DEVELOPMENT AND APPLICATION OF NEW APPROACHES FOR STUDIES OF INFLUENZA INDUCED INFLAMMATION AND DNA DAMAGE

... Investigation of DNA damage as a potential mechanism of influenza pathogenesis 66 3.1.2 Phosphorylation of H2AX (Ser-139) during DNA strand breaks 69 3.1.3 DNA damage is triggered by influenza. .. to sense for DNA damages, activate cell signaling and repair DNA lesions DNA damage are surveyed via DNA damage sensors, Mre11-Rad50-Nbs1 (MRN) complex, and replication protein A (RPA) and Rad9-Rad1-Hus1... more harmful DNA strand breaks Taken together, DNA damage (in the form of DNA strand breaks) can be induced directly by ROS / RNS, as well as indirectly during the repair of ROS / RNS -induced base

DEVELOPMENT AND APPLICATION OF NEW APPROACHES FOR STUDIES

OF INFLUENZA-INDUCED INFLAMMATION AND

DNA DAMAGE

Li Na

B.Sc (Hons), NUS

A THESIS SUBMITTED FOR THE

DEGREE OF DOCTOR OF PHILOSOPHY IN SCIENCE

DEPARTMENT OF MICROBIOLOGY

YONG LOO LIN SCHOOL OF MEDICINE

NATIONAL UNIVERSITY OF SINGAPORE

2015

i Declaration:

I hereby declare that the thesis is my original work and it has been written by me

in its entirety I have duly acknowledged all the sources of information which have been used in the thesis This thesis has not been submitted for any degree

in any university previously

_

Li Na

ii Acknowledgements:

First and foremost, I would like to express my sincere gratitude to my supervisors Professor Bevin Engelward and Professor Vincent Chow, for offering me the opportunity to be a part of their labs, for their intellectual and moral support, for their inspiration in doing good science, and for teaching me how to present myself and my work I would like to show my appreciation to my thesis advisory committee members Professor Fred Wong and Professor Marie Clement for their invaluable comments and suggestions for my thesis, and to my collaborators, Dr Damien Thévenin and Professor Donald Engelman

Many thanks to my lab mates and colleagues in Singapore – Dr Yamada Yoshiyuki, Prashant, Tze Khee, Kai Sen, Anandi and Dr Orsolya Kiraly- for lending me support in my projects, for giving me constructive suggestions, and most of all, for your friendship I will like to specially mention my appreciation towards Dr Yamada, who had mentored me for 3 years and had taught me useful skills and good lab practices In addition, my project would not have been possible if not for Mrs Phoon who painstakingly propagated all the viruses To my lab mates in MIT- Marcus, Jenny, Shelly, Lizzie, Yang Su, Jing- thank you for being such wonderful colleagues and for the good scientific exchanges we have had To all my SMART friends, the “Aunties of SMART”, the lung repair group, and my fellow SMA3 students, thank you for being great colleagues and amazing friends too, making

my time in SMART an extremely enjoyable one I am going to miss you guys!

And most importantly, I would like to thank my parents for believing in me and for being my pillar of support And last but not least, to my boyfriend, for being there with me, always, to listen, to share, to advice and to empathize Thank you all for

giving me the strength, courage and wisdom needed to complete this long and demanding journey

Some passages and images in chapter 2 and chapter 4 are quoted verbatim or reprinted from “Na Li, Yin Lu, Thevenin Damien, Yamada Yoshiyuki, Limmon Gino,

et al (2013) Peptide targeting and imaging of damaged lung tissue in infected mice Future Microbiol 8: 257-269.” with permission from Future Microbiology

influenza-Some passages and images from chapter 2 and chapter 5 are quoted and reprinted from “Sukup-Jackson R Michelle, Kiraly Orsolya, Kay Jennifer, Na Li, Rowland, Kelly E Winther Elizabeth A., Chow Danielle N., Kimoto Takafumi, Matsuguchi Tetsuya, Jonnalagadda Vidya S., Maklakova Vilena I., Singh Vijay R., Wadduwage Dushan N., Rajapakse Jagath, So Peter T C., Collier Lara S., Engelward Bevin P (2014) Rosa26-GFP direct repeat (RaDR-GFP) mice reveal tissue- and age-dependence of homologous recombination in mammals in vivo PLoS Genet 10: e1004299.” PLoS Genetics is an open-access journal, so no permission from the publisher was needed for using the materials Passages quoted in this thesis were contributed to the journal by the author of this thesis

iii Funding:

This thesis is supported by the Singapore National Research Foundation (NRF), Ministry of Education (MOE) and in part by the National Institute of Environmental Health Sciences (P01-ES006052) The views expressed herein are solely the responsibility of the author and do not necessarily represent the official views of the funding bodies

iv Publications:

Yoshiyuki Yamada, Gino V Limmon, Dahai Zheng, Na Li, Liang Li, Lu Yin, Vincent

T K Chow, Jianzhu Chen, Bevin P Engelward (2012) Major shifts in the temporal distribution of lung antioxidant enzymes during influenza pneumonia PLoS One 7: e31494

spatio-Na Li, Lu Yin, Damien Thévenin, Yoshiyuki Yamada, Gino Limmon, Jianzhu Chen,

Vincent TK Chow, Donald M Engelman, Bevin P Engelward (2013) Peptide targeting and imaging of damaged lung tissue in influenza-infected mice Future

Sukup-Jackson R Michelle, Kiraly Orsolya, Kay Jennifer, Na Li, Rowland, Kelly E

Winther Elizabeth A., Chow Danielle N., Kimoto Takafumi, Matsuguchi Tetsuya, Jonnalagadda Vidya S., Maklakova Vilena I., Singh Vijay R., Wadduwage Dushan N., Rajapakse Jagath, So Peter T C., Collier Lara S., Engelward Bevin P (2014) Rosa26-GFP direct repeat (RaDR-GFP) mice reveal tissue- and age-dependence

of homologous recombination in mammals in vivo PLoS Genet 10: e1004299

Na Li, Marcus Parrish, Tze Khee Chan, Lu Yin, Prashant Rai, Yamada Yoshiyuki,

Nona Abolhassani, Kong Bing Tan, Orsolya Kiraly, Vincent T K Chow, Bevin P Engelward Influenza infection induces host DNA damage and dynamic DNA damage responses during tissue regeneration Cell Mol Life Sci 2015 Mar 26

[Epub ahead of print]

v Conference Abstracts:

Na Li, Liang Li, Damien Thévenin, Yoshiyuki Yamada, Yin Lu, Gino Limmon, Vincent TK Chow, Donald M Engelman, Bevin P Engelward Peptide targeting and imaging of damaged lung tissue in influenza-infected mice Poster session presented at: Asia- Pacific Congress of Medical Virology; 2012, 6-8 June; Adelaide, Australia

Na Li, Damien Thévenin, Liang Li, Yoshiyuki Yamada, Yin Lu, Gino Limmon, Vincent TK Chow, Donald M Engelman, Bevin P Engelward Peptide targeting and imaging of damaged lung tissue in influenza-infected mice Poster session presented at: Singapore - Japan Joint Forum On Emerging Concepts In

Microbiology; 2011, 15-16 November; Singapore

vi Abbreviations:

8-OH-G / 8-OH-dG 8-hydroxyguanosine/ 8-hydroxy-deoxyguanosine

ARDS Acute respiratory distress syndrome

AEI/ AEII Alveolar epithelial type I/II cells

DNA-PK DNA- dependent protein kinase

DNA-PKcs DNA- dependent protein kinase catalytic subunit

iNOS Inducible nitric oxide synthase

MDCK cells Madin-Darby canine kidney cells

NAD+ Nicotinamide adenine dinucleotide

PARP-1 Poly (ADP-ribose) polymerase 1

PCNA Proliferating cell nuclear antigen

PIKK Phophatidylinositol-3-kinase-like kinases

RaDR-GFP Rosa26 Direct Repeat-Green Fluorescent Protein

Pro-SPC Pro- surfactant protein C

ROS/RNS Reactive oxygen and nitrogen species

TUNEL Terminal deoxynucleotidyl transferase dUTP nick end labeling

TABLE OF CONTENTS

I DECLARATION: I

II ACKNOWLEDGEMENTS: II

III FUNDING: IV

IV PUBLICATIONS: V

V CONFERENCE ABSTRACTS: VI

VI ABBREVIATIONS: VII

TABLE OF CONTENTS IX

SUMMARY 1

LIST OF FIGURES 4

LIST OF TABLES 7

CHAPTER 1 BACKGROUND 8

1.1 Influenza virus 8

1.1.1 Life cycle of Influenza virus 10

1.1.2 Antigenic drift and antigenic shift 14

1.1.3 Global burden imposed by influenza outbreaks 17

1.2 Influenza infection induces acute lung injury 19

1.2.1 Influenza-induced cytopathy and lung injury 20

1.2.2 Inflammatory responses and lung injury 20

1.2.3 Reactive oxygen and nitrogen species during influenza infection 21

1.3 ROS and RNS can damage genomic DNA 26

1.3.1 Cellular responses towards DNA damage 29

1.3.2 Base excision repair removes damaged bases 30

1.3.3 DNA damage can arise from BER intermediates 32

1.3.4 DNA double strand break (DSBs) repair 33

1.4 Studying significance of DNA repair pathway in vivo 37

1.4.1 Genetically engineered mouse that enable visualization of HR events 38

1.6 Thesis aims 41

CHAPTER 2 METHODS AND MATERIALS 42

2.1 Materials 42

2.1.1 Media, chemicals and reagents 42

2.1.2 Cell cultures 43

2.1.3 List of antibodies for western blot and immunofluorescence 44

2.1.4 List of antibodies for flow cytometry 45

2.1.5 Source of influenza viruses 45

2.1.6 pHLIP and pHLIP variant peptide sequences and conjugation 46

2.2 Methods and protocols 47

2.2.1 Infection of mice and tissue collection 47

2.2.2 Lung homogenization and virus titration 48

2.2.3 Lung histology, infiltration index calculation and pathology analysis 48

2.2.4 Measurement of ROS markers and TNF-α 49

2.2.5 Western blotting 50

2.2.6 Flow cytometry of immune cells 50

2.2.7 Immunofluorescence assay 51

2.2.8 Microscopy 52

2.2.9 Manual and semi-automated quantification of H2AX foci 53

2.2.10 Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) and quantification 54

2.2.11 Microarray data analysis 55

2.2.12 NU7441 treatment of infected mice 56

2.2.13 RaDR mice and single cell suspension preparation 56

2.2.14 RaDR cells RNA extraction and cDNA conversion 57

2.2.15 Direct PCR analysis using RNA transcripts 58

2.2.16 Nested PCR analysis for Full-length EGFP 59

2.2.18 In vitro experiment with pHLIP 62

2.2.19 Peptide injection in mice 62

2.2.20 Whole body and ex vivo whole organ bioimaging 62

2.2.21 Feature extraction and pHLIP quantification 63

2.2.22 Statistical analysis 65

CHAPTER 3 INFLUENZA INFECTION INDUCES DNA DAMAGE AND ROBUST DNA DAMAGE RESPONSES IN VIVO 66

3.1 Introduction 66

3.1.1 Investigation of DNA damage as a potential mechanism of influenza pathogenesis 66

3.1.2 Phosphorylation of H2AX (Ser-139) during DNA strand breaks 69

3.1.3 DNA damage is triggered by influenza infection in vitro 72

3.1.4 Aims of study 74

3.2 Results 76

3.2.1 Characterization of H1N1 murine model 76

3.2.2 Prolonged inflammation after suppression of virus load 76

3.2.3 Broad characterization of pulmonary inflammation 81

3.2.4 Oxidative stress is elevated during infection 84

3.2.5 Host responses induce DNA damage in lung epithelium after influenza infection 86 3.2.6 DNA damage occurs in immune cell populations 94

3.2.7 Influenza infection induces polymerization of poly (ADP- ribose) 96

3.2.8 Influenza infection leads to apoptosis 99

3.2.9 Infection increases DNA damage among dividing cells 102

3.2.10 Influenza infection induces DSB repair proteins during tissue regeneration 105

3.2.11 NHEJ component, DNA-PK, is dispensable for DNA repair during tissue regeneration 109

3.3 Discussion 113

3.4 Conclusion 120

3.5 Future studies 121

3.5.1 Verification of forms of DNA damage 121

3.5.2 DNA repair deficiencies on cell fate and disease outcome 122

CHAPTER 4 FLUORESCENT DETECTION OF HOMOLOGOUS RECOMBINATION EVENTS FOLLOWING INFLUENZA INFECTION 124

4.1 Introduction 124

4.1.1 Homologous recombination 124

4.1.2 Approach for studying HR during influenza infection in vivo 125

4.1.3 Aims of Study 129

4.2 Results and Discussion 131

4.2.1 PCR-based analysis of cDNA 131

4.2.2 Improved sensitivity with nested PCR 132

4.2.3 PCR analysis of HR mechanism in single cells 136

4.2.4 EGFP expression to monitor HR events in lung epithelial cells 140

4.3 Conclusion and Future studies 143

CHAPTER 5 IMAGING AND TARGETING OF PH (LOW) INSERTION PEPTIDE (PHLIP) AT INFLAMED LUNG TISSUE DURING INFLUENZA INFECTION 144

5.1 Introduction 144

5.1.1 Targeted drug delivery as a therapeutic strategy 144

5.1.2 pHLIP: pH (low) insertion peptide 147

5.1.3 Inflammation changes pH of tissue microenvironment 151

5.1.4 Aims of study 152

5.1.5 Specific aims 153

5.2 Results 154

5.2.1 pHLIP targets influenza-infected lungs 154

5.2.2 pHLIP also targets other acidic tissue 156

5.2.3 pHLIP has high retention in inflamed lungs as compared to other organs 158 5.2.4 Design of pHLIP variants 160

5.2.5 Evidence of pH-dependent pHLIP accumulation in infected lungs 165

5.3 Discussion 177

5.4 Conclusion 181

5.5 Future studies 181

5.5.1 pHLIP delivery of antioxidants 181

CHAPTER 6 CONCLUSIONS 183

REFERENCES: 188

Summary

Influenza viruses are a group of highly contagious pathogens that account for significant morbidity and mortality worldwide Extensive studies have suggested that dysregulated inflammatory responses including excessive production of reactive oxygen and nitrogen species (ROS / RNS) mediate lung injury in severe infections However, underlying mechanisms and effective treatment strategies for inflammation-induced injury are not fully established The aims of this thesis have been to broadly characterize an influenza-pneumonia mouse model, and to examine the impact of influenza infection on host genomes using the mouse model Using the mouse model, the potential of a low pH targeting peptide as a delivery agent that can be used to improve treatment for inflammation-induced lung injury

inflammation-genetic rearrangements and cytotoxicity that can impede recovery process Results show that DSB repair proteins Ku70 / Ku86 and Rad51 are upregulated during the later stage of infection, implicating DSB repair pathways in the resolution

of DNA damage during lung tissue recovery Taken together, the data demonstrate that DNA damage is associated with influenza pneumonia, and raise a possibility that DNA repair capacity may be a determinant for tissue recovery

While investigating disease mechanisms may set a foundation for identifying therapeutic targets for severe influenza infection, more effective treatment may be achieved by modifying delivery of existing treatments Optimized drug delivery in terms of dose and biodistribution are important challenges for treating inflammation-induced lung injury during influenza infections In the second part of this thesis, we investigated whether pH (Low) Insertion Peptide™ (pHLIP™), a peptide with high affinity for acidic microenvironments, can be used for specific delivery of drugs or imaging agents in influenza patients This study is designed based on the understanding that inflamed tissue is acidic while healthy tissues are slightly alkaline To test our hypothesis that pHLIP localize at sites of inflammation, fluorophore-conjugated pHLIP was injected into infected mice to track peptide

distribution in vivo Results show that pHLIP specifically targets inflamed lung

during the later stage of infection, when severe pneumonia manifest In addition, pHLIP-targeted lung tissues are injured and devoid of alveolar type I (AEI) and type II cells (AEII) that are found in healthy alveoli Interestingly, pHLIP-targeted lung tissue is also characterized by depletion of Peroxiredoxin 6 (Prdx6), an enzymatic antioxidant abundantly expressed in the lung Importantly, the specific distribution of pHLIP in inflamed tissue opens up opportunities for the delivery of pHLIP-drug conjugates to ameliorate lung injury Taken together, the results

provide new insights into the molecular pathology of influenza pneumonia, and offer opportunities to improve the management of influenza-induced lung disease

List of figures

Figure 1.1 Structure of an influenza virus 10

Figure 1.2 The life cycle of influenza virus 11

Figure 1.3 Inflammation-induced ROS / RNS damages cellular molecules 25

Figure 1.4 Structures of modified nucleobases upon exposure to ROS / RNS 28

Figure 1.5 Classical DNA double strand break repair pathways 36

Figure 3.1 PR8 infection of MDCK cells induced H2AX foci formation 73

Figure 3.2 Disease progression during sublethal PR8 infection 78

Figure 3.3 Characterization of inflammation kinetics 83

Figure 3.4 Real-time PCR analysis of IFN- in lung homogenate 84

Figure 3.5 Oxidative stress is elevated especially when viral load is suppressed 85

Figure 3.6 Reduction in superoxide dismutase (SOD) activity on 7 dpi 86

Figure 3.7 Western analysis of influenza antigen and H2AX 87

Figure 3.8 DSB markers, H2AX and 53BP1 foci, were induced in bronchiolar epithelium cells after active infection 89

Figure 3.9 H2AX foci were induced in alveolar epithelial type II cells in the recovery phase of influenza infection 91

Figure 3.10 Increased H2AX foci formation in both infected and uninfected cells 92

Figure 3.11 Pan-nuclear H2AX is found in the same regions as cleaved caspase 3 positive cells in sequential sections 93

Figure 3.12 H2AX foci formation increases in immune cells after infection 95

Figure 3.13 PARP mediated PAR polymerization increases after infection 97 Figure 3.14 PARP-1 is cleaved following influenza infection 99

Figure 3.15 DNA damage is initiated before, and proceeds after whole lung apoptosis 101

Figure 3.16 H2AX phosphorylation increases in Ki-67+ cells after

infection 104 Figure 3.17 NHEJ and HR-related genes are induced during tissue

regeneration 107 Figure 3.18 HR-related genes are induced during the later stage of

infection 108 Figure 3.19 DNA-PKcs activity is dispensable for DNA repair in bronchial epithelium on 9 and 13 dpi 111 Figure 3.20 PR8 infected mice injected with 2 doses of NU7441 at 9 dpi 112

Figure 4.1 RaDR-GFP HR substrate recombines to encode for full length EGFP sequences 127 Figure 4.2 PCR detection of HR that led to full length EGFP reconstitution in

RaDR-GFP pancreatic cells 135 Figure 4.3 Single strand annealing does not reconstitute RaDR-GFP 136 Figure 4.4 RaDR-GFP substrate yields different recombination products following gene conversion, sister chromatid exchange, and replication fork repair 138 Figure 4.5 Nested PCR analysis of single cells could identify cells that had undergone crossover associated-HR or gene conversion without

crossover 140 Figure 4.6 EGFP positive cells arise in the lung epithelial cells indicating HR

at Rosa26 locus 142 Figure 5.1 pHLIP peptide sequences 147 Figure 5.2 pHLIP exists in 3 states and facilitates bidirection transport of cargo molecules 150 Figure 5.3 pHLIP preferentially binds cell membrane at acidic pH in vitro 151 Figure 5.4 pHLIP targets infected but not uninfected mice lungs 155 Figure 5.5 pHLIP targets acidic tissue 157 Figure 5.6 Bioimaging of pHLIP-induced fluorescence on mice ventral

surface to estimate peptide clearance 159 Figure 5.7 pHLIP is retained in infected lungs 8 days post peptide

Figure 5.9 pH responsiveness of pHLIP variants, Leu26Gly, K-pHLIP and D3K-pHLIP 164 Figure 5.10 Evidence of pH-dependent targeting of pHLIP and Leu26Gly 167 Figure 5.11 pHLIP targets heavily infiltrated regions of lungs 170 Figure 5.12 pHLIP accumulates in lungs during the later phase of

infection 172 Figure 5.13 pHLIP accumulates in damaged lung tissue 174 Figure 5.14 pHLIP localizes at lung tissue depleted of Prdx6 176 Figure 5.15 Quantification of pHLIP-induced pixels in Prdx6 positive and negative regions 177

List of tables

Table 1.1 Licensed influenza anti-viral agents 16

Table 2.1 Commercial sources of media and reagents 42

Table 2.2 Formulations for solutions and buffers 43

Table 2.3 Sources and clones of antibodies used 44

Table 2.4 Panel 1 antibodies for myeloid cells 45

Table 2.5 Panel 2 antibodies for lymphoid cells 45

Table 2.6 Sequences and molecular weights of pHLIP and peptide variants 46

Table 2.7 Antigen retrieval methods 51

Table 2.8 Mean number of Ki-67+ and H2AX+ cells counted in 15 40x magnified images 54

Table 2.9 PCR primers to specifically amplify full length EGFP, Δ3egfp, and Δ5egfp 59

Table 2.10 External PCR primers designed to anneal upstream and downstream of the EGFP coding sequence 60

Table 2.11 Thermal cycler conditions and product sizes 60

Table 3.1 Summary of DNA damage in infectious diseases 68

Table 3.2 Interplay between viral infection and DDR pathway that is independent of DNA damage 69

Table 3.3 Pathology analysis of mice lung sections following PR8 infection 79

Table 5.1 Anti-inflammatory agents and their reported efficacy on influenza infection 146

Table 5.2 Alterations in sequences and Molecular weight of pHLIP variants 161

Table 5.3 Predicted aggregation score and reported pH-sensitivity 163

Chapter 1 Background

1.1 Influenza virus

Influenza viruses (commonly known as flu) are airborne viruses from the orthomyxoviridae family There are three genera (or antigenic type) of influenza viruses in this family -Influenza A, Influenza B and Influenza C – among which only influenza A and influenza B are known to cause severe respiratory illnesses and epidemic outbreaks in the world In contrast, Influenza C give rise to mild respiratory diseases in human and is not thought to contribute to annual epidemics (Zambon 1999)

Influenza viruses have been commonly found to be roughly spherical in shape, and are made up of lipid-bilayer envelopes derived from host plasma membrane Projecting out of the surface of each influenza virion, are approximately 500 spike – liked glycoproteins, haemagglutinin (HA) and neuraminidase (NA), that make up

~ 80% and ~ 17% of all viral envelope proteins respectively Around 16 to 20 molecules of another minor viral antigen, matrix 2 protein (M2), can also be found

on the envelope of each virion Right underneath the viral envelope, there is a spread - out distribution of matrix 1 proteins (M1), which are responsible for binding viral ribonucleoprotein (vRNP) complexes located in the core of each virus particle (Figure 1.1) (Samji 2009; Ruigrok et al 1984)

There are eight rod-shaped vRNP complexes in a virion, and each vRNP complex

is composed of a viral RNA associated with multiple nucleoproteins (NP) and a trimeric viral polymerase complex The 5’ and 3’ termini of viral RNA are partially complementary, allowing vRNP to form a helical corkscrew structure Influenza viruses possess eight segmented, single-stranded, negative sense RNA genomes

approximately 13 thousand bases in size Each RNA segment forms a single vRNP with NP monomers and viral polymerase complex in a mature virion These eight segments are historically known to encode for 10 viral proteins, namely three polymerases (PA, PB1, PB2), NP, M1 and M2, HA, NA and non-structural proteins (NS1 and NS2) (Figure 1.1) More recently, influenza has also been shown to express up to 16 viral proteins, which include PB1-F2 (encoded by the +1 alternate open reading frame within PB1 gene), a M2-related protein M42, N40 (encoded by PB1), PA-X, PA-N155 and PA-N182 (encoded by PA) (Shi et al 2014; Chakrabarti and Pasricha 2013)

Influenza viruses are highly diverse pathogens, with different viruses possessing varied transmissibility, pathogenicity and infectivity in hosts Many subtypes of Influenza A viruses exist, and they are classified based on their expression of different HA and NA antigens Up to early 2015, a total of 18 HA and 11 NA antigens have been identified among influenza A viruses isolated from humans, birds and bats (Freidl et al 2015) Based on the expression of different HA and NA antigens, Influenza A viruses can be broadly categorized into subtypes such as H1N1, H3N2 and H5N1, which can then be further subdivided into strains On the contrary, there is only one subtype of Influenza B, though it can also be further subdivided into lineages and strains, much like Influenza A (CDC1) Based on a commonly accepted naming convention published in the Bulletin of the World Health Organization (WHO) during 1980, individual Influenza strains are identified

by the antigenic type, the host of origin, the geographical origin where the strain was isolated, the strain number, the year of isolation, followed by the HA and NA

antigen description in parentheses for Influenza A viruses [e.g Influenza A/Puerto Rico/8/37 (H1N1) and Influenza B/Yamagata/16/88] (WHO 1980)

Figure 1.1 Structure of an influenza virus

Influenza is an enveloped virus composed of two major surface glycoproteins, haemagglutinin (HA) and neuraminidase (NA), and a minor component M2 The genome

of influenza consists of 8 RNA segments, which are folded into ribonucleoprotein complexes (RNP) and encode for nucleoprotein (NP), three polymerase proteins (PA, PB1, and PB2), matrix proteins (M1 and M2), nonstructural proteins (NS1 and NS2) and 2 glycoproteins (HA and NA) (Image is taken from Nelson MI, Holmes EC The evolution of

epidemic influenza.Nat Rev Genet 2007 Mar;8(3):196-205)

1.1.1 Life cycle of Influenza virus

The life cycle of an influenza virus occurs in several steps, which will be divided into five main stages here to facilitate description: (1) viral binding and entry into host cells, (2) uncoating and vRNPs translocation into the nucleus, (3) transcription and replication of viral RNA, (4) synthesis of new vRNPs and finally, (5) assembly and budding (Figure 1.2)

Figure 1.2 The life cycle of influenza virus

When influenza virus infects a host, viral envelope protein HA first binds to sialylated host receptors that facilitate the entry of the virus into host cells via receptor – mediated endocytosis Endosomal acidity then triggers the fusion of viral and endosomal membranes, resulting in uncoating of viral membrane and release of vRNP complexes into the cytosol Subsequently, vRNP complexes translocate into host cell nucleus, where viral RNA segments are transcribed and replicated In the cytoplasm, host translational machinery is exploited to synthesize viral proteins including NP, PA, PB1 and PB2, which are then translocated into the nucleus to assemble into novel vRNPs along with newly synthesized viral RNA These newly generated vRNPs are then transported out of the nucleus with the help of M1 and NS2 Finally, viral components are assembled at the cell membrane, where progeny virus buds out at the apical surface of host plasma membrane (Image is taken

from Shi Y, Wu Y, Zhang W et al Enabling the 'host jump': structural determinants of

receptor-binding specificity in influenza A viruses Nat Rev Microbiol 2014 Dec;12(12):822-31.)

step is thought to be an important determinant of host infectivity since avian influenza viruses generally binds to α 2, 3 - linked (avian-type) sialic acid receptors that are dominantly found in the avian respiratory and gastrointestinal tracts, while human influenza viruses preferentially bind to α 2,

6 - linked (human-type) sialic acid receptors that are prevalent in the human airways Other animals, such as the pigs, Japanese quail and mice possess both α 2, 3 - linked and α 2, 6 - linked sialic acid receptors, and hence, can be infected by both avian and human influenza viruses [Reviewed in (Kimble, Nieto, and Perez 2010; Ning et al 2009)] Following binding, Influenza virus enters the host cell inside an endosome via receptor-mediated endocytosis

(2) Acidity in the endosome (~pH 5 - 6) leads to a change in the conformation of

HA trimer, allowing HA2 (a subunit of HA) to facilitate fusion of the viral and endosomal membranes, leading to the formation of a pore which gives passage for vRNP complexes to enter the cytosol M2, a proton ion channel, further pumps H+ into the viral core, decreasing the pH of the viral core and disrupting the interactions between vRNP complexes and M1, so that vRNP particles can now freely move into the cytoplasm Nuclear localization signals found in vRNP proteins (NP, PA, PB1 and PB2) then mediates the entry of vRNP complexes into host nucleus using host cell’s nuclear import machinery

(3) In the nucleus, negative sense viral RNA are then transcribed into positive sense messenger RNAs (mRNAs) and complementary RNAs (cRNAs) using the trimeric viral RNA- dependent RNA polymerase (RdRp) composed of PA, PB1 and PB2 PB2, containing endonuclease activity, binds and cleaves 5’ methylated caps of host mRNAs via a “cap- snatching” mechanism to prime viral transcription As such, viral mRNAs will possess 5’ methylated caps even

though 5’ caps are not encoded in the viral genome Newly synthesized mRNAs are subsequently transported into the cell cytosol to be translated into viral proteins by exploiting the host translational machinery cRNAs remain in the nucleus as templates for synthesis of multiple copies of negative sense viral RNA

(4) Newly synthesized NP and viral polymerases are transported back into the nucleus, where they can assemble with negative sense viral RNA to form vRNPs These negative sense vRNPs are subsequently exported out of the nucleus through nuclear pores via exportin-1 (also known as CRM1) – mediated nuclear export, potentially via their interaction with M1 and NS2, which will then bind to exportin-1 (Samji 2009; Elton et al 2001)

(5) In order for budding to occur, influenza components must first assemble at the plasma membrane Following protein synthesis in the endoplasmic reticulum, envelope antigens HA, NA and matrix proteins are transported by the trans- Golgi network to host plasma membrane Newly synthesized vRNPs also bind

to M1 that are assembled at the inner surface of host membrane bilayer, preventing the re-entry of vRNPs into the nucleus If influenza replication occurs in the respiratory tract, viral components assemble at the apical side of polarized epithelial cells where progeny viruses will exclusively emerge from,

so that they can re-infect other epithelial cells in the airway, instead of entering the circulatory system After budding, progeny viruses are still bound to host cell membrane via HA and sialic-receptor interaction NA (and other host proteases) then cleaves sialic acid residue from surface glycoproteins and

1.1.2 Antigenic drift and antigenic shift

Influenza is a highly successful virus Based on historical evidences, influenza viruses have circulated in the human population since more than half a millennium ago (Taubenberger and Morens 2010), and yet, despite the advent in vaccines and anti- viral therapies, it is still impossible to eradicate influenza virus Influenza

A is commonly known to be the most virulent genus of influenza viruses, and has been found to infect many animals such as birds, human, swine, horses, whales, seals and bats, among which wild birds are thought to be the primary natural reservoir of influenza A virus (Spackman 2009) The success of influenza A virus can be partly attributed to its error-prone RdRp, which makes incorporation errors

at an estimated rate of 7.2×10−5 bp−1 per replication cycle (approximately 1 mistake per genome per cycle) (Drake 1993) High error rate in genome replication allows influenza A viruses to gradually mutate and develop modified antigens that escape recognition by pre-existing host immunity This process, called antigenic drift, is usually the cause of recurrent epidemics within communities which do not have specific immunity against the new virus

The ability of influenza A viruses to mutate also allows them to develop resistance towards currently available anti-viral agents, thereby creating challenges for treating influenza - infected patients There are currently only five licensed influenza anti-viral agents in the United States of America (CDC2), out of which only three NA inhibitors, Tamiflu® (Oseltamivir phosphate), Relenza® (Zanamivir) and Rapivab® (Peramivir), are recommended by U.S Food and Drug

2 Influenza Antiviral Medications: Summary for Clinicians (2015, February 25) Centers for Disease Control and Prevention.Retrieved on 30 May 2015, from

http://www.cdc.gov/flu/professionals/antivirals/summary-clinicians.htm

Administration (FDA) The other two M2 blockers, Symmetrel® and Flumadine®, have been in use for therapeutic purposes for a longer duration of time, but are no longer recommended as prescriptions for influenza therapy and prophylaxis This

is because amantadine and rimantadine, the compound names for Symmetrel® and Flumadine® respectively, are found to be ineffective during recent outbreaks with a widespread emergence of resistant H3N2 and 2009 pandemic H1N1 viruses, such that more than 99 % of all strains tested are resistant to both compounds In contrast, Tamiflu®, Relenza® and Rapivab® (recently approved agent) are still largely effective for the control of seasonal H3N2 and 2009 pandemic H1N1 viruses (more than 98% of viruses tested are susceptible) However, H1N1 viruses resistant towards Tamiflu® and Rapivab® due to single amino acid mutations (H274Y or N294S alterations) in the NA protein have also began to emerge (Hurt

et al 2009; Pizzorno et al 2011) (Table 1.1) Given the general effectiveness of Tamiflu® and Relenza® towards recent circulating strains, the Health Science Authority of Singapore recommends prescribing these two drugs for treating influenza infected patients

Table 1.1 Licensed influenza anti-viral agents

(Information taken from CDC3 and Health Science Authority4)

NA inhibitor Oral - Less than 2% of tested 2009 H1N1

viruses were resistant to H1N1

- Not effective towards Influenza A strains with H274Y or N294S alterations

- Effective with Influenza A H3N2 and Influenza B

- Approved for adults and children by Health Science Authority

Relenza®

(Zanamivir)

NA inhibitor Inhaled - 100% of 2009 H1N1 viruses, H3N2

and influenza B were susceptible to Zanamivir

- Only approved for adult treatment by Health Science Authority

3Influenza Antiviral Drug Resistance (2015, January 8) Centers for Disease Control and

Prevention.Retrieved on 07 June 2015, from

http://www.cdc.gov/flu/about/qa/antiviralresistance.htm

4Frequently Asked Questions (2014, April 22) Singapore Health Sciences Authority

(HSA).Retrieved on 07 June 2015 from

http://www.hsa.gov.sg/pub/faq/faq/faqcategory/antiviral-drugs-what-are-they-.aspx

Besides the occurrence of random mutations in influenza genome, Influenza virus possesses eight segregated RNA segments in its genome that enable genetic reassortment to take place This happens when two or more influenza subtypes co-infect a single host (e.g swine) that expresses both α 2, 3 - linked and α 2, 6 - linked sialic acid receptors, and their RNA segments mix to form new combinations for novel viruses Reassortment introduces drastic modification to the virus genome, and hence the phenotype, causing a process called antigenic shift By replacing the surface NA and HA antigens of influenza A viruses from an animal origin to those of a human origin, novel strains that can infect humans are created, and may lead to large scale pandemic outbreaks (Treanor 2004)

It is difficult to accurately predict which novel subtypes will emerge Hence, while annual vaccination of updated influenza vaccines is the best strategy of protecting oneself from influenza infection, these influenza vaccines may not confer enough protection against novel strains (Plans-Rubio 2012) Furthermore, it takes at least five to six months lead time to generate a customized vaccine after identifying the pandemic strain, thereby imposing a huge problem for health care systems when outbreaks occur (Haaheim, Madhun, and Cox 2009) Taken together, genetic variations that give rise to novel influenza strains are the greatest hurdle for effective vaccines and antiviral drugs With the continual circulation of influenza in the animal and human populations, this virus is expected to be a public health concern for a long time to come

1.1.3 Global burden imposed by influenza outbreaks

symptoms appear, to a week or more after they become sick (CDC5) If caution is not exercised, it is also possible to spread flu virus through contaminated objects Influenza usually manifests in the upper airway with mild cold-like symptoms that can subside with sufficient rest and fluids However, it can sometimes cause severe to life-threatening pneumonia and other complications that require intensive medical attention

Severe influenza infection is especially prevalent among children and elderlies, pregnant women and people with pre-existing medical conditions (Jain et al 2009; Ljungman et al 1993) There is also a particularly high risk of influenza - infected patients succumbing to bacteria co-infection (Chertow and Memoli 2013) Based

on an estimation made by the WHO, three to five millions of people worldwide suffer from severe debilitating influenza infections every year, while approximately 250-500 thousand people succumb and die from seasonal influenza and influenza-associated complications6 A more worrisome scenario arises with the emergence

of zoonotic influenza strains such as the H5N1 and H7N9 bird flu which have exceedingly high mortality rates of approximately 60% and 18.7% (Mei et al 2013)

It has been estimated that more than 60 million (up to 81 million) people may die today, should an influenza pandemic of the same scale as the 1918 Spanish flu outbreak occur (Murray et al 2006) Together with escalating costs in treatment fees, and monetary loss from reduced work productivity and absenteeism, influenza outbreaks continue to bring about drastic socio-economic impacts on individuals and society (Szucs 1999)

5 How Flu Spreads (2013, September 12) Centers for Disease Control and

Prevention.Retrieved July 22, 2014, from

http://www.cdc.gov/flu/about/disease/spread.htm

6Influenza (Seasonal) Fact Sheet (2014, March) World Health Organization.Retrieved

July 22, 2014, from http://www.who.int/mediacentre/factsheets/fs211/en/

1.2 Influenza infection induces acute lung injury

The onset of severe illnesses due to influenza infection is often due to infection spreading from the upper respiratory tract to the lower respiratory tract A proportion of patients with lower respiratory tract infection develop progressive pneumonia with gross lung damage and impaired oxygen intake This is medically known as acute lung injury (ALI) and acute respiratory distress syndrome (ARDS), characterized by acute infiltration of immune cells into lungs, and a reduction of oxygen level in the blood stream (hypoxemia) (Johnson and Matthay 2010) Influenza patients with ALI / ARDS are typically presented with diffused alveolar damage, bronchiolitis and organizing pneumonia In many cases, hemorrhage, edema, hyaline membrane formation and fibrosis can also be observed (Fujita et

al 2014; Nakajima et al 2012) As a result, lung functions are greatly compromised

Alveolar epithelial and endothelial cell death during the course of infection is a leading cause of lung failure The presence of infiltrated cells and fluid (edema) in the lung further diminishes air volume in the lungs, thereby enhancing hypoxemia Depletion of blood oxygen can cause multiple organs to fail since the entire body relies on oxygen to survive For instance, liver, which is positioned away from the site of infection (lungs), shows signs of hepatocellular injury during severe influenza insult, potentially due to hypoxia, and exposure to pro-inflammatory mediators in the systemic circulation (Papic et al 2012; Han et al 2014) Clinical statistics have shown that influenza patients with ALI / ARDS, such as those infected with pandemic H1N1 (2009), have a high reported mortality rate of 15%

in adults and children (Quispe-Laime et al 2010; Ali et al 2013), despite availability

that can be exploited to overcome the limitations of current influenza therapy, in order to reduce fatalities during severe influenza infections

1.2.1 Influenza-induced cytopathy and lung injury

Influenza virus is a causative agent for significant cell death In particular, programmed cell death (apoptosis) is considered to be a hallmark of influenza infection and is thought to contribute to lung injury during influenza pneumonia (Hashimoto et al 2007; Mori et al 1995; Short et al 2014) In cells, multiple components of influenza such as NP, M2, PB1-F2 and NS1 have been demonstrated to cause predominantly mitochondria-dependent apoptosis as a viral strategy to increase virus replication and release (Tripathi et al 2013; Zamarin

et al 2005; Schultz-Cherry et al 2001) Consistent with this finding, inhibition of pro-apoptotic factors such as BAD and Bax dramatically reduces viral titer in cell cultures (Tran et al 2013; McLean et al 2009), highlighting that the mechanisms involved in host cell apoptosis is exploited by influenza virus to propagate Viral-induced cell death can be said to be a direct cause of pathogenesis during influenza infection Hence, anti-viral agents that can block viral entry or replication, such as oseltamivir, remain a key strategy in containing the spread and detrimental effects of influenza

1.2.2 Inflammatory responses and lung injury

It is now clear that attributing influenza-induced lung injury solely to viral activity does not adequately describe the disease process In response to infection, influenza infection elicits a classical innate to adaptive inflammatory response (Buchweitz, Harkema, and Kaminski 2007; Pommerenke et al 2012) to increase host resistance towards influenza (Mordstein et al 2010), to suppress viral

replication (Tumpey et al 2005) and to remove viral-infected cells via host-cell apoptosis (Topham, Tripp, and Doherty 1997; Brincks, Katewa, et al 2008; Brincks, Kucaba, et al 2008; Ishikawa et al 2005) While the immune systems have evolved to protect hosts from pathogens, paradoxically, there is evidence that host responses can sometimes contribute to increased pulmonary injury and even quicken death in animal models of severe influenza infection

Existing data have shown that excessive inflammatory responses such as cytokine storm, high neutrophil influx, uncontrolled T cell cytotoxicity and unmodulated inflammatory signaling have been implicated with more severe disease outcomes (Brandes et al 2013; de Jong et al 2006; Sakthivel et al 2014) Although there is

a conflicting school of thought which proposed that heightened inflammation is merely a consequence of poorly contained viral load, and that lung injury is mainly attributable to high virus titers (Boon et al 2011), there are numerous animal data which support the hypothesis that excessive inflammation drives pathogenesis during infection Specifically, empirical studies demonstrated that attenuation of pro-inflammatory cells and molecules improves lung pathology and survival following influenza infection (Berri et al 2013; Snelgrove et al 2008; Walsh et al 2011; Lauder et al 2011) Taken together, these data underscore the detrimental effects of excessive inflammatory responses on tissue integrity during infection

1.2.3 Reactive oxygen and nitrogen species during influenza infection

Reactive oxygen species (ROS) and reactive nitrogen species (RNS) are an integral component of inflammation Activated neutrophils and macrophages

myeloperoxidase (abundant in neutrophils) and inducible nitric oxide synthase (iNOS) Respiratory burst results in the release of several ROS / RNS including superoxide (O2−•), hydrogen peroxide (H2O2) and nitric oxide (NO) O2−• can be further dismutated into H2O2, which can then be converted to highly reactive hypochlorous acid (HOCl) by myeloperoxidase (Robinson 2008; Lonkar and Dedon 2011; Buffinton et al 1992) In combination, these ROS / RNS exert strong microbicidal properties, and some of them (e.g NO and O2−•) can also serve as second messengers during signal transduction of normal physiological processes (Forman and Torres 2001) It was previously reported that leukocytes (Gr1+ and CD11b+ myeloid cells such as neutrophils and monocytes) extracted from influenza - infected mice lungs contain higher levels of ROS and nitrotyrosine (nitrosylated tyrosine) as compared to those taken from uninfected mice, suggesting an increase in ROS / RNS production by the myeloid cells (Lee et al 2013)

Under normal physiological condition, ROS / RNS can be removed rapidly by intracellular antioxidants and radical scavengers to dissipate their damaging effects However, evidence has shown that influenza infection causes downregulation of pulmonary antioxidants such as glutathione, superoxide dismutase (SOD), catalase and vitamin E, which can drive further redox imbalance (Hennet, Peterhans, and Stocker 1992; Kumar et al 2005; He et al 2013) In addition, other components of the inflammatory process such as tumor necrosis factor-α (TNF-α) and granzymes / perforin are also involved in augmenting intracellular oxidative stress in cells (Wheelhouse et al 2003; Suematsu et al 2003; Martinvalet et al 2008), and may contribute to elevated oxidative stress in lung tissue Indeed, direct measurement by electron spin resonance spectroscopy

revealed that higher levels of ROS / RNS and oxidized biomolecules (e.g malondialdehyde) are present in severely infected mouse lungs (H5N1 and H2N2 models) as opposed to uninfected mouse lungs (He et al 2013; Akaike et al 1996)

Although ROS / RNS are crucial molecules involved in regulatory cell signaling and anti-microbial activities, an excess of ROS / RNS can cause tissue injury (Chabot

et al 1998) In animals infected with influenza, it was found that the administration

of enzymatic (e.g SOD) or small molecule antioxidants suppresses lung injury (Oda et al 1989; He et al 2013; Snelgrove et al 2006; Akaike et al 1996), suggesting that excessive ROS / RNS leads to more severe outcomes Likewise, deficiency or inhibition of xanthine oxidase (XO, a O2−• generating enzyme), Nox2 (catalytic subunit of NADPH oxidase found on phagocytes), or iNOS (induced on activated phagocytes), have been shown to dramatically reduce immune cell infiltration, the levels of pro-inflammatory mediators, cell death and mortality rates (Vlahos et al 2011; Akaike et al 1990; Karupiah et al 1998) Surprisingly, treatment with antioxidants has little or no effect on virus titers despite dampening

inflammatory responses, and mice with iNOS or Nox2 gene knockout even have

reduced viral loads as compared to wildtype mice (Vlahos et al 2011; Karupiah et

al 1998; Snelgrove et al 2006) Interestingly, while deficiency in Nox2 confers benefits to hosts during influenza infection, some evidence suggests that deficiency in the Nox1 isoform of NADPH oxidase leads to a slight elevation of lung inflammation and increased weight loss The authors found small increases in neutrophil infiltration during the acute phase of inflammation and heightened levels

of pro-inflammatory cytokines in Nox1 knockout mice infected with influenza virus

Nox2 (also known as gp91phox) are homologs of the catalytic subunits found in non-phagocytic NOX (NOX1) and phagocytic NOX (NOX2) respectively While Nox1 and Nox2 are extremely similar in structure, Nox1 are found to be expressed

by lung epithelial and endothelial cells, whereas Nox2 are specifically expressed

by phagocytes (Selemidis et al 2013) Although it is not clear why Nox2 deficiency protects mice against influenza infection while Nox1 deficiency increases injury, one possibility could be that small amounts of ROS generation by NOX1 in lung epithelial and endothelial cells is necessary for modulating the expression of pro-inflammatory cytokines and Nrf2 (Selemidis et al 2013), such that when NOX1 is inhibited, phagocytes are primed to produce large amounts of ROS by NOX2-mediated respiratory burst Taken together, extensive experimental evidence suggests that while some level of ROS / RNS generated by lung cells is essential for maintaining normal lung physiology, and may be involved in controlling inflammation, excessive ROS / RNS induced by the inflammatory systems can exacerbate lung injury during influenza infections

It has been thought that oxidative / nitrosative stress can cause tissue injury by modifying biomolecules, which in turn alter cell signaling involved in regulating endothelium barrier, cell death and tissue reparation (Mittal et al 2014; Sunil et al 2012) However, the mechanisms underlying ROS / RNS-induced pathogenicity during influenza infections are not yet fully understood Under normal physiological conditions, the body produces low concentrations (nanomolar) of endogenous ROS / RNS (e.g O2−• and NO) that are short lived and have limited reactivity However, when high concentration of ROS / RNS are generated during inflammation, they may not be removed quickly, and can be converted to highly reactive molecules including hydroxyl radicals (OH•), lipid peroxides (ROOH) and

peroxynitrite (ONOO-) (Mittal et al 2014) (Figure 1.3) These toxic molecules can react with almost any biomolecule including lipids, proteins and nucleic acids in the body and are capable of altering cell physiology and causing cell damage (Beckman and Koppenol 1996; Lonkar and Dedon 2011; Dedon and Tannenbaum 2004) For instance, ROS / RNS can induce post-translational modifications such

as S-nitrosylation, tyrosine nitration, carbonylation and disulfide linkage to proteins One example of such modifications is the oxidative modifications to cofilin, which then promotes mitochondria swelling and cytochrome c release that eventually leads to apoptosis (Klamt et al 2009) Similarly, irreparable oxidative damage to other essential biomolecules, such as lipids and nuclei acids, can result in cell injury (Tribble, Aw, and Jones 1987; Lonkar and Dedon 2011) Given that oxidative stress is elevated during influenza infection, a potential mechanism of lung injury

is oxidative damage to cellular molecules during the inflammatory process Understanding the implications of influenza infection on the integrity of cellular molecules may shed light on the most fundamental mechanisms that contribute to tissue injury during influenza infection

Activated neutrophils and macrophages generate ROS / RNS that further react to form toxic compounds that modify biomolecules such as proteins and DNA (Image taken from Dedon PC, Tannenbaum SR Reactive nitrogen species in the chemical biology of inflammation Arch Biochem Biophys 2004 Mar 1;423(1):12-22.)

1.3 ROS and RNS can damage genomic DNA

Genomic DNA is an important target of intracellular ROS / RNS Every component

of the DNA, including its nucleobases, deoxyribonucleosides and its phosphate backbone can be damaged by reacting with various ROS / RNS Under normal physiological conditions, energy metabolism and other biochemical reactions contribute to continuous production of ROS / RNS that leads to a low steady-state level of DNA damage Accumulation of DNA damage over time has been proposed to play roles in promoting slow biological processes such as neurodegenerative diseases and aging (De Bont and van Larebeke 2004; Maynard

ribose-et al 2009) With the onsribose-et of inflammation, excessive formation of ROS / RNS including H2O2, HOCl, singlet oxygen, ONOO- and OH• will overwhelm antioxidant systems and exacerbate DNA damage

OH• radical is considered one of the most potent ROS that leads to DNA damage

by oxidizing bases and attacking DNA backbones to generate DNA strand breaks Due to its reactivity, OH• radical does not diffuse beyond one or two molecular diameter without reacting with a neighboring molecule Hence, OH• needs to be near to DNA in order to induce any modifications to the genomic molecules(Pryor 1986) It is thought that H2O2, a membrane permeable and more diffusible oxidant, can diffuse into nucleus to be close enough to DNA molecules H2O2, commonly generated under inflammatory conditions, then give rise to DNA-damaging OH• via Fenton reaction, although H2O2 itself does not appear to directly cause DNA damage The reduction of H2O2 to OH• takes place in the presence of transition

metals such as copper Cu(II) and iron Fe (III) found as co-factors in nuclear proteins (reviewed in (Cadet et al 1999; Wiseman and Halliwell 1996; Dizdaroglu 1992)), as shown in the equation for Fenton reaction below:

Fe2+ + H2O2 + H+  Fe3+ + H2O + OH•

To date, more than 100 types of oxidative nucleobases and deoxyribose modifications have been identified after DNA treatment with H2O2 and other ROS / RNS (Dizdaroglu 1992; Croteau and Bohr 1997) Modified nucleobases are highly diversified (some examples shown in Figure 1.4), among which 8-hydroxy-deoxyguanine (8-OH-dG) is one of the most prevalent and certainly, the most studied form of oxidative base lesion (Dizdaroglu 1992) In addition, other modifications exist to add on to the spectrum of DNA damage when DNA reacts with inflammation-induced reactive species and reactive intermediates For instance, DNA’s interaction with HOCl can give rise to halogenated nucleobases (e.g 8-chloro(2’-deoxy)guanosine, 5-chloro(2’-deoxy)cytidine, and 8-chloro(2’-deoxy)adenosine) (Masuda et al 2001), while reaction with dinitrogen trioxide, a RNS, has also been shown to cause base deamination to form 2’-deoxyxanthosine, 2’-deoxyoxanosine, 2’-deoxyinosine and 2’-deoxyuridine (Dong et al 2006) In addition, intermediate products of oxidation, such as lipid peroxides (e.g 4-hyrdoxynonenal and malondialdehyde) can also induce damage in DNA to form strongly mutagenic etheno-DNA adducts (Linhart, Bartsch, and Seitz 2014) Taken together, numerous forms of DNA modification can occur upon exposure to ROS / RNS, including abasic sites, bulky adduct, oxidized bases, deaminated bases, and DNA- DNA and DNA- protein cross-links (Lonkar and Dedon 2011) In addition,