DNA into nanoscale pores using a salt gradient. Nat

Nanotechnol 5, 160-165 (2010).

70. Henrickson, S.E., Misakian, M., Robertson, B. & Kasianowicz, J.J. Driven DNA

transport into an asymmetric nanometer-scale pore. Phys Rev Lett 85, 3057-3060

(2000).

71. Smith, D.E., Perkins, T.T. & Chu, S. Dynamical scaling of DNA diffusion

coefficients. Macromolecules 29, 1372-1373 (1996).

72. Oukhaled, A. et al. Dynamics of completely unfolded and native proteins through

solid-state nanopores as a function of electric driving force. ACS Nano 5, 3628-3638

(2011).

73. Mihovilovic, M., Hagerty, N. & Stein, D. Statistics of DNA capture by a solid-state

nanopore. Phys Rev Lett 110, 028102 (2013).

90

Chapter 4

Probing DNA 5’-Cytosine Methylation in Breast Cancer Cell Lines

4.1 Background ................................................................................................................................................. 91

4.2 Experimental Objectives ............................................................................................................................. 96

4.3 Results and Discussion ................................................................................................................................ 97

4.4 Conclusion................................................................................................................................................. 114

4.5 References ................................................................................................................................................. 116

Synopsis: Forkhead box (FOX) transcription factors, including FOXA1, play critical roles in cell

proliferation and act as either tumour suppressors or oncogenes. Epigenetic modification of the FOXA1 gene

such as DNA methylation has generated considerable interest as a biomarker for monitoring of breast cancer

development. In this study, solid-state nanopores were utilised as fast and robust biosensors to characterise

FOXA1 methylation. Using biological assays, the significance of FOXA1 methylation in various breast cancer

cell-lines is demonstrated. Subsequently, an in vitro methylated FOXA1 promoter is detected at single molecule

level with a Si3N4 nanopore, where the detection is enhanced by forming a complex with a 5-methylcytosine

antibody. Qualitative and quantitative analysis of the translocation process have provided valuable insight into

the characteristics of the methylated DNA and its transport mechanism as a complex through a narrow pore.

Chapter 4 Probing DNA Methylation in Breast Cancer Cell Lines

91

4.1 Background

Breast cancer is the most common female malignancy and the second most common cause of

death in women in the western world.1, 2 Most breast cancers occur as non-familial cases in

the population although a small proportion, about 10%, is inherited.3 Despite advances in

breast cancer treatment, metastasis causes > 90% of cancer deaths.4 Cancer can result from an

accumulation of genetic mutations leading to dysfunction of critical genes, including tumour

suppressor genes.3

4.1.1 Forkhead Box A1 (FOXA1)*

FOX proteins are a family of evolutionary conserved transcriptional regulators defined by a

common DNA-binding domain (DBD) termed the forkhead. FOX family members have been

shown to play roles in cell proliferation, differentiation and metabolic homeostasis and act as

either tumour suppressors or oncogenes.5-7 FOX proteins can both activate and repress gene

expression through the recruitment of co-factors or repressors as well as extensive

interactions with other factors such as p53 and estrogen receptor (ER). Deregulation of FOX

proteins activity directly or indirectly alters regulation of the target genes. Therefore,

characterisation of such molecules is critical for therapeutic purposes. The unique ability of

FOXA family transcription factors to bind to target sites in silent chromatin in a dominant

manner and initiate regulatory events distinguishes them from the rest of transcription factors

in the mammalian genome.7

FOXA1 has generated considerable interest as a biomarker for predicting and monitoring of

cancer development because of its role as a critical component in the hormonal signalling

network required for growth and differentiation of a specific subtype of breast epithelial cells

due to interaction with ER. In addition, high expression of FOXA1 is commonly observed in

tumours arising from these organs, including prostate and ERα-positive breast tumours. 7

FOXA1 binds to a target DNA sequence as a monomer, using a helix-turn-helix motif of 110

amino acids. Thus, FOXA1 is thought to contribute to gene regulation through its ability to

act as a pioneer factor binding to nucleosomal DNA. Williamson and co-workers showed that

* FOXA1also known as hepatocyte nuclear factor 3-alpha (HNF-3A).


92

FOXA1 expression is directly correlated with BRCA1† which was the first gene found to be

associated with breast cancer.6 Chromatin-modifying enzymes may also indirectly control

FOXA1 activity by redirecting FOXA1 to specific regions of the chromatin with histone

(H3K4) methylation and additional transcription factors may regulate the function of DNA-

bound FOXA1 at such sites.8

4.1.2 Epigenetic Modifications

All cells of a multicellular organism carry the same genetic material coded in their DNA

sequences. Nevertheless, due to differential expression of genes, a broad morphological and

diversity (heterogeneity) is displayed in cells. Inheritable modifications in gene activity that

are not caused by changes in the nucleotide sequence of the genetic code, are known as

epigenetic modifications.9, 10

On a molecular level, epigenetic phenomena are derived from a variety of mechanisms and

can be categorized into three main groups: histones modifications, DNA methylation and

changes in the positioning of nucleosomes. These alterations are fundamental in the

regulation of microRNA (miRNA) and small interference RNA (siRNA) expression, ATP-

dependent chromatin remodelling complexes, DNA-protein interactions and polycomb

complexes, X-chromosome inactivation, suppression of transposable element mobility and

embryogenesis. These diverse molecular mechanisms are employed to stabilise the faithful

propagation of epigenetic modification through cell division.9, 11 Epigenetic modifications

such as CpG island hypermethylation, loss of histone acetylation, silencing and mutation of

remodeler subunits implement key roles in cancer initiation and progression, in addition to

classical genetic mutation.11-13

A. Histone Modification

Histones are the main protein components of chromatin and are the key player of gene

regulation in eukaryotic cells. They function by packaging the DNA into an ordered structural

unit called nucleosome.14 The four core histones H2A, H2B, H3 and H4 are grouped into two

H2A-H2B dimers and one H3-H4 tetramer which show significant resemblance in structure

and are highly conserved through evolution. The main features of these similarities are a † BRCA1: Breast cancer associated gene 1


93

common “helix-turn-helix-turn-helix” which facilitates dimerisation and long “tails” at the

end of amino acids where they are subject to post-translation modification including

acetylation of lysines, methylation of lysine and arginine, phosphorylation of serine and

threonine15, 16 as well as ubiquitination, SUMOylation and ADPribosylation11, 17, 18. Histone

modifications have a crucial roles in DNA repair, DNA replication, transcriptional regulation

and alternative splicing.19, 20

B. Nucleosome Positioning

Nucleosomes are the basic unit of DNA packaging consisting of a DNA wrap around a

histone octamer complex, this process makes the DNA sequence inaccessible to binding of

transcription factors. Despite its stable structure, it is shown to undergo a structural re-

arrangement or so called nucleosome dynamics; hence, nucleosome positioning at the

promoter has a direct effect on regulation of transcription. Nucleosome positioning is

governed by three major contributions: i) the DNA sequence dependency the histone protein

core binding affinity, ii) Competitive or cooperative binding of transcription factors, iii) ATP-

dependent nucleosome re-modelling.21, 22

Nucleosome positioning not only determines accessibility of activator and inhibitor protein to

their target DNA sequence, but it is a key player in shaping the methylation landscape; for

instance, histone variants regulate nucleosome positioning and gene expression by protecting

genes against DNA methylation.11, 23 The genes encode subunits of this remodelling

machinery which are not only involved in transcriptional repression by hypermethylation in

cancer cell lines but which can themselves also be regulated by DNA methylation and histone

modification.11, 24

C. DNA Methylation

DNA methylation is the only genetically programmed DNA modification in mammals.9 The

most widely studied epigenetic modification in humans is methylation at the 5’ position of

pyrimidine cytosine rings in the context of CpG dinucleotide sequences which are clustered

in regions of approximately 1-4 kbp called CpG islands. Cytosine methylation is essential

during embryogenesis where methylation levels change dynamically.9,25 CpG islands are

highly abundant at the promoter and first exon of genes, therefore methylation of these


94

sequences can directly regulate gene expression and is associated with gene silencing.

Cancer is the best-studied disease with strong epigenetic modifications. In tumours, a global

hypomethylation of the genome is observed, as a result, transcriptional activation of

oncogenes and induction of chromosome instability is obtained. However this global loss of

methylation is usually accompanied with hypermethylation of CpG islands in the promoter

region of tumour suppressor genes and transcriptional silencing of carcinogenesis associated

genes (see Figure 4.1).26

Figure 4.1: Schematic illustration of the effect of a promoter’s CpG islands hypermethylation in gene

expression. (a) Essential level of cytosine methylation on the promoter segment is required for functionality and

expression of the gene (b) Hyper-cytosine-methylation of promoter results in repression of the gene.

DNA methylation is catalysed by the DNA methyltransferases family (DNTM) that mediates

the transfer of methyl group from a universal methyl donor, so called S-adenosyl methionine

(SAM), to DNA. CpG island hypermethylation can inhibit gene expression by various

mechanisms. It can directly inhibit the binding of transcription factors from their DNA target

site or methylated DNA can promote the recruitment of methyl-CpG-binding domain (MBD)

protein which in turn recruits transcriptional co-repressors such as histone-deacetyling

complexes, polycomb proteins, histone modifying and chromatin re-modelling complexes to

methylated sites.11,27-29

Hypermethylation patterns are tumour-type specific and it is not evident why specific regions

remain unmethylated whereas another region becomes hypermethylated. Esteller and co-

workers proposed that the inactivation of particular genes confer a growth advantage where

clonal selection is acquired.11, 27 In addition to DNA methylation, undoubtedly active global

or gene specific DNA demethylation occurs during development; however its exact

mechanism is still unclear. One possibility is the replacement of methylated cytosine through

an enzymatic cascade which involves glycolase; alternatively it could be due to the activation

of cytidine deaminase which deaminates 5’-methylcytosine.9, 30, 31

Promoter Gene

Methylated CpG islands

Me Me

Hypermethylated CpG islands

Promoter Gene

Gene expression repressedMe Me Me Me Me

(a) (b)


95

Current technologies in DNA methylation analysis:

Current methods for DNA methylation analysis include immunoprecipitation using a

methylcytosine antibody (MeDIP)32, bisulfite sequencing33, methylation-specific polymerase

chain reaction (PCR)34, combine bisulfite restriction enzyme (COBRA)35, hybridisation

arrays36, methylation-sensitive single-nucleotide primer extension (Ms-SNuPE)37, restriction

landmark genome scanning38. These methods require orders of 106 or 107 cells and extensive

preparation of materials prior to the experiments.39 Table 4.1 compares the main principles

of DNA methylation assays and outlines the limitations associated with these techniques.

Table 4.1: Comparison of the main methodologies and principles in DNA methylation analysis. 40-42

In recent years, single molecule sensing of methylated groups has attracted much interest and

has been demonstrated in single molecule real time (SMRT) DNA sequencing43, nanofluidic

channels44 and nanopore sensors45-48. In principle, these novel techniques are able to

overcome the limitations of conventional molecular biology methods mentioned above. In

Methodology Techniques Principles Throughput Advantages Limitations

Methylation

Sensitive

Restriction

Enzymes

• HpaII-MsPI cleavage

• SmaI-XmaI cleavage

• McrBC cleavage • HELP Assay • RLGS

Degradation of DNA by restriction enzymes to discriminate and/or enrich methylated or unmethylated DNA

• Individual gene/ locus

• Genomic (coupled to high throughput sequencing)

• High sensitivity

• Versatile restriction enzymes

• Providing methylation data only at restriction enzyme recognition site or adjacent region

• False positive caused by incomplete digestion

Affinity

enrichment

and

Purification

• MeDIP-PCR

• ChIP-Seq

• ChIP on Chip

• MIRA

The methylated or unmethylated fractions of genomic DNA can be immunoprecipitated by antibodies (e.g 5’mc Antibody) or proteins (e.g MBD2, MeCP2)

• Genomic • No chemical modification

• Rapid and straight forward experiment and analysis

• Exact methylation state of individual CpG sites cannot be determined

Bisulfite

Conversion

• Direct (sanger) Bisulfite sequencing

• MSP

• Pyrosequencing

Selective deamination of cytosine but not 5-methylcystosine with sodium biuslfite treatment, followed by sequencing

• Genomic

• Provide DNA methylation information at single nucleotide resolution

• Degradation of DNA fragments

• False positive caused by incomplete conversion of C to U

Mass

Spectrometry

• MassArrayEpiTyPER

Gene-specific amplification of bisulfite treated DNA, followed by in vitro transcription , base specific-RNA cleavage coupled to MALDI-TOF analysis

• Individual gene/locus

• Accurate quantitive result at multiple CpG dinulcotides

• High resolution survey is limited to throughput

• High cost for performing large scale DNA methylation


96

addition, no amplification and modification of nucleotides are required; yet they allow high

throughput DNA methylation analysis at the level of nucleotide bases.


Abnormal epigenetic changes, including DNA hypermethylation of FOXA1 gene have been

implicated in breast cancer development. Thereby, DNA methylation of this gene has

potential clinical utility in breast cancer diagnosis. In the present study, solid-state nanopores

were used to study the methylation status of FOXA1 promoter as an alternative approach to

conventional assays. Nanopore chips can be utilised as an inexpensive, robust, fast and

portable biosensor for single molecule epigenetic analysis, as well as benefiting us to observe

fine features that are masked or inadvertently biased in ensemble-averaged (bulk) studies. In

addition, they allow real-time monitoring of DNA methylation as a biomarker to screen for

the prognosis of breast cancer. However, sensing of small methyl groups on a DNA segment

with a 30-40 nm pore is not feasible, therefore, the 5’-methyl cytosine (5’-mc) antibody is

utilised as a label to enhance the electrical detection of CpG islands.

Figure 4.2: Hypothetical illustration of a current-time trace upon an electrokinetically driven of (translocation)

of (a) methylated (CH3) DNA with 3 methylated CpG regions and (b) CH3-DNA-5’-mc antibody complex

through a Si3N4 nanopore. The assigned sub-peaks represent the sites where an antibody is bound to methylated

CpGs.

- +

A

Si 3

N4

CG CG

GCGC

CG

GC

1 32

CG CG

GCGC

CG

GC

- +

A

1 32

Si 3

N4

Methylated DNA-antibody

1 32

Methylated DNA

(a)

(b)

I

I

t

t


97

The hypothetical electrophoretic transport of a methylated ds-DNA and (methylated) DNA-

antibody complex through a solid-state nanopore is illustrated in Figure 4.2. In this

technique, bisulfite conversion, sequencing or fluorescent tags are not required. The main

principle behind this methodology is similar to the “affinity enrichment and purification”

method presented in Table 4.1.

4.3 Results and Discussion

4.3.1 Effect of Cytosine Methylation on FOXA1 Expression

To begin with the methylation level of the FOXA1 promoter in breast cancer cell lines is

evaluated using the MedIP assay. The key steps of this technique are illustrated in Figure 4.3,

including, i) cell collection and lysis, ii) DNA extraction and purification, iii) DNA

sonication (100-600bp) and ds-denaturation to facilitate binding of antibodies to DNA

segments, iv) capturing of hypermethylated sheered DNA with magnetic beads and 5’-mc

antibodies, v) isolation of captured DNA by magnetic rack, vi) purification of DNA for PCR

analysis and other downstream applications.

Figure 4.3: Schematic of a MeDIP assay key stages: Following the cell lysis and DNA extraction, the genomic

DNA is sonicated to 100-600 bp fragments and then denatured at 95 ˚C to generate ss-DNA. Subsequently the

ss-DNA is incubated with the 5'-mc antibody which is already bound to specific magnetic beads. The enriched

DNA is precipitated and isolated by a magnet. At the end, the DNA is purified and prepared for PCR analysis.

Antibody specific

magnetic bead

Methyl group

Cell lysis & DNA

extraction

Sonication

Denaturation 5’mc Antibody

binding

Immunoprecipitation

CG CG

GCGC

CG

GC

Methylated

DNA Isolation

before PCR


98

In this study, the methylation level of the FOXA1 gene in two breast cancer cell lines of

MCF-7 and MLET-2 are compared. The MLET-2 cell line is an endocrine resistant cell line;

more specifically it is a tamoxifen‡ resistant clone of the MCF-7 cell line.

MeDIP was performed on 106 cultured cells, before incubation with the antibody. 10% of

sheered and denatured DNA were kept as a positive control (input). The rest of the samples

were enriched overnight with 5’-mc antibodies and then with antibody specific magnetic

beads at 4˚C. Furthermore, Taq-polymerase PCR was carried out on isolated DNA samples,

followed by an agarose gel electrophoresis. Two faint bands were observed on MeDIP DNA

samples (see Figure 4.4.a), which confirmed the binding of the 5’-mc antibody to the FOXA1

promoter. In addition, the corresponding band for FOXA1 is slightly more intense in MLET-2

compared to MCF-7 cells, which indicates a higher methylation level and perhaps,

consequently lower transcriptional activity of FOXA1 in endocrine resistant cell-lines. In

addition, gels analysis was carried out using ImageJ software, where the band-intensity of

MeDIP samples were measured and normalised to the corresponding “input” sample. The

image analysis confirmed that ~ 30% of MLET-2-FOXA1 promoter is methylated, which is

2.5 folds higher than MCF-7-FOXA1 with methylation level of ~ 12%. In this experiment,

GAPDH is used as a negative control as it is known to be unmethylated at the promoter

region.49

Figure 4.4: (a) MeDIP assay of FOXA1 in MCF-7 and MLET-2 cells, followed by ethidium bromide detection

on 1% agarose gel, GAPDH is a negative control gene. (b) qRT-PCR of mRNA of FOAX1 gene in MCF-7 and

MLET-2 cells. The Y-axis values are arbitrary and normalised to the L19 housekeeping gene. (c) Western-blot

analysis of FOXA1 protein in MCF-7 and MLET-2 cells. β-tubulin is a positive control protein.

‡ Tamoxifen is an antagonist of the ER in breast tissue which is currently used in breast cancer treatment.

MCF-7 MLET-20.0

0.2

0.4

0.6

0.8

1.0

1.2

FO

XA

1/L

19

mR

NA

level

(p

.d.u

.)

Cell Type

MC

F-7

ML

ET

-2

FOXA1

β-tubulin

MCF-7 MLET-2

FOXA1

GAPDH

(b)(a) (c)


99

To examine the impact of FOXA1 promoter methylation during the replication process, on

transcriptional and translation activities of this gene, mRNA and protein expression levels

were determined using qRT-PCR and Western-blots analysis respectively. In qRT-PCR

analysis (n = 3), the amount of FOXA1 mRNA was normalised to the L19 housekeeping gene

and, in western-blot analysis, the target protein was normalised to the structural protein, β-

tubulin. The results indicate that the MLET-2 FOXA1 mRNA level was reduced by ~97%

(Figure 4.4.b) and the protein’s expression was downregulated by ~37% (Figure 4.4.c)

compared to MCF-7 cells. Perhaps, the reduction in transcriptional activity and consequently

the reduction in translational activity were resulting from an ~18% rise in methylation levels

in tamoxifen resistant cells (MLET-2), due to transcriptional silencing of the FOXA1 gene by

hypermethylation of CpG islands at the promoter region. Moreover, the above findings show

that a small rise in promoter methylation level leads to a significant repression in

transcriptional activity.

4.3.2 In- vitro Methylation of FOXA1 Promoter

Prior to the detection of FOXA1-promoter methylation using solid-state nanopores, the

promoter must be isolated from the genomic DNA. In this study, long range PCR is used to

amplify the full length of the FOXA1 promoter (3.4 kbp) from MCF-7 genomic DNA

template. The resulting PCR product was not methylated anymore, as no methylated primers

were used. Nevertheless, it is not feasible to maintain the inherent cellular methylation status

with mentioned method. Therefore, following up the PCR purification, the newly synthesised

FOXA1 promoters are fully methylated in-vitro using the CpG methyltransferase (M.ssI)

enzyme. Figure 4.5 represents the reaction scheme of 5’ cytosine methylation using SAM as a

universal donor of methyl groups.

Figure 4.5: Scheme of 5’ cytosine (in-vitro) methylation reaction using SAM as a methyl donor and M.SssI

enzyme as a catalyst.

M.SssI


100

To investigate the efficacy of the methylation reaction, digestion with the methylation

sensitive restriction enzyme- HpaII was performed. The HpaII restriction site is shown in

Figure 4.6. The methyl group on CG dinucleotide prevents HpaII from accessing its

restriction site.

Figure 4.6: Restriction site of methyl sensitive HpaII enzyme.

The gel electrophoresis in Figure 4.7 compares HpaII digestion of unmethylated and

methylated FOXA1 promoter. As it is shown, methylated (CH3-) DNA (lane 5) was resistant

to HpaII and the corresponding band remained intact, whereas unmethylated DNA (lane 3)

was digested (fragmentised; < 1 kbp) with HpaII enzyme. The negative controls are Lane 2

and 4, representing the unmethylated and methyalted FOXA1 promoter respectively, where

no HpaII enzyme was added. Overall, this simple experiment provided sufficient information

confirming that full methylation of the target DNA had been achieved.

Figure 4.7: 1% agarose gel electrophoresis of HpaII digested in-vitro methylatated FOXA1 promoter: Lane1:1

kbp DNA ladder (New England BioLabs), 1 kbp and 3 kbp fragments are indicated. Lane 2: unmethylated

FOXA1 promoter (long range PCR product). Lane 3: unmethylated FOXA1 promoter + HpaII. Lane 4:

methylated FOXA1 promoter Lane 5: methylated FOXA1 promoter + HpaII.

5’...CCGG...3’

3’...GGCC...5’

3 kbp

1 kbp

1 2 3 4 5


101

4.3.3 Formation of DNA-Antibody Complex

In vitro methylated FOXA1 promoter was incubated with the 5’-mc monoclonal antibody§

(33D3 clone; Aviva systems biology) at 1:9 molar ratio of DNA: antibody in 100mM KCl**-

Tris-HCl (pH 8.5), for 2<= hr at 37 ˚C. The affinity of the antibody to 5’ methyl cytosine

groups has been already confirmed by MeDIP assay in section 4.3. In addition, an

electrophoretic mobility shift assay (EMSA) was performed to assess the specificity of the

DNA-antibody interaction. An overview of the detection procedure in a gel shift assay is

depicted in Figure 4.8.a. Briefly, the electrophoretic separation of a DNA-antibody mixture

was performed on a gel membrane for a short period. The speed at which different DNA-

Antibody complex molecules migrate through the gel is determined by their size, charge, and,

to a lesser extent, their confirmation and shape. Hence, assuming that the antibody is capable

of binding to the methylated DNA, the band representing the complex exhibits a lower

mobility (i.e. “shifted”) compared to free (unbound) DNA in solution.

Figure 4.8.b shows the EMSA of the CH3-FOXA1-5’mc antibody complex on a 0.4% agarose

gel, where a 2 V/cm electric field was applied for 4-5 hr. Lanes 1-3 are the controls:

unmethylated FOXA1, mixture of unmethylated FOXA1+ 5’-mc antibody and 5’-mc

antibody respectively. One band was detected in lane 1, representing the 3.4 kbp FOXA1

promoter. No shifted band was observed in lane 2, implying that there was no unspecific

interaction between 5’-mc and DNA fragments. As expected, no band was detected with

GelRed†† staining in lane 3, confirming that there was no nucleic acid cross-contamination.

The band-intensity analysis indicated that, there was ~15 ng DNA in each of the lanes 1 and

2. Lanes 4 and 5 contain CH3-FOXA1 promoter and mixture of CH3-FOXA1 promoter + 5’-

mc antibody respectively. As indicated with an arrow, a fraction of the DNA in lane 5

exhibits almost no mobility, which may represent the complex of DNA-antibody. Further

image analysis showed that the single band in lane 4 contains ~ 40 ng DNA, the lower band

of lane 5 contains ~21 ng unbound DNA and the upper band contains ~ 15 ng of bound DNA,

hence ~ 60% binding/complex formation in 100 mM KCl-10 mM Tris-HCl (pH 8.5) can be

§Mouse IgG1 isotype (150 kDa) ** Physiological ionic strength †† GelRed is an intercalating nucleic acid stain.


102

estimated. Nevertheless, the electrophoresis findings of CH3-FOXA1 promoter + 5’-mc

antibody mixture (lane 5) were counterintuitive, as the complex band (lane 5, the upper band)

showed no mobility even after a longer period or opposite bias electrophoresis. Besides, as

illustrated in Figure 4.8.a, multiple shifted-bands (blue bands) or a smear were expected, as

the number of antibodies bound to a DNA molecule could vary. Although, if an equilibrium

was reached in a way that the same number of antibodies were bound per DNA molecule, a

single complex band would be have been expected. Furthermore, the effective surface charge

and/or the total molecular weight may have significantly been altered upon binding of the

antibody to DNA, hence no electrophoretic mobility was observed.

Figure 4.8: Electrophoretic mobility shift assay (EMSA). (a) The schematic of EMSA with CH3-DNA

fragments and 5’mc antibodies. The first (left) lane: the electrophoresis of CH3-DNA (black) without antibody.

The second (right) lane: the electrophoresis of the CH3 after incubation with 5’mc antibody. The unbound DNA

fragments (grey) migrates at the same speed as the first lane and CH3 DNA-antibody complexes (blue) exhibit

lower mobility. Here 5 configurations of bindings are shown. (b) 0.4% agarose gel electrophoresis at 2V/cm for

4-5 hr on ice. Post-stained with 3x GelRed for 30 min. All samples incubated in 100mM KCl-Tris (pH 8.5), for

2<= hr at 37 ˚C. Lane 1: unmethylated FOXA1 promoter (3.4 kbp; long range PCR product). Lane 2: mixture of

unmethylated FOXA1 promoter +5’mc antibody. Lane 3: 5’mc antibody in (negative control). Lane 4: CH3-

FOXA1 promoter (3.4 kbp). Lane 5: mixture of CH3-FOXA1 promoter + 5’mc antibody. Lower band represents

the fraction of DNA that is not bound to antibody. The upper band (indicated with an arrow) represents the

fraction of DNA that formed a complex with the antibody.

(a) (b)

Dir

ecti

on

of

Mig

rati

on

Complex

1 2 3 4 5


103

In conjunction with the EMSA assay, atomic force microscopy (AFM) of the above samples

was carried out in air on a mica substrate to characterise the binding affinity of 5’-mc

antibody to CH3-DNA. Figure 4.9.a-c represent the topography images of the control samples

in the following order: CH3 FOXA1 promoter, 5’-mc antibody, mixture of FOXA1 promoter

+ 5’-mc antibody. As the latter shows, majority of the unmethylated species aggregated or

significantly folded upon incubation with 5’-mc antibody. The representative images of CH3-

FOXA1 promoter + 5’-mc antibody (the complex) are shown in Figure 4.9.d-f. The bound

sites are indicated by white arrows.

Figure 4.9: AFM topography images (flatten-3-order) of (a) CH3 FOXA1 promoter, (b) 5’-mc antibody

(appeared as small dots) (c) unmethylated FOXA1 promoter + 5’-mc antibody, (d)-(f) CH3-FOXA1 promoter +

5’-mc antibody (white arrows indicate the sites where an antibody is bound). The corresponding height (z) scale

bar is shown underneath of each image.

500 nm 200 nm 500 nm

200 nm300 nm 150 nm

-0.5 0 1.0 -0.2 0 0.6 -0.9 0 1.5

-0.8 0

nm nm nm

1.7nm

-0.5 0 1.2nm nm

-1.0 0 1.3

(a) (b) (c)

(d) (e) (f)


104

The histogram analysis of DNA and antibody sizes are plotted in Figure 4.10.a-c. AFM

images of 151 molecules of 5’-mc antibody determined the antibody’s height and diameter as

0.5 ± 0.1 nm and 21.0 ± 3 nm, respectively. DNA contour length of 1.0 ± 0.2 µm is estimated

from 55 individual fragments, which is in good agreement with predictions (bp = 0.34 nm).

Moreover, the number of antibodies bound to a DNA molecule was counted from 46

individual images of the complex and plotted in Figure 4.10.d. In summary, 7% of DNA

molecules were free (unbound), 11% were bound to one antibody, 24% to two antibodies,

11% to three antibodies, 13% to four antibodies, 13% to five antibodies, 7% to six antibodies

and 13% were bound to more than six antibodies.

Figure 4.10: AFM analysis: (a) Histogram analysis of the antibody height (z direction; n = 87), (b) Histogram

analysis of the antibody (5’-mc) diameter (x-y direction; n = 151) (c) Histogram analysis of DNA (CH3-FOXA1

promoter) contour length (n = 55), (d) Column bar of the number of bound antibodies per DNA molecule (n =

46). Histograms in (a-c) are fitted with Gaussian distribution indicated with blue curves.

0.0 0.3 0.6 0.9 1.2 1.5 1.8 2.10

5

10

15

20

25

30

35

40

45

50

Cou

nt

Antibody Height (nm)

0 5 10 15 20 25 30 35 40 45 500

5

10

15

20

25

30

35

40

45

50

C

ou

nt

Antibody Diamter (nm)

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.00

2

4

6

8

10

12

14

16

18

20

Cou

nt

DNA Counter Length (µµµµm)

(a)

(c)

(b)

(d)

0 1 2 3 4 5 6 >60

2

4

6

8

10

12

14

16

18

20

Co

un

t

Bound Antibodies per DNA Moelcule


105

Even though, AFM data could be informative with regards to studies on binding affinity of

the 5’-mc antibody to the CH3-DNA fragment, it is not possible to resolve the binding

mechanism of the antibody to a DNA duplex. For instance, in theory each monoclonal 5’-mc

antibody has two binding sites (Fab-arms) for 5’-methylcytidine bases, hence a bivalent

binding is expected per antibody. On the other hand, the majority of the AFM studies in the

literature were performed on a single stranded CH3-DNA,50 therefore it is a still unclear how

the antibody binds to the duplex methylated bases of a ds-DNA, which are located opposite

each other, i.e. the CG dinucleotides across each strand. Here, by simply comparing the

sizes of bound antibodies relative to free antibodies, we determined that only one antibody

can bind per duplex CpG site. In the following discussion, this observation, which may be

evident due to the steric effects arising from the binding of two antibodies so closely (2.2 nm

apart), has been taken into account.

Sequencing analysis (see Appendix I) revealed that there are 170 potential binding sites per

strand. For a DNA polymer (P) with multiple binding sites for a ligand (L) such as an

antibody, assuming the sites are independent, the association equilibria that characterise

DNA-antibody interaction can be written as 51

Ë " S< !] ÌËS<

ËS< " S= ]ÁÌ !ËS=

.

.

.

ËS£~< "!S£ ]ÍÌ!ËS£ (4. 1)

where,! is the (macroscopic) association constant , Î Q! Î , and = 170, the number of

(independent) binding sites. If < is described as

/¡0/¡0!/0 (4. 2)

where [P] represents the concentration of free DNA at equilibrium, [L] is the concentration of

free 5’-mc antibody at equilibrium and [PL] represents the concentration of the complex at

equilibrium. 5 can be given by


106

I /¡I0/¡0!/0I! (4. 3)

Assuming there is no cooperative binding‡‡ and ¤ < =Ï £ ,

/¡50/¡5~<0 ¤ /0! (4. 4)

where ¤ is the association constant. Given that DNA concentration is constant, the

concentration of bound antibodies can be found from multiplying [P] by the ratio of moles of

bound antibodies per moles of DNA. [L] can be obtained by subtracting the equilibrium

concentration of bound antibodies from the initial antibody concentration.52 The ratio of /¡50 to [¡5~<0 can be determined from Figure 4.10.d; hence, ¤ ! 3×108 M-1 is estimated,

corresponding to an affinity of 5’-mc antibody to CH3-DNA. Remarkably, this value is of the

same order of magnitude as values reported in the literature53, 54 and the manufacturer’s

specification§§. In general, the larger the association constant or the smaller the dissociation

constant, the more tightly bound the ligand is, or the higher the affinity between the ligand

and the polymer. However, it should be noted that this estimate is uncertain, due to a large

standard deviation associated with statistically small sample population (n = 46) and the

simplification of the binding model. In this analysis, we also assumed that the surface

composition is the same as the solution composition. Hence, there was no preference for

binding of the antibody to the DNA on the mica surface, compared to the solution.

4.3.4 Probing DNA Methylation Using Solid-State Nanopores

Having the specific formation of DNA-Antibody complex confirmed by AFM and EMSA

analysis, a Si3N4 nanopore sensing was employed to probe the methylation level of the

FOXA1 promoter. Figure 4.11 shows the conductance measurements (I-V curves) of a

nanopore fabricated on a ~ 60 nm thick Si3N4 membrane in 1 M and 0.1 M KCl solution.

‡‡ This assumption is in-line with the AFM images which showed that the bound antibodies were not bundled together. §§ According to the manufacturer’s (Aviva System Biology) specification, the binding constant was determined against a single stranded CH3-DNA. This may indicate that 5’-mc antibody exhibits relatively the same affinity towards both ss- and ds-DNA.


107

Figure 4.11: The conductance measurement (IV curve) of a ~ 40 nm pore fabricated on a Si3N4 membrane

(Lpore= ~60 nm) at two KCl concentration of 1 M (black) and 0.1 M (red). (Inset) The SEM image of the same

nanopore that was used in translocation experiments. The I-V measurement was performed at 50 mV/s scan rate

and bias of -500 -500 mV. The IV curve slope yields Gpore ≈ 338 nS (black; 1M KCl) and Gpore ≈ 47 nS (red; 0.1

M KCl).

The pore diameter was estimated by the conductance of the pore at 1 M KCl, Gpore ≈ 338 nS,

hence, based on the equation (3.5) dpore ≈ 40 nm, which was in-line with the SEM image of

the pore shown in the inset. In order to be consistent with the DNA-Antibody complex

binding buffer, 0.1 M KCl-Tris-HCl (pH 8.5) was used for the translocation experiment. At

0.1 M KCl, no current-rectification was observed and Gpore ≈ 47 nS, 7-fold lower than

conductance of 1 M KCl was obtained.

-600 -400 -200 0 200 400 600-200

-150

-100

-50

0

50

100

150

200

I (n

A)

Vbias (mV)

1M KCl

0.1M KCl


108

The typical I-t traces recorded during translocation of each analyte are displayed in Figure

4.12.

Figure 4.12: Detection of CH3-DNA, CH3-DNA-Antibody complex and 5’-mc antibody using a solid-state

nanopore. The figure displays the representative ionic current traces and the typical individual translocation

events observed during translocation of each analyte. I-t traces were recorded at 0.1 M KCl-Tris-HCl (pH 8.5),

at room temperature, sampled at 200 kHz and low-pass (Bessel) filtered at 10 kHz. (a) CH3-DNA molecules

were detected at 500 mV and current blockades observed. (b) CH3-DNA-Antibody complex molecules were

detected at -500 mV and current blockades observed. No translocation events were detected at 500 mV. (c)

Translocation of 5’-mc antibody molecules was only observed at -1000 mV with current enhancement

characteristics. No translocation events were detected at lower Vbias, including ± 500 mV.

The experimental setup and methodology of this translocation experiment are described in

chapter 2. Briefly, CH3-DNA was introduced to the cis chamber***. A 500 mV positive

voltage was applied to the trans side resulting in the passage of ds-DNA through the

nanopore. Current blockages were observed upon translocation of the DNA molecules. After

the DNA translocation, the nanopore, cell chambers and PDMS rings were disassembled and

*** RE was connected to the ground and inserted into the cis chamber. WE was inserted in the trans chamber and the potential bias was applied at the WE.

Vbias = 500 mV

IO ≈ 24 nA

Current blockade

Vbias = - 1000 mV

IO ≈ -48 nA

Current enhancement

Vbias = - 500 mV

IO ≈ - 24 nA

Current blockade

CH3-DNA

5’-mc antibody

DNA-Antibody complex


109

cleaned thoroughly with Piranha solution, followed by plasma cleaning to remove any trace

of DNA. Before addition of the new analyte, another I-t trace was monitored on the fresh

blank solution to ensure that there was no cross contamination or remaining DNA residues.

Then, the complex††† mixture was added to the cis chamber, which was already incubated in

100 mM KCl-Tris-HCl (pH 8.5), for 2<= hr at 37 ˚C. Surprisingly, no translocation events

were observed below -500 mV, perhaps due to the entopic cost associated with the larger

structure of the DNA-Antibody complex compared to ds-DNA. Lastly, the same cleaning

procedure as described above was repeated and the 5’-mC antibody was added to the cis

chamber and current-enhancements were detected at -1000 mV. No translocation events were

observed at applied positive bias or below -1000 mV.

Somewhat surprisingly was the sign of current modulation of 5’-mc antibody translocations.

In a simple volume-exclusion model, current blockade is expected due to the exclusion of

electrolyte solution by the (uncharged) analyte. On the other hand, current enhancement is

observed in translocation of highly charged analyte due to an increase in ionic flux upon

translocation. The 5’-mc antibody is an isotype of IgG1 group. The isoelectric point (pI) of

IgG1 antibodies spans the range of 6.1-8.5 (7.3±1.2),55 thus in electrolyte of pH 8.5, the 5’-

mc antibody net charge is predicted to be neutral or (slightly) negatively charged. Therefore,

in this study, the current-enhancement cannot have resulted from the high charge of the

analyte. Perhaps, a combination of transient effects, such as concentration polarisation,

(partial) separation of the antibody from its solvation ion cloud upon the entry to the pore or

the adsorption to the inner pore surface are the prevalent reasons for the current increase

observed.56

††† 1: 9 molar ratio of CH3- DNA to 5’-mc antibody


110

Figure 4.13: The histogram analysis of 5’-mc antibody translocations with a ~40 nm pore, at -1000 mV in 100

mM KCl-TrisHCl (pH 8.5). (a) Dwell time (τd) and (b) current-enhancement (∆I) distributions (n = 1001). The

(stretched) Gaussian fits are indicated with navy curves.

The histogram analysis of 5’-mc translocation is shown in Figure 4.13, where the most

probable τd = 0.21 ± 0.05 ms and the most probable ∆I = 430 ± 50 pA are obtained.

Furthermore, the event-rate analysis (normalised to the solution concentration) in Figure 4.14

shows that the frequency of 5’-cm antibody events (3×108 s-1M-1) to be an order of

magnitude lower than that of CH3-DNA (6×109 s-1M-1) and complex (3×109 s-1M-1). This is

against expectation, as a 2-fold larger electric-field was applied in the 5’-mc translocation

experiments, compared to the other two analytes.

Figure 4.14: Frequency of events (s.M)-1 analysis of 1000 events of each individual analyte. Translocation of

CH3-DNA performed at 500 mV, 5’-mc antibody at -1000 mV and the CH3-DNA-Antibody complex at -500

mV in 100 mM KCl-TrisHCl (pH 8.5).

The low event-rate can be rationalised by the fact that, at pH 8.5, the 5’-mc antibody

molecules are not highly charged species, therefore only a fraction of them were driven by

0 200 400 600 800 1000 1200 14000

20

40

60

80

100

120

140

160

180

200

Nu

mb

er o

f E

ven

ts

∆∆∆∆I (pA)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.00

20

40

60

80

100

120

140

160

180

200

Nu

mb

er o

f E

ven

ts

ττττd (ms)

(a) (b)

CH3-DNA Antibody Complex

1x109

2x109

3x109

4x109

5x109

6x109

7x109

8x109

F

req

uen

cy o

f E

ven

ts (

s M

)-1


111

the electric field. This effect also explains why such a large electric field (17 MV/m) was

required to capture the antibodies. In addition, depending on the ionic strength and pH of the

solution, nearly all (macro) biomolecules can adopt a variety of confirmations. Therefore, as

a result of low signal-to-noise ratio at lower ionic strengths and limited temporal resolution,

the fraction (conformations) of the molecules which traverse at high velocity cannot be

detected. This phenomenon has previously been reported for protein translocations.57

Event flux was lower for the complex species, in comparison with CH3-DNA molecules.

Considering the fact that in this study, a relatively large pore-dimension was utilised, a

diffusion-limited regime for the capture of the analytes can be speculated here. Since the flux

of analytes is determined by their diffusion coefficients, as well as the concentration gradient

a lower capture rate for larger structures such DNA-Antibody complex was expected, as their

diffusion coefficient is smaller than that of DNA molecules.58

Figure 4.15: The nanopore translocation data (n = 1001) of CH3 DNA and the complex in 100 mM KCl-

TrisHCl (pH 8.5). Event number density plots (2-D histogram of ∆I vs. τd ) of (a) CH3 DNA at 500 mV and (b)

the complex at -500 mV. 2-D histograms are normalised to 1 and the point densities are colour coded from blue

(low) to red (high). Comparison of (c) τd and (d) ∆I histograms of CH3-DNA (black) and the complex (blue).

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.00

200

400

600

800

1000

1200

1400

ττττd (ms)

∆∆ ∆∆I

(pA

)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.00

200

400

600

800

1000

1200

1400

ττττd (ms)

∆∆ ∆∆I

(pA

)

(b)(a)

(c) (d)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.00

20

40

60

80

100

120

140

160

180

200

Nu

mb

er o

f E

ve

nts

ττττd (ms)

CH3-DNA

Complex

0 200 400 600 800 1000 1200 14000

20

40

60

80

100

120

140

160

180

200 CH3-DNA

Complex

Nu

mb

er o

f E

ven

ts

∆∆∆∆I (pA)

CH3-DNA Complex


112

The different characteristics of the CH3-DNA and the complex current blockade events

translocations are evident in the event density plots (see Figure 4.15.a,b). The events

durations of DNA translocation events are highly clustered and the current modulations are

relatively distributed over a wider range. In contrast, the translocation events of the complex

showed a broader range of τd and more clustered ∆I, in relation to DNA. The most probable

(max) and mean τd and ∆I values obtained from the histogram analysis (Figure 4.15.c,d) are

presented in Table 4.2.

Table 4.2: Summary of blockade events parameters from the histogram analysis (n = 1001). The errors denote

the standard deviation resulting from the fitting procedure.

Analyte τd max (ms) τd mean (ms) ∆I max (pA) ∆I mean (pA)

CH3-DNA 0.1 ± 0.02‡‡‡ ~ 0.8 390 ± 60 ~ 510

Complex 0.30 ± 0.08 ~ 1 360 ± 50 ~ 420

Surprisingly, the magnitude of ∆I (max) was not affected (within the experimental error)

during translocation of the complex species. This observation differs from Shim et al.’s report

where, they recorded a ~3-fold increase in ∆I during translocation of CH3-DNA-protein

(MBD) complex.46 However, it should be noted that their experimental conditions differ

from those in this study.§§§ According to the AFM analysis (see Figure 4.10), the average

size-ratio of the antibody to DNA is 21/1000, implicating that perhaps current modulations

resulted from the complex translocation, mainly correspond to the pore-volume occupied by

the DNA rather than the antibody. In addition, above observation can be a manifestation of

the earlier hypothesis that the DNA and the complex event-rate is diffusion-limited, as the

bound antibodies to the DNA did not change the structural geometries of complex molecules

significantly, in comparison to free DNA molecules.

Moreover, as mentioned earlier, the Si3N4 pore walls are negatively charged, hence, in

addition to electrophoresis (EP), the electrosmosis (EO) contributes to the effective velocity

‡‡‡ The values obtained for the most probable dwell time, in particular in case of DNA are highly affected by the bandwidth (10 kHz low pass filter) of the instrument during recording, hence limited temporal resolution.

§§§ 871 bp ds-DNA (36 CpG sites), MBD (75 amino acids) as the methyl group label, ~12 nm Si3N4 pore , 600 mM KCl (pH 8.0) and 600 mV bias.


113

or even alters the passage direction. In the current study, the analyte was always added to the

cis chamber and the 5’-mc antibody and the complex translocation only occurred at negative

biases. As illustrated in Figure 4.16, at this potential polarity, EO-flow is in the same

direction as the translocation process, implying that the translocation of the antibody and the

complex might be induced by the EO transport, whereas DNA translocation is facilitated by

electrophoresis.

Figure 4.16: Hypothetical illustration of electrophoretic (EP) and electreoosmotic (EO) effects in nanopore

translocations in 100 mM KCl (pH 8.5). The Si3N4 pore walls are negatively charged. (a) EP governed

translocation of CH3-DNA at 500 mV. (b) EO governed translocation of CH3-DNA –Antibody complex at -500

mV. (c) EO governed translocation of 5’-mc antibody at -1000 mV. The nanopore and analyte sizes, as well as

the magnitude of the electokinetic forces are not to scale.

EO driven-translocation had already been observed for protein translocation in SiNx pore,

where Firnkes et al. proposed that the effective velocity (ª±± ) of an analyte inside a pore is

dependent on the zeta (ζ) potentials of the analyte and the pore,59

ª±± Ð "!ÐÑ xJx©b !ÒÓ¤®ÔÕª Ò§¨©ª (4. 5)

where, xJ is the vacuum permittivity,x© is the relative permittivity, b is the external electric

field and is the solution viscosity.

Si 3

N4

- + -+S

i 3N

4

-+

Si 3

N4

- - -

- - -

- - - - - -

- - -- - -

Cis Trans Cis Trans Cis Trans

Electrophoresis

Electroosmosis

5’-mc Antibody

CH3-DNA

+ 0.5 V - 0.5 V - 1.0 V

(a) (b) (c)


114

The translocation time analysis yielded ª±±!!≈ 11 mm.s-1 and ª±±!Ö¨§®ª×!≈ 4 mm.s-1,

assuming, the DNA was fully elongated during the translocation process. The effective

velocity of DNA is comparable with values reported in the literature.60, 61 Consequently, if

≈ 0.001 Pa.s, xJ ≈ 8.854×10-12 F.m-1, !x© ≈ 80 and b ≈ 8.3×106 V.m-1, ∆ζ (ÒÓ¤®ÔÕª Ò§¨©ª of ~2 mV and ~0.7 mV are estimated for the DNA and the complex, respectively. Depending

on the signs and the relative magnitude of ÒÓ¤®ÔÕª and!Ò§¨©ª, EO may enhance or counteract

EP. For an EP regime, ØÒÓ¤®ÔÕªØ \ ØÒ§¨©ªØ. If ØÒ§¨©ªØ \ ØÒÓ¤®ÔÕªØ1 the resulting translocation

direction is electrosomotic. In the case of ØÒÓ¤®ÔÕªØ ØÒ§¨©ªØ1 the voltage independent

diffusion is governed as a result of the concentration gradient between the cis and trans

chambers.59

4.4 Conclusion

This chapter presented a new nanopore-based electrical detection of 5’cytosine methylated

bases using 5’-mc antibody as a label. Here, the promoter region of the FOXA1 gene is

specifically chosen as the target. Using the MeDIP assay, the significance of methylation of

this gene in chemotherapy resistant cell-lines (MLET-2) is verified, where hypermethylation

resulted in transcriptional and translational repression of the gene. Prior to the nanopore

detection, the binding affinity and specificity of 5’-mc antibody to the duplex CH3-DNA was

evaluated by EMSA assay and AFM analysis. Subsequently, the Si3N4 nanopore detection

enabled the probing of the methylated FOXA1 promoter in a format of DNA-antibody

complex. Notably, the translocation of the labelled DNA was only achieved at the bias

polarity opposite to the unlabelled DNA. This change of direction in translocation will

provide a novel platform to separate the mixture of methylated and unmethylated DNA in a

faster and more robust manner compared to current technologies (e.g. MedIP assay) where

further downstream applications can be subjected. It is postulated that EO is the main driving

force of the complex, whereas in DNA translocation, EP is the dominant electokinetic flow.

However, this hypothesis needs to be investigated by direct ζ-potential measurement of the

nanopore and each individual analyte at 100 mM KCl-Tris-HCl (pH 8.5).

As a result of the limited temporal and spatial resolution of the current nanopore devices, it

was not possible to quantify and profile the CpG islands at the promoter regions. Meanwhile,


115

further studies are in progress to enhance the sensitivity of solid-state nanopores by

combination of atomic layer deposition of Al2O3, reducing the nanopore thickness,

introducing a mobile lipid layer, etc.62


116

4.5 References

1. Gonzalez, M.E. et al. Downregulation of EZH2 decreases growth of estrogen

receptor-negative invasive breast carcinoma and requires BRCA1. Oncogene 28, 843-

853 (2009).

2. Naderi, A. & Liu, J. Inhibition of androgen receptor and Cdc25A phosphatase as a

combination targeted therapy in molecular apocrine breast cancer. Cancer Letters

298, 74-87 (2010).

3. Catteau, A. & Morris, J.R. BRCA1 methylation: a significant role in tumour

development? Semin Cancer Biol 12, 359-371 (2002).

4. Song, Y., Washington, M.K. & Crawford, H.C. Loss of FOXA1/2 is essential for the

epithelial-to-mesenchymal transition in pancreatic cancer. Cancer Res 70, 2115-2125

(2010).

5. Shukla, V. et al. BRCA1 affects global DNA methylation through regulation of

DNMT1. Cell Res 20, 1201-1215 (2010).

6. Williamson, E.A. et al. BRCA1 and FOXA1 proteins coregulate the expression of the

cell cycle-dependent kinase inhibitor p27(Kip1). Oncogene 25, 1391-1399 (2006).

7. Myatt, S.S. & Lam, E.W. The emerging roles of forkhead box (Fox) proteins in

cancer. Nat Rev Cancer 7, 847-859 (2007).

8. Nakshatri, H. & Badve, S. FOXA1 in breast cancer. Expert Rev Mol Med 11, e8

(2009).

9. Tost, J. DNA Methylation Methods and Protocols, Vol. 507, Edn. 2nd. (Springer

Protocols, New York; 2009).

10. Mills, K.I. & Ramsahoye, B.H. DNA Methylation Protocols, Vol. 200, Edn. 1st.

(Humana Press Inc., USA; 2002).

11. Portela, A. & Esteller, M. Epigenetic modifications and human disease. Nat Biotech

28, 1057-1068 (2010).

12. Clapier, C.R. & Cairns, B.R. The biology of chromatin remodeling complexes. Annu.

Rev. Biochem. 78, 273-304 (2009).

13. Ho, L. & Crabtree, G.R. Chromatin remodelling during development. Nature 463,

474-484 (2010).


117

14. Lehninger, A.L., Nelson, D.L. & Cox, M.M. Lehninger principles of biochemistry /

David L. Nelson, Michael M. Cox, Edn. 4th edition. (W.H. Freeman, New York;

2004).

15. Alberts, B. Molecular biology of the cell, Edn. 5th. (Garland Science, New York;

2008).

16. Turner, B.M. Defining an epigenetic code. Nat Cell Biol 9, 2-6 (2007).

17. Kouzarides, T. Chromatin modifications and their function. Cell 128, 693-705 (2007).

18. Daujat, S., Zeissler, U., Waldmann, T., Happel, N. & Schneider, R. HP1 binds

specifically to Lys26-methylated histone H1.4, whereas simultaneous Ser27

phosphorylation blocks HP1 binding. J. Biol. Chem. 280, 38090-38095 (2005).

19. Huertas, D., Sendra, R. & Munoz, P. Chromatin dynamics coupled to DNA repair.

Epigenetics 4, 31-42 (2009).

20. Luco, R.F. Regulation of alternative splicing by histone modifications. Science 327,

996-1000 (2010).

21. Teif, V.B. & Rippe, K. Predicting nucleosome positions on the DNA: combining

intrinsic sequence preferences and remodeler activities. Nucleic Acids Res 37, 5641-

5655 (2009).

22. Henikoff, S. Nucleosome destabilization in the epigenetic regulation of gene

expression. Nat Rev Genet 9, 15-26 (2008).

23. Chodavarapu, R.K. et al. Relationship between nucleosome positioning and DNA

methylation. Nature 466, 388-392 (2010).

24. Wysocka, J. et al. A PHD finger of NURF couples histone H3 lysine 4 trimethylation

with chromatin remodelling. Nature 442, 86-90 (2006).

25. Reik, W., Dean, W. & Walter, J. Epigenetic reprogramming in mammalian

development. Science 293, 1089-1093 (2001).

26. Jones, P.A. & Baylin, S.B. The fundamental role of epigenetic events in cancer. Nat

Rev Genet 3, 415-428 (2002).

27. Esteller, M. Epigenetic gene silencing in cancer: the DNA hypermethylome. Hum.

Mol. Genet. 16, R50-R59 (2007).

28. Lopez-Serra, L. & Esteller, M. Proteins that bind methylated DNA and human cancer:

reading the wrong words. Br. J. Cancer 98, 1881-1885 (2008).


118

29. Kuroda, A. et al. Insulin gene expression is regulated by DNA methylation. PLoS One

4, e6953 (2009).

30. Barreto, G. et al. Gadd45a promotes epigenetic gene activation by repair-mediated

DNA demethylation. Nature 445, 671-675 (2007).

31. Morgan, H.D., Dean, W., Coker, H.A., Reik, W. & Petersen-Mahrt, S.K. Activation-

induced cytidine deaminase deaminates 5-methylcytosine in DNA and is expressed in

pluripotent tissues: implications for epigenetic reprogramming. J Biol Chem 279,

52353-52360 (2004).

32. Weber, M. et al. Chromosome-wide and promoter-specific analyses identify sites of

differential DNA methylation in normal and transformed human cells. Nat Genet 37,

853-862 (2005).

33. Zhang, X. et al. Genome-wide High-Resolution Mapping and Functional Analysis of

DNA Methylation in Arabidopsis. Cell 126, 1189-1201 (2006).

34. Fraga, M.F. & Esteller, M. DNA methylation: a profile of methods and applications.

Biotechniques 33, 632, 634, 636-649 (2002).

35. Xiong, Z. & Laird, P.W. COBRA: a sensitive and quantitative DNA methylation

assay. Nucleic Acids Res 25, 2532-2534 (1997).

36. Bibikova, M. et al. High-throughput DNA methylation profiling using universal bead

arrays. Genome Research 16, 383-393 (2006).

37. Gonzalgo, M.L. & Liang, G. Methylation-sensitive single-nucleotide primer extension

(Ms-SNuPE) for quantitative measurement of DNA methylation. Nat. Protocols 2,

1931-1936 (2007).

38. Ando, Y. & Hayashizaki, Y. Restriction landmark genomic scanning. Nat. Protocols

1, 2774-2783 (2007).

39. Cipriany, B.R. et al. Single molecule epigenetic analysis in a nanofluidic channel.

Anal Chem 82, 2480-2487 (2010).

40. Gupta, R., Nagarajan, A. & Wajapeyee, N. Advances in genome-wide DNA

methylation analysis. Biotechniques 49, iii-xi (2010).

41. Shanmuganathan, R. et al. Conventional and nanotechniques for DNA methylation

profiling. J Mol Diagn 15, 17-26 (2013).


119

42. Laird, P.W. Principles and challenges of genomewide DNA methylation analysis. Nat

Rev Genet 11, 191-203 (2010).

43. Flusberg, B.A. et al. Direct detection of DNA methylation during single-molecule,

real-time sequencing. Nat Methods 7, 461-465 (2010).

44. Fang Lim, S. et al. DNA methylation profiling in nanochannels. Biomicrofluidics 5,

34106-341068 (2011).

45. Clarke, J. et al. Continuous base identification for single-molecule nanopore DNA

sequencing. Nat Nanotechnol 4, 265-270 (2009).

46. Shim, J. et al. Detection and quantification of methylation in DNA using solid-state

nanopores. Sci Rep 3, 1389 (2013).

47. Wanunu, M. et al. Discrimination of methylcytosine from hydroxymethylcytosine in

DNA molecules. J Am Chem Soc 133, 486-492 (2011).

48. Li, W.W., Gong, L. & Bayley, H. Single-molecule detection of 5-

hydroxymethylcytosine in DNA through chemical modification and nanopore

analysis. Angew Chem Int Ed Engl 52, 4350-4355 (2013).

49. Clozel, T. et al. Mechanism-based epigenetic chemosensitization therapy of diffuse

large B-cell lymphoma. Cancer Discov 3, 1002-1019 (2013).

50. Zhu, R. et al. Nanomechanical recognition measurements of individual DNA

molecules reveal epigenetic methylation patterns. Nat Nanotechnol 5, 788-791 (2010).

51. Cantor, C.R. & Schimmel, P.R. Biophysical Chemistry: Part III: The Behavior of

Biological Macromolecules, Edn. 1st. (W. H. Freeman and Company, New York;

1980).

52. Uram, J. & Mayer, M. Estimation of solid phase affinity constants using resistive-

pulses from functionalized nanoparticles. Biosensors & Bioelectronics 22, 1556-1560

(2007).

53. Suter, J., Howard, D., Shi, H., Caldwell, C. & Fan, X. Label-free DNA methylation

analysis using opto-fluidic ring resonators. Biosensors & Bioelectronics 26, 1016-

1020 (2010).

54. Luo, Y., Lu, Z., Raso, S.W., Entrican, C. & Tangarone, B. Dimers and multimers of

monoclonal IgG1 exhibit higher in vitro binding affinities to Fcgamma receptors.

MAbs 1, 491-504 (2009).


120

55. Dang, H. & Harbeck, R.J. A comparison of anti-DNA antibodies from serum and

kidney eluates of NZB x NZW F1 mice. J Clin Lab Immunol 9, 139-145 (1982).

56. Japrung, D. et al. Single-molecule studies of intrinsically disordered proteins using

solid-state nanopores. Anal Chem 85, 2449-2456 (2013).

57. Plesa, C. et al. Fast translocation of proteins through solid state nanopores. Nano Lett

13, 658-663 (2013).


158 (2012).

59. Firnkes, M., Pedone, D., Knezevic, J., DoÉblinger, M. & Rant, U. Electrically

Facilitated Translocations of Proteins through Silicon Nitride Nanopores: Conjoint

and Competitive Action of Diffusion, Electrophoresis, and Electroosmosis. Nano

Letters 10, 2162-2167 (2010).

60. Steinbock, L.J., Otto, O., Chimerel, C., Gornall, J. & Keyser, U.F. Detecting DNA

folding with nanocapillaries. Nano Lett 10, 2493-2497 (2010).

61. Li, J., Gershow, M., Stein, D., Brandin, E. & Golovchenko, J. DNA molecules and

configurations in a solid-state nanopore microscope. Nature Materials 2, 611-615

(2003).

62. Bahrami, A., Doğan, F., Japrung, D. & Albrecht, T. Solid-state nanopores for

biosensing with submolecular resolution. Biochem Soc Trans 40, 624-628 (2012).

121

Chapter 5

Characterisation of Homologous Pairing in Closed Circular DNA

5.1 Background ............................................................................................................................................... 122

5.2 Experimental Objectives ........................................................................................................................... 123

5.3 Construction of the DNA Plasmids ........................................................................................................... 124

5.4 Results and Discussion .............................................................................................................................. 125

5.5 Conclusion................................................................................................................................................. 144

5.6 References ................................................................................................................................................. 147

Synopsis: The pairing of homologous ds-DNA is an important step during the early stages of meiosis and

DNA repair. While certain proteins are known to take part in this process at some stage, yet, there is no

explanation of how mutual recognition of intact homologues occurs initially at the chromosomal level. A theory

for the recognition of homologous DNA has been proposed, which is rooted in the electrostatics and sequence-

dependent structure of the DNA. Several works reported experimental indications that intact ds-DNA can indeed

distinguish homology without proteins. In this study, the presence of such homologue-pairing within plasmid

molecules in free solution was investigated using a combination of gel electrophoresis, DLS, AFM, and

nanopore translocation experiments. Two plasmids (cc-DNA) were constructed in such a way that one consisted

of two 1 kbp homologous segments and the other contained no homology in its sequence. However to this end,

the presence of the electrostatic interaction in the homologous regions was inconclusive. Based on contour

length analysis of the AFM data, the homologous sample was dimerised during cloning process and formed a

single loop of 12 kbp cc-DNA. In this chapter, the results obtained during characterisation of each of these

plasmids under various experimental conditions were surveyed. Moreover, further experiments to obtain a better

understating on possible mechanisms of the dimerisation in ds-DNA are proposed.

Chapter 5 Characterisation of Homologous Pairing in Circular DNA

122

5.1 Background

Paring of two homologous (identical sequence) molecules during genetic recombination in

cell division (meiosis) and DNA double strand break-repair mechanism is a well-known and

characterised process. In 1947, Joshua Lederberg demonstrated the first evidence of genetic

recombination in mixed cultures of E.coli mutants.1 In 1964, Robins Holliday developed a

model by proposing the formation of a Holliday junction during the genomic exchange and

cross-over in chromosome (more details in section 1.3.7-8).2 Successive studies developed

several different pathways to model the mechanism of homologous recombination which

involve a family of proteins known as recombinases (see section 1.5.2.D).3, 4 In all these

pathways, the recombination process is initiated by the alignment of two identical or similar

DNA molecules alongside each other. Nevertheless the very first step of “homologous-

sequence recognition” of two DNA segments in the vast network of genetic materials is

probably the least understood step.

It is mainly assumed that the mechanism by which nucleic acids recognize each other is

based on the sequence-complementarity (Watson-Crick model) between ss-DNAs.5-9 Within

this picture, gene-gene recognition should take place at the stage of broken strand exchange.

However, this mechanism is only efficient for fragments of about 10 bases. As a result,

frequent recognition and recombination errors leading to several mutations and

carcinogenesis would be prevalent.10, 11 Thus, one may rationalise the homology recognition

of the pairing process by direct (ds)DNA/(ds) DNA interaction.11

Direct interactions between DNA duplexes has already been shown in organisation and

packaging of chromosomal and viral genetic materials.12 These interactions can occur in a

simple electrolyte solution and seem to be sequence-dependent.13-17 Namely, DNA wraps

around histones at defined sequence tracks.18 It also been found that the nucleosomes

positioning is encoded and controlled by the sequence.19 In addition, biaxial correlation of

cholesteric liquid-crystalline of phage DNA also provides another indication of sequence

dependency of DNA-DNA interactions.20


123

5.1.1 Homology Recognition in DNA Duplexes

Over a decade ago, a theoretical model proposed that homology recognition between intact

double helices does not arise from the classical Watson-Crick base pairing model.21, 22 The

Kornyshev-Leikin (KL) theory of DNA-DNA interaction in solution was outlined in chapter

1 (see section 1.4.4). Briefly, KL theory proposed homology recognition between DNA

duplexes, resulting from sequence-dependent structure of the DNA backbone. The authors

hypothesised a protein-free "snapshot" recognition mechanism from a distance without DNA

unzipping, based on electrostatic interactions of matched patterns of phosphate charges for

homologous tracks in parallel juxtaposition. The resulting modulation of DNA surface

charge enables the homologous duplex longer than 50 bp to recognise each other. This

homology recognition is absent for non-homologous or anti-parallel homologous tracks due

to unfavourable elastic deformation.11

5.1.2 Reported Studies on Homologous DNA Segments Interaction

Several experimental studies have reported indications of electrostatic homology recognition.

For instance, Inoue et al. observed self-assembly of homologous DNA fragments by gel

electrophoresis.23 Baldwin et al. noted formation of cholesteric liquid crystalline aggregated

in fluorescently tagged homologous ds-DNAs (298 bp) in an electrolyte solution with minor

osmotic stress. They reported on the spontaneous segregations of homologues DNAs within

each cholesteric spherulite.24 Danilowicz et al. demonstrated homologous pairing of two ds-

DNA molecules with 5 kbp homology regions in accordance with theoretical predictions25

using a parallel single molecule magnetic tweezers assay. They observed the homology

pairing in the absence of proteins, divalent metal ions and crowding agents.26 Later, Wang et

al. reported structural transition in supercoiled plasmid molecules containing homologous

segments. Using AFM imaging, they observed a dumbbell structure in Paranemic crossover

(PX-) DNA27 molecules.28 However, PX-DNA base pairing has already occurred between the

double strands, presumably as a consequence of an initial electrostatic recognition step. Most

recently, also using AFM, Nishikawa and Ohyama showed that DNA sequence-based

selective associations occur between nucleosomes with identical DNA sequences in presence

of Mg2+ ions.29


124


Despite the above experimental findings, clear and unambiguous evidence of homology

recognition without base pairing for free DNA in solution was missing. This is needed to

exclude surface, matrix, or collective effects, thus establishing that such sequence-dependent

recognition is encrypted in the structure of duplex DNA.

Using a combination of gel electrophoresis, DLS, AFM, and nanopore translocation

experiments, the presence of any favourable and non-base-paired interaction between 1 kbp

long homologous regions of an engineered ds-DNA plasmid in simple electrolyte solutions

and in absence of proteins was assessed.

5.3 Construction of the DNA Plasmids

In order to test whether an attractive interaction between two segments of homologous DNA

exists in solution, two closed loop ds-DNA samples from pET-24a(+) plasmid (5.3 kbp,

Novagen) were first engineered. The full length Kanamycine gene ('Kan', 1 kbp) of the

above-named plasmid was amplified using PCR with specifically designed primers to

generate 3’ and 5’ EcoRI and BamHI overhangs. The amplified fragment was then purified

and inserted in pET-24a (+) vector where it was already digested with EcoRI and BamHI

restriction enzyme.

Figure 5.1: Schematic of the preparation of the parallel and anti-parallel DNA plasmids from the native pET-

24-a(+) plasmid. In the presence of homology recognition, the "parallel" sample is expected to have shape and

physical properties, compared to the control.

Pre-engineered

plasmid (5.3 kbp)

“cut”

Insert

1 kbp DNA (Kan gene) KL-Interaction

Linear DNA

(5.3 kbp)

?

?

“Parallel” DNA

(6.3 kbp)

“Anti-parallel” DNA

(6.3 kbp)


125

As illustrated in Figure 5.1, two recombinants were expected to be fabricated as a result of

the ligation process:

i. Parallel (p-) DNA: the second Kan gene was introduced in parallel to the original Kan

gene of the vector. This configuration would be compatible with the preferential

interaction of homology tracks.

ii. Anti-parallel (ap-) DNA: the second Kan gene was introduced in opposite orientation

(anti-parraell) to original Kan gene of the vector. In this configuration, no favourable

interaction between the two segments is expected

The newly inserted Kan-genes were sequenced and showed the expected configurations. All

samples were eluted in 10 mM Tris-HCl (pH 8.5). Topoisomerase I (Wheatgerm) treatment

was performed to compare the properties of the supercoiled DNA with relaxed samples. In

addition, as a control in mobility analysis, EcoRI digestion carried out to linearise the circular

DNA molecules. The purity of all samples, before and after enzymes treatments was

confirmed by gel electrophoresis and UV-Vis spectroscopy at various stages of the

experimental programme.

In accordance with the design described above, the structure of p-DNA in solution was

expected to be affected by the favourable interaction between the homologous (parallel)

segments. It should be elongated, somewhat ellipsoidal and more rigid when compared to ap-

DNA. The latter was expected to be looser, randomly coiled, on average more spherical.11, 30

This difference in geometry and rigidity of the p- and ap-samples should affect their physico-

chemical properties, which are investigated below.

5.4 Results and Discussion

5.4.1 Atomic Force Microscopy (AFM)

To visualise the structural differences of ap- and p-DNA, the AFM imaging (in air) was

conducted*. Representative images of the relaxed DNA samples, where the supercoiling is

* More than 500 images of ap(T)- and p(T)-DNA were collected by AFM. Collection of these images was carried out by the author, A. Rutkowska, I. Warych and W. Pitchford., the current and former members of the Albrecht group.


126

removed by TOPO treatment are shown in Figure 5.2. The upper panels, (a) and (b) show the

AFM data for ap- and p-DNA TOPO treated species which named apT and pT-DNA

respectively, on a Mg2+-modified mica. The lower panels (c) and (d) show the same DNA

samples but on a silanised (APTES†)-modified mica. Further images of relaxed and

supercoiled samples on both types of mica-modification are presented in Appendix III, which

confirmed that TOPO treated samples were indeed in a relaxed topology.

Figure 5.2: AFM imaging data for relaxed (TOPO treated) plasmids in air. (a) apT-DNA on Mg2+-modified

mica (image size: 2.5 µm by 2.5 µm), (b) pT-DNA on Mg2+-modified mica (image size: 5 µm by 5 µm), (c)

apT-DNA on APTES +-modified mica (image size: 1.5 µm by 1.5 µm), (d) pT-DNA on APTES +-modified mica

(image size: 1.0 µm by 1.0 µm).

Immobilisation of DNA molecules on the negatively charged surface of mica by means of

divalent ions was weaker than a silanised modified mica which had a positive surface charge.

The relaxed shape of DNA molecules on Mg2+ modified mica, relative to twisted shape of

† APTES:3-aminopropyl triethoxysilane

400 nm 800 nm

220 nm 160 nm

(a)

(c)

(b)

(d)

400 nm 800 nm

220 nm 160 nm


127

DNA molecules on APTES mica, implied the presence of two-dimensional diffusion over the

mica surface upon the adsorption of DNA, resulting in structural equilibrium. In APTES

mica, DNA molecules can be captured on the surface once they come in contact with it;

hence the AFM images of the DNA molecules of this surface would be comparable with

DNA structures in solution.31-33 In our study, APTES mica resulted in very dense DNA

structures on the surface and it was difficult to analyse or observe any systematic interactions

in homologous DNA segments.

Before quantitative analysis, AFM data on Mg2+ modified mica confirmed that the

differences observed between ap- and p-DNA had indeed not originated from the formation

of PX-DNA28 or Holliday junctions34, 35 or catenanes36. It should be noted that characteristic

differences between p- and ap-DNA were observed before and after TOPO treatment. Under

these conditions, it was found that of the apT-DNA 41% in an open-loop configuration, 26%

showed a single crossing of the DNA strands, 22% two crossings and 10% three crossing or

more (276 samples). For pT-DNA, 66% showed open-loop, 22% single crossing, 3% two

crossings and 9% three crossings or more (68 samples). Importantly, the crossings did not

occur at particular locations along the DNA strand. Hence, there was no indication of

systematic base pairing in the homologous regions.

Figure 5.3: Histogram analysis of the contour length (LDNA) of apT-DNA (blue; n=62) and pT-DNA (red; n=41)

using ImageJ software. The Gaussian fits are indicated with black lines.

Furthermore, a contour length (LDNA) analysis was performed on more than 40 fully relaxed

individual DNA molecules (see Figure 5.3), showing LDNA of ap-DNA is 2.2 ± 0.34 µm,

0 1 2 3 4 5 60

5

10

15

20

25

30 ap-DNA

p-DNA

Cou

nt

LDNA(µµµµm)


128

whereas p-DNA is 3.9 ± 0.34 µm. Presumably, p-DNA was dimerised at some point during

cloning/amplification and then formed a single loop of twice the size of the original DNA

(see Figure 5.4). Therefore, to interpret the presented data in following sub-sections, one

should consider the impact of the larger molecular weight of p(T)-DNA in comparison of

with ap(T)-DNA, as well as the presence of duplex homologues regions in p(T)-DNA

Figure 5.4: Hypothetical schematic of ap- and p- DNA structures after cloning and amplification. According to

AFM analysis, ap-DNA remained as a 6.3 kbp circular DNA, while p-DNA dimerised after cloning and

amplification and formed a 12.6 kbp single loop circular DNA.

5.4.2 Gel Electrophoresis

Gel electrophoresis was used to study the differences in electrophoretic mobilities of p- and

ap-DNA plasmids. The experiments were performed in different gel compositions, sample

incubation conditions and running buffers, each repeated at least three times. Figure 5.5

shows a typical agarose gel (0.8%) electrophoresis of ap- and p-DNA in the three

conformations of supercoiled, linear and relaxed topologies.

Lanes 1 + 2 show ap- and p-DNA before topoisomerase (TOPO) treatment. The DNA is

predominantly supercoiled with minor contributions from more relaxed and linear DNA

species. Within a given lane, the supercoiled DNA were the fastest in line with

expectations.37, 38 Lanes 3 + 4 contain linearised ap- and p-DNA which moved at the same

speed as the 6 kbp reference DNA (lane 7). Lanes 5 + 6 contain apT- and pT-DNA. After

TOPO treatment, supercoiled species disappeared, i.e. both DNA samples were fully relaxed.

Cloning and

amplification

Cloning and

amplification

?

?

?

“Parallel” DNA

(12.6 kbp)


(6.3 kbp)

Pre-engineered

plasmid (5.3 kbp)

“cut”

Insert

1 kbp DNA (Kan gene) KL-Interaction

Linear DNA

(5.3 kbp)

?

?

“Parallel” DNA

(6.3 kbp)


(6.3 kbp)


129

As expected, the most relaxed topomer before TOPO treatment (lane 1+2) moved at the same

speed as the majority of species after TOPO treatment (lane 5+6). However, as for the

supercoiled DNA, ap-DNA moved faster than p-DNA.

Figure 5.5: 0.8% agarose gel electrophoresis (1× TAE, 5 V/cm, 1 hr) of ap- and p-DNA in supercoiled,

linearised (EcoRI) and relaxed (TOPO treated) forms. Lanes 1-7: 1) ap-DNA, 2) p-DNA, 3) linear ap-DNA, 4)

linear p-DNA, 5) apT-DNA, 6) pT-DNA and 7) 1kb DNA ladder (New Englan BioLabs; 3 kb band is indicated).

The majority species of circular ap-DNA always moved faster than p-DNA. As expected, the most relaxed

topomer before topoisomerase treatment moved at the same speed as majority species after topoisomerase

treatment.

A. Ferguson Analysis:

In order to investigate the effect of the gel matrix on DNA mobility, we performed the gel

electrophoresis at various concentrations of agarose gel (see Figure 5.6). The higher the gel

percentage, the smaller is the average pore diameter in the gel and the slower the DNA.39, 40

As expected, the DNA moved slower in the higher percentage gels, due to the smaller pore

size. At the same time, the resolution decreased. In the 2% gel, it can clearly be seen that the

majority of the pT-DNA is being linearised on the gel, as confirmed by several independent

experiments. apT-DNA was significantly less sensitive in this respect, potentially indicating

that apT-DNA adapts more easily to the smaller pore size. This observation can be explained

by the fact that p(T)-DNA is twice the size of ap(T)-DNA, hence is less flexible compared to

apT-DNA and only passes through the pores after decomposition.

1 7

3 kb

32 654


130

Figure 5.6: A typical image of supercoiled, linear and relaxed ap- and p-DNA samples electrophoresis at

various agarose gel percentages. (a) 0.5% (w/v), (b) 0.8% (w/v), (c) 1.0% (w/v), (d) 2.0% (w/v), (e) 3.0% (w/v).

The lanes1-7 are the same in each gel: 1) 1 kb DNA ladder, 2) ap-DNA (supercoiled) 3) p-DNA (supercoiled) 4)

linear ap-DNA (see above) 5) linear p-DNA 6) apT-DNA 7) pT-DNA. Electrophoresis was conducted in 1×

TAE buffer at 23 ˚C with an applied field of 5 V/cm for 1.5 hr, except (b) 2% gel which carried out for 2.5 hr.

Gels were post stained with 3× GelRed DNA stain for 30 min before taking the images.

A Ferguson analysis41 (see Figure 5.7) was performed on the measured-mobility (µe) of DNA

samples as a function of the gel concentration. In Ferguson’s equation log (µe) is given by

3¦ ½o 3¦½¨ Ù+ (5. 1)

where µo is the electrophoretic mobility in free solution, Ù is the retardation coefficient and

+ is the gel concentration.

Figure 5.7: Ferguson plot of DNA mobility as a function of gel percentage (gel pore size). As the gel

percentage decreases, the mobility increases; extrapolation towards 0% yielded the mobility values for the gel-

free case. The slope determines the retardation coefficient. Linearised DNA (black starsc), ap- DNA (red, open

circles); p-DNA (blue squares); apT- DNA (purple triangles) and pT-DNA (green crosses).

(a) (b) (c) (d) (e)

0.0 0.5 1.0 1.5 2.0 2.5 3.010

-11

10-10

10-9

10-8

10-7

µµ µµe

(m2 V

-1 s

-1)

Gel concentration (% w/v)


131

In our study, for all DNA species, the relation between log (µe) and gel percentage was

linear, implying that the pore size was comparable to or larger than the characteristic size of

the molecule (Ogston sieving regime).37

At low gel percentages (0.5-1% w/v), the DNA mobility was of the following order: ap-

DNA > linear DNA > p-DNA ≈ ap-T DNA > pT-DNA. This order changed at high gel

percentages (2-3% w/v) where, linear DNA > ap-DNA > p-DNA > apT-DNA > pT-DNA. At

smaller gel pore sizes, the two linearised DNA samples moved at the same speed, but faster

than the circular samples, due to smaller steric constraints when passing through the gel

pores. The mobility of circular DNA depends on a range of factors, including compactness,

conformational flexibility and interactions with the gel matrix.38, 42 To this end, impaling of

ring-shaped DNA by gel fibres has been proposed as an explanation why especially large

plasmids can get stuck in the gel at high (constant) electric fields.43, 44 Release of impaled

DNA rings is then thought to depend on thermal fluctuations.44 Under the conditions used

here, supercoiled DNA moved faster than relaxed DNA, due to a smaller effective diameter

in the direction of transport (orientation effects) and a smaller tendency to explore the

complex pore network during transport. Generally, ap(T)-DNA samples moved faster than

their parallel counterparts (both before and after topoisomerase treatment). This is in contrast

to predictions, as homology pairing would simply lead to a compaction of DNA. On the other

hand, this effect is in-line with our AFM data (see section 5.4.1) where it showed that p(T)-

DNA was two-fold larger than ap(T)-DNA.

By extrapolating the data towards 0 %, we can obtain µo. The linear DNA was slower than all

circular DNA samples (µo = (4.0 ± 1.0) ×10-8 m2V-1s-1), in agreement with the literature.45

For the plasmid samples, the following were obtained: µo = (4.9 ± 1.1) ×10-8 m2V-1s-1 and

(6.5 ± 1.0) ×10-8 m2V-1s-1 for p- and ap-DNA, and (5.9 ± 2.0) ×10-8 m2V-1s-1 and (8.1 ± 1.3)

×10-8 m2V-1s-1 for pT- and apT-DNA, respectively. The mobility ratios are thus 0.75 ± 0.06

before and 0.73 ± 0.12 after topoisomerase treatment. In both cases, p(T)-DNA moved

approximately 25% slower than ap(T)-DNA, in a solution of TAE buffer, pH 8.3 (ionic

strength, I = 0.02 M). These results seem unexpected, as Stellwagen and Stellwagen (1997)

demonstrated that, in free solution (TAE buffer), ds-DNA molecules larger than ∼400 bp

exhibit an electrophoretic mobility of (3.75 ± 0.04) × 10-8 m2V-1s-1, independent of DNA


132

concentration, size and electric field strength. This phenomenon has been explained by the

fact that the charge/unit mass is the same for all DNA molecules.46, 47 However, the

difference we obtained between µo of ap(T)- and p(T)-DNA from the Ferguson analysis can

be resulted from the errors associated with inter and intra experiment variations.

The slope of the Ferguson plot yields Ù, which is a measure for the DNA's response to

changes in the pore size distribution of the gel. It was found it to be the same, within

experimental error, for p- and ap-DNA (-0.8 ± 0.1), for pT- and apT-DNA (-1.1 ± 0.1), and

for linearised DNA (-0.5 ± 0.1), respectively.

B. Testing the Presence of single stranded DNA in Plasmid Samples:

In this experiment, the aim was to find out whether any region of the DNA duplexes had been

inadvertently been damaged during the sample preparation or analysis process, thus eliminate

the single-stranded (ss) factor in the observed structural differences. For this purpose, the

samples were subjected to endonuclease S1 treatment, which selectively hydrolyses and

cleaves ds-DNA at the single-stranded region caused by a nick, gap, mismatch or loop.48 All

the samples in agarose gel in Figure 5.8 were incubated in 200 mM sodium acetate, 300 mM

NaCl and 10 mM ZnSO4, hence final I of 0.54 M.

Figure 5.8: S1 digestion.0.8% agarose gel electrophoresis (1× TAE, 5 V/m, 1 hr). Lanes 1-11: 1 kb DNA

ladder (0.5 kb band is indicated); 2) ss-DNA M13mp18 (control); 3) ssDNA M13mp18 + S1 endonuclease; 4)

ap-DNA; 5) ap-DNA + S1; 6) p-DNA; 7) p-DNA + S1; 8) apT-DNA; 9) apT-DNA + S1; 10) pT-DNA; 11) pT-

DNA + S1.


133

Accordingly, as a positive control, M13mp18 ss-DNA (7.2 kb, New England Biolabs) was

completely digested upon addition of S1 enzyme (cf. lanes 2 + 3). This confirmed that S1

nuclease is functional under the experimental conditions used above. On the other hand,

supercoiled and relaxed ap-, p-, apT- and pT-DNA remained unaffected (cf. lanes 5, 7, 9, 11

+ 4, 6, 8, 10, respectively), implying that the designed constructs were structurally intact,

before and after TOPO treatment.

Some degree of supercoiling in apT (multiple fragments; lanes 8+9) and pT (faster migration;

lanes 10+11) samples was observed. Perhaps, this phenomenon resulted from condensation

and compaction of DNA species due to the high ionic strength of incubation buffer and

presence of Zn2+ ions.49-51 Ionic strength of these samples were 108 fold greater than the

DNA samples in Figure 5.5, where I = 0.005 M.

C. The Ionic Strength-Dependence of ap- and p-DNA Structural Properties:

The proposed origin of the homology recognition effect is electrostatically based.11, 30

Therefore assuming that the deviation observed in p-DNA mobility is governed by the

homology interaction, a set of experiments at different ionic-strength conditions was

conducted to assess the salt dependency of p-DNA conformation.

i. Effect of Metal Chlorides on Electrophoretic Mobility of Plasmids.

To examine the tendency or reluctancy of the homologues DNA samples in changing of

direction and degree of supercoiling, we conducted an experiment according to Xu and

Bremer’s study.50 In 1997, they observed that when a 4.2 kbp supercoiled DNA was relaxed

by TOPO I in presence of 40 mM metal chloride, the resulting distributions of topoisomers

had positive supercoils (overwound), i.e. an increase in ΔLk value. In this study, based on

their findings and in order to compare the effect of divalent ions on ap- and p-DNA, the

purified DNA of supercoiled ap and p-DNA was treated with TOPO I at 37˚C (overnight) in

the presence of 40 mM of various metal ions, including MgCl2, CaCl2, MnCl2, NiCl2 and

CoCl2. As a control, the purified supercoiled and linear samples were also incubated under

the above conditions with no addition of TOPO enzyme. The resulting (incubated) DNAs

were then analysed by 1 hr electrophoresis on 0.8% agarose gels containing 1× TAE (pH


134

8.3), at 5 V/m applied field. Figure 5.9 shows a typical gel image of the plasmid samples

incubated in presence of Mg2+ and Ca2+ (inset) ions. Lanes 2-7 were the negative controls

where no metal chloride was added during incubation. The samples in lanes 9-11 and i-iv

(inset) were incubated with MgCl2 and CaCl2, respectively.

Figure 5.9: Effect of metal chlorides on mobility. 0.8% agarose gel electrophoresis (1× TAE, 5 V/m, 1 hr) of

ap- and p-DNA in presence of MgCl2 and CaCl2 (inset). Lanes 1,8 and 15 are the 1 kb DNA ladders (the 1 kb

bands are indicated). Lanes 2-7 are the negative control where no MgCl2 was added during incubation, same as

Figure 5.5: 2) ap-DNA, 3) p-DNA, 4) linear ap-DNA, 5) linear p-DNA, 6) apT-DNA, 7) pT-DNA. Samples in

lanes 9-14 were incubated with 40 mM MgCl2 overnight at 37˚C (lanes13+14 were relaxed by TOPO in

presence of this metal chloride): 9) ap-DNA+ Mg2+, 10) p-DNA+ Mg2, 11) linear ap-DNA + Mg2+, 12) linear p-

DNA + Mg2+, 13) ap-T DNA + Mg2+, 14) pT-DNA + Mg2+. The inset represents a typical 0.8% gel of

supercoiled and relaxed samples in presence of 40 mM CaCl2 in the same condition as above. Lanes i-iv: i) ap-

DNA + Ca2+, ii) p-DNA + Ca2+, iii) apT-DNA + Ca2+, iv) pT-DNA + Ca2+. apT + ion2+ topoisomers (smears) are

indicated by yellow (dashed) ellipses.

The presence of Mg2+ ions during incubation did not have any effect on the mobility of the

supercoiled or linearised DNA samples. It did have a structural effect on apT-DNA (smear),

albeit not on pT-DNA. The smear indicates lowering of the effective diameter and presence

of various topoisomers as reported in literature,50 as well as suggesting that pT-DNA is

stabilised, relative to apT-DNA. As is indicated in the inset, this presence of a smear was

consistent in apT- DNA with addition of Ca2+ ions. Nevertheless, qualitative analysis

indicated that the degree of supercoiling in apT + Ca2+ is higher than apT + Mg2+ (cf. lanes

iii+13). It is likely that the stronger effect of Ca2+ ions on coiling of the DNA topoisomers in


135

comparison to Mg2+ ions, is related to higher charge density and hence stronger binding of

Ca2+ ions to the negatively charged DNA molecules.50, 52

Furthermore, the effect of other metal chlorides under the same experimental conditions was

investigated. The presence of Mn2+ or Ni2+ or Co2+ at 37˚C-overnight incubation resulted in

immediate aggregation of all DNA samples, including supercoiled, linear and relaxed forms,

hence no band was detected during electrophoresis (data not shown). This observation is in-

line with other reports in literature.53-55

ii. Increasing the Ionic Strength of the Running Buffer and Gel during Electrophoresis.

In this experiment, the ionic strength of the running buffer was increased by addition of 0.1

and 0.5 M KCl solution (see Figure 5.10).

For experimental reasons, it was not possible to increase the KCl concentration beyond 0.5

M. At high ionic strength, the electrical conductivity is also high, leading to the generation of

a significant amount of heat in the system and gels tended to melt.

Figure 5.10: Increasing ionic strength of the running buffer and the gel matrix during electrophoresis. 0.8%

agarose gel electrophoresis (2.5 V/m, 1.5 hr) in (a) 1× TAE (pH 8.3), (b) 1× TAE + 0.1M KCl (pH 8.3), (c) 1×

TAE + 0.5M KCl (pH 8.3). Lanes 1-7 are the same in all gels: 1) ap-DNA, 2) p-DNA, 3) linear ap-DNA, 4)

linear p-DNA, 5) apT- DNA, 6) pT-DNA, 7) 1 kp DNA ladder (3 kb band is indicated).

In all gels, ap-DNA moved faster than p-DNA, and apT-DNA faster than pT-DNA,

respectively. Thus, it can be concluded that the difference in mobility is also preserved at

higher ionic strength of the running buffer and the gel. For 1× TAE + 0.1 M KCl (Figure

5.10.b) and 1× TAE + 0.5 M KCl (see Figure 5.10.c), apT- and pT-DNA exhibited the same

mobility as the supercoiled samples indicating that under these solution conditions the DNA

(a) (b) (c)


136

coils up again. This effect may be rationalised by electrostatic shielding of DNA strands at

such high ionic strength, resulting in coiling up of the relaxed samples. In addition, as Figure

5.10.c clearly shows, the mobility of all DNA samples, including the DNA marker, was

significantly reduced at 1× TAE + 0.5 M KCl, perhaps as a result of deformation of gel

matrix and/or the reduction of the voltage gradient in the gel.46, 47, 50, 56

In summary, addition of KCl in the running buffer and the gel matrix did not have any

structural effect on p(T)-DNA in comparison with ap(T)-DNA, hence no difference in their

relative mobilities was observed.

iii. Increasing the Ionic Strength of the DNA Buffers‡

In this set of experiments, the ionic strength of the DNA buffer was increased gradually by

addition of KCl before and after TOPO treatment. The purified DNA samples were originally

eluted in 10mM Tris-HCl, pH 8.5. Following addition of KCl , the samples were incubated

for 30 min at room temperature, then analysed by 0.8% agarose gel electrophoresis in 1×

TAE (pH 8.3) for 1 hr.

The final KCl concentration of each sample used in the electrophoresis experiment is

indicated in the caption of Figure 5.11, ranging from 0.04 to 0.3 M. Both gels show that

increasing the samples’ ionic strengths, did not result in qualitative differences in the mobility

of DNA molecules in the gel, as opposed to increasing the running buffer/gel matrix’s ionic

strength. The DNA bands in Figure 5.11.b are slightly tilted, perhaps as a result of high salt

concentration in the samples. This effect was minimal for supercoiled samples (see Figure

5.11.a).

‡ In parallel to this experiment, using the same experimental condition, Dynamic Light Scattering was employed to monitor the ionic strength dependence of ap(T)- and p(T)-DNA diffusion coefficients (see section 5.4.3).


137

Figure 5.11: Increasing ionic strength of the DNA samples buffers. 0.8% agarose gel electrophoresis (1× TAE,

5 V/m, 1 hr). DNAs were initially eluted in 10 mM Tris-HCl (pH 8.5) before 30 min incubation with KCl. (a)

Increasing KCl concentration of supercoiled DNA buffers. Linearised ap and p-DNAs are used as the controls

(reference bands). (b) Increasing KCl concentration of relaxed (TOPO treated) DNA buffers. Supercoiled ap

and p-DNAs are used as the controls (reference bands). Lane 1-15 in (a): 1) 1 kb DNA ladder ( 2 kb band

indicated), 2) ap-DNA with no KCl, 3) ap-DNA + 38 mM KCl, 4) ap-DNA + 75 mM KCl, 5) ap-DNA + 113

mM KCl, 6) ap-DNA + 150 mM KCl, 7) ap-DNA + 300 mM KCl, 8) p-DNA with no KCl, 9) p-DNA + 38 mM

KCl, 10) p-DNA + 75 mM KCl, 11) p-DNA + 113 mM KCl, 12) p-DNA + 150 mM KCl, 13) p-DNA + 300

mM KCl, 14) linear ap-DNA (no KCl), 15) linear p-DNA (no KCl). Lanes 16-30 in (b): 16) 1 kb DNA ladder (

2 kb band indicated), 17) supercoiled ap-DNA (no KCl), 18) apT-DNA with no KCl, 19) apT-DNA + 38 mM

KCl, 20) apT-DNA + 75 mM KCl, 21) apT-DNA + 113 mM KCl, 22) apT-DNA + 150 mM KCl, 23) apT-

DNA + 300 mM KCl, 24) supercoiled p-DNA (no KCl), 25) pT-DNA with no KCl, 26) pT-DNA + 38 mM KCl,

27) pT-DNA + 75 mM KCl, 28) pT-DNA + 113 mM KCl, 29) pT-DNA + 150 mM KCl, 30) apT-DNA + 300

mM KCl.


138

5.4.3 Dynamic Light Scattering (DLS)

In order to determine ap- and p-DNA diffusion coefficients, DLS§ was employed to

characterise the DNA samples with regards to their translational (centre-of-mass) diffusion.

In Figure 5.12, the ratio of translational diffusion coefficients, Dt (p) / Dt (ap) is plotted

against I, where the supercoiled and relaxed forms are presented with open squares and filled

triangles, respectively.

Figure 5.12: DLS study of ap- and p-DNA ionic strength dependence (n = 3). Ratio of translational diffusion

coefficients of p- and ap-DNA as a function of ionic strength on a semi-log graph. Open squares: supercoiled

DNA, filled triangles: relaxed (TOPO treated) DNAs the error bars denote three independent measurements with

three repeats each. (Source: courtesy of W. Pitchford).

For the supercoiled samples at low I (10 mM Tris-HCl), p-DNA appeared to move faster than

ap-DNA (i.e. Dt(p) > Dt(ap)). The ratio then decreased with increasing I, to ∼1 at I = 0.05 M

and ∼0.6 at I = 0.12 M. At I = 0.16 M and above, it increased to ∼1, the reason for this is

currently unknown. For relaxed DNA, Dt (pT)/ Dt (apT) was close to 1 (0.90) at low I and

then decreased by 11% at I = 0.16 M. The change in relaxed DNA was smaller than for

supercoiled DNA, but still significant.

The translational diffusion coefficients for 1.9, 2.7 and 5.2 kbp supercoiled plasmid DNA and

2.3 kbp relaxed, nicked plasmid DNA have previously been studied as a function of ionic

strength and were found to be independent of I within experimental error.57-59 These plasmids

did not have any specially engineered homologous regions, therefore it was expected that the

§ DLS experiments and analysis were performed by William Pitchford, a PhD candidate in the Albrecht group.

Dt(p

) /

Dt(a

p)

I (M)


139

ap-DNA would behave in the same way. On the other hand, if the interaction between the

homologous regions was indeed electrostatic, the p-DNA magnitude should be affected by I.

Accordingly, I would affect the shape of p-DNA in solution and thus its diffusional

properties, relative to ap-DNA. Assuming, Dt of a random sequence 12 kbp plasmid would

also be independent of I, perhaps what was observed in the change of Dt (p) / Dt (ap) as a

function of I implies the presence of an electrostatic interaction in p(T)-DNA according to

predictions.30, 60 However, further control experiments are required to rationalise this

suggestion.

It should be noted, at all ionic strengths, the same concentration (ng/µl) of ap(T)- and p(T)-

DNA was used, which was measured by UV-Vis spectroscopy. Given the fact that p-DNA is

twice as large as ap-DNA at any specific concentration (ng/µl), the number of p-DNA

molecule existing in the solution, was less than ap-DNA. The impact of this effect in the data

will be discussed in section 5.5.

5.4.4 Nanopore Translocation

To probe the structural and conformational properties of the plasmids at the single molecule

level, a Si3N4 nanopore was used to compare ap- and p-DNA translocation dynamics at 1 M

KCl. Prior to presenting the nanopore data, the gel electrophoresis image of the DNA samples

that incubated in 1 M KCl for 2 hr at room temperature is shown in Figure 5.13. This

experimental incubation condition mimics the the nanopore sensing condition.

Figure 5.13. Effect of 1 M KCl on supercoiled plasmids (buffer) during incubation for 2 hr, 0.8% agarose gel

electrophoresis (1× TAE, 5 V/m, 1 hr). All DNAs were initially eluted in 10 mM Tris-HCl (pH 8.5) before

incubation. Lanes 1-5: 1) DNA ladder (3 kb band is indicated), 2) ap-DNA+1 M KCl , 3) p-DNA + 1 M KCl, 4)

ap-DNA with no KCl 5) p-DNA with no KCl.


140

Lanes 2+3 are ap- and p-DNA in presence of 1 M KCl and lanes 3+4 are ap- and p-DNA with

no KCl. Electrophoresis data showed that, a KCl concentration of 1 M during sample

incubation results in a small decrease of the gel mobility of the supercoiled samples.

However, the mobility differences between ap- and p-DNA remained almost the same. No

KCl was added to the running buffer in this experiment and in all lanes, the supercoiled

topology was the most dominant conformation.

At 1M KCl, Rg is 90-100 nm for 6.3 kbp and 130-150 nm for 12.3 kbp supercoiled DNA,

respectively.61, 62 These estimates are in good agreement with the AFM imaging data on

APTES-modified mica (see Appendix III). In the nanopore experiments, based on the

measured pore conductance (cylindrical geometry), the dpore was ~ 44 nm (thickness of ~ 70

nm) and in-line with SEM imaging data, Figure 5.14.b (inset). This implied that the DNA can

only enter the pore after some deformation and partial unravelling,63, 64 as displayed in the

schematic, Figure 5.14.a.

Here, cc-DNA translocation was in a voltage range of 150-300 mV. Below 150 mV,

translocation events were not observed, presumably due to a finite activation barrier for

entering the pore; above 300 mV, events were too fast to be resolved. Examples of the

typical events observed during translocation of ap and p-DNA at 150 mV bias are shown in

Figure 5.14.c.


141

Figure 5.14: Nanopore translocation data at 1 M KCl-Tris-HCl (pH 8.5) (a) Schematic of the nanopore setup

(cross-sectional view). (b) Ion current/voltage trace for the pore used (conductance §¨©ª= 305.4 nS, solution

conductivity σs = 10.98 Ω-1m-1; pore channel length Lpore = 70 nm; estimated pore diameter dpore = 44 nm

assuming cylindrical geometry); inset: SEM image of the pore utilised. (c) Examples of DNA translocation

events for p- and ap-DNA at 150 mV bias. ΔI vs. τd event number density plots of (d) ap-DNA and (e) p-DNA

at Vbias of (i) 150 mV, n = 748 (ii) 200 mV, n = 1254 and (iii) 300 mV, n = 995. The histograms are normalised

to 1, colour code in panel e.iii.

Figure 5.14.d-e presents the event number density plots (2D histograms) of ΔI vs. τd. In each

panel, the translocation time τd is plotted on the abscissa, the corresponding current

0 5 10 15 20 25 30

-5

-4

-3

-2

-1

0

ττττd(ms)

∆∆ ∆∆I

(nA

)

0 5 10 15 20 25 30

-5

-4

-3

-2

-1

0

∆∆ ∆∆I

(nA

)

ττττd(ms)

0 5 10 15 20 25 30

-5

-4

-3

-2

-1

0

∆∆ ∆∆

I (n

A)

ττττd(ms)

0 5 10 15 20 25 30

-5

-4

-3

-2

-1

0

ττττd(ms)

∆∆ ∆∆I

(nA

)

0 5 10 15 20 25 30

-5

-4

-3

-2

-1

0

ττττd(ms)

∆∆ ∆∆I

(nA

)

0 5 10 15 20 25 30

-5

-4

-3

-2

-1

0

ττττd(ms)

∆∆ ∆∆I

(nA

)(a) (b) (c)

(d.i) (d.ii) (d.iii)

(e.i) (e.ii) (e.iii)

-600 -400 -200 0 200 400 600-200

-100

0

100

200

I (n

A)

Vbias

(mV)


142

modulation ΔI on the ordinate. The colour code represents the point density (from high/red to

low/blue, normalized to 1). The 2D histograms show that, with increasing voltage for both

samples, the distributions are compressed towards smaller dwell times; the translocation

process becomes faster, as expected for larger driving forces. In addition these histograms

demonstrate that ap-DNA exhibited a broader distribution of the translocation times and

current amplitudes (see Figure 5.14.d.i-iii) relative to p-DNA, implying the presence of

diversity of confirmations and flexibility or perhaps impurities during translocation process in

non-homologous DNA molecules. On the other hand, p-DNA dwell time and current

modulations distributions were highly clustered (see Figure 5.14.e.i-iii) in comparison with

ap-DNA. Surprisingly, ΔI magnitude of p-DNA was on average ~2.5 folds lower than ap-

DNA at all voltages. This observation is in-line with structural predictions for homologous

DNA (compact ellipsoidal confirmation) but in contrast to expectations for translocation of a

2-fold larger p-DNA plasmid relative to ap-DNA, unless the effective surface charges of ap-

and p-DNA and/or confirmations and folding generated due threading and unravelling of

plasmids were different.

The values of the most probable current blockade amplitudes, as well as the most probable

and mean translocation times (τd), are presented in Table 5.1.

Table 5.1 Summary of nanopore data at three applied potentials. The most probable (max) ΔI and τd values

obtained from the dwell time-histograms fitted with the (skewed) Gaussian distribution. (see Appendix II ). The

error associated with each data point, denotes the standard deviation resulted from the fitting procedure.

Vbias (mV)

ap-DNA p-DNA

τd max (ms) τd mean (ms) ∆I max (nA) τd max (ms) τd mean (ms) ∆I max (nA)

150 0.12 ± 0.01 ~ 19.3 -1.60 ± 0.11 0.71 ± 0.01 ~ 2.40 -0.54 ± 0.02

200 0.06 ± 0.01 ~ 8.3 -1.20 ± 0.02 0.31 ± 0.01 ~ 1.04 -0.63 ± 0.01

300 0.03 ± 0.01 ~ 4.4 -2.19 ± 0.06 0.27 ± 0.01 ~ 0.34 -0.82 ± 0.02

The most probable translocation times for a given voltage were smaller for ap-DNA than for

p-DNA, i.e. ap-DNA translocated faster than p-DNA. However, ap-DNA also exhibited a

much broader translocation time distribution; hence the mean values were larger for ap-DNA,

too. This reflected the greater structural diversity of ap-DNA and perhaps some

contaminations in its solution, compared to p-DNA.


143

Figure 5.15: A semi-logarithmic plot of the most probable translocation time τd max for ap-DNA (open

squares) and p-DNA (closed circles) vs. applied bias voltage (Vbias). The error bars were estimated from the

fitting procedures. The linear fit for ap-DNA is indicated with a dashed line.

The τd max of each DNA sample against the applied potential is plotted in Figure 5.15. In the

voltage range studied, the plot of ln(τmax) vs. Vbias is linear for ap-DNA (dashed line, R2 =

0.96; slope = -8.9 ± 1.7 V-1; intercept = -0.9 ± 0.4). This semi-log relationship is indicative of

the presence of DNA-pore interaction and an activated translocation process i.e. the DNA

chain has to be stretched to traverse through the pore, which involves with an extra energy

barrier. This is the same functional behaviour typically found for linear and circular ds-

DNA.65-67 On the other hand, as expected for larger molecules, τd max is significantly larger

for p-DNA relative to ap-DNA. Interestingly, τd max for p-DNA is less dependent on Vbias as

the ln(τd max) vs. Vbias relationship appears to be non-linear.

Furthermore, taking the length of the plasmid passing through the pore, as half its contour

length, according to AFM (see section 5.4.1) and τd (max) analysis, the effective translocation

speed of 0.8-3.3 cm/s for ap-DNA and 0.3-0.7 cm/s for p-DNA were estimated. These values

are in accordance with values reported previously, for comparable circular and linear DNA,

after correcting for differences in the local electric field.63, 64 The ratio of the most probable

translocation times, τmax (p-DNA) / τmax (ap-DNA) ≈ 3.3 on average for all Vbias used. This is

comparable to the gel electrophoresis results at 2% gel percentages (small pore size) whereas

differences between p- and ap-DNA are rather smaller in free solution, e.g. in gel

electrophoresis at 0% gel (see section 5.4.2.A) and in DLS (see section 5.4.3 ).

100 200 300-5

-4

-3

-2

-1

0

ap-DNA

p-DNA

ln[ ττ ττ

d m

ax (

ms)

]

Vbias(mV)


144

Assuming a linear potential drop across the pore, the average local electric field between 2.1

and 4.3 MV/m was obtained for Vbias = 150 mV and 300 mV, respectively, hence the average

effective mobility (µ eff.) of ~ 6×10-9 m2V-1s-1 for ap-DNA and ~ 2×10-9 m2V-1s-1 for p-DNA

was estimated. By comparing these values with those extrapolated to 0% gel, it was noticed

that the values obtained from the nanopore experiment were much smaller than for gel

electrophoresis measurements. This difference had already been reported for linear DNA, and

may be attributed to the influence of the pore wall on the drag coefficient and EO-flow, as

well as the viscous drag on the DNA moving outside the pore.68, 69

5.5 Conclusion

To summarise, the AFM analysis confirmed that ap-DNA (non-homologous) is a 6.3 kbp, and

p-DNA (homologous) is a 12.6 kbp plasmid. AFM data also suggested that the p-DNA was

dimerised at some point during cloning procedures and then formed a single loop, rather than

a catenane. The gel electrophoresis data showed that ap-DNA had higher electrophoretic

mobility than p-DNA before and after TOPO treatment at all gel percentages and ionic

strengths. This observation was in-line with the above AFM data, as p(T)-DNA has a higher

molecular weight in comparison to ap(T)-DNA, resulting in a lower mobility. However, why

divalent ions were ineffective in the changing of the superhelical density of p-DNA upon

TOPO treatment, compared to ap-DNA where a smear of topoisomers was observed is still

unclear. Presumably, the critical concentration of metal chlorides is different for larger DNA

molecules (p-DNA). On the other hand, using AFM analysis, Hansma and co-workers

reported on sequence-dependent DNA condensation which in line with KL theory.70 However

to confirm the latter hypothesis, one would need to test the effect of divalent ions on a variety

of molecular weights of the plasmids, in order to eliminate the size dependency of cc-DNA

winding by multivalent ions.

Furthermore, in contrast to the gel electrophoresis data, the DLS study showed that at low I ,

p-DNA has higher Dt than p-DNA. This trend was reversed at ~ 0.8 M I and then gradually

restored back to the original, as the I was increased. As mentioned earlier, this phenomenon is

counterintuitive, as several reports already showed the I independency of Dt in cc-DNA of

different sizes.57-59 Perhaps, the trend we observed in our experimental results, simply

occurred because of aggregation or precipitation of p-DNA at a slightly lower I compared to


145

ap-DNA. Moreover in this DLS experiment, ap-DNA (molar) concentration was 2-fold

higher than ap-DNA. The DLS concentration dependence has already been well established

in literature: high concentration of analyte, results in crowding effect and DNA interactions,

leading to biasing the scattering intensity and decreasing the diffusion coefficient.71, 72 To this

end, independent experiments are further required to investigate the effect of DNA

concentration and ionic strength for difference sizes of plasmids, in order to determine the

concentration- and ionic strength- independent regimes in this DLS study.

Lastly, the nanopore data showed the most probable translocation time is larger for the p-

DNA species compared to ap-DNA and the translocation time distribution is broader for ap-

DNA, in comparison with p-DNA. In addition, p-DNA showed smaller conductance changes

during translocation, relative to ap-DNA. These observations do not discount the

monomer/dimer scenario, as at this stage, the presence of sample contamination cannot be

ruled out. In addition in nanopore sensing, other factors such as surface charge,

hydrodynamic interactions and flexibility of the analyte play critical roles.

Combining all data and findings, it is indeed clear that p-DNA was in dimeric form and ap-

DNA was in monomeric form throughout all the experiments. However, the main question

here is, why did the dimer form in one case, but not in the other? Perhaps, these

serendipitous results provide a mechanism for dimerisation as a result of homologous

interactions in vivo. This hypothesis can be supported by several reports in literature: Tilly et

al. observed high frequency and stable dimeres of a 26 kbp cc-DNA. They confirmed that in

transformation process, the transforming DNA recombines with resident DNA, rather than

displacing it.73 Moreover, Bedbrook and co-workers showed multimer formation of E.coli

pMB9 (5.5 kbp) and pML21 (7.3 kbp) plasmids. They reported that multimers are most

probably generated by a single reciprocal recombination process occurring at regions of

homology between plasmid strands rather than by replication mechanism.74 In line with their

work, using hybrid constructed plasmids, Chang and Cohen showed that site-specific genetic

recombination can be promoted in vivo by EcoRI restriction endonuclease, in conjunction

with E.coli DNA ligase.75 The latter is comparable to the experimental conditions in this

study, as the EcoRI restriction site was employed for construction and cloning of ap- and p-

DNA from the pET-24a(+) pre-engineered plasmid.


146

To this end, the effect of sequence –dependent electrostatic homology interaction in some of

the findings cannot be dismissed, nor can the homology dependence of dimerisation be

confirmed. Future control experiments with comparable length plasmids are essential to gain

a better understanding of the transformation mechanism of homologous DNA.

Lastly, given the fact that there was a 2-fold size difference between ap- and p-DNA and

considering the challenges we encountered to reveal this structural diversity, we suggest to

improve some of methodologies of our choice in future studies. For instance, solely with DLS

or nanopore sensing experiments, we were unable to determine any size difference between

the plasmids. Therefore, in order to be able to compare the plasmids of the same size as it was

suggested above, we propose to employ techniques that exhibit a higher sensitivity and

resolution during detection of structural and conformational diversities, Namely, 2D-gel

electrophoresis, fluorescent labelling and optical detection, X-ray scattering techniques would

be more suited to study our current objective: to study the ability of duplex DNA to recognise

sequence homology recognition within the framework of the KL theory.


147

5.6 References

1. Lederberg, J. & Tatum, E.L. Gene recombination in Escherichia coli. Nature 158, 558

(1946).

2. Holliday, R. The Induction of Mitotic Recombination by Mitomycin C In Ustilago

and Saccharomyces. Genetics 50, 323-335 (1964).

3. Sung, P. & Klein, H. Mechanism of homologous recombination: mediators and

helicases take on regulatory functions. Nat Rev Mol Cell Biol 7, 739-750 (2006).

4. Leach, D. Genetic recombination. (Blackwell Science, Oxford; 1996).

5. Lewin, B. Genes VI. (Oxford University Press, Oxford; 1997).

6. Crick, F. General model for the chromosomes of higher organisms. Nature 234, 25-27

(1971).

7. Sobell, H.M. Molecular mechanism for genetic recombination. Proc Natl Acad Sci U

S A 69, 2483-2487 (1972).

8. Wagner, R.E. & Radman, M. A mechanism for initiation of genetic recombination.

Proc Natl Acad Sci U S A 72, 3619-3622 (1975).

9. Forsdyke, D.R. A stem-loop "kissing" model for the initiation of recombination and

the origin of introns. Mol Biol Evol 12, 949-958 (1995).

10. CameriniOtero, R. & Hsieh, P. Homologous recombination proteins in prokaryotes

and eukaryotes. Annual Review of Genetics 29, 509-552 (1995).

11. Kornyshev, A. & Leikin, S. Sequence recognition in the pairing of DNA duplexes.

Physical Review Letters 86, 3666-3669 (2001).

12. Minsky, A. Information content and complexity in the high-order organization of

DNA. Annual Review of Biophysics and Biomolecular Structure 33, 317-342 (2004).

13. Livolant, F. Cholesteric organization of DNA in vivo and in vitro. Eur J Cell Biol 33,

300-311 (1984).

14. Livolant, F. Ordered phases of DNA in vivo and in vitro. Physica A: Statistical

Mechanics and its Applications 176, 117-137 (1991).

15. Rau, D. & Parsegian, V. Direct measurement of the intermolecular forces between

counterion-condensed DNA double helices. Evidence for long-range attractive

hydration forces. Biophysical Journal 61, 246-259 (1992).


148

16. Gelbart, W., Bruinsma, R., Pincus, P. & Parsegian, V. DNA-inspired electrostatics.

Physics Today 53, 38-44 (2000).

17. Kornyshev, A.A., Lee, D.J., Leikin, S. & Wynveen, A. Structure and interactions of

biological helices. Reviews of Modern Physics 79, 943-996 (2007).

18. Battistini, F., Hunter, C., Moore, I. & Widom, J. Structure-Based Identification of

New High-Affinity Nucleosome Binding Sequences. Journal of Molecular Biology

420, 8-16 (2012).

19. Gaffney, D. et al. Controls of Nucleosome Positioning in the Human Genome. Plos

Genetics 8 (2012).

20. Leforestier, A. & Livolant, F. Structure of toroidal DNA collapsed inside the phage

capsid. Proceedings of the National Academy of Sciences of the United States of

America 106, 9157-9162 (2009).

21. Kornyshev, A. & Leikin, S. Theory of interaction between helical biomolecules.

Biophysical Journal 72, 3656-3674 (1997).

22. Kornyshev, A. & Leikin, S. Electrostatic zipper motif for DNA aggregation. Physical

Review Letters 82, 4138-4141 (1999).

23. Inoue, S., Sugiyama, S., Travers, A. & Ohyama, T. Self-assembly of double-stranded

DNA molecules at nanomolar concentrations. Biochemistry 46, 164-171 (2007).

24. Baldwin, G.S. et al. DNA double helices recognize mutual sequence homology in a

protein free environment. J Phys Chem B 112, 1060-1064 (2008).

25. Kornyshev, A. Physics of DNA: unravelling hidden abilities encoded in the structure

of 'the most important molecule'. Physical Chemistry Chemical Physics 12, 12352-

12378 (2010).

26. Danilowicz, C. et al. Single molecule detection of direct, homologous, DNA/DNA

pairing. Proc Natl Acad Sci U S A 106, 19824-19829 (2009).

27. Shen, Z., Yan, H., Wang, T. & Seeman, N. Paranemic crossover DNA: A generalized

Holliday structure with applications in nanotechnology. Journal of the American

Chemical Society 126, 1666-1674 (2004).

28. Wang, X., Zhang, X., Mao, C. & Seeman, N.C. Double-stranded DNA homology

produces a physical signature. Proc Natl Acad Sci U S A 107, 12547-12552 (2010).


149

29. Nishikawa, J. & Ohyama, T. Selective association between nucleosomes with

identical DNA sequences. Nucleic Acids Res 41, 1544-1554 (2013).

30. Cortini, R., Kornyshev, A., Lee, D. & Leikin, S. Electrostatic Braiding and

Homologous Pairing of DNA Double Helices. Biophysical Journal 101, 875-884

(2011).

31. Rivetti, C., Guthold, M. & Bustamante, C. Scanning force microscopy of DNA

deposited onto mica: Equilibration versus kinetic trapping studied by statistical

polymer chain analysis. Journal of Molecular Biology 264, 919-932 (1996).

32. Shlyakhtenko, L. et al. Silatrane-based surface chemistry for immobilization of DNA,

protein-DNA complexes and other biological materials. Ultramicroscopy 97, 279-287

(2003).

33. Lyubchenko, Y. Preparation of DNA and nucleoprotein samples for AFM imaging.

Micron 42, 196-206 (2011).

34. Grainger, R., Murchie, A. & Lilley, D. Exchange between stacking conformers in a

four-way DNA junction. Biochemistry 37, 23-32 (1998).

35. McKinney, S., Declais, A., Lilley, D. & Ha, T. Structural dynamics of individual

Holliday junctions. Nature Structural Biology 10, 93-97 (2003).

36. Yamaguchi, H., Kubota, K. & Harada, A. Preparation of DNA catenanes and

observation of their topological structures by atomic force microscopy. Nucleic Acids

Symp Ser, 229-230 (2000).

37. Rodbard, D. & Chrambach, A. Unified theory for gel electrophoresis and gel

filtration. Proc Natl Acad Sci U S A 65, 970-977 (1970).

38. Viovy, J.L. Electrophoresis of DNA and other polyelectrolytes: Physical mechanisms.

Reviews of Modern Physics 72, 813-872 (2000).

39. Narayanan, J. et al. Determination of agarose gel pore size: Absorbance

measurements vis a vis other techniques. International Conference on Materials for

Advanced Technologies (ICMAT 2005) 28, 83-86 (2006).

40. Pernodet, N., Maaloum, M. & Tinland, B. Pore size of agarose gels by atomic force

microscopy. Electrophoresis 18, 55-58 (1997).

41. Ferguson, K.A. Starch-gel electrophoresis--application to the classification of

pituitary proteins and polypeptides. Metabolism 13, SUPPL:985-1002 (1964).


150

42. Alon, U. & Mukamel, D. Gel electrophoresis and diffusion of ring-shaped DNA.

Physical Review E 55, 1783-1793 (1997).

43. Mickel, S., Arena, V. & Bauer, W. Physical properties and gel electrophoresis

behavior of R12-derived plasmid DNAs. Nucleic Acids Res 4, 1465-1482 (1977).

44. Levene, S.D. & Zimm, B.H. Separations of open-circular DNA using pulsed-field

electrophoresis. Proc Natl Acad Sci U S A 84, 4054-4057 (1987).

45. Robertson, R.M., Laib, S. & Smith, D.E. Diffusion of isolated DNA molecules:

dependence on length and topology. Proc Natl Acad Sci U S A 103, 7310-7314

(2006).

46. Stellwagen, N.C., Gelfi, C. & Righetti, P.G. The free solution mobility of DNA.

Biopolymers 42, 687-703 (1997).

47. Stellwagen, N. & Stellwagen, E. Effect of the matrix on DNA electrophoretic

mobility. Journal of Chromatography a 1216, 1917-1929 (2009).

48. Wiegand, R.C., Godson, G.N. & Radding, C.M. Specificity of the S1 nuclease from

Aspergillus oryzae. J Biol Chem 250, 8848-8855 (1975).

49. Bloomfield, V.A. DNA condensation by multivalent cations. Biopolymers 44, 269-

282 (1997).

50. Xu, Y. & Bremer, H. Winding of the DNA helix by divalent metal ions. Nucleic Acids

Research 25, 4067-4071 (1997).

51. Manning, G.S. The persistence length of DNA is reached from the persistence length

of its null isomer through an internal electrostatic stretching force. Biophys J 91,

3607-3616 (2006).

52. Anderson, P. & Bauer, W. Supercoiling in closed circular DNA: dependence upon ion

type and concentration. Biochemistry 17, 594-601 (1978).

53. Sarma, M.H. & Sarma, R.H. Structure & Expression: DNA & its drug complexes,

Vol. 2. (Adenine Press, 1988).

54. Andrushchenko, V., Van De Sande, H. & Wieser, H. DNA interaction with Mn2+

ions at elevated temperatures: VCD evidence of DNA aggregation. Biopolymers 69,

529-545 (2003).

55. Ma, C. & Bloomfield, V.A. Condensation of supercoiled DNA induced by MnCl2.

Biophys J 67, 1678-1681 (1994).


151

56. Stellwagen, E. & Stellwagen, N. Probing the electrostatic shielding of DNA with

capillary electrophoresis. Biophysical Journal 84, 1855-1866 (2003).

57. Langowski, J. Salt effects on internal motions of superhelical and linear pUC8 DNA.

Dynamic light scattering studies. Biophys Chem 27, 263-271 (1987).

58. Hammermann, M. et al. Salt effects on the structure and internal dynamics of

superhelical DNAs studied by light scattering and Brownian dynamics. Biophys J 73,

2674-2687 (1997).

59. Newman, J., Tracy, J. & Pecora, R. Dynamic light scattering from monodisperse 2311

base pair circular DNA: ionic strength dependence. Macromolecules 27, 6808-6811

(1994).

60. Cortini, R., Lee, D. & Kornyshev, A. Chiral electrostatics breaks the mirror symmetry

of DNA supercoiling. Journal of Physics-Condensed Matter 24 (2012).

61. Latulippe, D. & Zydney, A. Radius of Gyration of Plasmid DNA Isoforms From

Static Light Scattering. Biotechnology and Bioengineering 107, 134-142 (2010).

62. Manning, G. A procedure for extracting persistence lengths from light-scattering data

on intermediate molecular weight DNA. Biopolymers 20, 1751-1755 (1981).

63. Fologea, D., Brandin, E., Uplinger, J., Branton, D. & Li, J. DNA conformation and

base number simultaneously determined in a nanopore. Electrophoresis 28, 3186-

3192 (2007).

64. Uplinger, J. et al. K plus , Na+, and Mg2+on DNA translocation in silicon nitride

nanopores. Electrophoresis 33, 3448-3457 (2012).


158 (2012).

66. Chen, Q., Diao, S. & Wu, C. How does a supercoiled DNA chain pass through a small

conical glass pore? Soft Matter 8, 5451-5458 (2012).

67. Wanunu, M., Sutin, J., McNally, B., Chow, A. & Meller, A. DNA translocation

governed by interactions with solid-state nanopores. Biophys J 95, 4716-4725 (2008).

68. Li, J. & Talaga, D. The distribution of DNA translocation times in solid-state

nanopores. Journal of Physics-Condensed Matter 22 (2010).

69. Chen, P. et al. Probing Single DNA Molecule Transport Using Fabricated Nanopores.

Nano Letters 4, 2293-2298 (2004).


152

70. Sitko, J.C., Mateescu, E.M. & Hansma, H.G. Sequence-dependent DNA condensation

and the electrostatic zipper. Biophys J 84, 419-431 (2003).

71. Tscharnuter, W. in Encyclopedia of Analytical Chemistry (John Wiley & Sons, Ltd,

2006).

72. Szydlowski, J. & Van Hook, W. Concentration and temperature dependence of

dynamic light scattering for some polystyrene solutions in toluene, cyclohexane,

methylcyclohexane and deuteromethylcyclohexane. Macromolecules 31, 3266-3274

(1998).

73. Tilly, K., Lubke, L. & Rosa, P. Characterization of circular plasmid dimers in Borrelia

burgdorferi. J Bacteriol 180, 5676-5681 (1998).

74. Bedbrook, J.R. & Ausubel, F.M. Recombination between bacterial plasmids leading

to the formation of plasmid multimers. Cell 9, 707-716 (1976).

75. Chang, S. & Cohen, S.N. In vivo site-specific genetic recombination promoted by the

EcoRI restriction endonuclease. Proc Natl Acad Sci U S A 74, 4811-4815 (1977).

153

Chapter 6

Conclusion and Outlook

Synopsis: This chapter presents a summary of the work that has been discussed in this thesis. An overview of

the motivations, challenges and key findings of each chapter is presented below, followed by a short discussion

and recommendations for future work.

Chapter 6 Conclusion and Outlook

154

The primary focus of this thesis was the study of biological, chemical and physical properties

of DNA at the single molecule level in both in vivo and in vitro regimes. In particular, this

research project attempted to address two current challenges in the field, including: i)

ultrafast sensing of DNA methylation with implications in cancer diagnosis and ii)

investigation of the homology recognition in cc-DNA.

In this project, solid-state nanopore sensors were employed as an alternative method to study

the above modifications and features at the nanoscale. The fabrication process of a nano-scale

pore in a Si3N4 membrane and the operational set-up of these sensors were outlined in

chapter 3. In order to gain a better understanding of the functionality, sensitivity and

efficiency of these devices, nanopore chips were examined by translocation of a sonicated

genomic DNA extracted from MCF-7 breast cancer cell lines, through a sub-20 nm pore.

Sonication parameters were set in such away as to create sub-3 kbp DNA fragments, in order

to be consistent with the size-range of the DNA under investigation in later studies. This

experiment allowed the optimisation of the operational set-up and handling procedures, as

well as of the data acquisition parameters. Moreover, the analysis of the data, namely

blockade event characteristics and capture rate, confirmed that entropic factors were

important in the system and that the translocation process was driven by electrophoresis.

Current modulation and the dwell time analyses provided some information regarding the

folding features, effective charge and the effective velocity of DNA, while it was passing

through the pore. To this end, it was not possible to resolve the translocation dynamics of

each length of DNA fragment due to the broad structural diversity of our sample, as well as

lack of temporal and spatial resolutions. Hence, further experiments are required to obtain a

better understanding of the patterns and characteristics of sonicated DNA translocation,

including translocation of fixed–length DNA fragments ( i.e. 0.5, 1, 2, 3 kbp) individually

and in mixture, in order to determine the corresponding sub-cluster and sub-peaks in the

present data analyses.

The significance of DNA methylation sensing in breast cancer cell lines was outlined in

chapter 4. In particular, the study focussed on the methylation level of the FOXA1 promoter

due to its potential role as a biomarker for monitoring the progress and development of breast

cancer. FOXA1 protein expression and mRNA levels were compared in MCF-7 and MLET-2

(chemotherapy resistant) cell lines. It was found out that in MLET-2 cells, FOXA1 mRNA


155

and proteins levels were down-regulated. Furthermore using a MeDIP assay, the methylation

level of this gene was quantified in those cell lines, showing that the FOXA1 promoter

methylation level is indeed higher in MLET-2 compared to MCF-7 cells. This finding

confirmed that, as a result of hypermethylation in the chemotherapy restraint cell line, the

transcriptional and translational activity of this gene was silenced. Thereafter, solid-state

nanopore sensors were utilised to detect methylated regions of an in vitro methylated DNA

sample. MeDIP assay already showed that by exploiting DNA-protein (antibody)

interactions, the detection of methylated regions of DNA can be facilitated.

Prior to nanopore detection, the binding affinity and specificity of 5’-mc antibody to the

duplex (methylated) DNA was evaluated by EMSA assay and AFM imaging. AFM data

Furthermore, from the AFM analysis, an association constant of Ka 3×108 M was

extracted. This estimate was in good agreement with values reported in the literature.

Subsequently, Si3N4 nanopore-based sensing was used to probe the in vitro methylation of the

FOXA1 promoter, again as a DNA-antibody complex. Notably, translocation of the antibody-

bound DNA was only achieved at the bias polarity opposite to that of the unlabelled DNA.

We speculated that this phenomenon was potentially due to the fact that electroosmosis was

the main driving force of the complex translocation, whereas in the DNA translocation,

electrophoresis was the most dominant electokinetic flow. As a result, we obtained a different

regime for translocation of the complex versus DNA. However, this hypothesis needs to be

investigated by ζ-potential measurements of the nanopore and each individual analyte.

Meanwhile, this change of direction in the translocation process can potentially provide a

new platform to separate mixtures of methylated and unmethylated DNA in a faster and more

robust manner compared to current technologies (e.g. MedIP assay).

Moreover, as a result of the limited temporal and spatial resolution of the current nanopore

devices, it was not possible to quantify and profile the CpG islands at the promoter regions.

Meanwhile, further studies to enhance the sensitivity of solid-state nanopores by the

combination of atomic layer deposition of Al2O3, reducing the nanopore thickness,

introducing a mobile lipid layer, etc. are in progress. The latter is of particular interest, as

specific antibodies or proteins can be incorporated into a lipid bilayer, allowing for

improvement of the specificity and affinity of the interaction as well as overcoming the


156

limitations of solid-state nanopores such as fast translocation processes, pore clogging and

hydrodynamic interaction of the analyte with the pore walls.

Finally, in chapter 5, an investigation into the presence of an electrostatic recognition step in

homologous segments of a bacteria plasmid, within the framework of Kornyshev-Leikin

theory, was attempted. In this study, two plasmids (cc-DNA) were constructed in such a way

that one consisted of two 1 kbp homologous segments and the other one contained no

homology in its sequence. However to this end, the verification of the electrostatic zipper

model was inconclusive. Based on AFM contour length analysis of the samples, it was

discovered that the plasmid with homologous regions was dimerised resulting in a single loop

of twice the expected size. This did not occur for the non-homologous cc-DNA. Accordingly,

a lower electrophoretic mobility was observed for the larger plasmid in gel electrophoresis

experiments, compared to the shorter, non-homologous sample.

In contrast to the gel data, DLS results revealed that, at low ionic strength, the

homologous/dimerised DNA exhibited a higher translation diffusion coefficient. This finding

was in contrast to expectations, where lower translational diffusion coefficients are predicted

for larger molecules. This deviation might be explained by the fact that during the DLS

experiment, the molar concentration of homologous/dimerised DNA was 2-fold lower than

the smaller plasmid. As aforementioned, a high concentration of analyte may result in

crowding effects, effectively decreasing the observed diffusion coefficient. Thereby,

independent experiments are required to examine the effect of DNA concentration and ionic

strength for difference sizes of plasmids, in order to determine the concentration- and ionic

strength- independent regimes in the present DLS study.

In addition to the DLS data, the nanopore translocation study was not fully in line with the

dimerisation scenario as a smaller conductance modulation was observed for the larger

plasmid. Nevertheless, one needs to recall that in a nanopore experiment other factors such as

surface charge effect, hydrodynamic interactions and flexibility of the analyte play critical

roles in the characteristics and features of the translocation events.

To this end, combining all the data, it was not possible to dismiss the effect of sequence –

dependent electrostatic homology interaction in some of our findings, nor confirm the

homology dependence of dimerisation. Further experiments with comparable length plasmids


157

are essential to potentially identify an electrostatic mechanism of homology recognition. In

conjunction with the methodologies that were conducted, X-ray scattering techniques and

optical studies would be beneficial to probe structural and physical properties of each type of

DNA sample used in this study.

Overall in this research project, by studying some of parts of the recombination process, as

well as genes and proteins regulations, we gained a better understanding on biophysical

properties of DNA-DNA and DNA-protein (antibody) interactions at single molecule level

within the framework of central dogma of molecular biology.

158

Appendices

Appendix I: Sequencing Data ............................................................................................................................. 159

Appendix II: Nanopore Data .............................................................................................................................. 161

Appendix III: AFM Data .................................................................................................................................... 162

Appendix IV: Copyright Permissions ................................................................................................................. 164

159

Appendix I: Sequencing Data

Sequencing data of FOXA1 promoter prepared by long range PCR (3388 nt)

TCTTTGTGTNNAAGCGTGCATTGCCAGTCCCTCCCCTAGGCCCCTAGCCCTAGGGGTTTTAAAGTAGAGCGGGAA

AGCCCGAGGATCCTTTCAGCAGCACAGAGCAGAGACCGCGCTCCCCAGGAGGGGGATCGGCTGGATGAGGAGGGA

CCGGCTGGGGCTGAGCACGTCCTCCAACAGGCGCTGACCGGTGTGAACAAAAGTCAGTAACTCGGGGGCCACAAC

GGCGGGAGACGCGCGCGGCCGCGGGGACCCAGGTCTGGGCGGGGTCGCCATATCGGTTCCGGGGGCCTCGAGGGA

GGTTTGTTCCCGTTTCCGCGAGGCCCTAGGGGGCACTTCTTCCTTCCAACGCTGTCGACTGCCCAGCAGGGGAAA

TCGCCCTTCTCAGTCTTTCTCATTTGAATACATGAAAATGCGAGTTGATTTTGGCAAGGCGATGCTCTCCCGTGG

CCAGAGGGACGGTTTTGTCGCCCGCGGGCGAGGCCGGGTGGGGAGCAGCTCCGGCTCCGACATCCGGCTGCGGGT

GAGCTAGGTCCGCGCCCGGAGCAGCTCAGAAACCCGGCGGCGCTCAGGCAGGAGTAGGGGAAANNACAAACCCAG

NNGNTGTTTTNTTGTTTGCTTCANNAAAATAAGTGAAANCCCTGGGTATGTTTATGCGTTGCTTTTACACACAAG

NCACACAGACACAAGCACACGCACACACAGCCGGGGTTGAGGCCCGTTACGGGTTAGGACTCTGTACAGAACTGC

CTGCAAAGGCCCGTCAAGTGGTCCCAGGGCCCCTACCTACTACAGAGTTGTGAATTAATCCCTTAAAGATTTAAT

GATGGAACCAAGGGAGAGAGGAAGGAAGATGAGGAAGAAAAATAAAAGGAAGGCAGCGAAGGAACGGAGTTCGAA

CCCTGGCCAGGCCCATACTTCTCTCTCGCGGGGGACCGCAGTCTCGCGGCTCCCTGGCCCCTTCCCCAGCTCGGC

GCACACATCCACCTGCGAAGGCCAGGCACACGCACGTGTGACCCCTTGTTTCCAGCGTCCCAGGCCTTCCTCACT

TCTAAGCTGAGCCGTGTCCCAGTTGGGGTCGGCTTTGAACCGCGTTAACGACCACAGGTCTTAGTTTCATTGCAT

CCGATTCTCCCCTGCAGAAGGAGGACTTAAGCTTCTTATCTGGGTGGGGGGTGGGGGCCCGCGGAGGAACTCGAA

CCACTCCACACCAAGCATGGCTGGGTTGCTGAAGCCGTCACCTGACAACAACCCCGTCCAGGGAAGGGTAGGCAA

GGAAAGGGGGCAGGCGAGGGACAGCCACAGAGGGAAGCCCTGGAGAGACTCGAATGGGCACCGCAGAGCCCCGCC

GGCCCAGGCCCTGCTGTCTGCCAACCCTGCGGTCCTGCTAAATAGGAAGGACCCTGTCGCCGCGAGAGGCCGCCG

TCTGGGCCCGGGGCCCGAGGCCCGAGGCCGCGGCCTCTCCTTCCACCTTTCGCCGTCGTCTCCCTCTGCTTCTAC

CTGCCTGGTGAATATCTGAGGAAGGGGCTCCCAGCCAGATCCTGCTGGGAAGTGTAAGCTGAGGCAGCCAAGCGT

GGTCAANGACNGAGTGCCGNNCAGNANTGNGGGANTGCGCCNGGAGNNNCCCCNCGGGTGTAGCAAGGGNAGCTT

CCGGGAAGGGNGGAGCCGCTTCCTCCGCTGGCGGCTNCCCGGAGTACCCGCGCACCCTACAGTCCTCACTGCCGC

GGATCCCCCACTTGAGGAGCCCGCCCTCCCCGCGACNTGGACCGCCCCAAGCTGTCGCAGACCCGTNTTAAACAC

AGGCAAGTTTAACCCGGGACACCGCAGGAGCCGCCACGTGCCTGCCCTCGCGGGTACCCTGCCCCGAGTCCACCC

ATCCTCCCCACAGCTGAAGACAGGCCTGGTTTCCTCCAACGAGCAGCACAATCCTTGCAAAGCACAGTTTCCAAT

GGTGTAGGTGCCTATTTGGGAAAAGAGGTTGGAAAAGAGCAGGCTGCAGCCGCTGGACCTGGCTACCGACTGGCC

AGCAACTCCCAGACAGCAGCATTACTTTATGTATGTATGTATGCATTTCCAAGTCAAAGACCCAACACGACCCTG

CTGATTCGCTTTTAAACCTCAGGGAAAGTGACTGGCTGGCATCTGGGTACACTAGAACCTCACCTCCTAAATTTG

TTAGGTATCTTGCAATGCGGCCACTGAGATGGAGAGAGGCTGGGAGCTGGACGAACGGCCATCTCTCCTTCCATC

TTGGCCTCGGCTCTAAGAACCTAAATCCCTGACTGGGATGCGTGCGGAGTTCAATCCAGTATCGCCTGGCGGTGC

TCCCCGGAGGCGCCTGCGGGAAAATTCAGCTTTCCCTCTCCACGACAGGAGGCGCTGTTTTCGTGGAAATCCCCC

GGCTGCGAACCTGGGATCCCTGACCTGGATCAAGTCTCCGAAGCTGGCAGAGTCCATTCTGCATCACCGGTCTTG

GGCTTTGAAGAAGCCTAGGAGAAATTCCGCTTCGGCCATCACGCTATGAAAAGTGGATTTTTTTTTCTTAAGTCA

ATTTTTTTTTTTGAAAATATGAGACTTAGTAGGTTTGGGAAGTGGGCTAAAAGAACATTTGATATTGTAATTGAC

CCCCCCTCCTTCCNTTCNNGANGGGGGGNNGTTTCCCCCCCCCCCCCCTTTNNAAAAAAAAAAAAAAAAAAATGT

AATNCAATGCTATTATCTTTTATTATNTCCTTAACNNNACATCTTCNCCTTNTCTGCTCNCCTGATGACTTAAGA

GATTTGTNTGGTTCAGGGATNTAAAGTGATTCTCTTGCCAGATTTCAAGTAAAAGAGATTTAAAAGAACAAAGCA

CAGGGAAAAAGGTATTTGTTTGGAAGACAAGTTAAAACACATTTCTTAAAATGAGATTAATAACATTTTAAAAAC

TTTGCAAAACAAGATTTTGCGGATTCTTAATTACTTTAGATTTTATTTTATTGTTACTTAAGGAAACCTAGTGGT

TCTACAGGCAGTACAACAAACACATGGTCACAGACACTCAGAAACACACACAGTCACACATGCTCAGAAATATAC

AAACGGTCACACACTCCAAACACACGCACCATCTCCCAATCTATCACCAACTAATTGCCTATCACCCGGTCACTT

160

CAGTTGTTCTCTCTCCCAGGACAAGTGGGCAACCACCACCCAGGGCCTCATGAGAGTAAAGAGACTTTGCGTTGG

GAAGACTCTCCCACCGACGTCCAGGACCGATTAGGAACCAAGCGAGCCCCTGAGATCTCAGCCCGGACCCTCCGG

GTCACAGACAGGACCAAGCGCCGCTGCGGGCAGTACTCGGGCTTCTGCCTGGCATCTCTTTCGCAGTGCAGATGC

GNNCCCGCGCCCA

Sequencing data of Kan insert in ap-DNA (1013 nt)

GGATCCGAAAAACTCATCGAGCATCAAATGAAACTGCAATTTATTCATATCAGGATTATCAATACCATATTTTTG

AAAAAGCCGTTTCTGTAATGAAGGAGAAAACTCACCGAGGCAGTTCCATAGGATGGCAAGATCCTGGTATCGGTC

TGCGATTCCGACTCGTCCAACATCAATACAACCTATTAATTTCCCCTCGTCAAAAATAAGGTTATCAAGTGAGAA

ATCACCATGAGTGACGACTGAATCCGGTGAGAATGGCAAAAGTTTATGCATTTCTTTCCAGACTTGTTCAACAGG

CCAGCCATTACGCTCGTCATCAAAATCACTCGCATCAACCAAACCGTTATTCATTCGTGATTGCGCCTGAGCGAG

ACGAAATACGCGATCGCTGTTAAAAGGACAATTACAAACAGGAATCGAATGCAACCGGCGCAGGAACACTGCCAG

CGCATCAACAATATTTTCACCTGAATCAGGATATTCTTCTAATACCTGGAATGCTGTTTTCCCGGGGATCGCAGT

GGTGAGTAACCATGCATCATCAGGAGTACGGATAAAATGCTTGATGGTCGGAAGAGGCATAAATTCCGTCAGCCA

GTTTAGTCTGACCATCTCATCTGTAACATCATTGGCAACGCTACCTTTGCCATGTTTCAGAAACAACTCTGGCGC

ATCGGGCTTCCCATACAATCGATAGATTGTCGCACCTGATTGCCCGACATTATCGCGAGCCCATTTATACCCATA

TAAATCAGCATCCATGTTGGAATTTAATCGCGGCCTAGAGCAAGACGTTTCCCGTTGAATATGGCTCATAACACC

CCTTGTATTACTGTTTATGTAAGCAGACAGTTTTATTGTTCATGACCAAAATCCCTTAACGTGAGTTTTCGTTCC

ACTGAGCGTCAGACCCCGTAGAAAAGATCAAAGGATCTTCTTGAGATCCTTTTTTTCTGCGCGTAATCTGCTGCT

TGCAAACAAAAAAACCACCGCTACCAGCGGTGGAATTC

Sequencing data of 1kbp Kan insert in p-DNA (1013 nt)

GGATCCCACCGCTGGTAGCGGTGGTTTTTTTGTTTGCAAGCAGCAGATTACGCGCAGAAAAAAAGGATCTCAAGA

AGATCCTTTGATCTTTTCTACGGGGTCTGACGCTCAGTGGAACGAAAACTCACGTTAAGGGATTTTGGTCATGAA

CAATAAAACTGTCTGCTTACATAAACAGTAATACAAGGGGTGTTATGAGCCATATTCAACGGGAAACGTCTTGCT

CTAGGCCGCGATTAAATTCCAACATGGATGCTGATTTATATGGGTATAAATGGGCTCGCGATAATGTCGGGCAAT

CAGGTGCGACAATCTATCGATTGTATGGGAAGCCCGATGCGCCAGAGTTGTTTCTGAAACATGGCAAAGGTAGCG

TTGCCAATGATGTTACAGATGAGATGGTCAGACTAAACTGGCTGACGGAATTTATGCCTCTTCCGACCATCAAGC

ATTTTATCCGTACTCCTGATGATGCATGGTTACTCACCACTGCGATCCCCGGGAAAACAGCATTCCAGGTATTAG

AAGAATATCCTGATTCAGGTGAAAATATTGTTGATGCGCTGGCAGTGTTCCTGCGCCGGTTGCATTCGATTCCTG

TTTGTAATTGTCCTTTTAACAGCGATCGCGTATTTCGTCTCGCTCAGGCGCAATCACGAATGAATAACGGTTTGG

TTGATGCGAGTGATTTTGATGACGAGCGTAATGGCTGGCCTGTTGAACAAGTCTGGAAAGAAATGCATAAACTTT

TGCCATTCTCACCGGATTCAGTCGTCACTCATGGTGATTTCTCACTTGATAACCTTATTTTTGACGAGGGGAAAT

TAATAGGTTGTATTGATGTTGGACGAGTCGGAATCGCAGACCGATACCAGGATCTTGCCATCCTATGGAACTGCC

TCGGTGAGTTTTCTCCTTCATTACAGAAACGGCTTTTTCAAAAATATGGTATTGATAATCCTGATATGAATAAAT

TGCAGTTTCATTTGATGCTCGATGAGTTTTTCGAATTC

161

Appendix II: Nanopore Data

Histogram analysis of nanopore data of ap and p-DNA.

Figure App. 1: (a) Bloackage current (∆I) and (b) dwellt time (d ) histogram anaalysis of p- and ap-DNA

translocation through a ~ 44 nm pore at voltage bias of (i) 150 mV, (ii) 200 mV and (iii) 300 mV. The ap-DNA

is blue column bars and p-DNA is black columns bars. The histograms are fitted with (skewed) Guaasian

distrubutions and colour coded with blue (ap-DNA) and black. p-(DNA) curves. The peak value of each curve

is presneted as the mot probale value in Table 5.1, section 5.4.3.

(b.i)

-5 -4 -3 -2 -1 00

20

40

60

80

100

Cou

nt

∆∆∆∆I (nA)

-5 -4 -3 -2 -1 00

20

40

60

80

100

Co

un

t

∆∆∆∆I (nA)

-5 -4 -3 -2 -1 00

20

40

60

80

100

Co

un

t

∆∆∆∆I (nA)

0 5 10 15 20 25 300

20

40

60

80

100

Cou

nt

ττττd (ms)

0 5 10 15 20 25 300

100

200

300

400

500

600

Co

un

t

ττττd (ms)

0 5 10 15 20 25 300

100

200

300

400

500

600

Co

un

t

ττττd (ms)

(a.i)

(a.ii)

(a.iii)

(b.ii)

(b.iii)

162

Appendix III: AFM Data

AFM data in air for supercoiled and relaxed ap- and p- DNA on Mg2+

modified mica

Figure App. 2: AFM data in air on Mg2+ modified mica. (a) Supercoiled ap-DNA, 2.5 µm × 2.5 µm scan (b)

supercoiled p-DNA, 2.5 µm × 2.5 µm scan, (c) relaxed apT-DNA, 5.0 µm × 5.0 µm scan, (d) relaxed pT-DNA,

10 µm × 10 µm scan.

400 nm 400 nm

(a)

(c)

(b)

(d)

1600 nm800 nm800 nm

163

AFM data in air for supercoiled and relaxed ap- and p- DNA on APTES modified mica

Figure App. 3: AFM data in air on silinised (APTES) modified mica. (a) Supercoiled ap-DNA, 1.0 µm × 1.0

µm scan (b) supercoiled p-DNA, 2.65 µm × 2.65 µm scan, (c) relaxed apT-DNA, 1.5 µm × 1.5 µm scan, (d)

relaxed pT-DNA, 1.0 µm × 1.0 µm scan.

(a)

(c)

(b)

(d)

160 nm 500 nm

250 nm 160 nm

164

Appendix IV: Copyright Permissions

Table App. 1: Summary of permissions for third party copyright works.

Chapter and

page

number

Type Source Copyright

holder Permission

Licence

number

Chapter 1, Page 2

Figure 1.1.(a)

Nature (1953), vol 171, p 740-741

© 1953, Nature Publishing Group

Granted 3310781191367

Chapter 1, Page 2

Figure 1.1.(b)

Nature (1953), vol 171, p 737-738

©1953, Nature Publishing Group

Granted 3310780241604

Chapter 1, Page 4

Figure 1.2

Molecular Biology (2008), 4th Ed, ISBN: 978-0-07-

110216-2, chapter 2, p 24

© 2008, McGraw-Hill

Granted AZA37745

Chapter 1, Page 30

Figure 1.13

Soft Matter (2010), vol 6, p 3402-3429

© 2010, Royal Society of Chemistry

Granted 3310820233895

Chapter 1, Page 32

Figure 1.14

Biophysical Journal (2011), vol 101, p 875-884

© 2011, Elsevier Granted 3310820770863

Chapter 3, Page 60

Figure 3.1.(a)

Chemical Society Reviews (2011), vol 38, p 2360-

2384

© 2010, Royal Society of Chemistry

Granted 3310821164024

Chapter 3, Page 60

Figure 3.1.(b)

Biophysical Journal (1999), vol 77, p 3227-3233

© 1999, Elsevier Granted 3310830225419

Date post:	03-Aug-2020
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

Biophysical study of DNA at single molecule level using solid … · 2017-02-18 · Commercial No...

Documents