Characterisation of Bilirubin Metabolic Pathway in Hepatic Mitochondria Siti Nur Fadzilah Muhsain M.Sc

Characterisation of Bilirubin Metabolic Pathway in Hepatic Mitochondria Siti Nur Fadzilah Muhsain M.Sc. (Medical Research) 2005 Universiti Sains Malaysia Postgrad. Dip. (Toxicology) 2003 University of Surrey B.Sc.(Biomed. Sc.) 2000 Universiti Putra Malaysia

A thesis submitted for the degree of Doctor of Philosophy at The University of Queensland in 2014 School of Medicine

ABSTRACT

Bilirubin (BR), a toxic waste product of degraded haem, is a potent antioxidant at physiological concentrations. To achieve the maximum benefit of BR, its intracellular level needs to be carefully regulated. A system comprising of two enzymes, haem oxygenase-1 (HMOX1) and cytochrome P450 2A5 (CYP2A5) exists in the endoplasmic reticulum (ER), responsible for regulating BR homeostasis. This system is induced in response to oxidative stress. In this thesis, oxidative stress caused accumulation of these enzymes in mitochondria — major producers and targets of reactive oxygen species (ROS) — is demonstrated. To understand the significance of this intracellular targeting, properties of microsomal and mitochondrial BR metabolising enzymes were compared and the capacity of mitochondrial CYP2A5 to oxidise BR in response to oxidative stress is reported.

Microsomes and mitochondrial fractions were isolated from liver homogenates of DBA/2J mice, administered with sub-toxic dose of pyrazole, an oxidant stressor. The purity of extracted organelles was determined by analysing the expressions and activities of their respective marker enzymes. HMOX1 and CYP2A5 were significantly increased in microsomes and even more so in mitochondria in response to pyrazole-induced oxidative stress. By contrast, the treatment did not increase either microsomes or mitochondrial Uridine-diphosphate-glucuronosyltransferase 1A1 (UGT1A1), the sole enzyme that catalyses BR elimination through glucuronidation.

In response to pyrazole-induced oxidative stress, BR oxidation was enhanced not only in microsomes but also in mitochondria. The reaction in both control and pyrazole treated microsomes were blocked by up to 70% with anti-CYP2A5 antibody. By contrast, no CYP2A5 antibody inhibition on BR oxidation was observed in control mitochondria but was 55% inhibited in treated mitochondria. Antibody against adrenodoxin reductase (AdxR) (a component of mitochondrial monooxygenase system) had no effect on the microsomal BR oxidation, but the reaction was inhibited by up to 50% in mitochondria

i after pyrazole treatment. As in microsomes, the antibody had no effect in control mitochondria.

Inhibition study with ascorbic acid — a scavenger of ROS — was undertaken to determine the contribution of ROS to BR oxidation. In control microsomes, a slight inhibition of 5% in BR degradation was noted which was augmented to 22% after pyrazole treatment. The inhibition was 100% in control mitochondria but limited to 50% after pyrazole treatment.

Mass spectrometry analysis reveals that in microsomes, biliverdin (BV), the main metabolite of CYP2A5-dependant BR oxidation was formed. Meanwhile, mitochondria produced predominantly dipyrroles, the products of BR scavenging ROS. Pyrazole treatment caused mitochondria to yield relatively similar amount of BV and dipyrroles. Enzyme kinetic analysis showed that mitochondrial and microsomal CYP2A5 have an equally strong affinity towards BR. Additionally, no initiation of apoptosis, as indicated by the absence of cytosolic cytochrome C was observed in treated liver.

Collectively, these observations suggest that: i) both HMOX1 and CYP2A5 are recruited into mitochondria to regulate BR metabolism during oxidative stress; (ii) CYP2A5- dependent BR oxidation in mitochondria is driven by the mitochondria-specific electron transfer chain; (iii) constitutive mitochondrial BR oxidation is driven un-enzymatically yet equally driven by ROS and CYP2A5 in response to oxidative stress; (iv) BR affinity to mitochondrial CYP2A5 is in the same range to that of microsomal CYP2A5; (v) mitochondrial BR is not eliminated by glucuronidation but recycled by CYP2A5 oxidation to BV; and (vi) mitochondrial targeting of enzymes crucial for BR homeostasis is not associated with initiation of mitochondria specific apoptosis. It is thus submitted that targeting of key BR regulatory enzymes to mitochondria is an adaptive response to oxidative stress, aimed at mitochondrial protection against oxidative damage.

DECLARATION BY AUTHOR

This thesis is composed of my original work, and contains no material previously published or written by another person except where due reference has been made in the text. I have clearly stated the contribution by others to jointly-authored works that I have included in my thesis.

I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance, survey design, data analysis, significant technical procedures, professional editorial advice, and any other original research work used or reported in my thesis. The content of my thesis is the result of work I have carried out since the commencement of my research higher degree candidature and does not include a substantial part of work that has been submitted to qualify for the award of any other degree or diploma in any university or other tertiary institution. I have clearly stated which parts of my thesis, if any, have been submitted to qualify for another award.

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and, subject to the General Award Rules of The University of Queensland, immediately made available for research and study in accordance with the Copyright Act 1968.

I acknowledge that copyright of all material contained in my thesis resides with the copyright holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright holder to reproduce material in this thesis.

iii

PUBLICATIONS DURING CANDIDATURE

Peer-Reviewed Journal Article in Submission

Muhsain, S.N.F., Lang, M.A., and Abu-Bakar, A. (2014). Mitochondrial targeting of bilirubin regulatory enzymes: An adaptive response to oxidative stress. Toxicology and Applied Pharmacology (under review), Appendix 8.

Peer-Reviewed Conference Proceedings

Muhsain, S.N.F., Lang, M.A., and Abu-Bakar, A. (2013). Bilirubin regulation in hepatic mitochondria. World Academy of Science, Engineering and Technology 74, 545.

Conference Presentations

Muhsain, S.N.F., Lang, M.A., Abu-Bakar, A. Bilirubin regulation in hepatic mitochondria. International Conference on Pharmacology and Pharmaceutical Sciences, 14th- 15th Feb 2013. Kuala Lumpur, Malaysia.

Muhsain, S.N.F., Lang, M.A., Abu-Bakar, A. Cytochrome P450 2A5 a mitochondrial bilirubin oxidase? Annual Scientific Meeting of the Australasian Society of Clinical and Experimental Pharmacologists and Toxicologists (ASCEPT), 1st- 4th Dec 2013. Melbourne, Australia.

PUBLICATIONS INCLUDED IN THIS THESIS

Chapter 5

Muhsain, S.N.F., Lang, M.A., and Abu-Bakar, A. (2013) Mitochondrial targeting of bilirubin regulatory enzymes: An adaptive response to oxidative stressToxicology and Applied Pharmacology (under review as of 22/09/2014)

Contributor Statement of contribution

Designed experiments (20%) Muhsain, S.N.F., (Candidate) Conducted experiments (100%) Data analysis (60%) Wrote the paper (50%) Statistical analysis (100%)

Project conceptualisation (20%) Lang, M.A. Designed experiments (40%) Wrote and edited paper (25%)

Project conceptualisation (80%) Abu-Bakar, A. Designed experiments (40%) Data analysis and interpretation (40%) Wrote and edited paper (25%)

CONTRIBUTIONS BY OTHERS TO THE THESIS

Significant contributions to this project were made by the following:

Dr A’edah Abu Bakar and Professor Matti Lang – in the conceptualisation and design of the project based on their research works, as well as critically revising the thesis so as to contribute to the interpretation.

Dr Wasantha Wickramasinghe – in the assistance of non-routine technical work.

Mr Geoff Eaglesham – in the analysis of HPLC/MS-MS work.

STATEMENT OF PARTS OF THE THESIS SUBMITTED TO QUALIFY FOR THE AWARD OF ANOTHER DEGREE

None

ACKNOWLEDGEMENTS

First and foremost, thanks to God Almighty who has given me the strength and will to complete this thesis. I would also wish to express my gratitude to the many individuals who have supported me throughout this study. I am greatly indebted to my principal supervisor, Dr A’edah Abu Bakar, for her guidance, support and attention during the course of my PhD candidature. I truly appreciate the invaluable time, effort and patience that she has spent in supervising me. I would also like to thank my associate supervisor, Professor Matti Lang, who provided great ideas, assistance and encouragement.

A special thanks to Universiti Teknologi Mara (UiTM), Malaysia for their financial assistance, without which I would not have been able to pursue my study. My gratitude also goes to National Research Centre for Environmental Toxicology (Entox) as well as Faculty of Health Sciences, University of Queensland, Australia for granting a scholarship during the final semester of my study.

I am also indebted to Dr Wasantha Wickramasinghe (Entox, Coopers Plains Australia) for his advice and assistance in the laboratory. Thanks is also due to Mr Geoff Eaglesham (Entox, Coopers Plains Australia), without whose help in HPLC/MS-MS and Q-tof analysis, this thesis would not be complete. My sincere appreciation is extended to my many colleagues and friends, especially Nur Syafawati ,Amalina and Jarina for their endless support and belief in my potential.

Last but not least, my eternal gratitude goes to my parents, Salmah Md Nor and Muhsain Mohamed, as well as my siblings, Mohd Fahmi and Siti Nur Fathini, for their never ending encouragement to strive for educational and personal excellence.

vii

KEYWORDS mitochondria, oxidative stress, haem oxygenase-1, biliverdin reductase, cytochrome P450 2A5 (CYP2A5), bilirubin, bilirubin oxidation, liver

AUSTRALIAN AND NEW ZEALAND STANDARD RESEARCH CLASSIFICATIONS (ANZSRC) 060107, Enzymes, 70% 060199, Biochemistry and Cell Biology not elsewhere classified, 20% 060104, Cell Metabolism, 10%

FIELDS OF RESEARCH (FOR) CLASSIFICATION 0601, Biochemistry and Cell Biology, 100%

viii

TABLE OF CONTENTS

Page Abstract i Declaration by author iii Publications during candidature iv Publications included in this thesis v Contributions by others to the thesis vi Statement of parts of the thesis submitted to qualify for the vi award of another degree Acknowledgements vii Keywords viii Australian and New Zealand Standard Research Classifications viii (ANZSRC) Fields of Research (FOR) Classification viii Table of Contents ix List of Figures and Tables xii List of Abbreviations xiv

1. INTRODUCTION 1

2. LITERATURE REVIEW 3 2.1 Reactive Oxygen Species (ROS) and Oxidative Stress 3 2.1.1 Sources of cellular ROS 4 i. Mitochondrial sources of ROS 7 ii. Endoplasmic reticulum sources of ROS 10 2.2 Adaptive Response to Oxidative Stress 12 2.2.1 Cytoprotection of haem oxygenase against oxidative 13 damage 2.2.2 Regulation of bilirubin levels 15

Page 2.2.3 Concerted induction of bilirubin production and oxidation 18

3. RESEARCH PROBLEMS 22 3.1 Research Questions 23 3.2 Working Hypotheses 24 3.3 Research Strategy 25

4. METHODS AND MATERIALS 27 4.1 Chemical Reagents and Antibodies 27 4.2 Laboratory Animals and Treatment 27 4.3 Biochemical and Molecular Biology Procedures 28 4.3.1 Preparation of subcellular fractions 28 4.3.2 Enzyme activity assays 29 4.3.3 Western blot 31 4.3.4 Bilirubin disappearance activity 31 4.4 Screening of Bilirubin Oxidative Metabolites (BOMs) 32 4.4.1 HPLC/MS-MS analysis of BOMs 32 4.5 Statistical Analysis 33

5. RESULTS AND DISCUSSION 34 5.1 Purity of Microsomal and Mitoplasts Fractions 34 5.2 Bilirubin Metabolising Enzymes in Mitochondria 36 5.2.1 Effects of pyrazole on HMOX1 and CYP2A5 protein 36 expression 5.2.2 Effects of pyrazole on BVR and UGT1A1 expression 39 5.3 Bilirubin Degradation in Mitochondria 42 5.3.1 Contribution of ROS to bilirubin degrading activity 43 5.3.2 Contribution of CYP2A5 to bilirubin degrading activity 45 5.3.3 Microsomal and mitochondrial CYP2A5 enzymes have 53 similar affinity to bilirubin

Page 5.3.4 Effect of pyrazole on the release of cytochrome C from 56 mitochondria 5.4 Conclusions 59

6. GENERAL CONCLUSIONS AND FUTURE DIRECTIONS 63 6.1 General Conclusions 63 6.2 Future Directions 64

7. BIBLIOGRAPHY 66

8. APPENDICES 83 Appendix 1: List of chemicals, reagents and instruments 83 Appendix 2: Preparation of liver microsomal, cytosolic and 86 mitoplasts fractions Appendix 3: Coumarin hydroxylase activity 87 Appendix 4: Western immunoblotting 88 Appendix 5: HPLC/MS-MS conditions used to identify bilirubin 89 and its oxidative metabolites Appendix 6: HPLC/MS-MS chromatograms of bilirubin, biliverdin 90 and various dipyrroles Appendix 7: Effect of ascorbic acid on Fenton oxidation of 91 bilirubin (BR) Appendix 8: Manuscript: Mitochondrial targeting of bilirubin regulatory enzymes: An adaptive response to 95 oxidative stress

LIST OF FIGURES AND TABLES

Page Figure 2.1 Reactive oxygen species (ROS) homeostasis in cell 6

Figure 2.2 Generation of ATP and ROS by mitochondrial 9 electron transport chain (ETC)

Figure 2.3 Bilirubin metabolism pathways in the liver 14

Figure 2.4 Proposed regulation of key enzymes in bilirubin metabolism in endoplasmic reticulum and 21 mitochondria during oxidative stress

Figure 5.1 Evaluation of purity of microsomal and mitoplasts 35 isolation using subcellular enzymes markers

Figure 5.2 Effects of pyrazole on treatment on HMOX1 and 36 CYP2A5 protein expression in liver of DBA/2J mice

Table 5.1 Haem oxygenase (HMOX), biliverdin reductase (BVR) and coumarin 7-hydroxylase (COH) activities 38 of liver cytosol, microsomes and mitoplasts from DBA/2J mice

Figure 5.3 Effect of pyrazole treatment on BVR protein 40 expression in the liver of DBA/2J mice

Figure 5.4 Effect of pyrazole on UGT1A1 protein expression 41

Table 5.2 Bilirubin degrading activity in liver microsomes and 42 mitoplasts of DBA/2J mice

xii

Page Figure 5.5 Effect of 200mM ascorbic acid on bilirubin degradation in liver microsomes and mitoplasts of 45 DBA/2J mice

Figure 5.6 Effect of anti-CYP2A5 antibody on bilirubin (BR) degradation activity of control and pyrazole-treated 47 microsomes and mitoplasts fractions

Figure 5.7 Effect of anti-adrenodoxin reductase (AdxR) antibody on bilirubin (BR) degradation activity in 49 microsomes and mitoplasts fractions

Figure 5.8 Summary of the oxidative products of bilirubin 51 formed in the chemical and enzymic systems

Figure 5.9 Effect of pyrazole on microsomal and mitoplasts 52 BOMs

Figure 5.10 Effect of bilirubin on coumarin 7-hydroxylase 55 activity in pyrazole treated mitoplasts

Figure 5.11 Effect of pyrazole on cytochrome C protein 57 expression

Figure 5.12 Proposed protective mechanism through bilirubin regulation during oxidative stress induced by 62 pyrazole intoxication.

xiii

LIST OF ABBREVIATIONS

AdxR Adrenodoxin reductase ALA Aminolevulinic acid AMPK Adenine monophosphate-activated protein kinase ANOVA Analysis of variance Asn Asparagine ATP Adenosine triphosphate BOMs Bilirubin oxidative metabolites BR Bilirubin BSA Bovine serum albumin BV Biliverdin BVR Biliverdin reductase Ca2+ Calcium ion CO Carbon monoxide

CO2 Carbon dioxide COH Coumarin 7-hydroxylase COX Cytochrome C oxidase CRC Colorectal cancer Cu Copper CYP Cytochrome P450 enzymes CYP1A Cytochrome P450 1A subfamily CYP2A5 Mouse cytochrome P450 2A5 CYP2A6 Human cytochrome P450 2A6 CYP2B4 Mouse cytochrome P450 2B4 CYP2E1 Human or rat cytochrome P450 2E1 CYP3A Human or rat cytochrome P450 3A subfamily DMSO Dimethyl sulfoxide DNA Deoxyribonucleic acid DTT Dithiothreitol e- Electron eV Electronvolt EDTA Ethylenediamine tetra acetic acid ER Endoplasmic reticulum ETC Electron transport chain FAD Flavin adenine dinucleotide

FADH2 Reduced flavin adenine dinucleotide xiv

Fe2+ Ferrous ion Fe3+ Ferric ion G6P Glucose 6-phosphate G6PDH Glucose-6-phosphate dehydrogenase GPx Glutathione peroxidase GRP Glucose-regulated protein GRx Glutathione reductase GS Gilbert’s Syndrome GSH Glutathione GSSG Oxidised glutathione GST2A Glutathione S-transferase 2A

H2O2 Hydrogen peroxide HeLa Human cervical cancer cell line Hepa 1c1c7 Murine hepatoma cells HepG2 Human liver carcinoma cell line HEK 293 Human embryonic kidney cells HMOX1 Haem oxygenase-1 HMOX2 Haem oxygenase-2 HPLC High performance liquid chromatography HPLC-MS/MS High performance liquid chromatography/mass spectrometry

IC50 Concentration of inhibitor at which 50% of enzyme activity is inhibited IMM Inner mitochondrial membrane

Ki The amount of inhibitor required for 50% of saturation of the binding sites

Km Michaelis-Menten constant m/z Mass-to-charge ratio MAM Mitochondria-Associated Membranes MgCl2 Magnesium chloride Mn Manganese MPT Mitochondrial permeability transition mRNA Messenger RNA MRP2 Multidrug resistance-associated protein 2 mtDNA Mitochondrial Deoxyribonucleic acid NADH Reduced nicotinamide adenine dinucleotide NADP+ Nicotinamide adenine dinucleotide phosphate NADPH Reduced nicotinamide adenine dinucleotide phosphate xv

NaOH Sodium hydroxide ND Not detected NF-E2 Nuclear factor-erythroid 2 NH2 Amine NIH National Institutes of Health NPR NADPH-cytochrome P450 reductase NQO1 NADPH quinone oxidoreductase Nrf2 Nuclear factor erythroid 2-related factor 2

O2 Molecular oxygen .- O2 Superoxide anions OH.- Hydroxyl radical PCB Polychlorinated biphenyl PGC-1α Peroxisome proliferator activated receptor-gamma coactivator 1alpha PMSF Phenylmethylsulfonylfluoride PP-IX Protoporphyrin-IX Prx Peroxiredoxin PVDF Polyvinylidene fluoride RNA Ribonucleic acid ROS Reactive oxygen species rpm Revolutions per minute SD Standard deviation SDS-PAGE Sodium dodecylsulfate-polyacrylamide gel electrophoresis siRNA Small interfering RNA SOD Superoxide dismutase TBS Tris-buffered saline TCA Tetraacetic acid TCDD 2,3,7,8-tetrachlorodibenzo-p-dioxin Thr Threonine 305 Trx2 Thioredoxin UGT1A1 Uridine-diphosphate-glucuronosyltransferase 1A1 UPR Unfolded protein response UQ Ubiquinone UQH2 Reduced ubiquinones Vmax Maximum velocity VDAC Voltage-dependent anion channels XME Xenobiotic metabolising enzymes Zn Zinc

xvi

1. INTRODUCTION

Mitochondria are essential organelles in regulating the life and death of cells. They produce energy through oxidative phosphorylation of the electron transfer chain (ETC), as well as regulate apoptotic proteins, which are essential for cells survival in critical conditions. In producing energy, mitochondria release reactive oxygen species (ROS), which if not controlled can lead to oxidative damage of the ETC and mitochondrial DNA (mtDNA). This can promote a vicious loop, where inefficient ETC will generate more ROS. Some of these ROS may permeate out of mitochondria and spread to the whole cell (Zorov et al., 2000; Zorov et al., 2006). As such, mitochondrial oxidative stress has been implicated in the progression of chronic diseases, such as cancers, neurodegenerative diseases, atherosclerotic diseases, and cardiovascular diseases, as well as in aging processes (Kidd, 2005; Sullivan and Brown, 2005; Fukui and Moraes, 2008; Seo et al., 2010). Thus, to counteract the deleterious effects of ROS and to maintain cellular homeostasis, mitochondria contain a number of antioxidant systems, such as glutathione, superoxide dismutase (SOD), glutathione peroxidase (GPx) (Tilak and Devasagayam, 2006; Mari et al., 2009) and potentially the bile pigment, bilirubin (BR), which is the focus of this thesis.

BR is a product of haem degradation, catalysed initially by haem oxygenase (HMOX) and subsequently by biliverdin reductase (BVR). Though toxic at high concentrations, BR has been regarded as a potent cellular antioxidant at physiological concentrations, (Stocker et al., 1987a; Stocker et al., 1987b; Tomaro and Battle, 2002) and therefore, its levels require careful maintenance. In scavenging ROS, BR is broken down to more polar metabolites ranging from tripyrroles to dipyrroles that are readily excreted (Yamaguchi et al., 1995; Kunikata et al., 2000; Yamaguchi et al., 2002; Arthur et al., 2012; Abu-Bakar et al., 2013). Recently, Abu-Bakar et al. (2011) has demonstrated a major role for microsomal CYP2A5 as BR oxidase, metabolising BR to predominantly biliverdin (BV). During oxidative stress, microsomal BR oxidation can be induced following increased hepatic BR levels due to HMOX1 accumulation (Abu- Bakar et al., 2005; Abu-Bakar et al., 2011). This induction is achieved as BR can

1 upregulate its own oxidation at sub-toxic concentrations by delaying the degradation of labile CYP2A5 enzyme that would stabilise the protein (Abu-Bakar et al., 2005; Abu- Bakar et al., 2011; Abu-Bakar et al., 2012; Abu-Bakar et al., 2013).

Accordingly, these findings are consistent with a proposition that HMOX1 and CYP2A5 work together for a rapid increase of BR to scavenge ROS and for a rapid removal of any excess BR amount by oxidising it to nontoxic BV, once the stress has subsided. The high affinity of BR to CYP2A5 with Km of 1-2 μM (Abu-Bakar et al., 2005; Abu-Bakar et al., 2012) seems to be ideal for this purpose. Consequently, these adaptive responses aim at maintaining intracellular BR homeostasis, keeping BR at sufficient cytoprotective concentrations as well as at safe tissue levels (Abu-Bakar et al., 2005; Abu-Bakar et al., 2012; Abu-Bakar et al., 2013).

Interestingly, CYP2A5 enzyme is also found in mitochondria, and can be upregulated in response to 2, 3, 7, 8-tetrachlorodibenzo-p-dioxin (TCDD) and pyrazole- induced stress (Honkakoski et al., 1988; Genter et al., 2006). However, the function of mitochondrial CYP2A5 under oxidative stress needs to be clarified. It is possible that mitochondrial CYP2A5 can function as BR oxidase to circumvent its toxicity in this organelle. In this thesis, key enzymes that involved in mitochondrial BR metabolism were identified and compared to those in microsomes. Furthermore, the importance of CYP2A5-BR oxidation, plausibly for mitochondrial protection under oxidative stress condition, was also investigated. Pyrazole, a strong oxidant in endoplasmic reticulum (ER), as well as in mitochondria, was used in the study to induce the oxidative stress in both organelles (Lämsä et al., 2010; Bae et al., 2012).

2. LITERATURE REVIEW

2.1 REACTIVE OXYGEN SPECIES (ROS) AND OXIDATIVE STRESS

Radical oxygen species (ROS) are small, highly reactive molecules, with one or more unpaired electrons, derived from molecular oxygen (O2). They include superoxide .- .- anion (O2 ), hydroxyl radical (OH ) and hydrogen peroxide (H2O2). They are by-products of normal cellular functions of various organelles, such as mitochondria, peroxisomes and ER (microsomes) (Jezek and Hlavata, 2005). Although they are essential for some physiological functions, excessive amounts can cause cellular damage as they readily attack macromolecules, such as lipids, proteins, carbohydrates, and DNA. Therefore, cellular integrity is achieved through normal “ROS homeostasis” that is essentially maintained by a battery of cellular antioxidant systems (Jezek and Hlavata, 2005; Alfadda and Sallam, 2012).

Exposure to toxic chemicals, radiation and pathogenic microorganisms often lead to formation of ROS which can react with proteins, lipids and nucleic acids leading to toxic / mutagenic effects (Nebert and Russell, 2002). Sustained detoxification of the increased ROS by systems that maintain ROS homeostasis may lead to insufficient antioxidant capacity to protect cells against ROS generated by normal physiological or pathological processes. Thus, cells enter a state of oxidative stress. If ROS homeostasis and cellular antioxidant capacity are not restored, cells will eventually undergo necrotic cell death, which in turn propagates the pathogenesis of chronic diseases, such as cancers, artherosclerotic diseases and cardiovascular diseases (Diplock, 1994; Willcox et al., 2004).

2.1.1 SOURCES OF CELLULAR ROS

There are various sources of ROS in mammalian cells. One of the many sources .- of O2 — the primary ROS molecule — is the catalytic activity of oxidoreductases, a group of enzymes that catalyse transfer of electrons from the donor (reductant) to the acceptor (oxidant).

Generally, superoxide is a by-product of a specific oxidoreductase catalytic activity, except for reduced nicotinamide adenine dinucleotide phosphate (NADPH) oxidase whose primary function is to catalyse the transfer of electrons from NADPH to molecular oxygen resulting in the generation of superoxide (Griendling et al., 2000). Other oxidoreductases that are potential sources of superoxide in mammalian cells include:

 cyclooxygenase (Kukreja et al., 1986), an enzyme involved in biosynthesis of prostaglandins (Ricciotti and FitzGerald, 2011);

 nitric-oxide synthase (Pou et al., 1992), the catalyst for production of nitric oxide, an important cellular signalling molecule vital for many biological processes, such as vasodilation and immune response (Gkaliagkousi and Ferro, 2011);

 cytochrome P450 (CYP) enzymes (Zangar et al., 2004), catalysts for phase I metabolism of xenobiotics / endobiotics; and

 mitochondrial reduced nicotinamide adenine dinucleotide (NADH):ubiquinone oxidoreductase, a component of ETC (Lambert and Brand, 2004).

.- Invariably, clearance of O2 leads to formation of other ROS. For example, .- .- conversion of O2 to water involves a two-step reaction where O2 is first converted to

H2O2 by SOD. In turn, H2O2 is metabolised by catalase and GPx to water (Figure 2.1) (Dawes, 2006). Alternatively, in the presence of reduced transition metals, such as 2+ .- ferrous ion (Fe ), H2O2 can give rise to a highly reactive OH through catalytic ion .- oxidation in Fenton reaction. This reaction is aggravated by O2 in Haber-Weiss reaction by reducing ferric (Fe3+) to Fe2+ (Droge, 2002; Valko et al., 2007).

Plasma membrane

Endoplasmic

reticulum Mitochondria

Cytosol Cytochrome P450 enzymes

O2 - e

GRx

Haber-Weiss reaction .- O2 O2

GSH GSSG SOD Fe3+ Fe2+

.- H2O H2O2 OH + H2O GPx Fenton reaction

Catalase PRx / Trx2

LIPID / PROTEIN / DNA DAMAGES

O2 + H2O

Figure 2.1 Reactive oxygen species (ROS) homeostasis in cell. During mitochondrial and endoplasmic .- .- reticulum (ER) metabolisms, leakages of electrons generate superoxide (O2 ). O2 can be dismutated to

H2O2 by superoxide dismutase (SOD). H2O2 can be detoxified by catalase and glutathione peroxidase (GPx). Oxidative perturbation of glutathione (GSH): glutathione disulphide (GSSG) ratio indicates oxidative stress in cell. Other critical ROS scavengers include peroxiredoxin (PRx) and thioredoxin (Trx2).

i. Mitochondrial sources of ROS

Mitochondria can generate ROS during oxidative phosphorylation that involves a series of protein complexes in the inner mitochondrial membrane (IMM), called electron transport chain (ETC) (Figure 2.2). These processes are essential to generate metabolic energy from the breakdown of carbohydrates and fatty acids (Cooper and Hausman, 2009).

The continuous reduction of molecular oxygen (O2) to water (H2O) by ETC .- generates O2 as an inevitable by-product and primary ROS (Fridovich, 1995; Droge, .- 2002). The O2 accumulates when electron flow slows down and when concentration of

O2 increases (Turrens, 2003). About 1 to 4% of O2 in ETC is reduced to ROS (Chance et al., 1979; Nohl et al., 2005; Tilak and Devasagayam, 2006). They are mostly formed at Complexes I and III (Turrens et al., 1985; Turrens and Boveris, 1980; Sugioka et al., .- 1988; Chen et al., 2003) (Figure 2.2). In Complex I, O2 radicals are formed in the matrix side of the IMM (Figure 2.2). Succinate, the substrate of complex II, can also increase the production of radicals by complex I (de Vries, 1986; Liu, et al., 2002). In complex III, . - O 2 are produced on both side of IMM (matrix and intermembrane space) (Muller et al., 1 2004). Oxidation of reduced ubiquinones (UQ), (UQH2) in Q cycle is regarded as an important process in formation of radicals at this site (Turrens et al., 1985; Kowaltowski et al., 2009). Presumably the intermembrane space ROS have easier access to cytosol through voltage dependent anion-selective channel (VDAC), as compared to matrix ROS, in which they have to cross the outer membrane and IMM (Sena and Chandel, 2012).

1 Q cycle: This reaction creates semiquinone radical (UQ.-) at the Qp site of complex III, that faces intermembrane space. An electron of UQ.- at Qp is then transferred to Qn site (faces the mitochondrial matrix), to produce Qn UQ.-. .- .- The UQ is changed to a reduced form (UQH2) by a second UQ that formed at Qp site (faces intermembrane space). . - .- O 2 radicals may be formed from unstable UQ /UQ pair (Kowaltowski et al., 2009).

Production of mitochondrial superoxide is estimated to be 5 to 10-fold higher than that of cytosol and nucleus (Cadenas and Davies, 2000). In the mitochondria matrix, superoxide is further transformed into a relatively stable H2O2 by manganese (Mn)-SOD . - (Starkov and Wallace, 2006). In the intermembrane space, the conversion of O 2 to H2O2 is catalysed by copper (Cu) and zinc (Zn)-SOD (Weisiger and Fridovich, 1973; Okado- Matsumoto and Fridovich, 2001). Production of mitochondrial ROS can initiate damaging reactions to mitochondrial macromolecules, as well as, to macromolecules outside of mitochondria as ROS can diffuse through mitochondrial membrane (Tilak and Devasagayam, 2006). It is noted that mitochondrial DNA (mtDNA) is especially susceptible to oxidative damage due to close proximity to ETC in the matrix, as well as lacking the protective DNA-associated proteins such as histones. Additionally, mitochondria need a sufficient supply of iron to maintain ETC complexes. Ineffectively metabolised iron overload by Mn-SOD can produce highly reactive OH.- radical through Fenton reaction that can damage biological systems (Toyokuni, 1996; Atamna, 2004; Levi and Rovida, 2009).

H+ .- H+ Mitochondrial H+ O2 H+ + + + + H H + H + + intermembrane H H H H space H+

Cyt C I Q Inner III mIV ATP mitochondrial Q- synthase e membrane II

NADH NAD+ O2 H2O FADH2 ADP+Pi ATP + FAD H

Mitochondrial

matrix .- O2

Figure 2.2 Generation of ATP and ROS by electron transport chain (ETC) in mitochondria. Electrons (e-), donated by NADH and

FADH2, are passed through a series of protein complexes embedded in the inner membrane, via redox reactions to molecular oxygen. In the process, protons are pumped from the mitochondrial matrix to the intermembrane space, generating a proton gradient, and oxygen is reduced to form water. The gradient is used by ATP synthase to synthesize ATP. ROS are produced as a result of e- escaping mostly from complex I and III, forming superoxide radicals in the matrix and intermembrane space. 9

Mitochondria are also the site for haem biosynthesis, during which, free radicals may also be generated by oxidation and photochemical reactions of haem precursors — aminolevulinic acid (ALA) and protoporphyrin-IX (PP-IX) — which are both synthesised in the mitochondria (Ryter and Tyrrell, 2000). Under physiological condition, the breakdown of free haem catalysed by haem oxygenase-2 (HMOX2) releases Fe2+, which amplifies Fenton reaction to convert peroxide to hydroxyl radical (Gozzelino et al., 2010) (Figure 2.1).

In addition, mitochondria are critical regulators of intracellular calcium (Ca2+) levels to stimulate and control oxidative phosphorylation and induce permeability of mitochondrial inner membrane (Gunter et al., 2004). The accumulation of Ca2+ has been shown to increase production of mitochondrial ROS. The mechanism underlying this is uncertain but has been linked to mitochondrial permeability transition (MPT)2. For example, Hansson et al (2008) showed that brain-derived mitochondria and human liver mitochondria have a sustained increment of ROS production following calcium-induced MPT. It was thought that the bioenergetic consequences of MPT leading to loss of detoxification function mainly caused the increased ROS.

ii. Endoplasmic Reticulum Sources of ROS

Endoplasmic reticulum (ER) can generate ROS during metabolism of exogenous and endogenous compounds, as well as, in cellular protein synthesis. The metabolism of xenobiotic is performed by cytochrome P450 (CYP) enzymes, a family of haemoproteins localised in the ER, predominantly in the liver (Josephy, 1997; Parkinson, 2001).

2 MPT: increase permeability of mitochondrial membrane caused by opening of a large pore in the inner membrane. The membrane is normally closed and can be opened by Ca2+ overloading, oxidative stress and other factors. This result in degeneration of mitochondrial outer membrane and released of cytochrome C into cytosol (Tilak and Devasagayam, 2006).

In humans and other mammals, CYP families 1-4 are generally regarded as xenobiotic-metabolising enzymes (XMEs), responsible for the oxidative clearance of drugs and environmental chemicals. They exhibit a broad substrate selectivity that is important in protection against damages by these chemicals. Accordingly, these XMEs form part of a coordinated system for xenobiotic clearance, which also includes other enzymes such as glucuronosyltransferases and a series of nuclear receptors/transcription factors that regulate the response to chemical challenge (Josephy, 1997).

CYP enzymes commonly catalyse the oxidation of carbon and heteroatom centres on lipophilic chemicals to make them more hydrophilic by using molecular oxygen and NADPH as a reducing cofactor. However, catalytic cycle of CYP enzymes can generate .- O2 and H2O2 through a process called uncoupling (Josephy, 1997; Schlezinger et al., 1999; Parkinson, 2001) . When substrate and CYP enzymes are poorly coupled, these ROS can be produced in excess and leak subcellularly, hence rendering the cycle futile (Bondy and Naderi, 1994). According to Shertzer et al (2004), certain known CYP enzymes that are affected by this process include CYP1A, CYP2E1, CYP2B4 and CYP3A. Uncoupling is aggravated when CYP enzymes-substrate specificity is promiscuous. For example, much of the toxicities of environmental pollutants such as TCDD and polychlorinated biphenyls (PCBs) depend on ROS generated when CYP1A cycle is uncoupled (Schlezinger et al., 1999; Schlezinger and Stegeman, 2001; Schlezinger et al., 2006; Lu et al., 2009). Interestingly, subsequent altered redox status caused by these ROS can lead to oxidation of certain target molecules, particularly BR (De Matteis et al., 1991; Zaccaro et al., 2001; Pons et al., 2003; De Matteis et al., 2006). Finally, even in the absence of substrates, CYP can oxidise NADPH and subsequent uncoupling can persist, thus significantly promoting the overall formation of cellular ROS.

ER is also the major site for cellular protein synthesis (Chakravarthi et al., 2006). To accommodate protein folding, ER lumen possesses oxidising condition for disulphide bond formation which may also contribute to the formation of ROS that leads to ER stress, a condition where ER loses its protein folding capacity (Bhandary et al., 2012).

Unfolded proteins can accumulate as a result of the oxidising condition, leaking Ca2+ from lumen into cytosol, and increase ROS production in mitochondria (Leem and Koh, 2012; Bravo-Sagua et al., 2013). This in turn activates adaptive unfolded protein response (UPR) to restore protein folding and ER homeostasis. However, unresolved persistent stress and ER homeostasis will shift UPR towards apoptosis to protect the organism (Chakravarthi et al., 2006).

2.2 ADAPTIVE RESPONSE TO OXIDATIVE STRESS

In normal cells, ROS homeostasis is maintained by antioxidant enzymes as depicted in Figure 2.1, and non-enzymatic, low molecular mass ROS scavengers, such as ascorbic acid, α-tocopherol, and most importantly glutathione (Leibovitz et al., 1980; Valko et al., 2007).

Glutathione is a primary low-molecular weight thiol in all aerobic cells (Jezek and Hlavata, 2005). In normal cells, more than 90% of glutathione exist in the reduced form (GSH) and less than 10% as glutathione disulfide (GSSG), the oxidised state. The GSH: GSSG ratio is critical to ensure cell’s capability of immediately neutralising any damaging ROS. Upon oxidative stress, GSH is readily reactive and capable of reducing ROS with the help of glutathione peroxidase (GPx) and in turn produce GSSG. Glutathione reductase (GRx) can regenerate GSH from GSSG. These enzymes as well as SOD represent the main mitochondrial defence system in maintaining ROS homeostasis (Baranano et al., 2002; Mari et al., 2009).

Alteration of the GSH:GSSG ratio through exposure to toxic chemicals, pathogenic microorganisms and radiation, amongst others, is indicative of oxidative stress, which in turn invokes activation of cytoprotective genes to counteract oxidative damage. For example, ROS generated by xenobiotics can be detoxified through the reaction of GSH. ROS may also be formed through the catalytic activity of CYPs, further

increasing the consumption of GSH. This may lead to alterations in the ratio of GSH to GSSG, leaving insufficient GSH to protect the cell against ROS generated by normal physiological processes (Dalton et al., 1999; Valko et al., 2007).

At this point, the cells invoke a second tier of damage control, where the stress- responsive transcription factor NF-E2 related factor-2 (Nrf2) is activated via a redox sensitive signalling cascade (Kang et al., 2005). In turn, Nrf2 regulates the coordinated activation of a battery of genes that encode certain phase II drug-metabolising enzymes that include NAD(P)H quinone oxidoreductase (NQO1) and glutathione S-transferase A2 (GST2A) (Kang et al., 2005), HMOX1 (Alam et al., 1999), and CYP2A5 (Abu-Bakar et al., 2007). These enzymes serve to neutralise ROS and electrophiles, biosynthesise glutathione, direct xenobiotic efflux and remove oxidised proteins. The net result is to scavenge ROS and conjugate electrophiles, which serve to limit oxidative damage and to detoxify cells.

2.2.1 CYTOPROTECTION OF HAEM OXYGENASE AGAINST OXIDATIVE DAMAGE

The microsomal haem oxygenase (HMOX), also known as heat shock protein 32, catalyses the initial and rate-limiting step in haem degradation. It exists in two forms: an inducible enzyme, HMOX1, and a constitutive form, HMOX2 (Ryter and Tyrrell, 2000; Maines, 2005). HMOX1, a 32kDa protein, is conventionally localised in ER of many tissues and cell types. It can be induced in wide variety of tissues, particularly the liver by a variety of physiological stimuli, such as hyperoxia, shear stress, and non- physiological stimuli, such as heavy metals, ultraviolet light, and H2O2 (Maines, 1997; Agarwal and Nick, 2000). By contrast, HMOX2, a 36 kDa protein, is constitutively expressed in mitochondria of brain, testes, nephron, endothelium and liver (Agarwal and Nick, 2000). Because its expression is not inducible, it is assumed to regulate haem degradation during normal physiologic conditions.

HMOX catalyses the first step in haem degradation, where it is converted to BV. In this reaction, free Fe2+ is liberated from haem and carbon monoxide (CO) is released as by-product (Figure 2.3) (Gozzelino et al., 2010). BV is then reduced to lipophilic BR by BVR. These catalytic products, if present at above physiological concentrations, are cytotoxic. For instance, free Fe2+ can potentiate oxidative stress through ROS generation via the Fenton reaction (Gozzelino et al., 2010). CO can bind to haemoglobin and cytochrome C oxidase (COX), thus decreasing O2 in arterial blood and inhibiting mitochondrial energy metabolism, respectively (Piantadosi, 2008). BR may physically interact with cell membranes which in turn, disturbs lipid polarity, redox states and causes neonatal jaundice as well as brain damage (Tomaro and Batlle, 2002; Tell and Gustincich, 2009).

NADPH NADP+ ROS Fe2+ BOMs H2O CO NADPH NADP+ Biliverdin ROS reductase (BVR) Haem Biliverdin Bilirubin Excretion Haem oxygenase UGT1A1 (HMOX) CYP2A5

glucuronides

Figure 2.3 Bilirubin metabolism pathways in the liver. The major pathway of BR degradation is via UGT1A1 enzyme glucuronidation. This diagram also shows that BR can be oxidised to BV by the enzyme CYP2A5. The non-enzymatic pathway involves BR oxidation by ROS to produce oxidative metabolites.

However, at physiological concentrations, these compounds have important biological functions. For instance, the iron may be utilised in haemoglobin synthesis as evident in HMOX1 deficient mice that developed iron deficiency anaemia as a result of low serum iron levels (Poss and Tonegawa 1997; Yachie et al. 1999). On the other hand, CO has recently been acknowledged as a second messenger of the anti-inflammation signalling transduction pathways in various tissues that can consequently promote overall anti-apoptotic effect (Maines 1997; Brouard et al., 2002; Micheau and Tschopp, 2003; Arruda et al., 2004; Wegiel et al., 2008; De Backer et al., 2009; Gozzelino et al., 2010). Additionally, physiological concentrations of CO are important in maintaining vasodilation (Ndisang et al., 2004).

BR, the final product of haem degradation is known for its anti-mutagenic and anti-oxidative properties (Stocker and Peterhans, 1989; Bulmer et al., 2008b). Evidence for antioxidant effects of BR in vivo has been obtained from a HMOX1 deficient patient with low serum BR who is vulnerable to oxidative stress-mediated endothelial cell injury (Yachie et al., 1999). Additionally, epidemiology studies linking BR to low risk of oxidative stress-related chronic diseases have also been reported (discussed further in section 2.2.2) (Heyman et al., 1989; Schwertner et al., 1994; Djousse et al., 2001; Temme et al., 2001).

Given their toxicity at high concentrations, it is evident that the products of haem degradation need to be tightly regulated to maintain homeostasis, which is crucial for cellular survival. This thesis will focus on the regulation of intracellular BR especially in cellular response to oxidative stress.

2.2.2 REGULATION OF CELLULAR BILIRUBIN

In the blood, the highly hydrophobic BR is bound to serum albumin to be delivered to the liver where it is conjugated with glucuronic acid to form soluble mono-

and di-glucuronides. This process is solely catalysed by the microsomal Uridine diphosphate glucuronyltransferase 1A1 (UGT1A1) enzyme (Tukey and Strassburg, 2000; Vitek and Ostrow, 2009) (Figure 2.3). BR glucuronides are then excreted into the bile by a multidrug resistance-associated protein 2 (MRP2) (Kamisako et al., 2000). In the gastrointestinal tract, the glucuronides can be metabolised by colonic bacteria to urobilinogen, stercobilinogen and stercobilin (Kamisako et al., 2000; McDonagh, 2010).

Serum concentrations of BR exceeding the binding capacity of albumin are associated with brain damage in neonatal jaundice and in individuals with the rare congenital condition of Crigler–Najjar syndrome (deletion of UGT1A1 activity) (Ihara et al., 1999; Kaplan and Hammerman, 2011). However, serum BR concentrations rarely approach these toxic levels and approximate 10 µM in the general population, which is 30-60 times below the reported toxic concentration of BR (Bulmer et al., 2008a).

Interestingly, mildly elevated serum BR (2-9 times above the concentrations of the general population) has been shown to be associated with reduction in the prevalence and incidence of cancers and cardiovascular disease (Rigato et al., 2005; Vitek and Schwertner, 2008; Horsfall et al., 2011). These protective effects are also reported in individuals with Gilbert's Syndrome (GS), a congenital condition characterised by mild and chronic hyperbilirubinemia unconnected with either haemolysis or liver disease (Vitek et al., 2002; Jiraskova et al., 2012).

GS patients carry the genotype UGT1A1*28/*28, a homozygous insertion of one base pair that resulted in a decrease in BR glucuronidation activity, which leads to an increase in the level of serum BR, ranging from 20 - 90 µM (Gil and Sasiadek, 2012). The UGT1A1*28 allele has been associated with a reduction of the prevalence of chronic diseases. For example, a Czech Republic exploratory case-control study on 777 colorectal cancer (CRC) patients and 986 controls indicated that UGT1A1*28 allele is associated with a 20% decrease in risk of CRC and each 1 µM decrease in serum BR is associated with a 7% increase of CRC risk (Jiraskova et al., 2012). Similarly, in a three- year follow-up of healthy GS subjects, none of them developed new-onset symptomatic 16

ischemic heart disease, whilst 4.3% of control subjects with normal serum BR levels developed the disease (Vitek et al., 2002). In a Framingham Heart Study population, UGT1A1 mutation with increased serum BR concentration showed a strong correlation to lower risk of cardiovascular diseases (Lin et al., 2006). These studies indicate preventive effects of BR on cancer and cardiovascular disease development. Additionally, BR protective role has also been found in end stage kidney disease, diabetes mellitus and diabetic nephropathy (Chin et al., 2009; Han et al., 2010). However, the mechanism by which BR confers such protection is still obscure.

It has been established that BR can act as an efficient antioxidant. One hypothesis for this efficiency is that BR can be recycled through a cyclic redox cycle in which BR is oxidised to BV when it scavenges ROS, a mechanism that is similar to the GSH/GSSG regeneration. BV, in turn, is reduced to BR by BVR, an abundant and ubiquitous enzyme with high turnover rate. In this way, BR could be recycled as antioxidant many times over and capable of protecting against greater concentrations of ROS, even though its cellular concentrations is approximately 10,000 times lower than that of GSH, the principal cellular antioxidant (Baranano et al., 2002; Sedlak and Snyder 2004; Sedlak et al., 2009).

However, in 2004, McDonagh argued against the concept of BV being the main product of ROS attack on BR. It has been demonstrated that once BR is oxidised by ROS, it is broken down to more polar di- and tripyrroles that are readily excreted in the urine (Kobayashi et al., 2003; Arthur et al., 2012; Abu-Bakar et al., 2013), unlike that of GSH system. It has been proposed that rather than being eliminated, any excess of BR is constantly monitored and oxidised to BV by CYP2A5 (further discussed in section 2.2.3). Indeed, the significance of CYP2A5 in cytoprotection has been indicated in vitro, where overexpression of CYP2A5 in a human liver cell line reduced BR- induced apoptosis (Kim et al., 2013). As BV is a more hydrophilic compound as compared to BR, it is possible for it to be present at a higher concentration without causing cytotoxicity. In this regard, BV could be reduced back to BR by BVR, depending on the ROS levels. Additionally, BV has also been shown to possess reducing ability (Stocker, 2004; Jansen

et al., 2010) and thus could provide a surplus of antioxidant capacity. Finally, it is also possible that BV could act as a negative feedback regulator for CYP2A5 activity, once BV level reached saturation. Whatever the fate of BV is, the proposed adaptive response to cellular stress aims to provide a rapid availability of BR antioxidant capacity whilst curbing its toxicity.

In this way, the proposed BR-BV model would parallel the recycling of GSH, another chain-breaking antioxidant. Both systems involve a single chemical that exists in two states: an oxidised form and a reduced form. In both cases, the BV and reduced GSH predominate under healthy conditions and are part of oxidative stress response which is induced when cells are challenged with a wide range of toxic insults (discussed in 2.2). When either system is disrupted, oxidative stress increases and cells die.

Importantly, an in vitro study using human embryonic kidney cells (HEK293) and HMOX knock-out mice indicated that BR complements GSH cytoprotection: the water-soluble GSH primarily protects water soluble proteins and the lipophilic BR protects lipids (Sedlak et al., 2009) such as membrane structures from oxidation . The study also showed that mice with deletion of HMOX2, which generates BV, display greater lipid than protein oxidation, while the reverse holds for GSH depletion. Furthermore, RNA interference depletion of BVR increases oxidation of lipids more than protein and depletion of BVR or GSH augments oxidative-stress related cell death (Sedlak et al., 2009).

2.2.3 CONCERTED INDUCTION OF BILIRUBIN PRODUCTION AND OXIDATION

The concerted induction of microsomal HMOX1 and CYP2A5 in response to oxidative stress is thought to regulate BR homeostasis, in order to protect ER against oxidative damage (Abu-Bakar et al., 2005; Abu-Bakar et al., 2013). In response to

chemical-induced oxidative stress, Nrf2 is activated to induce the expression of HMOX1 and CYP2A5 in the mouse liver (Alam et al., 1999; Abu-Bakar et al., 2007; Lämsä et al., 2012). In turn, HMOX1 increased BR production to scavenge elevated ROS. Upon oxidation by ROS, BR is broken down to polar BR oxidative metabolites (BOMs). These BOMs include dipyrroles, tripyrroles and propentdyopent (De Matteis et al., 2006; Abu- Bakar et al., 2011; Arthur et al., 2012), and are excreted in the urine (Kobayashi et al., 2003; Arthur et al., 2012).

The excess of intracellular BR is predominantly oxidised by CYP2A5 to BV instead of by UGT1A1-catalysed glucuronidation (Arthur, 2010). The role of CYP2A5 and its human orthologue CYP2A6 as BR oxidase was recently demonstrated in vitro using liver microsomes and the recombinant enzymes (Abu-Bakar et al., 2011; Abu- Bakar et al., 2012). These studies indicated that coumarin (the diagnostic substrate of CYP2A5 and CYP2A6) and antibodies raised against these metabolising enzymes suppressed BR by about 90% (Abu-Bakar et al., 2005; Abu-Bakar et al., 2012). BR also significantly inhibited 7-hydroxylation of coumarin by CYP2A5 and CYP2A6, with an

IC50 almost equal to the substrate concentration (Abu-Bakar et al., 2005; Abu-Bakar et al., 2012). The affinity of BR for the CYP2A5 and CYP2A6 is in the same range of their diagnostic substrate coumarin (Ki(BR) = 0.7-2.2 µM; Km(coumarin) = 0.5-2.0 µM) (Abu- Bakar et al., 2005; Abu-Bakar et al., 2012). Furthermore, in vitro incubation of recombinant CYP2A5 and CYP2A6 with BR produced predominantly BV and minor amounts of dipyrroles (Abu-Bakar et al., 2011; Abu-Bakar et al., 2012). In support of the experimental data, docking analysis using the crystal structure of CYP2A6 indicated that the enzymatic BR oxidation occurs at the CYP2A6 active site via a mechanism producing maximal amounts of BV (Abu-Bakar et al., 2012).

Collectively, these observations presented BR as a high affinity endogenous substrate for microsomal CYP2A5 and CYP2A6 and that its oxidation by these enzymes generated primarily BV. This, in turn, lends support to the proposition that the CYP2A5 is one of the enzymes driving the BR redox cycle, crucial for maintaining optimal BR antioxidant activity (discussed in section 2.2.2).

The role of CYP2A5 as a part of the HMOX1 antioxidant system is further reinforced by the observation that the Nrf2-dependent CYP2A5 induction paralleled but followed the induction of HMOX1 by 5-10 hours interval (Abu-Bakar et al., 2005). Under this condition, the CYP2A5-dependent BR oxidation was significantly increased (Abu-Bakar et al., 2005). At elevated concentrations, BR induced its own oxidation through stabilising the labile CYP2A5 protein. This is evidenced when the CYP2A5 protein half-life was increased as a result of BR co-treatment with cycloheximide, a protein synthesis inhibitor (Abu-Bakar et al., 2011). In fact, Aida and Negishi (1991) demonstrated that CYP2A5 mRNA is also stabilised when mice were treated with pyrazole. Collectively, these observations would be consistent with CYP2A5 serving to clear the BR generated by HMOX1 to ensure that levels of this essential but toxic compound do not exceed a safe threshold. This system is highly efficient as BR, a high affinity substrate with an ideal Km of 1-2 µM, is able to regulate its own oxidation in order to maintain the optimal intracellular levels well below the toxic levels of >10 µM.

Further evidence demonstrating CYP2A5 cytoprotective role include: (a) the mouse strain DBA/2 known to have high constitutive activity of CYP2A5 is resistant to metal-induced oxidative stress (Hata et al., 1980; Kershaw and Klaasen, 1991; Liu, et al., 1992; Sendelbach et al., 1992); (b) the constitutive expression of CYP2A5 is maintained by the Nrf2 pathway (Lämsä et al., 2010), suggesting a major role in homeostatic function; and (c) the overexpression of CYP2A5 in mouse liver cell lines reduced BR- mediated apoptosis (Kim et al., 2013).

An interesting preliminary observation that is crucial to this thesis is that HMOX1 and CYP2A5 can also be induced in the liver mitochondria in response to chemical-induced oxidative stress (Genter et al., 2006). Furthermore, mitochondrial CYP2A5 induction is associated with Nrf2 pathway (Lämsä et al., 2012). The role of such induction is unknown. It is plausible that HMOX1 and CYP2A5 could potentially play a role in maintaining BR homeostasis in mitochondria, as observed in ER (Figure 2.4).

Extracellular matrix

Endoplasmic reticulum (ER)

Haem Bilirubin BR excretion Biliverdin glucuronides Bile 1

excretion dipyrroles Urine 2

ROS

CYPs CYTOPLASM

MITOCHONDRIA Drugs ER Xenobiotics References: 1McDonagh et al., 2010 2Arthur et al., 2012 3 Abu Bakar et al ., 2012 4 Abu Bakar et al., 2011

Figure 2.4 Proposed regulations of key enzymes in bilirubin metabolism in mitochondria and endoplasmic reticulum (ER) during oxidative stress. During oxidative stress, induced microsomal (ER) HMOX1 and CYP2A5, and cytosolic BVR are targeted to mitochondria, presumably to induce antioxidant bilirubin regulation (red arrows). The microsomal UGT1A1, the sole enzyme that catalyses bilirubin glucuronidation for excretion is absent in mitochondria. The CYP2A5 induction paralleled but followed the HMOX1 induction by several hours (Abu-Bakar et al., 2005). 21 88 followed the induction of microsomal HMOX1 by several hours

3. RESEARCH PROBLEMS

Bilirubin production from haem in mitochondria is maintained by the resident non-inducible HMOX2 and the cytosolic BVR that is targeted to mitochondria (Agarwal and Nick, 2000; Converso et al., 2006). In response to oxidative stress, BR production in mitochondria is heightened by the reaction catalysed by the HMOX1 and BVR. Ten to 20% of the inducible microsomal HMOX1 and cytosolic BVR are targeted to the inner mitochondrial membrane under oxidative stress conditions (Srivastava and Pandey, 1996 ; Converso et al., 2006; Bindu et al., 2011). On the other hand, high BR levels can inhibit COX and induce mitochondrial membrane permeability that eventually leads to mitochondrial apoptosis (Rodrigues et al., 2002; Malik et al., 2010). Therefore, why is there a need for HMOX1 and BVR to be targeted to mitochondria? Is it possible that BR is also an antioxidant in mitochondria, as observed in microsomes? Since UGT1A1, the main conjugation enzyme and detoxification pathway of BR is not present in mitochondria, how is BR maintained at a physiological concentration in this organelle? As it has been shown that BR oxidation can provide cytoprotection when its glucuronidation pathway is deficient (Cuperus et al., 2009; Tell and Gustincich, 2009), could the mitochondrial cytochrome P450 system play a role in BR oxidation, and in turn, maintain local BR homeostasis in protecting mitochondria against BR toxicity?

The microsomal CYP enzymes can be imported into inner mitochondrial membrane by way of: (i) direct canonical mitochondrial-targeting signals after synthesis in cytosol; or (ii) localisation of microsomal CYPs in mitochondria after amine (NH2)- terminus processing that cause diminished CYPs incorporation and retention in ER (Ahn and Yun, 2010). Consequently, these allow mitochondrial CYP enzymes to adopt the mitochondrial monooxygenase system, which include a protein adrenodoxin reductase (AdxR). Most mitochondrial CYPs are involved in the metabolism of steroids (Hanukoglu, 1996).

The existence of BR oxidase in mitochondrial membrane was first proposed by Brodersen and Bartels (1969). This was further supported by studies that showed that significant BR oxidation in extrahepatic tissues, such as the intestine (Yokosuka and Billing, 1987) and brain (Hansen et al., 1999), occurred in the inner mitochondrial membrane. Although its activity seemed dependent on cytochrome C and O2, the responsible enzyme could not be confirmed as a CYP oxidase. Subsequently, enzymatic oxidation of BR in hepatic mitochondria was also demonstrated by Cardenas-Vazquez et al (1986).

Interestingly, CYP2A5 a known microsomal BR oxidase, is constitutively expressed in mitochondria (Genter et al., 2006) and the CYP2A5-dependent coumarin 7- hydroxylation (COH activity) is almost as high as in microsomes (Honkakoski et al., 1988). In response to chemically-induced ER stress, mitochondrial COH activity increased by 4-fold (Honkakoski et al., 1988). Furthermore, the expression of CYP2A5 in mitochondria is dependent on the Nrf2 pathway (Lämsä, 2012; Lämsä et al., 2012), suggesting protective role of CYP2A5 in mitochondria. These observations lend support to my hypothesis that mitochondrial CYP2A5 may be part of the machinery that protects mitochondria against oxidative stress.

3.1 RESEARCH QUESTIONS

It has been shown that the CYP2A5 enzyme can be targeted to mitochondria during oxidative stress. However, the exact role of CYP2A5 in mitochondria, particularly in relation to BR metabolism is currently unknown. The question then arises as whether key enzymes regulating BR in microsomes, the HMOX1, CYP2A5, UGT1A1 and BVR are targeted to mitochondria when challenged with oxidative stress. Is BR a substrate for the CYP2A5 in mitochondria? Is BR oxidation increased in mitochondria during oxidative stress? What are the products of mitochondrial CYP2A5-mediated BR oxidation? This thesis addresses the above questions and investigates the role of CYP2A5

in mitochondria pertaining to BR metabolism under the condition of pyrazole-induced oxidative stress.

3.2 WORKING HYPOTHESES

Considering the absence of UGT1A1 in mitochondria and that the CYP2A5 enzyme has been indicated in microsomal BR regulation, the CYP2A5 constitutive presence and intracellular targeting during oxidative stress to mitochondria might also be linked to BR regulation in this organelle. The capacity of mitochondrial CYP2A5 to oxidise BR has never been reported. Thus, this thesis will address the following working hypotheses:

1) Enzymes regulating microsomal BR metabolism, HMOX1 and CYP2A5 are targeted to mitochondria during oxidative stress to regulate BR levels. 2) Mitochondrial BR metabolism is comparable to that of microsomes during oxidative stress. 3) Mitochondrial CYP2A5-mediated BR oxidation activity is driven by a different monooxygenase system than that of ER. 4) Mitochondrial BR oxidation products are comparable to those of microsomes. 5) Upregulation of mitochondrial BR oxidation can confer mitochondrial cytoprotection.

Addressing these research questions will give insight to new therapeutic target given that mitochondrial oxidative damage is a crucial event in many pathologies. Expectantly, this mitochondria-targeted antioxidant therapy may offer a better clinical intervention and improved prognosis in these diseases.

3.3 RESEARCH STRATEGY

To test the proposed working hypotheses the following strategy was adopted:

A) Identification and characterisation of key BR regulation enzymes in mitochondria

Aim: i) Establishment of isolation method for the preparation of highly enriched organelles; cytosol, microsomes and mitochondria. ii) To compare the effect of pyrazole on hepatic microsomal and mitochondrial expressions of HMOX1, CYP2A5, BVR and UGT1A1 proteins and enzyme activities.

Experimental procedure: Subcellular fractionations by differential centrifugation methods were compared so as to come up with the best protocol. This is crucial so that presence of a given enzyme cannot be explained on the basis of cross organelle contamination. Microsomes and mitochondria were extracted from treated and non-treated mice livers using the pre-established protocols. Their purity was determined at protein and activity levels of respective marker enzymes. The identity of each protein of interest was confirmed by Western immunoblotting and their relative enzyme activities were measured spectrometrically (procedures are outlined in Chapter 4). Microsomal and mitochondrial protein expressions and enzyme activities were appropriately compared.

B) Study on effects of pyrazole treatment on mitochondrial BR degrading activity

Aim: i) To investigate the effect of pyrazole on BR degrading activity in both microsomes and mitochondria.

ii) To investigate the role of CYP2A5 in the regulation of mitochondrial BR metabolism, in comparison to microsomes.

Experimental procedure: The BR degradation activities in isolated microsomal and mitochondrial were investigated and compared by determining the rate of substrates disappearance (spectrometrically) and metabolites production (HPLC-MS/MS). BR oxidation inhibition with anti-CYP2A5 antibody, anti-AdxR antibody and ascorbic acid were performed in order to determine the involvement of mitochondrial CYP2A5 and ROS in BR oxidation (procedure outlined in Chapter 4). The reactions were compared to its microsomal counterpart. Additionally, the extent of hepatic oxidative damage was assessed by mitochondrial cytochrome C release into cytosol, a key initiator event of apoptosis.

C) Study on the catalytic properties of mitochondrial CYP2A5 in oxidising BR

Aim: i) To determine whether BR can inhibit mitochondrial CYP2A5 activity. ii) To determine the affinity of BR to mitochondrial CYP2A5 enzyme. iii) To assess the metabolite profile of CYP2A5 catalysed BR metabolism.

Experimental procedure: In mitochondria, hydroxylation of different concentrations of coumarin (substrate specific for CYP2A5) was performed in the presence of increasing amounts of BR as inhibitor. Enzyme kinetic parameters was determined and evaluated against those in microsomes. Ascorbic acid was used to block the BR oxidation caused by ROS and to distinguish between the ROS and CYP2A5 dependant BR metabolite profile.

4. METHODS AND MATERIALS

4.1 CHEMICAL REAGENTS AND ANTIBODIES

All chemicals and reagents are listed in Appendix 1.

4.2 LABORATORY ANIMALS AND TREATMENT

The DBA/2J inbred strain of mouse was chosen as the constitutive expression of CYP2A5 in this strain is the highest amongst common laboratory strains (Wood and Conney, 1974; Wood and Taylor, 1979; Juvonen et al., 1985; Lang et al., 1989). Earlier studies linking CYP2A5 induction with oxidative stress used this strain (Abu-Bakar et al., 2004; Abu-Bakar et al., 2005; Abu-Bakar et al., 2007; Abu-Bakar et al., 2011; Abu- Bakar et al., 2012; Abu-Bakar et al., 2013). Additionally, this strain was known to be resistant to metal-induced oxidative stress (Hata et al., 1980; Kershaw and Klaasen, 1991; Liu et al., 1992; Shaikh et al., 1993), an attribute that was thought to be partly related to the high constitutive expression of CYP2A5 (Abu-Bakar, 2006).

In this study, twelve 7-9 weeks old DBA/2J male mice (Animal Resources Centre, Western Australia) were randomly divided into three groups of control animals and three groups of treated animals (2 mice per group). They were maintained at a 12 hour light / dark cycle, and had access to standard rodent chow ad libitum.

To invoke oxidative stress, these mice were given intra-peritoneal injection of 200 mg pyrazole / kg body weight once daily, for 3 consecutive days (Honkakoski et al., 1988; Gilmore and Kirby, 2004). The animals in the control groups were given normal saline only. The mice were sacrificed 24 hours after the last injection by CO2 overdose and their livers were excised. All the experimental procedures were approved by and

conducted in accordance with, National Health and Medical Research Council (NHMRC) Code of Practice for care and use of experimental animals, and approved by the Queensland Forensic and Scientific Services Animal Ethics Committee (FSS-AEC Approval No. 10P01).

The pyrazole doses used in this study were known to cause ER stress in the mouse liver, as evidenced by an increase in the amount of glucose-regulated protein 78 (GRP78), a marker for adaptive ER stress responses (Gilmore and Kirby, 2004; Nichols and Kirby, 2008). Pyrazole is also known to induce mitochondrial Ca2+ accumulation and to increase ROS production in mitochondria of human3 and mouse4 cell lines (Bae et al., 2012). These observations support the proposition that ER stress often results in the increased loading of Ca2+ in mitochondria (Gorlach et al., 2006; Deniaud et al., 2008) and sustained accumulation of Ca2+ in the matrix of mitochondria stimulates mitochondrial metabolism. This, in turn disrupts the ETC (Cederbaum and Rubin, 1974; Deniaud et al., 2008), which leads to the increased production of ROS in mitochondria (Gorlach et al., 2006; Malhotra and Kaufman, 2007). For these reasons, pyrazole was considered as a relevant oxidant stressor that affects redox status of ER and mitochondria: conditions that are suitable to address the objectives of this study.

4.3 BIOCHEMICAL AND MOLECULAR BIOLOGY PROCEDURES

4.3.1 Preparation of subcellular fractions

Microsomes and mitochondria were isolated from fresh livers at 0-4oC as described by Genter et al (2006) with some modifications (Appendix 2). Mitoplasts fraction (mitochondria void of the outer membrane) was then further prepared from

3 HeLa cells derived from human cervical cancer cells.

4 Hepa1c1c7 cells derived from mouse liver hepatoma cells.

mitochondria by digitonin treatment. The microsomes and mitoplasts preparations were aliquoted and stored at -80oC until required. Protein concentrations of the microsomal and mitoplasts fractions were determined by the Lowry method (Lowry et al., 1951). The average protein concentrations of microsomal and mitoplasts fractions (BSA as standard) were 11.3 ± 1.04 µg/µl (3.4 ± 0.3 mg/g wet liver weight) and 7.8 ± 1 µg/µl (0.78 ± 0.09 mg/g wet liver weight) respectively.

4.3.2 Enzyme activity assays

NADPH Cytochrome P450 Reductase Activity

NADPH-cytochrome P450 reductase (NPR) activity, a microsomal marker, was measured using an assay kit (Sigma-Aldrich, Sydney Australia), according to manufacturer’s instruction. NPR transfers electrons from NADPH to cytochrome C. The reduction of cytochrome C is monitored spectrophotometrically by the increase in absorbance at 550 nm. The rate of NPR activity in each sample was measured over 10 minutes. The addition of potassium cyanide to the mitoplasts reaction mixture eliminates any contribution of COX to the activity.

Cytochrome C Oxidase activity

Cytochrome C oxidase (COX) activity, an inner mitochondrial membrane marker, was measured using an assay kit (Sigma-Aldrich, Sydney Australia), according to manufacturer’s instruction. Reduced cytochrome C is re-oxidised by COX. The decrease in absorbance at 550 nm in each samples was observed.

Haem oxygenase activity

Haem oxygenase (HMOX) activity was measured by monitoring BR generation using the method previously described (Ryter et al., 2000). Incubation of 500 μl of reaction mixture5 was done in the dark at 37oC for 1 hour. The reaction was started by addition of substrate hemin (25 μM) and 1mM NADPH. The reaction was terminated by one volume of chloroform. The extracted BR was measured by difference in absorbance between 464 and 530 nm. BR concentration was calculated using ϵ (mM) of 40 cm-1 in chloroform.

Biliverdin Reductase activity

BVR activity was determined spectrophotometrically by monitoring the conversion of BV to BR. Assay conditions were carried out as previously described (Tenhunen et al., 1970; Ryter et al., 2000). Briefly, the reaction mixture6 was pre- incubated for 5 minutes at 37oC. The reaction was started with substrate BV (25 μM) and 100 µM NADPH. BR formation was monitored at 450 nm every 10 minutes for 1 hour. BR concentration was expressed as ρmol BR/mg protein/min using ϵ (mM) of 48.2 cm-1.

Coumarin 7-hydroxylase activity

Coumarin 7-hydroxylase (COH) activity was measured spectrofluorometrically by a modified method (Abu-Bakar et al., 2004) that was previously described (Burke and Mayer, 1974) using a final concentration of 0.1 mM coumarin. The detailed protocol is in Appendix 3.

5 Containing 2mg/ml mouse cytosol, 1mM NADPH, 2 mM glucose 6-phosphate, 1 U glucose 6-phosphate dehydrogenase, 25 µM hemin, 0.25 M sucrose, 20 mM Tris-HCl, pH 7.4.

6Containing 0.5mg/ml protein, 25 µM biliverdin, 100 µM NADPH and 0.1 M potassium phosphate buffer, pH 7.4. 30

In the enzyme kinetic study, treated mitoplasts fractions were incubated with various concentrations of substrate coumarin (0.5 μM to 50 μM) in the presence of increasing amounts of inhibitor BR (1.25 μM to 10 μM). This data was used to measure the enzyme velocity and to construct a Lineweaver–Burk plot to determine the type of inhibition BR presented. In all experiments, 50 μg of microsomes or mitoplasts proteins were used.

4.3.3 Western blot

The sodium dodecylsulfatepolyacrylamide gel electrophoresis (SDS-PAGE) technique was used to separate the proteins of interest by their molecular weight, followed by transferring the proteins onto PVDF membrane. Identification of specific protein was done by immunoblot analysis using mouse anti-HMOX1 monoclonal antibody, chicken anti-CYP2A5 monoclonal antibody, anti-goat polyclonal UGT1A1 antibody, rabbit anti-NPR, anti-cytochrome oxidase IV (COX IV) , anti-cytochrome C and anti-BVR polyclonal antibodies. The antibody-antigen complexes were detected by chemiluminescent substrate. The protocol is detailed in Appendix 4.

4.3.4 Bilirubin disappearance activity

The rate of BR degradation by microsomes and mitoplasts was monitored by incubating BR with these fractions by a method described previously by Abu-Bakar et al (2005). The rate of BR oxidation is expressed as ρmol BR degradation/min/mg protein, using a ϵ (mM) of 48.2 cm-1.

In the study of inhibitory effect of anti-CYP2A5 and anti-AdxR antibodies on BR degradation, the final concentration of microsomal or mitoplasts proteins used were 50 μg. Increasing concentrations of antibodies were added at the same time as 10 μM BR (in dimethyl sulfoxide (DMSO)) to obtain concentration ratios to proteins ranging from

approximately 0.0005 to 2.0. Inhibited BR oxidation rate in the presence of antibodies against uninhibited rate was calculated to determine the remaining percentage of BR oxidation rate. To study the contribution of ROS towards BR disappearance, the same condition was employed with different concentrations of ascorbic acid (ranging from 2 mM to 200 mM).

4.4 SCREENING OF BILIRUBIN OXIDATIVE METABOLITES (BOMs)

4.4.1 HPLC/MS-MS analysis of BOMs

Metabolites screening was performed as previously described (Abu-Bakar et al., 2011; Arthur et al., 2012). The reaction mixtures were incubated for 1 hour in the dark at 37oC after which samples were filtered using 0.45 µm filter membrane. For positive control, in vitro oxidation of BR was generated by incubating BR with Fe-EDTA/H2O2 for 1 hour. These mixtures were immediately injected to a HPLC-MS/MS (ABSciex QTrap 5500 mass spectrometer, USA) equipped with an electrospray (Turbo V) interface coupled to a HPLC system (Nexera, Shimadzu Corp., Kyoto, Japan). Separation was achieved using a 2.5 micron 50 x 2 mm Synergi MAX RP column (Phenomenex, Torrance, CA, USA) run at 40oC, and mobile phase flow rate of 0.5 mL/minute with a linear gradient starting at 5% B for 1.0 minute, ramped to 80% B in 10 minutes, held for 1 minute, to 5% B in 0.2 minute and equilibrated for 3 minutes (mobile phase A = 1% acetonitrile / HPLC grade water, mobile phase B = 95 % acetonitrile / HPLC grade water, both containing 5 mM ammonium acetate). The mass spectrometer was operated in the positive ion, multiple reactions monitoring mode using nitrogen as the collision gas under the following conditions: declustering potential was 75 or 110 (BV only), and the collision energy was between 25-83 eV. HPLC/MS-MS conditions are detailed in Appendix 5 and Appendix 6. The percentages of screened metabolites were calculated

based on the intensity of 10 μM BR at zero incubation time. The relative amounts of each dipyrroles were determined using area under curve (cps) data.

4.5 STATISTICAL ANALYSIS

Analysis of the enzyme kinetic parameters was determined by GraphPad Prism 5 kinetics programme from GraphPad Software Inc (La Jolla, CA, USA). It uses a non- linear regression method of curve and the runs test of residuals to determine statistically whether experimental data are randomly distributed around the curve with 95% confidence. The programme calculates Ki by plotting observed Km (Y-axis) vs BR concentrations (X-axis). For data other than enzyme kinetic parameters Student’s t-test was used for comparisons between two groups. Mean differences were considered significant when p < 0.05.

5. RESULTS AND DISCUSSION

5.1 Purity of Microsomal and Mitoplasts Fractions

The purity of the microsomal and mitoplasts fractions was assessed by measuring the presence of cytochrome C oxidase IV (COX IV) and NADPH P450 reductase (NPR): markers of mitochondrial inner membrane (IMM) and ER respectively. Western immunoblotting showed that a minimal amount of NPR was present in the mitoplasts (Figure 5.1A), which commensurate with the NPR activity (approximately 10% of the activity in microsomes) (Figure 5.1B). This suggests that nearly all of the ER associated with mitochondria has been removed.

On the other hand, no COX IV-reactive band was found in the microsomal fraction although a minimal decrease in absorbance of 550 nm (2-4% of mitoplasts activity) was detected, possibly due to unspecific reaction or normal fluctuation. This suggests a high microsomal enrichment with only minor mitochondrial contamination. Subsequent data analysis was adjusted against the 10% microsomal contamination and 2% to 4% mitoplasts contamination when relevant. It is noted that the COX activity in the treated mitoplasts was significantly reduced to about 25% of control mitoplasts.

(A) Microsomes Mitoplasts

control treated control treated

μg protein/well 2 4 2 4 2 4 2 4 +

NPR 78 kDa

COX IV 17 kDa

(B)

9000

8000

protein)

7000 6000 5000 4000 µmole µmole /min/mg 3000 2000 * 1000 0

enzyme enzyme activity( NADPH P450 reductase activity cytochrome C oxidase activity

treated treated

mitoplasts mitoplasts treated microsomes control mitoplasts control microsomes mitoplasts control microsomes mitoplasts treated microsomes microsomes control

Figure 5.1 Evaluation of purity of microsomal and mitoplasts isolation using subcellular enzymes markers. Male DBA/2J mice were given normal saline (control group) or 200 mg pyrazole/kg body weight for 3 consecutive days (treated group). (A) Western blot analysis of NADPH P450 reductase (NPR) and cytochrome C oxidase IV (COX IV) proteins of microsomes and mitoplasts fractions isolated from control and pyrazole treated mice. Positive control (+) = NPR and COX IV peptides. (B) NPR and COX IV activities in liver microsomes and mitoplasts from and pyrazole-treated mice. Results are given as means ± SD of three separate observations (n=3). *Significant different from control group, p < 0.05 (Student t-test). 35

5.2 Bilirubin Metabolising Enzymes in Mitochondria

5.2.1 Effects of Pyrazole on HMOX1 and CYP2A5 Protein Expression

Pyrazole treatment substantially increased HMOX1 and CYP2A5 immunoreactive bands in microsomes and mitoplasts (Figure 5.2). The intensities of respective bands were consistent with the amount of protein loaded in each well (Figure 5.2A).

Compared to controls, HMOX1 induction in the microsomes was greater than that of CYP2A5 (6-fold for HMOX1 and 3-fold for CYP2A5) (Figure 5.2B). The opposite was observed in mitoplasts where HMOX1 induction was about 2-fold less than the CYP2A5 induction (Figure 5.2B).

A) Microsomes Mitoplasts

control treated control treated

μg protein/well 5 10 5 10 5 10 5 10 +

HMOX1 32 kDa

CYP2A5 50 kDa

4500000

* control

4000000 treated

3500000

values)

3000000 * 2500000 2000000 * 1500000 * 1000000

BAND DENSITY (arbitrary DENSITY BAND 500000 0 microsomes mitoplasts microsomes mitoplasts HMOX1 CYP2A5

Figure 5.2 Effects of pyrazole treatment on HMOX1 and CYP2A5 protein expression in liver of DBA/2J mice A) Western blot analysis of HMOX1 and CYP2A5 proteins in microsomes and mitoplasts from livers of control and pyrazole-treated mice. HMOX1 peptide and recombinant yeast microsomes were used as positive controls (+) for HMOX1 and CYP2A5, respectively. B) Densitometric analysis of 10 µg of HMOX1 and CYP2A5 blots. Results are given as means ± SD of three separate observations (n=3). * Significant different from control group, p < 0.05 (Student t-test).

Although HMOX1 protein induction in mitoplasts was three times less than in microsomes (Figure 5.2B), the increment in catalytic activity of this enzyme was similar in both organelles (increased by 4-fold compared to control) (Table 5.1). This may suggest differences in the reactivity of anti-HMOX1 antibody vis-à-vis the mitochondrial and microsomal HMOX1.

On the other hand, the strong induction of microsomal and mitochondrial CYP2A5 protein was reflected in the CYP2A5-catalysed coumarin 7-hydroxylase (COH) activity 10-fold and 60-fold increase, respectively (Table 5.1).

Table 5.1 Haem oxygenase (HMOX), biliverdin reductase (BVR) and coumarin 7-hydroxylase (COH) activities of liver cytosol, microsomes and mitoplasts from DBA/2J mice. Mice were either untreated controls or were treated with 200 mg pyrazole/kg body weight for 3 consecutive days. Results are given as means ± SD of three separate observations (n=3).

Enzyme Cytosol Microsomes Mitoplasts Activity None Pyrazole None Pyrazole None Pyrazole

COH# ND ND 743 ± 99 7329 ± 500* 10 ± 0.5 622 ± 53*

HMOX## ND ND 260 ± 23 964 ± 63* 83 ± 10 295 ± 30*

BVR## 30 ± 2.5 36 ± 10 ND ND 23.8 ± 4.5 44.6 ± 1* # ρmol umbelliferone/min/mg protein ## ρmol BR/min/mg protein * p < 0.05, compared with corresponding values of un-induced fractions.

Pyrazole is known to induce oxidative liver injury in Nrf2 deficient mice (Lu et al., 2008). It is also known that pyrazole-induced microsomal HMOX1 and CYP2A5 are associated with elevated Nrf2 in the liver (Lu et al., 2008). Furthermore, Gilmore and Kirby, (2004) reported that induction of CYP2A5 by pyrazole is a consequense of ER stress. These published data together with previous observations that HMOX1 and CYP2A5 expressions are regulated by Nrf2 (Alam et al., 1999; Abu-Bakar et al., 2007) strongly suggest that the changes in CYP2A5 and HMOX1 expressions in mitoplasts and microsomes observed in the present study are caused by pyrazole-induced ER oxidative stress.

Given that haem is the substrate for HMOX, induction of this enzyme by pyrazole would result in depletion of cellular haem pool. This, could in turn, affect the activity of haemoproteins, such as CYP enzymes and mitochondrial COX. However, this is the opposite of what I observed with regard to CYP2A5 activity. Elevation of COH activity indicates a complete incorporation of limited free haem into the CYP2A5 38

apoprotein. This therefore indicates preferential haem incorporation into CYP2A5 apoprotein, as it has been acknowledged that induction of HMOX1 by any form of oxidative stress is associated with reduction in total CYP activity, and that CYP2A5 is the only CYP form that is induced under this condition (Beri and Chandra, 1993; Seubert et al., 2002; Abu-Bakar et al., 2004). These observations support the hypothesis that CYP2A5 maybe differently regulated compared to other CYP enzymes, and have a role in cytoprotection against oxidative damage.

As expected, the activity of mitochondrial COX was significantly reduced by pyrazole treatment (Figure 5.1B). This could be explained, in part, by the fact that the mitochondrial haem content is limited (Converso et al., 2006). The activities of both HMOX2 and HMOX1 during oxidative stress (Agarwal and Nick, 2000; Converso et al., 2006) could impact on the limited mitochondrial haem availability and consequently reduce the COX activity, which in turn may negatively affect mitochondrial respiration, increase local ROS and contribute to membrane damage.

These results suggest that the BR production and oxidation enzymes are targeted to mitochondria in response to pyrazole-induced ER oxidative stress. To further characterise the microsomal and mitochondrial BR metabolic systems, BVR and UGT1A1 protein levels were determined in the respective organelles.

5.2.2 Effects of pyrazole on BVR and UGT1A1 expression

The activity of cytosolic BVR, an enzyme that catalyses the reduction of BV to BR, was insignificantly affected by pyrazole treatment, although the protein expression was down regulated to about 60% of the control levels. By contrast, BVR protein levels in the treated mitoplasts increased by about 32% of control (Figure 5.3). The elevated protein levels corresponds well with the enzyme activity, where the induction was about 2-fold in treated mitoplasts (Table 5.1). Additionally, no BVR protein was detected in

both microsomal fractions which further confirm high microsomal enrichment, as shown in section 5.1.

A) Cytosol Microsomes Mitoplasts

control treated control treated control treated

μg protein/well 10 20 10 20 10 20 10 20 10 20 10 20 +

BVR 33 kDa

control 2500000

treated

2000000 *

values)

1500000

1000000

500000 BAND DENSITY (arbitrary DENSITY BAND 0 cytosol microsomes mitoplasts

Figure 5.3 Effect of pyrazole treatment on BVR protein expression in the liver of DBA/2J mice. (A) Western blot analysis of BVR protein in cytosol, microsomes and mitoplasts from livers of control and pyrazole-treated mice. BVR peptide was used as a positive control (+). (B) Densitometric analysis of 20 µg of BVR blots. The means ± SD are of three separate observations (n=3). * Significant different from control group, p < 0.05 (Student t-test).

By contrast, pyrazole totally repressed the expression of UGT1A1 in microsomes, suggesting that BR conjugation does not take place during ER stress (Figure 5.4). UGT1A1 protein was not detected in mitoplasts (Figure 5.4), confirming previous observation that UGT enzymes are not known to be present in the mitochondria (Radominska-Pandya et al., 2005).

Microsomes Mitoplasts

control treated control treated

μg protein/well 5 10 5 10 5 10 5 10 +

UGT1A1 53 kDa

Figure 5.4 Effect of pyrazole on UGT1A1 protein expression. Western blot analysis of microsomal and mitoplasts proteins from livers of control and pyrazole-treated DBA/2J mice probed with anti- UGT1A1 antibody. UGT1A1 peptide was used as a positive control (+).

As UGT1A1 is the sole enzyme that catalyses glucuronidation of BR, the present observations indicate that BR was not glucuronidated when the ER is under stress. This result supports the assumption that UGT1A1 may not be involved in regulating minute intracellular levels of BR, but rather important in the clearance of excessive amounts of BR from systemic circulation (McDonagh, 2010). The absence of this pathway in mitochondria further emphasise the importance of BR oxidation in managing intracellular BR.

Importantly, the evidence of recruitment of HMOX1 and BVR in mitochondria (Figure 5.2, 5.3 and Table 5.1), indicates heightened local BR production in response to oxidative stress. Because mitochondria are particularly sensitive to high BR concentrations that can eventually lead to apoptosis, the presence of CYP2A5 could

suggest that the local BR is maintained within safe levels by converting to nontoxic BV, possibly to be used when required (as discussed in section 2.2.2). However, the capacity of mitochondrial CYP2A5 to oxidise BR is not known and is addressed in the next section.

5.3 Bilirubin Degradation in Mitochondria

It has been established that the CYP2A5-dependent BR oxidation exists in microsomes and that the activity is substantially increased in response to chemical- induced oxidative stress (Abu-Bakar et al., 2005). In exploring the existence of this system in mitochondria and its inducibility in response to oxidative stress, BR degradation rate was initially measured spectrometrically in incubations with mitoplasts and microsomes. Table 5.2 shows that BR disappeared in the presence of control microsomes and mitoplasts, and that the disappearance rate increased in pyrazole-treated microsomes and mitoplasts. Interestingly, the rate in control microsomes and mitoplasts is about the same, but in treated microsomes it increased three times more than in treated mitoplasts (3-fold increase in treated microsomes; 1.7-fold increase in mitoplasts) (Table 5.2). The observations suggest that while a BR oxidation system exists in both organelles it differs in the kinetics of inducibility.

Table 5.2 Bilirubin degrading activity in liver microsomes and mitoplasts of DBA/2J mice. Mice were either untreated controls or were treated with 200 mg pyrazole/kg body weight for 3 consecutive days. Hepatic microsomal and mitoplasts fractions were isolated for the estimation of bilirubin degrading activities. Results are given as means ± SD of three separate observations (n=3).

Enzyme Activity MICROSOMES MITOPLASTS

None Pyrazole None Pyrazole Bilirubin degradation rate# 1175 ± 30 3783 ± 259* 940 ± 85 1580 ± 70*

#ρmol BR/min/mg protein *p < 0.05, compared with corresponding values of un-induced fractions.

It is known that in the microsomes, BR is subjected to random oxidation by ROS and to specific oxidation by mouse CYP2A5 and human CYP2A6 enzymes (De Matteis et al., 2006; Abu-Bakar et al., 2011; Abu-Bakar et al., 2012; Jansen and Daiber, 2012). The non-enzymatic oxidation of BR predominantly generates smaller metabolites, dipyrroles (De Matteis et al., 2006; Abu-Bakar et al., 2011), while the predominant product of enzymatic oxidation is BV (Abu-Bakar et al., 2011). As mitochondria are the major producers of intracellular ROS through leakage of their electron transfer system, it is thus plausible that the observed BR disappearance activity may be caused by both enzymatic and non-enzymatic oxidations.

5.3.1 Contribution of ROS to bilirubin degrading activity

To explore the possibility that ROS may be involved in the oxidation of BR, incubations of microsomal and mitoplasts fractions were done in the presence of ascorbic acid — a ROS scavenger — to exclude random oxidation by ROS (Zhao et al., 1990; Meister, 1992; Tsai et al., 2007). Accordingly, 200mM ascorbic acid was used in all reactions. This concentration has been shown to effectively inhibit BR oxidation by ROS (see Appendix 7).

Ascorbic acid did not significantly inhibit BR degradation in control microsomes but inhibited about 22% of the reaction in pyrazole-treated microsomes (Figure 5.5). This suggests that random oxidation of BR by ROS is a minor event in microsomes and most of the disappearance (about 78%) is catalysed by some other mechanism, most likely CYP2A5.

By contrast, ROS-driven BR disappearance seems to be the major, if not the only degradation pathway in the control mitoplasts, as 100% of the BR disappearance was blocked by ascorbic acid (Figure 5.5). In the treated mitoplasts however, only 54% of BR degradation was blocked by ascorbic acid (Figure 5.5), further suggesting that in

response to oxidative stress another pathway of BR degradation, independent of ROS, is as important as the ROS-mediated oxidation.

In addition, these results seem to suggest that there is a correlation between the CYP2A5 catalysed COH activity and the BR disappearance rate: whenever the COH activity is low, the relative contribution of the ROS to BR degradation seems to be high. In extreme case, in the control mitoplasts, where the COH activity was absent, the ascorbic acid blocked BR disappearance was 100%. Conversely, when the COH activity is increased, the proportion of ROS-driven BR oxidation is decreased.

Why the ROS-driven BR oxidation proportionally increased in microsomes despite the strong increased of the COH activity? This is not known but one could speculate that pyrazole treatment leads to disintegration of the membrane structure and increase leakage of ROS from the monooxygenase complex.

In the case of untreated mitoplasts, it could be that ROS is an inherent by- product of the electron transport chain which could explain the un-enzymatic oxidation of BR.

4500 0 mM ascorbic acid

4000 200 mM ascorbic acid 3500

3000

2500

mol BR/min/mg protein) BR/min/mgmol ρ 2000

1500

1000 *

500 BR degradation rate ( rate degradationBR * 0 control treated control treated Microsomes Mitoplasts

Figure 5.5 Effect of 200mM ascorbic acid on bilirubin (BR) degradation activity in liver microsomes and mitoplasts of DBA/2J mice. Mice were either untreated controls or were treated with 200 mg/kg body weight of pyrazole for 3 consecutive days. Hepatic microsomal and mitoplasts fractions were isolated for the estimation BR degradation activities. Results are given as means ± SD of three separate observations (n=3). * Significant different from control group, p < 0.05 (Student t- test).

5.3.2 Contribution of CYP2A5 to bilirubin degrading activity

To explore the relative contribution of CYP2A5 to BR oxidation in both organelles, studies with antibody directed against microsomal CYP2A5 were undertaken. It is known that this antibody can recognise its mitochondrial counterpart (Honkakoski et al., 1988).

The results show that increased concentrations of anti-CYP2A5 antibody were able to gradually inhibited BR disappearance. The maximum inhibitory effect was strong, 45

about 70% in both control and pyrazole-treated microsomes, respectively (Figure 5.6). The amount of CYP2A5 antibody needed to achieve maximum inhibition in treated microsomes was about 13-fold higher than needed in control fraction, indicating that larger amounts of CYP2A5 protein responsible for BR disappearance were present in this sample, in conformity with Western blot analysis and the COH activity (Figure 5.2 and Table 5.1).

As expected, a significant inhibition of about 55% could be reached in mitoplasts only after pyrazole treatment. This correlates well with the results of ascorbic acid incubation, whereby the inhibition was about 54% (Figure 5.5), and taken together this amounts to about 100% of BR oxidation. On the contrary, no significant inhibition could be seen in the control mitoplasts with only some non-dose response inhibition taking place, implying the CYP2A5 catalysed BR oxidation in this fraction is insignificant. By contrast, these results suggest that BR oxidation in treated mitoplasts is almost equally driven by CYP2A5 and ROS.

120 120 Microsomes control Microsomes pyrazole-treated

100 100

80 80

60 60

degradation degradation rate remaining)(%

20 20

BR BR

0 0 0 0.05 0.1 0.15 0 0.5 1 1.5 2 2.5

antibody (µg antibody/ug microsomal protein) antibody (µg antibody/ug microsomal protein)

120 120 Mitoplasts control Mitoplasts pyrazole-treated

100 100

80 80

60 60

40 40 degradation degradation rate remaining)(%

BR degradation degradation rate remaining)(% 20 20 BR 0 0 0 0.001 0.002 0.003 0.004 0.005 0 0.01 0.02 0.03 0.04

antibody (µg antibody/ug mitoplasts protein) antibody (µg antibody/ug mitoplasts protein)

Figure 5.6 Effect of anti-CYP2A5 antibody on bilirubin (BR) degradation activity of control and pyrazole-treated microsomes and mitoplasts fractions. The antibody was added in increasing amount so as to achieve the antibody to microsomal or mitoplasts protein concentrations indicated. In all fractions, 10 μM final concentration of BR was used. BR-degrading activity is expressed as a percentage of the corresponding values. Results are given as means ± SD of three separate observations (n=3).

To confirm that the enzymatic BR oxidation is catalysed by the respective monooxygenase complexes in each organelle, anti-adrenodoxin reductase (AdxR) antibody was added to the incubation mixtures. AdxR, is a key component of the mitochondrial monooxygenase system that provides electrons to the mitochondrial CYP catalysed reactions and is absent in microsomes.

Figure 5.7 shows that the antibody did not inhibit microsomal BR disappearances, suggesting that the microsomes are essentially free from mitoplast contamination (further confirming data presented in section 5.1), and confirming that the microsomal BR degradation is essentially driven by the microsomal monooxygenase system.

By contrast, the anti-AdxR antibody inhibited BR degradation by 50% in treated mitoplasts but not at all in control mitoplasts (Figure 5.7). This result is in agreement with the results in section 5.3.1 and Figure 5.6, and confirms that in control mitoplasts BR disappearance is predominantly driven by ROS-mediated oxidation and that in treated mitoplasts the metabolism is driven 50% by mitochondrial CYP2A5. Furthermore, this result also confirms the purity of microsomal fractions with least contamination from mitochondria.

(A)

160 Treated 140 Control 120 100

rate remaining)rate (%

60 40 20

BR degradation BR 0 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 μg anti-AdxR antibody/μg microsomal protein

(B)

200 180 Treated 160 Control 140 120 100 80 60 40 20 0 BR degradation rate remaining)rate (% degradation BR 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 μg anti AdxR antibody/ μg mitoplasts protein

Figure 5.7 Effect of anti-adrenodoxin reductase (AdxR) antibody on bilirubin (BR) degradation activity in microsomes and mitoplasts fractions. Inhibition of BR degrading activity of (A) control and pyrazole-treated microsomes (B) control and pyrazole-treated mitoplasts, by anti-adrenodoxin reductase antibody.The antibody was added in increasing amounts so as to achieve the antibody to microsomal or mitoplasts protein concentrations indicated. In both (A) and (B) 10 μM final concentration of BR was used. BR-degrading activity is expressed as a percentage of the corresponding values. Results are given as means ± SD of three separate observations (n=3).

It is pertinent to confirm these observations by screening the metabolites profile of BR oxidation. As BV is the predominant metabolite of CYP2A5-driven BR oxidation and dipyrroles are the major products of BR scavenging ROS, it is anticipated that the conditions in the treated microsomes would favour more BV than dipyrroles production due to high CYP2A5 expression, whilst dipyrroles would be the main metabolites produced in untreated mitoplasts, with increasing BV production when treated with pyrazole, due to induced CYP2A5 .

After one hour incubation, almost 100% BR was lost in the chemical oxidation (driven solely by ROS scavenging). The microsomal and mitoplasts system was comparatively equally efficient at BR oxidation, but less so than the chemical oxidation system (Figure 5.8). An average of 49% and 59% loss of BR in control and pyrazole- treated microsomes was observed respectively. The most significant difference between either incubations of microsomal and mitoplasts with chemical oxidation was in the level of BV produced.

No BV formation but only dipyrroles products were observed in the chemical oxidation system, confirming previous findings that dipyrroles are the major products of non-enzymatic BR-oxidation (De Matteis et al., 2006; Abu-Bakar et al., 2011; Abu- Bakar et al., 2012). The highest levels of BV proportionally to dipyrroles were produced by untreated microsomes followed by treated microsomes. On the other hand, most dipyrroles relative to BV was produced by control mitoplasts followed by treated mitoplasts (Figure 5.8). These results are in accordance with ascorbic acid inhibition studies (Chapter 5.3.1) as well as with studies using CYP2A5 antibody (Figure 5.6). The small amounts of dipyrroles showed in control microsomes and BV showed in control mitoplasts are potentially artefacts, i.e., marginal side products of unspecific reactions (De Matteis et al., 2006).

60 biliverdin

50 dipyrroles

intensity intensity

% of initial BR initial of % * 10 *

0 chemical microsomes microsomes mitoplast mitoplast oxidation control treated control treated

Figure 5.8 Summary of the oxidative products of bilirubin (BR) formed in the chemical and enzymic systems. Oxidative products formed in each of the systems are expressed as a percentage of the intensity of initial BR concentration. Results are given as means ± SD of three separate observations (n=3). *Significant different from control group, p < 0.05 (Student t-test).

Additionally, break down products of non-enzymatic microsomal BR oxidation were compared with those of mitochondria. Three smaller products (ions m/z 301,315 and 333) were identified in the BR oxidations of microsomes and mitoplasts as previously reported (De Matteis et al., 2006; Abu-Bakar et al., 2011; Arthur et al., 2012). The relative amount of each BOM product in the control and pyrazole-treated microsomes and mitoplasts are depicted in Figure 5.9. The major BOM in both cases was ion m/z 301, although almost as much of ion m/z 315 was formed in mitoplasts. The levels of these ions were increased after pyrazole exposure. According to Arthur et al (2012), the more hydrophilic dipyrrole (ion m/z 301) is predominantly produced, possibly to assist the excretion, as seen in urinary BOMs of oxidative stressed mice.

It has been purported that these BOMs result from the cleavage of BR central methylene bridge by ROS that give rise to BR peroxyradical (De Matteis et al., 2006; Abu-Bakar et al., 2012), which can then disintegrate into ions m/z 315 and 301. It is suggested that a relatively stable ion m/z 333 could be further formed when ion m/z 315 encounter hydroxyl radical (personal communication) (see Figure 1 in Appendix 7). As reviewed in Chapter 2.1.1, mitochondria are the primary source of ROS, which can be aggravated during oxidative stress. Thus, the higher level of ion m/z 333 in treated mitoplasts as compared to microsomes could owe to the increased ROS availability in this organelle.

ion m/z 301 ion m/z 315 900000 ion m/z 333 * *

800000 *

700000

600000

500000 *

under the curve (cps) the curve under 400000

Area 300000 * 200000

100000

0 microsomes control microsomes treated mitoplasts control mitoplasts treated

Figure 5.9 Effect of pyrazole on microsomal and mitoplasts BOMs. Relative amounts of each BOM formed in control and pyrazole-treated microsomes and mitoplasts were determined using area under the curve data obtained from HPLC/MS-MS. Results are given as means ± SD of three separate observations (n=3). * Significant different from control group, p < 0.05 (Student t-test). 52

5.3.3 Microsomal and mitochondrial CYP2A5 enzymes have similar affinity to bilirubin

It has been shown that both microsomal and mitochondrial CYP2A5 enzymes have a high affinity towards coumarin. Thus, coumarin can be used as diagnostic substrate for these enzymes (Honkakoski et al., 1988). To further characterize the interaction of BR with mitochondrial CYP2A5, 7-hydroxylation of coumarin in treated mitoplasts was inhibited with different concentrations of BR.

Figure 5.10A shows that the formation of 7-hydroxycoumarin by mitoplasts CYP2A5 followed simple Michaelis-Menten kinetics. Inhibition of mitochondrial coumarin 7-hydroxylation by BR is of a mixed type that affected both the Vmax and Km, where the Vmax is slightly reduced and the Km is increased. The Km was 6.14 ± 0.5 μM, which is 13 times more than that of microsomes, suggesting that coumarin affinity to the mitochondrial CYP2A5 is a lot less than to the microsomal CYP2A5 (Figure 5.10B). This aligns with previous observations that targeted CYPs are modified to depend on mitochondrial ETC for a supply of electrons for catalytic activity toward altered substrate specificity/affinity (reviewed by Kinonen et al., 1995). This in turn may explain the odd correlation between changes of mitochondrial COH activity and the corresponding mitochondrial CYP2A5 protein levels observed in section 5.2.1, Figure 5.2 and Table 5.1.

The plot Km observed vs BR concentrations (Figure 5.10B) yielded a linear graph, which confirmed the competitiveness of BR inhibition to the reaction with Ki = 2.67μM. This indicates that BR affinity for mitochondrial CYP2A5 is as high as the reported affinity for microsomal enzyme (ranging from 0. 7 to 2.2 μM) and comparable to coumarin affinity for microsomal CYP2A5 (ranging from 0.5 to 2.0 μM) (Kinonen et al., 1995; Draper et al., 1997; Abu-Bakar et al., 2005; Abu-Bakar et al., 2012). Importantly, BR affinity for mitochondrial/microsomal CYP2A5 (0.5-2.7 μM) appears therefore to be

similar (unlike that of coumarin) and optimal for maintaining the intracellular BR concentrations at sub-toxic levels of <10 μM.

According to Genter et al (2006), protein secondary structure of microsomal CYP2A5 is altered to allow mitochondrial targeting, as well as integration into mitochondrial monooxygenase-electron transfer system, although this would not be expected to alter its function. As a result, mitochondrial enzymes can show changes in substrate specificity (Kinonen et al., 1995; Sangar et al., 2010). This could then affect the accuracy of coumarin-CYP2A5 docking. It would seem that the interaction of BR- CYP2A5 was not affected by this, perhaps due to dissimilar active site. Indeed, docking analysis of CYP2A6 protein (the human orthologue of CYP2A5) has been analysed previously to simulate molecular docking of BR and coumarin on the CYP2A6’s active site. Both interact with a few amino acid residues: BR forming hydrogen bond at Asn297 and Thr305, while coumarin with Asn297, as well as a few phenylalanine residues (Abu- Bakar et al., 2012; Abu-Bakar et al., 2013). Additionally, considering the nature of BR that has a large structure, it is also plausible that the binding of BR to CYP2A5 would hinder the coumarin-CYP2A5 interaction. Thus, it would be interesting to predict the coumarin and BR mitochondrial CYP2A5 binding site to confirm these results.

(A)

Vmax (ρmol / mg protein / min) / protein mg / (ρmol Vmax

V K [BR] μM max m Observed (ρmol/mg protein/min) (µM coumarin)

0 711 ± 21 6.8 ± 0.5 1.25 688 ± 40 7.5 ± 1.0 2.5 665 ± 34 12.48 ± 1.4 10 601 ± 23 14.9 ± 1.2

(B)

20 18

μ ( 14 12

observed

m 8 K 6 Ki=-2.67 4 Km=6.14 2 0 -3 -2 -1 -2 0 1 2 3 4 [Bilirubin] μM

Figure 5.10 Effect of bilirubin (BR) on coumarin 7-hydroxylase activity in pyrazole-treated mitoplasts. (A) Mitoplasts was incubated with various concentrations of coumarin with increasing amounts of BR. The rates of 7-hydroxylation were measured fluorometrically. The Km value for coumarin

7-hydroxylation was 6.14 ± 0.5 μM. BR increased the Km and slightly reduced the Vmax. (B) Km observed vs BR concentrations plot indicates that BR is a competitive inhibitor with Ki = 2.67 μM. 55

5.3.4 Effect of pyrazole on the release of cytochrome C from mitochondria

Cytochrome C is a protein bound to inner mitochondrial membrane and can be released when the membrane is damaged. Cytochrome C release is a pre-requisite condition for apoptosis and can be initiated by oxidative stress (Srinivasan and Avadhani, 2012). Analysis of this protein in mitoplasts and cytosol will give a glimpse on the extent of damage in this organelle; whether or not pyrazole treatment caused any damage on mitochondrial membrane. The result shows that cytochrome C was not detected in the cytosolic fractions of either group. On the other hand, mitoplasts fractions showed substantial accumulation, with no significant difference between the two groups (Figure 5.11). This suggests that the mitochondrial membrane is intact, and that the organelles are not damaged.

A) Cytosol Mitoplasts

control treated control treated

μg protein/well 15 30 15 30 15 30 15 30 +

cytochrome C 15kDa

4000000

control 3500000 pyrazole 3000000

2500000

2000000

1500000

1000000 BAND DENSITY (arbitrary values) (arbitrary DENSITY BAND 500000

0 cytosol mitoplasts

Figure 5.11 Effect of pyrazole on cytochrome C protein expression. (A) Western blot analysis of cytosolic and mitoplasts proteins from livers of control and pyrazole-treated DBA/2J mice probed with anti-cytochrome C antibody. Cytochrome C peptide was used as a positive control (+). (B) Densitometric analysis of 30 µg of cytochrome C blots. The means ± SD are of three separate observations (n=3).

It has been suggested that reduction of mitochondrial COX level is accompanied by release of cytochrome C into cytosol (Sanchez-Alcazar et al., 2000). This is in contrast to the present results (Figure 5.1 and Figure 5.11). This could be due to:

1) methodological limitations: whether the cytochrome C antibody applied in the Western blotting was enough to detect low levels of cytochrome C released into cytosol. Thus, it would be interesting to employ a more sensitive technique such as immunoelectron microscopy to study the microstructure in mitochondria and detect apoptotic proteins; or

2) mitochondrial integrity was uncompromised and apoptotic signals were not amplified due to increased antioxidant capacity via recruitment of the HMOX1 / CYP2A5 antioxidative system. The results suggest that although with limited COX activity, the mitochondria were not overly stressed so as to initiate apoptosis. Previous findings have shown that despite a strong correlation between upregulated CYP2A5 and hepatocellular damage, and that some CYP2A5 inducers are pro-apoptotic (Habeebu et al., 1998; Shi et al., 1998), no link between CYP2A5 build-up and apoptosis can be established. Recently, CYP2A5 overexpression has been shown to confer cytoprotection against BR-induced apoptosis (Kim et al., 2013), which is in accordance with the current result. It is likely that the cells are in damage control mode and that local ATP pool in mitochondria and in ER may be dedicated to importing necessary cytoprotective proteins, as reflected in the changed activities of BR regulating enzymes in microsomes as well as in mitochondria. Additionally, HMOX1 catabolism product, CO can bind to COX. Such binding would also supress COX activity, which in turn may degrade pro-apoptotic pathway that prevents mitochondrial membrane permeability and apoptosis (Gozzelino et al., 2010; Queiroga et al., 2011).

It has been shown that mitochondria defects (increasing ROS levels and decreasing ATP levels) can regulate adenine monophosphate-activated protein-activated kinase (AMPK) and peroxisome proliferator-activated receptor γ coactivator 1α (PGC- 1α) (Adom and Nie, 2013), that are important in energy homeostasis and liver functions

(Yang et al., 2010; Liu and Lin, 2011). These inductions can lead to ROS detoxifications, promotion of catabolic pathway such as glycolysis to maintain ATP, as well as mitochondrial biogenesis to increase mitochondrial content and activity that would allow tissue adaptation to stress. Additionally, defective mitochondria can be eliminated by mitophagy through AMPK pathway, to protect cells from apoptosis (Nadege et al., 2009; Liu and Lin, 2011; Wu and Wei, 2012; Adom and Nie, 2013).

Therefore, it is of particular interest that pyrazole treatment is associated with hepatic glycogen depletion and that CYP2A5 can be induced by fasting and glucagon (hormone that promotes breakdown of glycogen) through PGC-1α. (Arpiainen et al., 2008; Nichols and Kirby, 2008; Abu-Bakar et al., 2013). Nevertheless, the PGC-1α- CYP2A5 induction is thought to prevent BR cytotoxicity as fasting can upregulates serum BR levels whilst the reduced hepatic glycogen content would be due to increased glucose 6-phosphate production through glycogenolysis to replenish NADPH. NADPH in turn is used to regenerate GSH (Gilmore et al., 2003; Kirby et al., 2011), which is important in protecting water soluble proteins from oxidative damage (Sedlak et al., 2009). Then again, inhibition of AMPK during pyrazole intoxication has been shown to downregulate CYP2A5 mRNA expression (reviewed by Kirby et al., 2011). Given the role of AMPK and PGC-1α in maintaining cellular energy homeostasis, mitochondrial generation and liver functions, could it be possible that their interactions with CYP2A5 are also in part to restore ATP homeostasis, whilst cells are geared towards combating oxidative damage? Indeed, it would be attractive to consider a few synchronised damage control mechanism regulated by CYP2A5 to restore BR and energy homeostasis, as depicted in Figure 5.12.

5.4 CONCLUSIONS

It has been shown that mitochondria are particularly sensitive to oxidative stress and local antioxidant capacity including the BR is needed to be present to protect its structures against ROS. On the other hand, high BR concentrations can also be toxic and

can eventually lead to apoptosis (Rodrigues et al., 2002; Malik et al., 2010). As has been shown (Figure 5.2 and 5.3, Table 5.1), critical BR producing enzymes, HMOX1 and BVR are targeted to mitochondria during pyrazole-induced oxidative stress. However, it is not known how mitochondria are protected against high BR levels. Mitochondrial BR oxidase has been previously proposed yet never been confirmed. My initial results clearly demonstrate the recruitment of mitochondrial CYP2A5 under oxidative stress conditions (Figure 5.2 and Table 5.1). However, its role in this organelle, let alone pertaining to BR metabolism is unknown.

It would seem that the microsomal defence system alone is insufficient to manage intracellular oxidative stress but needs to be recruited into mitochondria too. In response to pyrazole-induced ER stress, the enzymes crucial in regulating BR were recruited into mitochondria, which indicate increased local BR production capacity and elevated BR levels, which in turn could damage mitochondrial membrane. Nevertheless, the UGT1A1 enzyme that is responsible to eliminate systemic excess of BR was not targeted into this organelle. Interestingly enough, the mitochondrial membrane was intact, as indicated by no leakage of cytochrome C into cytosol. The current findings also present primary evidence that microsomal CYP2A5 is used to constantly monitor excessive cellular BR levels and prevent its accumulation, whereas the mitochondrial CYP2A5 becomes significant under strong oxidative stress for local control of BR levels, when mitochondrial defence capacity against oxidative stress is upregulated by induction of HMOX1.

In summary, the present findings demonstrate:

i) key enzymes regulating intracellular BR (HMOX1, BVR and CYP2A5) are targeted to mitochondria during oxidative stress;

ii) during oxidative stress, mitochondrial BR degradation is equally driven by ROS and CYP2A5, in contrast to microsomes, in which the degradation is driven mainly by CYP2A5;

iii) mitochondrial CYP2A5-catalysed BR degradation is driven by the mitochondrial monooxygenase system;

iv) during oxidative stress, mitochondrial BR degradation produce equal amounts of BV and dipyrroles, whilst in microsomes, the product is mainly BV.

v) microsomal and mitochondrial CYP2A5 have equally strong affinity to BR;

vi) targeting of key enzymes that regulate BR levels to mitochondria is not associated with initiation of apoptosis;

Collectively, these results suggest that targeting of key BR regulatory enzymes to mitochondria confer their protection against oxidative stress.

PYRAZOLE ? 2+ Ca release (Bae et al., 2012) ER stress Mitochondrial stress Gorlach et al., 2006; Deniaud et al., 2008) COXIV Ca2+accumulation HMOX1 BVR CYP2A5

UGT1A1 ATP Mitochondrial swelling

cytochrome C release bilirubin Restore ATP ?

ER / mitochondria stress lipid

peroxidation

APOPTOSIS

Figure 5.12 Proposed protective mechanism through bilirubin regulation during oxidative stress induced by pyrazole intoxication. Increased microsomal and mitochondrial bilirubin metabolism enzymes activities (HMOX1, BVR and CYP2A5) levels ameliorate lipid peroxidation, preventing mitochondrial cytochrome C release. Reduced cellular ATP presumably increased glycolysis through PGC-1α - CYP2A5 induction to restore ATP homeostasis (red arrows). These concerted activities are potentially in part to rescue cells from oxidative damage and apoptosis.

6. GENERAL CONCLUSIONS AND FUTURE DIRECTIONS

6.1 GENERAL CONCLUSIONS

The findings in this thesis contribute to the advancement of current knowledge in the research field of mitochondrial CYP enzymes as well as on the molecular events of hepatic cytoprotective mechanism in response to oxidative stress. Essentially, this study potentially describes a novel antioxidant mechanism that implicates a CYP enzyme as part of the mitochondrial adaptive response to cellular stress. In this regard, it demonstrates:

i) the involvement of CYP2A5 as a major catalyst for mitochondrial BR oxidation, to facilitate BR maintenance for cellular protection during ER stress;

ii) that mitochondrial CYP2A5 and microsomal CYP2A5 are functional BR oxidases; and

iii) non-enzymatic mitochondrial BR oxidation by ROS is dominant constitutively whilst the enzymatic, CYP2A5-catalysed become crucial during pyrazole-mediated ER stress.

Given that mitochondrial oxidative damage is a crucial event in many pathologies, these molecular events may offer a potential therapeutic target for clinical intervention.

6.2 FUTURE DIRECTIONS

The importance of mitochondrial oxidative damage in clinical conditions makes preventing it an attractive clinical approach that calls for fundamental and comprehensive knowledge on mitochondrial antioxidative regulation. This thesis has resolved but concurrently raised a few research questions that include:

a) Is the temporal activation of the mitochondrial CYP2A5 and HMOX1 enzymes coordinated or controlled in the same way as in microsomes?

b) What are the sequential events leading to cellular apoptosis when mitochondrial BR homeostasis is lost?

c) How is microsomal CYP2A5 directed to mitochondria during oxidative stress?

d) How does coumarin and BR interact with mitochondrial CYP2A5 protein?

e) Is the recruitment of BR regulating enzymes: the HMOX1, BVR and CYP2A5 a universal event in all organs or is it just liver specific?

f) How much does BR contribute to overall mitochondria protection against oxidative stress?

These questions, apart from narrowing the knowledge gap that will help to provide a complete map on adaptive response towards oxidative stress, is further significant given that oxidative stress has been implicated in pathological conditions as diverse as malignancies, cardiovascular diseases, neurodegenerative disorders and autoimmune diseases. It is therefore pertinent to understand fundamental cellular defence mechanisms particularly those at subcellular level notably mitochondria. This may in

turn help in developing more effective, novel clinical management and practices with improved targeting selectivity to impact the oxidative damages and disease outcomes.

7. BIBLIOGRAPHY

Abu-Bakar, A. (2006). The Cytochrome P450 2A5: Induction by cadmium and its role as hepatic bilirubin oxidase. School of Population Health. University of Queensland, Brisbane, Australia, pp. 121.

Abu-Bakar, A., Arthur, D.M., Aganovic, S., Ng, J.C., and Lang, M.A. (2011). Inducible bilirubin oxidase: a novel function for the mouse cytochrome P450 2A5. Toxicol Appl Pharmacol 257, 14-22.

Abu-Bakar, A., Arthur, D.M., Wikman, A.S., Rahnasto, M., Juvonen, R.O., Vepsalainen, J., Raunio, H., Ng, J.C., and Lang, M.A. (2012). Metabolism of bilirubin by human cytochrome P450 2A6. Toxicol Appl Pharmacol 261, 50-58.

Abu-Bakar, A., Hakkola, J., Juvonen, R., Rahnasto-Rilla, M., Raunio, H., and Lang, M.A. (2013). Function and regulation of the Cyp2a5/CYP2A6 genes in response to toxic insults in the liver. Curr Drug Metab 14, 137-150.

Abu-Bakar, A., Lämsä, A., Arpiainen, A., Moore, M.R., Lang, M., and Hakkola, J. (2007). Regulation of CYP2A5 gene by the transcription factor nuclear factor (erythroid-derived 2)-like 2. Drug Metab Dispos 35, 787-794.

Abu-Bakar, A., Moore, M.R., and Lang, M.A. (2005). Evidence for induced microsomal bilirubin degradation by cytochrome P450 2A5. Biochem Pharmacol 70, 1527-1535.

Abu-Bakar, A., Satarug, S., Marks, G.C., Lang, M.A., and Moore, M.R. (2004). Acute cadmium chloride administration induces hepatic and renal CYP2A5 mRNA, protein and activity in the mouse: involvement of transcription factor NRF2. Toxicol Lett 148, 199-210.

Adom, D., and Nie, D. (2013). Regulation of autophagy by short chain fatty acids in colon cancer cells. In Autophagy-A Double Edge Sword-Cell Survival or Death (Y. Bailly, Ed.), pp. 235-247. InTech, doi: 10.5772/50855. Available from: http://www.intechopen.com/books/autophagy-a-double-edged-sword- cell-survival-or-death-/natural-compounds-and-their-role-in-autophagic cell- signaling-pathways.

Agarwal, A., and Nick, H.S. (2000). Renal response to tissue injury: lessons from heme oxygenase-1 gene ablation and expression. J Am Soc Nephrol 11, 965-973.

Ahn, T., and Yun, C.H. (2010). Molecular mechanisms regulating the mitochondrial targeting of microsomal cytochrome P450 enzymes. Curr Drug Metab 11, 830-838.

Aida, K., and Negishi, M. (1991). Posttranscriptional regulation of coumarin 7- hydroxylase induction by xenobiotics in mouse liver: mRNA stabilization by pyrazole. Biochemistry 30, 8041-8045.

Alam, J., Stewart, D., Touchard, C., Bionapally, S., Choi, A.M.K., and Cook, J.L. (1999). Nrf2,a Cap'n'Collar transcription factor, regulates induction of the heme oxygenase-1 gene. J Biol Chem 274, 26071-26078.

Alfadda, A.A., and Sallam, R.M. (2012). Reactive oxygen species in health and disease. J Biomed Biotechnol 2012, doi: 10.1155/2012/936486.

Arpiainen, S., Jarvenpää, S.M., Manninen, A., Viitala, P., Lang, M.A., Pelkonen, O., and Hakkola, J. (2008). Coactivator PGC-1alpha regulates the fasting inducible xenobiotic-metabolizing enzyme CYP2A5 in mouse primary hepatocytes. Toxicol Appl Pharmacol 232, 135-141.

Arruda, M.A., Rossi, A.G., de Freitas, M.S., Barja-Fidalgo, C., and Graca-Souza, A.V. (2004). Heme inhibits human neutrophil apoptosis: involvement of phosphoinositide 3-kinase, MAPK, and NF-kappaB. J Immunol 173, 2023- 2030.

Arthur, D. (2010). Bilirubin metabolism during arsenic toxicity of the liver: involvement of murine cytochrome P450 2a5 and its human orthologue CYP2A6. School of Medicine, Faculty of Health Sciences. The University of Queensland, Brisbane, Australia, pp. 104.

Arthur, D.M., Ng, J.C., Lang, M.A., and Abu-Bakar, A. (2012). Urinary excretion of bilirubin oxidative metabolites in arsenite-treated mice. J Toxicol Sci 37, 655- 661.

Atamna, H. (2004). Heme, iron, and the mitochondrial decay of ageing. Ageing Res Rev 3, 303-318.

Bae, S.H., Sung, S.H., Lee, H.E., Kang, H.T., Lee, S.K., Oh, S.Y., Woo, H.A., Kil, I.S., and Rhee, S.G. (2012). Peroxiredoxin III and sulfiredoxin together protect mice from pyrazole-induced oxidative liver injury. Antioxid Redox Signal 17, 1351-1361.

Baranano, D.E., Rao, M., Ferris, C.D., and Snyder, S.H. (2002). Biliverdin reductase: a major physiologic cytoprotectant. Proc Natl Acad Sci USA 99, 16093-16098.

Beri, R., and Chandra, R. (1993). Chemistry and biology of heme. Effect of metal salts, organometals, and metalloporphyrins on heme synthesis and catabolism, with special reference to clinical implications and interactions with cytochrome P- 450. Drug Metab Rev 25, 49-152.

Bhandary, B., Marahatta, A., Kim, H.R., and Chae, H.J. (2012). An involvement of oxidative stress in endoplasmic reticulum stress and its associated diseases Int J Mol Sci 14, 434-456.

Bindu, S., Pal, C., Dey, S., Goyal, M., Alam, A., Iqbal, M.S., Dutta, S., Sarkar, S., Kumar, R., Maity, P., and Bandyopadhyay, U. (2011). Translocation of heme oxygenase-1 to mitochondria against non-steroidal anti-inflammatory drug- induced mitochondrial oxidative stress, apoptosis, and gastric mucosal injury. J Biol Chem 286, 39387-39402.

Bondy, S.C., and Naderi, S. (1994). Contribution of hepatic cytochrome P450 systems to the generation of reactive oxygen species. Biochem Pharmacol 48, 155-159.

Bravo-Sagua, R., Rodriguez, A.E., Kuzmicic, J., Gutierrez, T., Lopez-Crisoto, C., Quiroga, C., Diaz-Elizondo, J., Chiong, M., Gillette, T.G., Rothermel, B.A., and Lavandero, S. (2013). Cell death and survival through the endoplasmic reticulum-mitochondrial axis. Curr Mol Med 13, 317-329.

Brodersen, R., and Bartels, P. (1969). Enzymatic oxidation of bilirubin. Eur J Biochem 10, 468-473.

Brouard, S., Berberat, P.O., Tobiasch, E., Seldon, M.P., Bach, F.H., and Soares, M.P. (2002). Heme oxygenase-1-derived carbon monoxide requires the activation of transcription factor NF-kappa B to protect endothelial cells from tumor necrosis factor-alpha-mediated apoptosis. J Biol Chem 277, 17950-17961.

Bulmer, A.C., Blanchfield, J.T., Toth, I., Fassett, R.G., and Bulmer Coombes, J.S. (2008a.) Improved resistance to serum oxidation in Gilbert’s syndrome: a mechanism for cardiovascular protection. Atherosclerosis 199, 390-396.

Bulmer, A.C., Ried, K., Blanchfield, J.T., and Wagner, K.H. (2008b). The anti- mutagenic properties of bile pigments. Mutat Res 658, 28-41.

Burke, M.D., and Mayer, R.T. (1974). Ethoxyresorufin: direct fluorimetric assay of a microsomal O-dealkylation which is preferentially inducible by 3- methylcholanthrene. Drug Metab Dispos 2, 583-588.

Cadenas, E., and Davies, K.J. (2000). Mitochondrial free radical generation, oxidative stress, and aging. Free Radic Biol Med 29, 222-230.

Cardenas-Vazquez, R., Yokosuka, O., and Billing, B.H. (1986). Enzymic oxidation of unconjugated bilirubin by rat liver. Biochem J 236, 625-633.

Cederbaum, A.I., and Rubin, E. (1974). Effects of pyrazole, 4-bromopyrazole and 4- methylpyrazole on mitochondrial function. Biochem Pharmacol 23, 203-213.

Chakravarthi, S., Jessop, C.E., and Bulleid, N.J. (2006). The role of glutathione in disulphide bond formation and endoplasmic-reticulum-generated oxidative stress. EMBO Rep 7, 271-275.

Chance, B., Schoener, B., Oshino, R., Itshak, F., and Nakase, Y. (1979). Oxidation- reduction ratio studies of mitochondria in freeze-trapped samples. NADH and flavoprotein fluorescence signals. J Biol Chem 254, 4764-4771.

Chen, Q., Vazquez, E.J., Moghaddas, S., Hoppel, C.L., and Lesnefsky, E.J. (2003). Production of reactive oxygen species by mitochondria: central role of complex III. J Biol Chem 278, 36027-36031.

Chin, H.J., Cho, H.J., Lee, T.W., Na, K.Y., Oh, K.H., Joo, K.W., Yoon, H.J., Kim, Y.S., Ahn, C., Han, J.S., Kim, S., Jeon, E.S., Jin, D.C., Kim, Y.L., Park, S.H., Kim, C.D., Song, Y.R., Kim, S.G., Kim, Y.G., Lee, J.E., Oh, Y.K., Lim, C.S., Lee, S.K., Chae, D.W., Cho, W.Y., Kim, H.K., Jo, S.K., and The Progressive REnal disease and Medical Informatics and gEnomic Research (PREMIER) members. (2009). The mildly elevated serum bilirubin level is negatively associated with the incidence of end stage renal disease in patients with IgA nephropathy. J Korean Med Sci 24, S22-S29.

Converso, D.P., Taille, C., Carreras, M.C., Jaitovich, A., Poderoso, J.J., and Boczkowski, J. (2006). HO-1 is located in liver mitochondria and modulates mitochondrial heme content and metabolism. FASEB J 20, 1236-1238.

Cooper, G.M., and Hausman, R.E. (2009). Bioenergetics and Metabolism: Mitochondria, Chloroplasts, and Peroxisome. In The Cell: A Molecular Approach, Fifth Edition, pp. 433-451. ASM Press, Washington, D.C. and Sinauer Associates, Inc., Sunderland.

Cuperus, F.J., Hafkamp, A.M., Hulzebos, C.V., and Verkade, H.J. (2009). Pharmacological therapies for unconjugated hyperbilirubinemia. Curr Pharm Des 15, 2927-2938.

Dalton, T.P., Shertzer, H.P., and Puga, A. (1999). Regulation of gene expression by reactive oxygen. Annu Rev Pharmacol Toxicol 39, 67-101.

Dawes, I.A. (2006). Cellular Responses to Reactive Oxygen Species. In Oxidative Stress, Disease and Cancer (K.K. Singh, Ed.), pp. 281-283. Imperial College Press, London.

De Backer, O., Elinck, E., Blanckaert, B., Leybaert, L., Motterlini, R., and Lefebvre, R.A. (2009). Water-soluble CO-releasing molecules reduce the development of postoperative ileus via modulation of MAPK/HO-1 signalling and reduction of oxidative stress. Gut 58, 347-356.

De Matteis, F., Dawson, S.J., Boobis, A.R., and Comoglio, A. (1991). Inducible bilirubin-degrading system of rat liver microsomes: role of cytochrome P450IA1. Mol Pharmacol 40, 686-691.

De Matteis, F., Dawson S.J., and Gibbs, A. (1993). Two pathways of iron-catalysed oxidation of bilirubin: Effect of desferoxamine and Trolox, and comparison with microsomal oxidation. Free Rad Biol Med 15, 301-309.

De Matteis, F., Lord, G.A., Kee Lim, C., and Pons, N. (2006). Bilirubin degradation by uncoupled cytochrome P450. Comparison with a chemical oxidation system and characterization of the products by high-performance liquid chromatography/electrospray ionization mass spectrometry. Rapid communications in mass spectrometry. Rapid Commun Mass Spectrom 20, 1209-1217.

Deniaud, A., Sharaf el dein, O., Maillier, E., Poncet, D., Kroemer, G., Lemaire, C., and Brenner, C. (2008). Endoplasmic reticulum stress induces calcium-dependent permeability transition, mitochondrial outer membrane permeabilization and apoptosis. Oncogene 27, 285-299. de Vries, S. (1986). The pathway of electron transfer in the dimeric QH2: cytochrome c oxidoreductase. J Bioenerg Biomembr 18, 195-224.

Diplock, A.T. (1994). Antioxidants and disease prevention. Mol Aspects Med 15, 293- 376.

Djoussé, L., Levy, D., Cupples, L.A., Evans, J.C., D'Agostino, R.B., and Ellison, R.C. (2001). Total serum bilirubin and risk of cardiovascular disease in the Framingham offspring study. Am J Cardiol 87, 1196-1200.

Draper, A.J., Madan, A., and Parkinson, A. (1997). Inhibition of coumarin 7-hydroxylase activity in human liver microsomes. Arch of Biochem Biophys 341, 47-61.

Droge, W. (2002). Free radicals in the physiological control of cell function. Physiol Rev 82, 47-95.

Fridovich, I. (1995). Superoxide radical and superoxide dismutases. Annu Rev Biochem 64, 97-112.

Fukui, H., and Moraes, C.T. (2008). The mitochondrial impairment, oxidative stress and neurodegeneration connection: reality or just an attractive hypothesis? Trends Neurosci 31, 251-256.

Genter, M.B., Clay, C.D., Dalton, T.P., Dong, H., Nebert, D.W., and Shertzer, H.G. (2006). Comparison of mouse hepatic mitochondrial versus microsomal cytochrome P450 following TCDD treatment. Biochem Biophysi Res Commun 342, 1375-1381.

Gil, J., and Sasiadek, M.M. (2012). Gilbert syndrome: the UGT1A1*28 promoter polymorphism as a biomarker of multifactorial diseases and drug metabolism. Biomark Med 6, 223-230.

Gilmore, W.J., Hartmann, G., Piquette-Miller, M., Marriott, J.,and Kirby, G.M. (2003). Effects of lipopolysaccharide-stimulated inflammation and pyrazole- mediated hepatocellular injury on mouse hepatic Cyp2a5 expression. Toxicology 184, 211-226.

Gilmore, W.J., and Kirby, G.M. (2004). Endoplasmic reticulum stress due to altered cellular redox status positively regulates murine hepatic CYP2A5 expression. J Pharmacol Exp Ther 308, 600-608.

Gkaliagkousi, E., and Ferro, A. (2011). Nitric oxide signalling in the regulation of cardiovascular and platelet function. Front Biosci 16, 1873-1897.

Gorlach, A., Klappa, P., and Kietzmann, T. (2006). The endoplasmic reticulum: folding, calcium homeostasis, signaling, and redox control. Antioxid Redox Signal 8, 1391-1418.

Gozzelino, R., Jeney, V., and Soares, M.P. (2010). Mechanisms of cell protection by heme oxygenase-1. Annu Rev Pharmacol Toxicol 50, 323-354.

Griendling, K.K., Sorescu, D., and Ushio-Fukai, M. (2000). NAD(P)H oxidase: role in cardiovascular biology and disease . Circ Res 86, 494–501.

Gunter, T.E., Yule, D.I., Gunter, K.K., Eliseev, R.A., and Salter, J.D. (2004). Calcium and mitochondria. FEBS Lett 567, 96-102.

Habeebu, S.S., Liu, J., and Klaassen, C.D. (1998). Cadmium-induced apoptosis in mouse liver. Toxicol Appl Pharmacol 149, 203-209.

Han, S.S., Na, K.Y., Chae, D.W., Kim, Y.S., Kim, S., and Chin, H.J. (2010) High serum bilirubin is associated with the reduced risk of diabetes mellitus and diabetic nephropathy. Tohoku J Exp Med 221, 133-140.

Hansen, T.W., Allen, J.W., and Tommarello, S. (1999). Oxidation of bilirubin in the brain-further characterization of a potentially protective mechanism. Mol Genet Metab 68, 404-409.

Hansson, M.J., Mansson, R., Morota, S., Uchino, H., Kallur, T., Sumi, T., Ishii, N., Shimazu, M., Keep, M.F., Jegorov, A., and Elmer, E. (2008). Calcium- induced generation of reactive oxygen species in brain mitochondria is mediated by permeability transition. Free Rad Biol Med 45, 284-294.

Hanukoglu, I. (1996). Electron transfer proteins of cytochrome P450 systems. Adv Mol Cell Biol 14, 29-56.

Hata, A., Tsunoo, H., Nakajima, H., Shintaku, K., and Kimura, M. (1980). Acute cadmium intoxication in inbred mice: a study on strain differences. Chem Biol Interact 32, 29-39.

Heyman, E., Ohlsson, A., and Girschek, P. (1989). Retinopathy of prematurity and bilirubin. N Engl J Med 320, 256.

Honkakoski, P., Kojo, A., Raunio, H., Pasanen, M., Juvonen, R., and Lang, M.A. (1988). Hepatic mitochondrial coumarin 7-hydroxylase: comparison with the microsomal enzyme. Arch Biochem Biophys 267, 558-567.

Horsfall, L.J., Rait, G., Walters, K., Swallow, D.M., Pereira, S.P., Nazareth, I., and Petersen, I. (2011). Serum bilirubin and risk of respiratory disease and death. JAMA 305, 691-697.

Ihara, H., Hashizume, N., Shimizu, N., Aoki, T. (1999). Threshold concentration of unbound bilirubin to induce neurological deficits in a patient with type I Crigler–Najjar syndrome. Ann Clin Biochem 36, 347-352.

Jansen, T., and Daiber, A. (2012). Direct antioxidant properties of bilirubin and biliverdin. Is there a role for biliverdin reductase? Front Pharmacol 3, doi: 10.3389/fphar.2012.00030.

Jansen, T., Hortmann, T., Oelze, M., Opitz, B., Steven, S., Schell, R., Knorr, M., Karbach, S., Schuhmacher, S., Wenzel, P., Münzel, T., and Daiber, A. (2010). Conversion of biliverdin to bilirubin by biliverdin reductase contributes to endothelial cell protection by heme oxygenase-1-evidence for direct and indirect antioxidant actions of bilirubin J Mol Cell Cardiol 49, 186-195.

Jezek, P., and Hlavata, L. (2005). Mitochondria in homeostasis of reactive oxygen species in cell, tissues, and organism. Int J Biochem Cell Biol 37, 2478-2503.

Jiraskova, A., Novotny, J., Novotny, L., Vodicka, P., Pardini, B., Naccarati, A., Schwertner, H.A., Hubacek, J.A., Puncocharova, L., Smerhovsky, Z., and Vitek, L., (2012). Association of serum bilirubin and promoter variations in HMOX1 and UGT1A1 genes with sporadic colorectal cancer. Int J Cancer 131, 1549-1555.

Josephy, P.D. (1997). Cytochrome P450. In Molecular Toxicology. pp. 209-247. Oxford University Press Inc, New York.

Juvonen, R.O., Kaipainen, P.K., and Lang, M.A. (1985). Selective induction of coumarin 7-hydroxylase by pyrazole in D2 mice. Eur J Biochem 152, 3-8.

Kamisako, T., Kobayashi, Y., Takeuchi, K., Ishihara, T., Higuchi, K., Tanaka, Y., Gabazza, E.C., and Adachi, Y. (2000). Recent advances in bilirubin metabolism research: the molecular mechanism of hepatocyte bilirubin transport and its clinical relevance. J Gastroenterol 35, 659-664.

Kang, K.W., Lee, S.J., and Kim, S.G. (2005). Molecular mechanism of nrf2 activation by oxidative stress. Antioxid Redox Signal 7, 1664-1673.

Kaplan, M., and Hammerman, C. (2011). Neonatal screening for glucose-6-phosphate dehydrogenase deficiency: biochemical versus genetic technologies. Semin Perinatol 35, 156-161.

Kershaw, W.C., and Klaasen, C.D. (1991). Cadmium-induced elevation of hepatic isometallothionein concentrations in inbred strains of mice. Chem Biol Interac 78, 269-282.

Kidd, P.M. (2005). Neurodegeneration from mitochondrial insufficiency: nutrients, stem cells, growth factors, and prospects for brain rebuilding using integrative management. Altern Med Rev 10, 268-293.

Kim, S.D., Antenos, M., Squires, E.J., and Kirby, G.M. (2013). Cytochrome P450 2A5 and bilirubin: mechanisms of gene regulation and cytoprotection. Toxicol Appl Pharmacol 270, 129-138.

Kirby, G.M., Nichols, K.D., and Antenos, M. (2011). CYP2A5 induction and hepatocellular stress: an adaptive response to perturbations of heme homeostasis. Curr Drug Metab 12, 186-197.

Kinonen, T., Pasanen, M., Gynther, J., Poso, A., Jarvinen, T., Alhava, E., and Juvonen, R. (1995). Competitive inhibition of coumarin 7-hydroxylation by pilocarpine and its interaction with mouse CYP2As and human CYP2A6. Br J Pharmacol 116, 2625-2630.

Kobayashi, A., Takahashi, T., Sugai, S., Miyakawa, Y., Iwatsuka, H., and Yamaguchi, T. (2003). Urinary excretion of oxidative metabolites of bilirubin in fenofibrate- treated rats. J Toxicol Sci 28, 71-75.

Kowaltowski, A.J., de Souza-Pinto, N.C., Castilho, R.F., and Vercesi, A.E. (2009). Mitochondria and reactive oxygen species. Free Rad Biol Med 47, 333-343.

Kukreja, R.C., Kontos, H.A., Hess, M.L., and Ellis, E.F. (1986). PGH synthase and lipoxygenase generate superoxide in the presence of NADH or NADPH. Circ Res 59, 612–619.

Kunikata, T., Itoh, S., Ozaki, T., Kondo, M., Isobe, K., and Onishi, S. (2000). Formation of propentdyopents and biliverdin, oxidized metabolites of bilirubin, in infants receiving oxygen therapy. Pediatr Int 42, 331-336.

Lambert, A.J., and Brand, M.D. (2004). Superoxide production by NADH:ubiquinone oxidoreductase (complex I) depends on the pH gradient across the mitochondrial inner membrane. Biochem J 382, 511–517.

Lämsä, V. (2012). Regulation of murine hepatic cytochrome P450 2a5 expression by transcription factor Nuclear factor (erythroid-derived 2)-like factor. Institute of Biomedicine, Faculty of Medicine. University of Oulu, Finland, pp.86.

Lämsä, V., Levonen, A.L., Leinonen, H., Yla-Herttuala, S., Yamamoto, M., and Hakkola, J. (2010). Cytochrome P450 2A5 constitutive expression and induction by heavy metals is dependent on redox-sensitive transcription factor Nrf2 in liver. Chem Res Toxicol 23, 977-985.

Lämsä, V., Levonen, A.L., Sormunen, R., Yamamoto, M., and Hakkola, J. (2012). Heme and heme biosynthesis intermediates induce heme oxygenase-1 and cytochrome P450 2A5, enzymes with putative sequential roles in heme and bilirubin metabolism: different requirement for transcription factor Nuclear factor erythroid-derived 2-like 2. Toxicol Sci 130, 132-144.

Lang, M.A., Juvonen, R., Jarvinen, P., Honkakoski, P., and Raunio, H. (1989). Mouse liver P450Coh: genetic regulation of the pyrazole-inducible enzyme and comparison with other P450 isoenzymes. Arch Biochem Biophys 271, 139- 148.

Leem, J., and Koh, E.H. (2012). Interaction between mitochondria and the endoplasmic reticulum: implications for the pathogenesis of type 2 diabetes mellitus. Exp Diabetes Res 2012, doi: 10.1155/2012/242984.

Leibovitz, B.E., and Siegel, B.V. (1980). Aspects of free radical reactions in biological systems: aging. J Gerontol 35, 45-56.

Levi, S., and Rovida, E. (2009). The role of iron in mitochondrial function. Biochim Biophys Acta 1790, 629-636.

Lin, J.P., O'Donnell, C.J., Schwaiger, J.P., Cupples, L.A., Lingenhel, A., Hunt, S.C., Yang, S., and Kronenberg, F. (2006). Association between the UGT1A1*28 allele, bilirubin levels, and coronary heart disease in the Framingham Heart Study. Circulation 114, 1476-1481.

Liu, C., and Lin, J.D. (2011). PGC-1 coactivators in the control of energy metabolism. Acta Biochim Biophys Sin (Shanghai) 43, 248-257.

Liu, J., Kershaw, W.C., Liu, Y.P., and Klaassen, C.D. (1992). Cadmium-induced hepatic endothelial cell injury in inbred strains of mice. Toxicology 75, 51-62.

Liu, Y., Fiskum, G., and Schubert, D. (2002). Generation of reactive oxygen species by the mitochondrial electron transport chain. J Neurochem 80, 780-787.

Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. (1951). Protein measurement with the Folin phenol reagent. J Biol Chem 193, 265-275.

Lu, C.F., Wang, Y.M., Peng, S.Q., Zou, L.B., Tan, D.H., Liu, G., Fu, Z., Wang, Q.X., and Zhao, J. (2009). Combined effects of repeated administration of 2,3,7,8- tetrachlorodibenzo-p-dioxin and polychlorinated biphenyls on kidneys of male rats. Arch Environ Contam Toxicol 57, 767-776.

Lu, Y., Gong, P., and Cederbaum, A.I. (2008). Pyrazole induced oxidative liver injury independent of CYP2E1/2A5 induction due to Nrf2 deficiency. Toxicology 252, 9-16.

Maines, M.D. (1997). The heme oxygenase system: a regulator of second messenger gases. Ann Rev Pharmacol Toxicol 37, 517-554.

Maines, M.D. (2005). The heme oxygenase system: update 2005. Antioxid Redox Signal 7, 1761-1766.

Malhotra, J.D., and Kaufman, R.J. (2007). Endoplasmic reticulum stress and oxidative stress: a vicious cycle or a double-edged sword? Antioxid Redox Signal 9, 2277-2293.

Malik, S.G., Irwanto, K.A., Ostrow, J.D., and Tiribelli, C. (2010). Effect of bilirubin on cytochrome c oxidase activity of mitochondria from mouse brain and liver. BMC Res Notes 3, doi: 10.1186/1756-0500-3-162.

Mari, M., Morales, A., Colell, A., Garcia-Ruiz, C., and Fernandez-Checa, J.C. (2009). Mitochondrial glutathione, a key survival antioxidant. Antioxid Redox Signal 11, 2685-2700.

McDonagh, A.F. (2004). Bilirubin the beneficient. Pediatrics 114, 1741-1742.

McDonagh, A.F. (2010). Controversies in bilirubin biochemistry and their clinical relevance. Semin Fetal Neonatal Med 15, 141-147.

Meister, A. (1992). On the antioxidant effects of ascorbic acid and glutathione. Biochem Pharmacol 44, 1905-1915.

Micheau, O., Tschopp, J. (2003). Induction of TNF receptor I-mediated apoptosis via two sequential signaling complexes. Cell 114, 181-190.

Muller, F.L., Liu, Y., and Van Remmen, H. (2004). Complex III releases superoxide to both sides of the inner mitochondrial membrane. J Biol Chem 279, 49064- 49073.

Nadege, B., Patrick, L., and Rodrgue, R. (2009). Mitochondria: from bioenergetics to the metabolic regulation of carcinogenesis. Front Biosci 14, 4015-4034.

Ndisang, J.F., Tabien, H.E., and Wang, R. (2004). Carbon monoxide and hypertension. J Hypertens 22, 1057-1074.

Nebert, D.W., and Russell, D.W. (2002). Clinical importance of the cytochromes P450. Lancet 360, 1155-1162.

Nichols,K.D., and Kirby, G.M. (2008). Microarray analysis of hepatic gene expression in pyrazole-mediated hepatotoxicity: identification of potential stimuli of Cyp2a5 induction. Biochem Pharmacol 75, 538-551.

Nohl, H., Gille, L., and Staniek, K. (2005). Intracellular generation of reactive oxygen species by mitochondria. Biochem Pharmacol 69, 719-723.

Okado-Matsumoto, A., and Fridovich, I. (2001). Subcellular distribution of superoxide dismutases (SOD) in rat liver: Cu, Zn-SOD in mitochondria. J Biol Chem 276, 38388-38393.

Parkinson, A. (2001). Biotransformation of xenobiotics. In Casarette & Doull's Toxicology: The Basic Science of Poison. (C.D. Klaassen, Ed.), pp.133-224. McGraw-Hill, Inc., Kansas.

Piantadosi, C.A. (2008). Carbon monoxide, reactive oxygen signaling, and oxidative stress. Free Rad Biol Med 45, 562-569.

Pons, N., Pipino, S., and De Matteis, F. (2003). Interaction of polyhalogenated compounds of appropriate configuration with mammalian or bacterial CYP enzymes. Increased bilirubin and uroporphyrinogen oxidation in vitro. Biochem Pharmacol 66, 405-414.

Poss, K.D., and Tonegawa, S. (1997). Reduced stress defense in heme oxygenase 1- deficient cells. Proc Natl Acad Sci USA 94, 10925-10930.

Pou, S., Pou, W.S., Bredt, D.S., Snyder, S.H., and Rosen, G.M. (1992). Generation of superoxide by purified brain nitric oxide synthase. J Biol Chem 267, 24173- 24176.

Prousek, J. (2007). Fenton chemistry in biology and medicine. Pure Appl Chem 79, 2235-2338.

Queiroga, C.S., Almeida, A.S., Alves, P.M., Brenner, C., and Vieira, H.L. (2011). Carbon monoxide prevents hepatic mitochondrial membrane permeabilization. BMC Cell Biol 12 , doi: 10.1186/1471-2121-12-10.

Radominska-Pandya, A., Ouzzine, M., Fournel-Gigleux, S., and Magdalou, J. (2005). Structure of UDP-glucuronosyltransferases in membranes. Methods Enzymol 400, 116-147.

Ricciotti, E., and FitzGerald, G.A. (2011). Prostaglandins and inflammation. Arterioscler Thromb Vasc Biol 31, 986-1000.

Rigato, I., Ostrow, J.D., and Tiribelli, C. (2005). Bilirubin and the risk of common non- hepatic diseases. Trends Mol Med 11, 277-283.

Rodrigues, C.M., Sola, S., Brito, M.A., Brites, D., and Moura, J.J. (2002). Bilirubin directly disrupts membrane lipid polarity and fluidity, protein order, and redox status in rat mitochondria. J Hepatol 36, 335-341.

Ryter, S.W., Kvam, E., and Tyrrell, R.M. (2000). Heme oxygenase activity. Current methods and applications. Methods Mol Biol 99, 369-391.

Ryter, S.W., and Tyrrell, R.M. (2000). The heme synthesis and degradation pathways: role in oxidant sensitivity. Heme oxygenase has both pro- and antioxidant properties. Free Rad Biol Med 28, 289-309.

Sanchez-Alcazar, J.A., Ault, J.G., Khodjakov, A., and Schneider, E. (2000). Increased mitochondrial cytochrome c levels and mitochondrial hyperpolarization precede camptothecin-induced apoptosis in Jurkat cells. Cell Death Diff 7, 1090-1100.

Sangar, M.C., Bansal, S., and Avadhani, N.G. (2010). Bimodal targeting of microsomal cytochrome P450s to mitochondria: implications in drug metabolism and toxicity. Expert Opin Drug Metab Toxicol 6, 1231-1251.

Schlezinger, J.J., and Stegeman, J.J. (2001). Induction and suppression of cytochrome P450 1A by 3,3',4,4',5-pentachlorobiphenyl and its relationship to oxidative stress in the marine fish scup (Stenotomus chrysops). Aquat Toxicol 52, 101- 115.

Schlezinger, J.J., Strunz, W.D.J., Goldstone, J.V., and Stegeman, J.J. (2006). Uncoupling of cytochrome P450 1A and stimulation of reactive oxygen species production by co-planar polychlorinated biphenyl congeners. Aquat Toxicol 77, 422-432.

Schlezinger, J.J., White, R.D., and Stegeman, J.J. (1999). Oxidative inactivation of cytochrome P-450 1A (CYP1A) stimulated by 3,3',4,4'-tetrachlorobiphenyl: production of reactive oxygen by vertebrate CYP1As. Mol Pharmacol 56, 588-597.

Schwertner, H.A., Jackson, W.G., and Tolan, G. (1994). Association of low serum concentration of bilirubin with increased risk of coronary artery disease. Clin Chem 40, 18-23.

Sedlak, T.W., Saleh, M., Higginson, D.S., Paul, B.D., Juluri, K.R., and Snyder, S.H. (2009). Bilirubin and glutathione have complementary antioxidant and cytoprotective roles. Proc Natl Acad Sci USA 106, 5171-5176.

Sedlak, T.W., and Snyder, S.H. (2004). Bilirubin benefits: cellular protection by a biliverdin reductase antioxidant cycle. Pediatrics 113, 1776-1782.

Sena, L.A., and Chandel, N.S. (2012). Physiological roles of mitochondrial reactive oxygen species. Mol Cell 48, 158-167.

Sendelbach, L.E., Kershaw, W.C., Cuppage, F., and Klaassen, C.D. (1992). Cd- metallothionein nephrotoxicity in inbred strains of mice. J Toxicol Environ Health 35, 115-126.

Seo, A.Y., Joseph, A.M., Dutta, D., Hwang, J.C., Aris, J.P., and Leeuwenburgh, C. (2010). New insights into the role of mitochondria in aging: mitochondrial dynamics and more. J Cell Sci 123, 2533-2542.

Seubert, J.M., Webb, C.D., and Bend, J.R. (2002). Acute sodium arsenite treatment induces Cyp2a5 but not Cyp1a1 in the C57B1/6 mouse in a tissue (kidney) selective manner. J Biochem Mol Toxicol 16, 96-106.

Shaikh, Z.A., Jordan, S.A., and Tewari, P.C. (1993). Cadmium disposition and metallothionein induction in mice: strain-, sex-, age- and dose-dependent differences. Toxicology 80, 51-70.

Shertzer, H.G., Clay, C.D., Genter, M.B., Chames, M.C., Schneider, S.N., Oakley, G.G., Nebert, D.W., and Dalton, T.P. (2004). Uncoupling-mediated generation of reactive oxygen by halogenated aromatic hydrocarbons in mouse liver microsomes. Free Radic Biol Med 36, 618-631.

Shi, J., Aisaki, K., Ikawa, Y., and Wake, K. (1998). Evidence of hepatocyte apoptosis in rat liver after the administration of carbon tetrachloride. Am J Pathol 153, 515-525.

Srinivasan, S., and Avadhani, N.G. (2012). Cytochrome c oxidase dysfunction in oxidative stress. Free Radic Biol Med 53, 1252-1263.

Srivastava, P., and Pandey, V.C., (1996). Mitochondrial heme oxygenase of Mastomys coucha. Int J Biochem Cell Biol 28, 1071-1077.

Starkov, A., and Wallace, K.B. (2006). Yin and yang of mitochondrial ROS. In Oxidative Stress, Disease and Cancer (K.K. Singh, Ed.), pp. 1-7. Imperial College Press, London.

Stocker, R. (2004). Antioxidant activities of bile pigments. Antioxid Redox Signal 6, 841- 849.

Stocker, R., Glazer, A.N., and Ames, B.N. (1987a). Antioxidant activity of albumin- bound bilirubin. Proc Natl Acad Sci USA 84, 5918-5922.

Stocker, R., Yamamoto, Y., McDonagh, A.F., Glazer, A.N., and Ames, B.N. (1987b). Bilirubin is an antioxidant of possible physiological importance. Science 235, 1043-1046.

Stocker, R., and Peterhans, E. (1989). Antioxidant properties of conjugated bilirubin and biliverdin: biologically relevant scavenging of hypochlorous acid. Free Rad Res Commun 6, 57-66.

Sugioka, K., Nakano, M., Totsune-Nakano, H., Minakami, H., Tero-Kubota, S., and Ikegami, Y. (1988). Mechanism of O2- generation in reduction and oxidation cycle of ubiquinones in a model of mitochondrial electron transport systems. Biochim Biophys Acta 936, 377-385.

Sullivan, P.G., and Brown, M.R. (2005). Mitochondrial aging and dysfunction in Alzheimer's disease. Prog Neuropsychopharmacol Biol Psychiatry 29, 407- 410.

Tell, G., and Gustincich, S. (2009). Redox state, oxidative stress, and molecular mechanisms of protective and toxic effects of bilirubin on cells. Curr Pharm Des 15, 2908-2914.

Temme, E.H., Zhang, J., Schouten, E.G., and Kesteloot, H. (2001). Serum bilirubin and 10-year mortality risk in a Belgian population. Cancer Causes Control 12, 887-894.

Tenhunen, R., Ross, M.E., Marver, H.S., and Schmid, R. (1970). Reduced nicotinamide- adenine dinucleotide phosphate dependent biliverdin reductase: partial purification and characterization. Biochemistry 9, 298-303.

Tilak, J.C., and Devasagayam, T.P.A. (2006). Oxidative damage to mitochondria. In Oxidative Stress, Disease and Cancer (K.K. Singh, Ed.), pp. 91-94. Imperial College Press, London.

Tomaro, M.L., and Batlle, A.M. (2002). Bilirubin: its role in cytoprotection against oxidative stress. Int J Biochem Cell Biol 34, 216-220.

Toyokuni, S. (1996). Iron-induced carcinogenesis: the role of redox regulation. Free Radic Biol Med 20, 553-566.

Tsai, H.L., Chang, S.K., and Chang, S.J. (2007). Antioxidant content and free radical scavenging ability of fresh red pummelo [Citrus grandis (L.) Osbeck] juice and freeze-dried products. J Agric Food Chem 55, 2867-2872.

Tukey, R.H., and Strassburg, C.P. (2000). Human UDP-glucuronosyltransferases: metabolism, expression, and disease. Annu Rev Pharmacol Toxicol 40, 581- 616.

Turrens, J.F. (2003). Mitochondrial formation of reactive oxygen species. J Physiol 552, 335-344.

Turrens, J.F., Alexandre, A., and Lehninger, A.L. (1985). Ubisemiquinone is the electron donor for superoxide formation by complex III of heart mitochondria. Arch Biochem Biophys 237, 408-414.

Turrens, J.F., and Boveris, A. (1980). Generation of superoxide anion by the NADH dehydrogenase of bovine heart mitochondria. Biochem J 191, 421-427.

Valko, M., Leibfritz, D., Moncol, J., Cronin, M.T., Mazur, M., and Telser, J. (2007). Free radicals and antioxidants in normal physiological functions and human disease. Int J Biochem Cell Biol 39, 44-84.

Vitek, L., Jirsa, M., Brodanova, M., Kalab, M., Marecek, Z., Danzig, V., Novotny, L., and Kotal, P. (2002). Gilbert syndrome and ischemic heart disease: a protective effect of elevated bilirubin levels. Atherosclerosis 160, 449-456.

Vitek, L., and Ostrow, J.D. (2009). Bilirubin chemistry and metabolism; harmful and protective aspects. Curr Pharm Des 15, 2869-2883.

Vitek, L., and Schwertner, H.A. (2008). Protective effects of serum bilirubin on peripheral vascular disease. Ann Hepatol 7, 94-95.

Wegiel, B., Chin, B.Y., and Otterbein, L.E. (2008). Inhale to survive, cycle or die? Carbon monoxide and cellular proliferation. Cell Cycle 7, 1379-1384.

Weisiger, R.A., and Fridovich, I. (1973). Mitochondrial superoxide simutase. Site of synthesis and intramitochondrial localization. J Biol Chem 248, 4793-4796.

Willcox, J.K., Ash, S.L., and Catignani, G.L. (2004). Antioxidants and prevention of chronic disease. Crit Rev Food Sci Nutr 44, 275-295.

Wood, A.W., and Conney, A.H. (1974). Genetic variation in coumarin hydroxylase activity in the mouse (Mus musculus). Science 185, 612-614.

Wood, A.W., and Taylor, B.A. (1979). Genetic regulation of coumarin hydroxylase activity in mice: evidence for single locus controlon chromosome 7. J Biol Chem 254, 5647-5651.

Wu, S.B ., and Wei, Y.H. (2012). AMPK-mediated increase of glycolysis as an adaptive response to oxidative stress in human cells: implication of the cell survival in mitochondrial diseases. Biochim Biophys Acta 1822, 233-247.

Yachie, A., Niida, Y., Wada, T., Igarashi, N., Kaneda, H., Toma, T., Ohta, K., Kasahara, Y., and Koizumi, S. (1999). Oxidative stress causes enhanced endothelial cell injury in human heme oxygenase-1 deficiency. J Clin Invest 103, 129-135.

Yamaguchi, T., Horio, F., Hashizume, T., Tanaka, M., Ikeda, S., Kakinuma, A., and Nakajima, H. (1995). Bilirubin is oxidized in rats treated with endotoxin and acts as a physiological anti-oxidant synergistically with ascorbic acid in vivo. Biochem Biophys Res Commun 214, 11-19.

Yamaguchi, T., Shioji, I., Sugimoto, A., Komoda, Y., and Nakajima, H. (2002). Physiological stress increases bilirubin metabolites in human urine. Biochem Biophys Res Commun 293, 517-520.

Yang, Y.M., Han, C.Y., Kim, Y.J., and Kim, S.G. (2010). AMPK-associated signaling to bridge the gap between fuel metabolism and hepatocyte viability. World J Gastroenterol 16, 3731-3742.

Yang, M., Kabulski, J.L., Wollenberg, L., Chen, X., Subramanian, M., Tracy, T.S., Lederman, D., Gannett, P.M., and Wu, N. (2009). Electrocatalytic drug metabolism by CYP2C9 bonded to a self-assembled monolayer-modified electrode. Drug Metab Dispos 37, 892-899.

Yokosuka, O., and Billing, B. (1987). Enzymatic oxidation of bilirubin by intestinal mucosa. Biochim Biophys Acta 923, 268-274.

Zaccaro, C., Sweitzer, S., Pipino, S., Gorman, N., Sinclair, P.R., Sinclair, J.F., Nebert, D.W., and De Matteis, F. (2001). Role of cytochrome P450 1A2 in bilirubin degradation Studies in Cyp1a2 (-/-) mutant mice. Biochem Pharmacol 61, 843-849.

Zangar, R.C., Davydov, D.R., and Verma, S. (2004). Mechanisms that regulate production of reactive oxygen species by cytochrome P450. Toxicol Appl Pharmacol 199, 316-331.

Zhao, B.L., Li, X.J., Liu, G.T., Jia, W.Y., and Xin, W.J. (1990). Scavenging effect of schizandrins on active oxygen radicals. Cell Biol Int Rep 14, 99-109. 81

Zorov, D.B., Filburn, C.R., Klotz, L.O., Zweier, J.L., and Sollott, S.J. (2000). Reactive oxygen species (ROS)-induced ROS release: a new phenomenon accompanying induction of the mitochondrial permeability transition in cardiac myocytes. J Exp Med 192, 1001-1014.

Zorov, D.B., Juhaszova, M., and Sollott, S.J. (2006). Mitochondrial ROS-induced ROS release: an update and review. Biochim Biophys Acta 1757, 509-517.

8. APPENDICES

Appendix 1

List of Chemicals, Reagents and Instruments

Chemical / Reagent Manufacturer / Supplier

Pyrazole, coumarin, umbelliferone (7- Sigma-Aldrich, Sydney, Australia hydroxycoumarin),reduced nicotinamide adenine dinucleotide phosphate (β-NADPH), ascorbic acid, glycerol, glycine, mannitol, leupeptin, Hepes, phenylmethylsulfonyl-fluoride (PMSF), digitonin, bovine serum albumin (BSA), ethylenediamine tetraacetic acid (EDTA), DMSO, hemin, Glucose-6-phosphate (G6P),Glucose-phosphate dehydrogenase (G6PDH),Tween 20, sucrose

Bilirubin and biliverdin Frontier Scientific, Inc. ,Utah, USA Dithiothreitol (DTT) Applichem, Germany

Goat anti-rabbit, goat anti-mouse, Thermo Fisher Scientific Inc., Victoria, Australia mouse anti-goat and rabbit anti-

chicken secondary antibodies

conjugated with horseradish-

peroxidase

Mouse monoclonal anti-HMOX1, Abcam, Australia rabbit polyclonal anti-BVR and anti- cytochrome C antibodies

Goat polyclonal anti-UGT1A1 Santa Cruz Biotechnology, INC antibody

Rabbit polyclonal anti-COX IV Cell Signalling Technology, USA antibody

Rabbit polyclonal anti-NPR antibody Assay Designs, New York, USA

Chicken anti-CYP2A5 monoclonal Kind gift from Dr Risto Juvonen, University of Kuopio, Finland antibody

Anti-AdxR antibody Kind gift from Dr Alexey Kluchenovich, National Academic

of Sciences, Republic of Belarus

Instruments Manufacturer / Supplier

Potter-Elvehjem Homogeniser Sartorius, Goettingen, Germany

HIMAC CP80 WX Preparative Hitachi Koki Co. Ltd., Japan Ultracentrifuge

Allegra X-15R Centrifuge Beckman Coulter Inc, USA

Fluorescence Spectrometer LS55 Perkin Elmer Ltd., Beaconsfield, Buckinghamshire, England

xMarkTM Microplate Bio-Rad Laboratories Pty. Spectrophotometer Ltd.,Sydney Australia

Versadoc™ Molecular Imaging Bio-Rad Laboratories Pty. Ltd., System Sydney Australia

QTrap 5500 mass spectrometer AB Sciex, USA equipped with an electrospray (Turbo Shimadzu Corp., Kyoto, Japan V) interface coupled to a HPLC system (Nexera)

Cary Series UV-Vis Agilent Technologies, Australia Spectrophotometer

Appendix 2

Preparation of Liver Microsomal, Cytosolic and Mitoplasts Fractions

Fresh livers pooled from 2 mice (about 3 g) were homogenised in isolation buffer (2 mM Hepes, 70 mM sucrose, 220 mM mannitol, 2 mM EDTA, 1 mM DTT, 0.5 mM PMSF, 0.01 µg/µl leupeptin) using a motor-driven Potter-Elvehjem homogeniser (Sartorius, Goettingen, Germany). The homogenate was centrifuged at 2,000 x g for 10 min. The supernatant was aspirated into a clean centrifuge tube and further centrifuged at 14,000 x g for 15 min. The supernatant (post-mitochondrial fraction) was reserved to prepare microsomes. The mitochondrial pellet was washed by resuspending in isolation buffer and centrifuged again at 14,000 x g for 15 min. Resulting pellet was resuspended in isolation buffer, underlay with 27% sucrose using a 20-gauge blunt-end needle and centrifuged for 15 min at 14,000 x g. The pellet was washed and resuspended in buffer. An appropriate amount of 0.13 mg digitonin / mg protein was added to the mitochondrial suspension to remove outer mitochondrial membrane, thus producing mitoplasts. This mixture was incubated on ice with occasional agitation for 15 min. The reaction was stopped by resuspending the mixture with isolation buffer and centrifuged at 12,000 x g for 15 min. The resulting pellet was suspended in the buffer and centrifuged again for 15 min at 12,000 x g. The mitoplasts was resuspended in dilution buffer (2 mM Tris-HCl, pH 7.4, 1 mM EDTA, 20% glycerol) and sonicated (4 x 30 seconds with 30 seconds intervening cooling times).

Meanwhile, the post mitochondrial supernatant was centrifuged for 1 h at 100,000 x g. The supernatant (cytosolic fraction) was removed and the pellet was resuspended in 3 ml to 5 ml of buffer and centrifuged again for 1 h at 100,000 x g. The microsomal pellet was collected and resuspended in dilution buffer. All the centrifugation procedures were done at 4oC. All samples were aliquoted and kept at - 80oC.

Appendix 3

Coumarin Hydroxylase Activity

Briefly, 300 μl of sample buffer [160 mM Potassium phosphate buffer (pH 7.4),

8 mM MgCl2, 0.16 mM substrate coumarin] and 100 μl of microsomal / mitoplasts samples (containing 50 μg of protein) were placed in a glass tube. Reactions were started by the addition of 100 μl 7.5 mM NADPH and incubated with shaking for 20 min at 37oC. Reactions were stopped by the addition of 500 μl of 6.5% trichloroacetic acid (TCA) and centrifuged for 5 min at 2800 rpm. An aliquot of the supernatant (500 μl) was removed to a 2 ml quart cuvette and mixed thoroughly with 2 ml 1.6 M glycine / NaOH, pH 10.3. Hydroxycoumarin formation was then determined spectrofluorometrically at excitation wavelength of 390 nm and mission wavelength of 440 nm. A linear standard curve was achieved when the slit widths of the spectrofluorometer were set at 2.5 nm and 5.0 nm for excitation and emission, respectively.

Appendix 4

Western Immunoblotting

Liver microsomal and mitoplasts proteins were separated by electrophoresis through a 4-12% Mini-Protean® TGX™ (Bio-Rad Laboratories Pty. Ltd., Sydney Australia), under denaturing condition and blotted overnight onto 0.2 µm PVDF membranes (Bio-Rad Laboratories Pty. Ltd., Sydney Australia). The blots were blocked in Tris-buffered saline (TBS) containing 0.1% Tween 20 and 5% non-fat milk for at least 6 hours at room temperature. Immunoblot analysis was performed by incubating with anti-HMOX1 antibody, anti-CYP2A5 antibody, anti-UGT1A1 antibody, anti-NPR antibody, anti-COX IV, anti-BVR antibody and anti-cytochrome C antibody in dilutions of 1:250, 1:1000, 1:500, 1:5000, 1:1000, 1:5000, 1:200 respectively. All solutions contained 5% non-fat dairy milk. Blots were then incubated for 2 hours with horseradish- peroxidase-conjugated secondary antibodies at room temperature. Detection was performed with commercial chemoluminescence reagents (Bio-Rad Laboratories Pty. Ltd., Sydney Australia). The intensity of each band was detected using Versadoc™ Molecular Imaging System. Finally, for the densitometric analysis, quantification was conducted using the software Image Lab™ 3.0 (Bio-Rad Laboratories Pty. Ltd., Sydney Australia).

Appendix 5

HPLC/MS-MS conditions used to identify bilirubin and its oxidative metabolites

Transitions Declustering Collision Retention monitored (%) potential Energy time (mins) Bilirubin 585.5/299.2 (100) 25 9.33 to 75 585.5/271.2 (10) 60 10.37 Biliverdin 583.5/297.2 (100) 48 5.8 to 110 583.5/209.2 (20) 83 6.16 Metabolite 301.1/256.1 (100) 27 2.54 to m/z 301 301.1/242.1 (30) 75 27 3.1 301.1/228.1 (25) 37 Metabolite 315.1/287.1 (100) 30 5.52 to m/z 315 75 315.1/174.1 (60) 32 5.8 Metabolite 333.1/174.1 (100) 30 3.34 to m/z 333 333.1/231.1 (70) 75 25 3.7 333.1/191.1 (70) 30

Appendix 6

HPLC/MS-MS chromatograms of bilirubin, biliverdin and various dipyrroles. a) bilirubin; ions m/z 585 and biliverdin; ions m/z 583 b) dipyrroles: ions m/z 301; ions m/z 333; ions m/z 315 detected in order of their elution time.

Appendix 7

Effect of ascorbic acid on Fenton oxidation of bilirubin (BR)

Aim: To determine the concentration of ascorbic acid that inhibits ROS in Fenton- driven bilirubin (BR) oxidation

Experiment

To test whether or not ascorbic acid can scavenge ROS that drive the chemical oxidation of BR, increasing concentrations of ascorbic acid (ranging from 2mM to

200mM) were added to the incubation mixtures that comprise of 8 μmol H2O2, 0.26

µmol FeCl3 and 10 µM BR. BR oxidation rate was calculated by monitoring absorbance at wavelength 440nm for each minute for 20 minutes.

Results and Discussion

Fe-H2O2 (Fenton reagent) is an efficient oxidant for various organic substrates

(Prousek, 2007). The oxidation of BR catalysed by H2O2 has been demonstrated previously (De Matteis et al., 1993; De Matteis et al., 2006). Ascorbic acid is a known scavenger of ROS (Tsai et al., 2007; Zhao et al., 1990). In the Fenton reaction of H2O2,

0.26 µmol FeCL3 8 µmol H2O2 was used to produce ROS as below:

2+ 3+ - Fe + H2O2 → Fe + ·OH + OH

3+ 2+ + Fe + H2O2 → Fe + ·OOH + H

These ROS, in particular ·OH and ·O2, are involved in the fractionation of BR to its smaller metabolites, dipyrroles (see Figure 1). At lower concentration, ascorbic acid did not seem to inhibit BR disappearance. However, addition of 200mM ascorbic acid effectively inhibited the rate of Fenton-driven BR disappearance to almost 100% (Figure 2). This suggests that ascorbic acid is effective in scavenging ROS produced in the reaction. 200mM ascorbic acid will be used in subsequent experiments to determine the contribution of ROS to BR disappearance in mitochondria and microsomes.

Conclusion

It is concluded that 200mM ascorbic acid is an effective antioxidant to scavenge ROS in the Fenton-driven BR oxidation. This concentration is used to determine the contribution of ROS in microsomal and mitochondrial BR oxidations.

Figure 1 Hypothetical mechanism for formation of BR oxidative metabolites (BOMs). Direct interaction of ROS with BR will produce BR peroxyradical and is suggested to be fragmented into ions m/z 301, 315 and 333. 93

300

250

200

mol BR/min) mol ρ

150

100 BR degradation rate( degradationBR

* 0 0 2 20 200 Ascorbic acid (mM)

Figure 2 Effect of ascorbic acid on bilirubin (BR) degradation rate in the chemical oxidation system. Activities were observed in the presence of different concentrations of ascorbic acid. The chemical oxidation is driven by ROS (mainly peroxyl radical) produced by

Fenton reaction of H2O2. BR level was monitored by recording ΔA440 for 20 minutes. Results are given as means ± SD of three separate observations (n=3). *Significant different from control group, p < 0.05 (one-way ANOVA).

Appendix 8

MANUSCRIPT

Elsevier Editorial System(tm) for Toxicology and Applied Pharmacology Manuscript Draft

Manuscript Number: TAAP-D-14-00875

Title: Mitochondrial Targeting of Bilirubin Regulatory Enzymes: An Adaptive response to Oxidative stress.

Article Type: Full Length Article

Keywords: Mitochondria; Endoplasmic reticulum; CYP2A5; HMOX1 Bilirubin; Oxidative stress

Corresponding Author: Dr. A'edah Abu-Bakar,

Corresponding Author's Institution: The University of Queensland

First Author: Siti Nur F Muhsain, MSc

Order of Authors: Siti Nur F Muhsain, MSc; Matti A Lang, PhD; A'edah Abu-Bakar

Abstract: The intracellular level of bilirubin (BR), an endogenous antioxidant that is cytotoxic at high concentrations, is tightly controlled within the optimal therapeutic range. We have recently described a concerted intracellular BR regulation by two microsomal enzymes: haem oxygenase 1 (HMOX1), essential for BR production and cytochrome P450 2A5 (CYP2A5), a BR oxidase. Herein, we describe targeting of these enzymes to hepatic mitochondria during oxidative stress. The kinetics of microsomal and mitochondrial BR oxidation were compared. Treatment of DBA/2J mice with 200 mg pyrazole/kg/day for 3 d increased hepatic intracellular protein carbonyls content and induced nucleo- translocation of Nrf2. HMOX1 and CYP2A5 proteins and activities were elevated in microsomes and mitoplasts but not the UGT1A1, a catalyst of BR glucuronidation. A CYP2A5 antibody inhibited 75% of microsomal BR oxidation. The inhibition was absent in control mitoplasts but elevated to 50% after treatment. An adrenodoxin reductase antibody did not inhibit microsomal BR oxidation but inhibited 50% of mitochondrial BR oxidation. Ascorbic acid inhibited 5% and 22% of the reaction in control and treated microsomes, respectively. In control mitoplasts the inhibition was 100%, which was reduced to 50% after treatment. Bilirubin affinity to mitochondrial and microsomal CYP2A5 enzyme is equally high. Lastly, the treatment did not induce the release of cytochrome c into cytoplasm, indicating absence of mitochondrial membrane damage. Collectively, the observations suggest that BR regulatory enzymes are recruited to mitochondria during oxidative stress and BR oxidation by mitochondrial CYP2A5 is supported by mitochondrial mono-oxygenase system. The induced recruitment potentially confers membrane protection.

Cover Letter

Dear Dr. Burchiel,

Ms No: TAAP-D-14-00773

We submitted the above mentioned manuscript in July 2014, which was considered by one of your Associate Editors as an “innovation and priority publication” in TAAP provided that we include further evidence on relevant marker(s) that oxidative stress actually took place in pyrazole treated mice.

Herein, we submit a revised manuscript that includes additional data on protein oxidation (indicated by elevated protein carbonyls content) and Nrf2 nuclear translocation as generally accepted markers of oxidative stress. In agreement with previous findings, the treatment regimen used in the present study evoked oxidative stress as demonstrated by; (i) elevated protein carbonyls content, particularly in the microsomes; and (ii) induced translocation of the key regulator of antioxidant response, Nrf2, from the cytoplasm to the nucleus. The Nrf2 nuclear translocation indicates initiation of a downstream protective mechanism, which is supported by a small but significant reduction of mitochondrial protein carbonyls content in the treated animals.

We could show that under this condition the CYP2A5 (a bilirubin oxidase) and other key enzymes regulating the intracellular bilirubin levels are targeted into mitochondria. However, UGT1A1 — the sole microsomal enzyme that catalyses bilirubin glucuronidation for elimination from the body — is not targeted into mitochondria. Our observations support a hypothesis that the purpose for mitochondrial targeting of bilirubin regulatory enzymes during oxidative stress is to increase the local capacity to eliminate oxygen radicals and to protect mitochondria against excessive bilirubin concentration. This is supported by the observed lack of cytochrome c leakage into the cytosol, indicating absence of mitochondrial membrane damage.

We believe that our findings contribute significantly to our understanding on the mechanism by which bilirubin levels in mitochondria are regulated and on its biological significance.

The authors declare that there is no conflict of interest. All authors have read the manuscript, agree it is ready for submission and take full responsibility for its contents. The manuscript has not been submitted elsewhere at present nor previously published. Please note that the corresponding author of this manuscript is:

A’edah Abu-Bakar National Research Centre for Environmental Toxicology The University of Queensland 39 Kessels Road, Coopers Plains 4108 Queensland, Australia Phone: +617 3274 9147 Fax: +617 3271 9003 E-mail: [email protected] *Manuscript Click here to view linked References

1 MITOCHONDRIAL TARGETING OF BILIRUBIN REGULATORY ENZYMES: AN 2 ADAPTIVE RESPONSE TO OXIDATIVE STRESS 3 Siti Nur Fadzilah Muhsaina,b, Matti A. Langa, A’edah Abu-Bakara* 4 5 aThe University of Queensland, National Research Centre for Environmental Toxicology 6 (Entox), 4072 Brisbane, Queensland, Australia; [email protected], 7 [email protected], [email protected]. 8 9 bFaculty of Pharmacy, University Teknologi Mara, Malaysia 10 11 *Address correspondence to: A’edah Abu-Bakar, Entox, The University of Queensland, 39 12 Kessels Road, Coopers Plains, 4108 Queensland, Australia. Phone: +617 0419 301 740; Fax: 13 +617 3274 9003; E-mail: [email protected] 14 15 Abbreviations used: BR, bilirubin; ROS, reactive oxygen species; CYP, cytochrome P450; 16 CYP2A5, cytochrome P450 2A5; HMOX1, haem oxygenase-1; UGT1A1, UDP- 17 glucuronosyltransferase 1A1; BVR, biliverdin reductase. 18 19 20 21 22

32 Abstract 33 The intracellular level of bilirubin (BR), an endogenous antioxidant that is cytotoxic at high 34 concentrations, is tightly controlled within the optimal therapeutic range. We have recently 35 described a concerted intracellular BR regulation by two microsomal enzymes: haem 36 oxygenase 1 (HMOX1), essential for BR production and cytochrome P450 2A5 (CYP2A5), a 37 BR oxidase. Herein, we describe targeting of these enzymes to hepatic mitochondria during 38 oxidative stress. The kinetics of microsomal and mitochondrial BR oxidation were compared. 39 Treatment of DBA/2J mice with 200 mg pyrazole/kg/day for 3 d increased hepatic 40 intracellular protein carbonyls content and induced nucleo-translocation of Nrf2. HMOX1 41 and CYP2A5 proteins and activities were elevated in microsomes and mitoplasts but not the 42 UGT1A1, a catalyst of BR glucuronidation. A CYP2A5 antibody inhibited 75% of 43 microsomal BR oxidation. The inhibition was absent in control mitoplasts but elevated to 44 50% after treatment. An adrenodoxin reductase antibody did not inhibit microsomal BR 45 oxidation but inhibited 50% of mitochondrial BR oxidation. Ascorbic acid inhibited 5% and 46 22% of the reaction in control and treated microsomes, respectively. In control mitoplasts the 47 inhibition was 100%, which was reduced to 50% after treatment. Bilirubin affinity to 48 mitochondrial and microsomal CYP2A5 enzyme is equally high. Lastly, the treatment did not 49 induce the release of cytochrome c into cytoplasm, indicating absence of mitochondrial 50 membrane damage. Collectively, the observations suggest that BR regulatory enzymes are 51 recruited to mitochondria during oxidative stress and BR oxidation by mitochondrial 52 CYP2A5 is supported by mitochondrial mono-oxygenase system. The induced recruitment 53 potentially confers membrane protection. 54 55 Keywords: Mitochondria, CYP2A5, HMOX1, Bilirubin, Oxidative stress, Endoplasmic 56 reticulum 57 58 59 60 61 62 63 64 65

66 Introduction 67 In mitochondria, bilirubin (BR) is constitutively generated from haem in two successive 68 reactions: haem oxygenase-catalysed degradation to biliverdin (BV), followed by biliverdin 69 reductase-catalysed reduction to BR (Converso et al., 2006). At low constitutive level the 70 free-radical quencher BR prevents accumulation of radical oxygen species (ROS) in 71 mitochondria (Jansen et al., 2010), but at above 10 µM it disrupts mitochondrial membrane 72 permeability, which in turn causes apoptosis (Rodrigues et al., 2002). Therefore, 73 mitochondrial BR needs to be well regulated especially under oxidative stress conditions 74 when BR production is markedly elevated following the induction of haem oxygenase -1 75 (HMOX1). However, little is known about the regulatory mechanism of mitochondrial BR. 76 The normal route of BR excretion in mammals is via the bile and involves the 77 conversion of BR into its glucuronides, a reaction solely catalysed by the microsomal uridine- 78 diphosphate-glucuronosyltransferase 1A1 (UGT1A1) (Tukey & Strassburg, 2000). This 79 enzyme however is not present in mitochondria (Radominska-Pandya et al., 2005). In 1969, 80 Brodersen & Bartels described a BR oxidase in guinea-pig brain mitochondria. The reaction 81 products include BV (Brodersen & Bartels, 1969). Subsequently, similar BR oxidase activity 82 was described in rat liver mitochondria (Cardenas-Vazquez et al., 1986). Bilirubin oxidation 83 was thus proposed as the main metabolic pathway for mitochondrial BR. However, the brain 84 and liver enzymes were not identified. 85 Importantly, we have recently discovered that the liver microsomal cytochrome P450 86 2A5 (CYP2A5) can function as an inducible BR oxidase that catalyses BR oxidation to BV 87 (Abu-Bakar et al., 2011). In response to oxidative stress CYP2A5 induction paralleled but 88 followed the induction of microsomal HMOX1 by several hours (Abu-Bakar et al., 2005). 89 The CYP2A5 induction was delayed until cellular BR concentrations were elevated above the

90 Km of 1-2 µM. At these concentrations BR was shown to stabilise the labile CYP2A5 protein 91 by delaying its degradation (Abu-Bakar et al., 2011; Abu-Bakar et al., 2012). These 92 observations would be consistent with BR inducing its own metabolism and CYP2A5 serving 93 to clear BR generated by HMOX1 to ensure the levels of this essential but toxic compound 94 do not exceed a safe threshold. 95 We have proposed that the significance of this regulatory mechanism is to ensure a 96 rapid increase in intracellular antioxidant capacity during oxidative stress, by increasing the 97 levels of BR, followed by an efficient elimination of any excess of BR as soon as cellular 98 ROS reached physiological level (Abu-Bakar et al., 2013). The kinetics of CYP2A5

99 catalysing BR oxidation seem ideal for this purpose as the Km for BR is roughly one tenth of 100 the cytotoxic concentrations (Abu-Bakar et al., 2005; Abu-Bakar et al., 2012). 101 Earlier observations that a fraction of microsomal HMOX1 is targeted to mitochondria 102 in response to oxidative stress (Converso et al., 2006; Srivastava and Pandey, 1996) indicate 103 an increased mitochondrial BR production. The fact that the CYP2A5 is also targeted to 104 mitochondria in response to oxidative stress (Genter et al., 2006; Honkakoski et al., 1988) 105 raises an interesting question if mitochondria are equipped with a similar system to that of 106 microsomes in regulating local BR homeostasis. 107 In addressing this question, we explored the induction profile of key BR regulatory 108 enzymes in the microsomes and mitochondria of the mouse liver in response to oxidative stress. 109 We also investigated the contribution of mitochondrial CYP2A5 to BR oxidation. We observed 110 that oxidative stress — indicated by elevated intracellular protein carbonylation and nucelo-Nrf2 111 levels — markedly increased HMOX1 and CYP2A5 proteins and activities in the microsomes 112 and mitochondria. The induction of CYP2A5 in mitochondria was stronger than in the 113 microsomes. We also observed that mitochondrial CYP2A5 oxidised BR with an affinity similar 114 to that of the microsomal enzyme but required the mitochondrial electron transfer chain for its 115 activity. During oxidative stress the mitochondrial CYP2A5-catalysed BR oxidation was highly 116 significant, whilst under normal condition oxidation by ROS was the dominant pathway. By 117 contrast, the microsomal BR oxidation was mainly driven by the CYP2A5 both under normal 118 and oxidative stress conditions. Additionally, protein carbonyls content in microsomes was 119 increased by 200% but in mitochondria it was reduced by about 25% of control. This suggests 120 marked oxidative damage to microsomal proteins, whilst the mitochondrial proteins seemed to 121 be well protected, which is further supported by observed lack of mitochondrial cytochrome c 122 leakage into the cytosol in treated mice. Collectively, our observations indicate that induced 123 recruitment of key bilirubin regulatory enzymes to mitochondria potentially confers 124 mitochondrial protection against severe oxidative damage. 125 126 Materials and Methods 127 Chemicals and antibodies. Coumarin, mesoporphyrin, umbelliferone (7-hydroxycoumarin), 128 pyrazole, β-NADPH, glycerol, mannitol, leupeptin, Hepes, phenylmethylsulfonyl-fluoride 129 (PMSF), digitonin, Percoll, bovine serum albumin (BSA), ethylene glycol tetraacetic acid 130 (EGTA), ethylenediamine tetraacetic acid (EDTA), dimethyl sulfoxide (DMSO) and Tween 131 20 were purchased from Sigma-Aldrich (Sydney, Australia). Bilirubin, mesobilirubin, 132 biliverdin and sucrose were from Frontier Scientific, Inc. (Utah, USA) and Ajax Chemical

133 (Sydney, Australia), respectively. Dithiothreitol (DTT) was purchased from Applichem 134 (Germany). Goat anti-rabbit, goat anti-mouse, mouse anti-goat and rabbit anti-chicken 135 antibodies conjugated with horseradish-peroxidase were from Thermo Fisher Scientific Inc. 136 (Victoria, Australia). Rabbit polyclonal anti-COX IV antibody was from Cell Signalling 137 Technology (NY, USA) whereas anti-cytochrome P450 reductase antibody was from Assay 138 Designs (NY, USA). Rabbit polyclonal anti-Nrf2 (ab92946), anti-biliverdin reductase 139 (ab19260), anti-β-actin (ab8227), and mouse monoclonal anti-HMOX1 (ab13248) antibodies 140 were sourced from Abcam (Cambridge, UK). Rabbit polyclonal anti-Lamin B1 (sc20682) 141 and goat polyclonal anti-UGT1A1 (sc27419) antibodies were purchased from Santa Cruz 142 Biotechnology Inc. (Texas, USA). Rabbit polyclonal anti-CYP2A5 antibody and anti- 143 adrenodoxin reductase antibody was sourced from Dr Risto Juvonen, University of Kuopio, 144 Finland and Dr Sergej Usanov, National Academic of Sciences, Republic of Belarus, 145 respectively. The specificity of these antibodies in inhibiting CYP2A5 activity (coumarin 7- 146 hydroxylase) and adrenodoxin reductase activity, respectively, have been well documented in 147 various models (Abu-Bakar et al., 2005; Asikainen et al., 2003; Chernogolov et al., 1997; 148 Guzov et al., 1993; Honkakoski et al., 1988; Honkakoski et al., 1989; Honkakoski et al., 149 1993; Juvonen et al., 1988; Kaipainen et al., 1984; Usanov et al., 1984; Usanov et al., 1989; 150 Usanov et al., 1990). 151 152 Treatment of animals. All experimental procedures involving handling of laboratory animals 153 were conducted in accordance with the Australian National Health and Medical Research 154 Council (NHMRC) Code of Practice for care and use of experimental animals, and approved 155 by the Queensland Forensic and Scientific Services Animal Ethics Committee (FSS-AEC 156 Approval No. 10P01). Male DBA/2J mice were purchased from the Animal Resource Centre, 157 Western Australia. They were group-housed, maintained at a 12 hour light/dark cycle, and 158 had access to standard rodent chow ad libitum. The mice were randomly allocated into 159 treated and control groups (6 mice per group). Induction of oxidative stress was achieved by 160 intraperitoneal administration of 200 mg pyrazole/kg bwt/day for three days (Bae et al., 2012; 161 Gilmore and Kirby, 2004). The animals in control group were given normal saline only.

162 Twenty-four hours after the final dose the animals were euthanised with CO2/O2 overdose 163 and their livers excised. 164 165 Preparations of subcellular fractions. Microsomes and mitochondria were prepared from 166 fresh liver at 0-4oC as described by Genter et al. (2006) with some modifications. Briefly,

167 fresh livers were homogenised in isolation buffer (2mM Hepes, 70mM sucrose, 220mM 168 mannitol, 2mM EDTA, 1mM DTT, 0.5mM PMSF, 0.01 µg/µl leupeptin) using a motor- 169 driven Potter-Elvehjem homogenizer (Sartorius, Goettingen, Germany). The homogenate was 170 centrifuged (2,000 x g) for 10 minutes at 4oC. The supernatant was aspirated into a clean tube 171 and further centrifuged (14,000 x g) for 15 minutes at 4oC. The supernatant (post- 172 mitochondrial fraction) was used to prepare microsomes. The mitochondrial pellet was 173 washed by resuspending in isolation buffer and centrifuged again at 14,000 x g for 15 174 minutes (4oC). Resulting pellet was resuspended in isolation buffer, underlay with 27% 175 sucrose using a 20-gauge blunt-end needle and centrifuged for 15 minutes at 14,000 x g 176 (4oC). The pellet was washed and resuspended in isolation buffer. To increase the purity of 177 mitochondrial fractions the pellet was treated with an appropriate volume of 0.13 mg 178 digitonin/mg protein to strip off the outer membrane that is associated with microsomes. This 179 mixture was incubated on ice with occasional agitation for 15 minutes at 4oC. The reaction 180 was stopped by resuspending the mixture with isolation buffer and centrifuged (12,000 x g) 181 for 15 minutes at 4oC. The resulting pellet was suspended in the buffer and centrifuged again 182 for 15 minutes at 12,000 x g. The mitoplasts were resuspended in dilution buffer (2mM Tris- 183 HCl, pH 7.4, 1mM EDTA, 20% glycerol), aliquoted and stored at -80oC until required. 184 Enzyme activity assays and Western Immunoblot were performed with mitoplasts as 185 the enzymes studied (HMOX1, CYP2A5, UGT1A1), and cytochrome c are integrated in the 186 inner mitochondrial membrane (Cortese et al., 1998; Gallet et al., 1997; Genter et al., 2006; 187 Honkakoski et al., 1988; Rytomaa et al., 1995), which is intact in the mitoplasts. Prior to 188 enzyme activity assays the purified mitoplasts were brought down by centrifugation, 189 suspended in dilution buffer and sonicated (4 X 30 s with 30 s intervening cooling time) to 190 disrupt the inner membrane to allow NADPH and substrates to reach the enzymes of interest. 191 Nuclei fractions were prepared in the cold (2-4oC) as previously described (Wang, 192 1967) with modifications. Briefly, fresh livers were rinsed in ice-cold PBS. The livers were 193 cut into small pieces in 6 ml of homogenisation buffer (HB) [containing final concentrations 194 of 10 mM Hepes pH 7.6, 15 mM KCl, 1mM EDTA, 1M sucrose, 10% glycerol, 0.5 mM 195 spermidine, 0.15 mM spermine, 0.5 mM DTT, 0.5 mM PMSF, and 10 μg/ml leupeptin] and 196 homogenized in Potter-Elvehjem homogenizer (Sartorius, Goettingen, Germany). The liver 197 homogenate then was gently poured into a centrifuge tube containing 5 ml of the HB and 198 centrifuged at 50,000 x g, 4oC for 50 min. The white pellet obtained was dissolved in 1 ml of 199 TGEM [containing final concentrations of 50 mM Tris pH 8.0, 40% glycerol, 5 mM EDTA, o 200 and 5 mM MgCl2] and centrifuged at 600 x g, 4 C for 5 min. The pellet (nuclei) was

201 resuspended in 0.7 ml TGEM, aliquoted and stored in –80oC. When needed nuclear proteins 202 were extracted by dialysis. The purified nuclei was brought down by centrifugation (600 x g, 203 4oC for 5 min) and resuspended in 2 volumes of suspension buffer [20 mM Hepes pH 7.6,

204 420 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 25% glycerol, 0.5 mM DTT, and 0.5 mM 205 PMSF]. The nuclei suspension was stirred with a magnetic flea for 30 min at 4oC and 206 centrifuged at 1200 x g, 4oC for 5 min. The supernatant was dialyzed overnight on a 0.025 207 μm nitrocellulose membrane (Millipore, Massachusetts, USA) against 100 volumes of 208 dialysis buffer [20 mM Hepes pH 7.6, 100 mM KCl, 0.2 mM EDTA, 20% glycerol, 0.5 mM 209 DTT, and 0.5 mM PMSF] at 4oC. The dialyzed supernatant was centrifuged at 1200 x g, 4oC 210 for 5 min to bring down impurities. The supernatant, which contained nuclear proteins were 211 aliquoted and stored at –80oC. 212 213 Protein analysis. Protein concentrations of the subcellular fractions were determined by the 214 Lowry method (Lowry et al., 1951). Protein concentrations of the cytosol, microsomes, 215 mitoplasts, and nuclei ranged from 7.8±1 µg/µl (0.78±0.09 mg/g wet liver weight) to 216 16.3±1.04 µg/µl (3.4±0.3 mg/g wet liver weight). Protein fractions were separated by 217 electrophoresis through a 4-12% Mini-Protean® TGX™ (Biorad Laboratories, USA), under 218 denaturing condition and blotted overnight onto 0.2 µm PVDF membranes (Biorad 219 Laboratories, USA). The blots were blocked overnight in Tris-buffered saline (TBS) 220 containing 0.1% Tween 20 and 5% non-fat milk at 4oC. To assess the purity of mitochondrial 221 fraction, the blots were then incubated with the following antibodies overnight at 4oC: anti- 222 COX IV antibody and anti-P450 reductase antibody (1:1000 dilutions). To determine the 223 levels of key BR regulatory proteins the blots were incubated with HMOX1, biliverdin 224 reductase (BVR), UGT1A1, CYP2A5, or antibodies for 3 hours at room temperature (1:500 225 to 1:2000 dilutions). Detection of a key regulator of antioxidant response — the nuclear 226 factor (erythroid-derived 2)-like 2 (Nrf2) — as indicator of oxidative stress was by incubating 227 the blots with anti-Nrf2 antibody. The blots were then incubated for 1 hour with horseradish- 228 peroxidase-conjugated secondary antibodies (1:5000 dilutions). After further washing with 229 TBS containing 0.05% Tween 20 and 5% non-fat milk, blots were incubated in 230 chemiluminescence reagents (Biorad Laboratories, USA). Finally, bands’ intensities were 231 detected using Versadoc™ Molecular Imaging System (Biorad Laboratories, USA). The 232 results were expressed as relative density where the signals of a specific antibodies were

233 normalised against the loading control signals: β-Actin for cytosolic and microsomal 234 fractions, and COXIV for mitochondrial fractions. 235 236 Protein carbonyls assay. Protein carbonyl groups that are formed when proteins are 237 oxidatively damaged have been increasingly used as biomarkers for oxidative stress (Dalle- 238 Donne et al., 2003). The carbonyls were quantitated spectrometrically using 2,4- 239 dinitrophenylhydrazine (DNPH) as previously described (Reznick & Packer, 1994; Fagan et 240 al., 1999) with modifications. Cytosolic, microsomal and mitochondrial proteins (1- 3 241 mg/assay) were precipitated with 10 volumes of HCl-acetone (3:100) (v/v). The proteins 242 were centrifuged down at 6000 x g for 10 min. The pellets were washed with 2 x 2 ml of 243 HCl-acetone to remove chromophores, which may interfere with the assay. The protein 244 pellets were then washed twice with 2 ml of 10% TCA. After the final wash, protein pellets 245 were resuspended in 500 μl of 2 M HCl to which 500 μl of 10 mM DNPH (in 2 M HCl) was 246 added and vortexed every 5 min for 20 min at room temperature. Protein blanks were 247 prepared by adding 500 μl of 2 M HCl instead of DNPH to the assay tubes containing protein 248 sample. After mixing, 500 μl of 30% TCA was added to each tube, the samples vortexed then 249 placed on ice for 10 min. Following centrifugation (3 min at 11,000 x g), the supernatant was 250 discarded and the pellets were subjected to extensive washing with 2 x 2 ml of 20% TCA 251 followed by 3 x 2 ml of ethanol-ethylacetate (1:1) (v/v) washes to remove any unreacted 252 DNPH. The final precipitates were dissolves in 1 ml of 6 M guanidine hydrochloride in 20 253 mM potassium dihydrogen phosphate (pH 2.3) and left for 30 min at 37oC with vortex 254 mixing. Any insoluble materials were removed by additional centrifugation (6000 x g for 10 255 min). The protein carbonyl content was calculated from the absorbance measurement at 380 256 nm and extinction coefficient ε = 22,000 M-1cm-1 (Johnson, 1953). Protein samples with 257 A280/A260 nm ratio of less than 1 were incubated with streptomycin sulphate (final 258 concentration of 1%) for 15 min at room temperature to remove nucleic acids that may 259 erroneously contribute to higher estimation of carbonyls. 260 261 Enzyme activity measurement. Coumarin 7-hydroxylase (COH) — indicator of CYP2A5 262 activity — was determined spectrofluorometrically by assessing 7-hydroxylase activity 263 towards the substrate coumarin (100 µM final concentration) using 0.1 mg of mitoplast or 264 microsomal proteins incubated for 20 minutes as previously described (Abu-Bakar et al., 265 2004).

266 Haem oxygenase activity was measured by detecting BR generation as previously 267 described (Ryter et al., 2000). Briefly, microsomes or mitoplasts were added to a reaction 268 mixture and mouse liver cytosol was used as a source of biliverdin reductase. The reaction 269 was done in the dark. BR concentration was calculated as the difference in absorbance 270 between 464 and 530 nm, using an extinction coefficient of 40 mM-1cm-1. 271 NADPH P450 reductase and cytochrome c oxidase activities were determined with the 272 respective assay kits (Cat Nos: CY0100 and CYTOCOX 1, Sigma Aldrich, Sydney, 273 Australia) in accordance with the manufacturer’s instruction. 274 Biliverdin reductase (BVR) activity was measured according to published method 275 (Tenhunen et al., 1970) with modifications. In brief, the activity was assayed in buffer

276 containing 100 mM KPO4 pH 7.4, 25 μM biliverdin and 0.2 mg/ml protein. The mixture was 277 incubated for 5 minutes at 37oC and the reaction was initiated with the addition of 100 µM 278 NADPH. Bilirubin formation was determined spectrophotometrically at 450 nm, using the 279 extinction coefficient 56 mM-1cm-1. 280 281 Bilirubin oxidation analysis. Bilirubin oxidation was assessed by determining the rate of 282 bilirubin disappearance spectrometrically as previously described (Abu-Bakar et al., 2005; 283 Arthur et al., 2012). In the study of the inhibitory effects of anti-CYP2A5 antibody, the final 284 concentration of microsomal or mitoplast protein used was 5 µg/ml. Increasing 285 concentrations of the antibody were then added to obtain concentration ratios to 286 microsomal/mitoplast protein ranging from approximately 0.0005 to 2.0. Twenty µM BR (in 287 DMSO) was added and the reaction was initiated by addition of NADPH and incubated at 288 37oC for 20 minutes. The same condition was employed when the inhibitory effects of 200 289 mM ascorbic acid and anti-adrenodoxin reductase (AdxR) antibody on the bilirubin- 290 degrading activity of the induced microsomes and mitoplast were studied. 291 Screening of putative dipyrroles was performed as previously described (Abu-Bakar et 292 al., 2011; Arthur et al., 2012). Briefly, the reaction mixtures were incubated for 1 hour after 293 which samples were filtered using 0.45µm filter membrane (Milipore) and immediately 294 injected to a HPLC-MS/MS (AB/Sciex AP14000Q mass spectrometer, USA) equipped with 295 an electrospray (TurboV) interface coupled to a HPLC system (Prominence Model XYZ, 296 Shimadzu Corp., Kyoto, Japan). Separation was achieved using a 2.5µm 50 x 2mm Synergi 297 MAX RP column (Phenomenex, Torrance, CA, USA) run at 40oC, and mobile phase flow 298 rate of 0.5ml/min with a linear gradient starting at 5% B for 1.0 min, ramped to 80% B in 10 299 min, held for 1 min, to 5% in 0.2 min and equilibrated for 3 min (mobile phase A=1%

300 acetonitrile/HPLC grade water, mobile phase B= 95% acetonitrile/HPLC grade water, both 301 containing 5mM ammonium acetate). The mass spectrometer was operated in the positive 302 ion, multiple reactions monitoring mode using nitrogen as the collision gas under the 303 following conditions: declustering potential was 75 or 110 (biliverdin only), and the collision 304 energy was between 25 and 83eV. 305 306 Kinetics and statistical analysis. Enzyme kinetic parameters were determined by GraphPad 307 Prism 5 kinetics programme from GraphPad Software Inc (La Jolla, CA, USA). It uses a non- 308 linear regression method of curve fitting and the Runs test of residuals to determine 309 statistically whether experimental data are randomly distributed around the curve with 95%

310 confidence. The Ki value was determined by plotting observed Km (Y-axis) vs bilirubin 311 concentrations (X-axis). For data other than enzyme kinetic parameters Student’s t-test was 312 used for comparisons between two groups. Mean differences were considered significant 313 when p < 0.05. 314 315 Results 316 Assessment of purity of microsomal and mitochondrial fractions. 317 The purity of the microsomal and mitoplast fractions was assessed by measuring the 318 presence of cytochrome C oxidase IV (COX IV) and NADPH P450 reductase (NPR), 319 markers of mitochondrial inner membrane and endoplasmic reticulum, respectively. Western 320 immunoblotting showed that a minimal amount of NPR was present in the mitoplast (Fig. 321 1A), which commensurate with the NPR activity (approximately 10% of the activity in 322 microsomes) (Fig.1B). This suggests that most but not all of the endoplasmic reticulum (ER) 323 associated with mitochondria has been removed. On the other hand, no COX IV-reactive 324 band was found in the microsomal fraction (Fig. 1A) although a minimal microsomal COX 325 activity (2-4% of mitoplast activity) was present (Fig 1B). This indicates a high microsomal 326 enrichment with only minor mitochondrial contamination. Subsequent data analysis was 327 offset against the 10% microsomal contamination and 2-4% mitoplasts contamination. 328 329 Pyrazole induces oxidative stress in the liver 330 Pyrazole (PYR) doses used in this study were previously shown to cause oxidative 331 stress in the mouse liver (Gilmore and Kirby, 2004; Lu et al., 2008). Particularly, PYR was 332 shown to induce endoplasmic reticulum stress and elevate protein carbonyls content in a 333 time- and dose-dependent manner (Gilmore and Kirby, 2004). The treatment indeed evoked

334 endoplasmic reticulum stress as indicated by a significant increase (200%) of protein 335 carbonylation in the microsomes (Fig. 2A). It is noted that the carbonyls content in control 336 mitoplasts was significantly higher (by about 40 - 50%) than in control cytosol and 337 microsomes (Fig. 2A). This is perhaps to be expected as mitochondria produce oxygen 338 radicals during cellular respiratory, moreso than the other organelles. Interestingly, PYR 339 reduced mitochondrial protein carbonyls content by 25% (Fig. 2A), indicating mitochondrial 340 proteins were protected against oxidative damage when microsomal proteins were 341 signficantly oxidised. This observation is paralleled with the PYR-induced accumulation of 342 Nrf2 — a key regulator of antioxidant response — in the nucleus (Fig. 2B & C). Previous 343 findings have also observed elevated Nrf2 levels in livers of mice treated with PYR 344 (Cederbaum, 2013; Lu et al., 2008). 345 346 Pyrazole accumulates BR regulatory enzymes in microsomes and mitoplasts 347 Importantly, during oxidative stress HMOX1 and CYP2A5 immunoreactive bands were 348 substantially increased in microsomes and mitoplasts (Fig. 3). Compared to controls HMOX1 349 induction in the microsomes was stronger than that of CYP2A5 (2-fold more than the 350 CYP2A5 induction) (Fig. 3B). By contrast, HMOX1 induction in the mitoplasts was 2-fold 351 less than the CYP2A5 induction (Fig. 3B). 352 Although HMOX1 protein induction in mitoplasts was two times less than in 353 microsomes (Fig. 3B), elevation of its catalytic activity was similar in both organelles 354 (increased by 4-fold compared to control) (Table 1). This suggests differences in the 355 reactivity of anti-HMOX1 antibody vis-à-vis the mitochondrial and microsomal HMOX1. 356 This is plausible as some microsomal proteins are targeted to mitochondria as truncated 357 proteins (Addya et al., 1997; Anandatheerthavarada et al., 1999; Anandatheerthavarada et al., 358 2009). In the case of HMOX1, the precise mitochondria targeting signal has not been 359 identified. However, recent study demonstrated that deletion of N-terminal of the microsomal 360 HMOX1 protein resulted in markedly increased mitochondrial targeting (Bansal et al., 2014). 361 On the other hand, the strong induction of microsomal and mitochondrial CYP2A5 protein 362 was reflected in 10-fold increase and 60-fold increase of the CYP2A5 catalysed coumarin 7- 363 hydroxylase (COH) activity, respectively (Table 1). 364 It is noted that UGT1A1, the sole enzyme that catalyses BR glucuronidation, was not 365 detected in mitoplasts (Fig. 3), confirming previous observation that UGT enzymes are not 366 known to be present in the mitochondria (Radominska-Pandya et al., 2005). PYR, however 367 reduced microsomal UGT1A1 by 40% (Fig. 3B), indicating that elimination of BR by

368 glucuronidation was significantly hampered during oxidative stress. Additionally, PYR did 369 not affect the activity and protein levels of cytosolic BVR (Fig. 3 and Table 1), the enzyme 370 that reduces BV to BR. By contrast, BVR protein and activity increased by 100% in treated 371 mitoplasts (Fig. 3 and Table 1). 372 Taken together, these observations indicate that in response to PYR-induced oxidative 373 stress BR producing, oxidising and reducing enzymes are recruited to mitochondria. The 374 absence of UGT1A1 in mitochondria further suggests that instead of being eliminated 375 through glucuronidation, local BR is metabolised by other pathway, perhaps through 376 oxidation by the CYP2A5. 377 378 Pyrazole treatment increases BR degradation in microsomes and mitoplasts 379 We have previously shown that CYP2A5-dependent BR oxidation exists in microsomes 380 and the activity is substantially increased in response to cadmium-induced oxidative stress 381 (Abu-Bakar et al., 2005). In exploring the existence of this system in mitochondria and its 382 inducibility in response to oxidative stress we measured BR degradation rate 383 spectrometrically in incubations with mitoplasts and microsomes. Fig. 4A shows that BR 384 disappears in the presence of control microsomes and mitoplasts, and that the disappearance 385 rate increased in PYR-treated microsomes and mitoplasts. Interestingly, the rate in control 386 microsomes and mitoplasts is similar but in treated microsomes it increased 3 times more 387 than in treated mitoplasts (200% increase in treated microsomes; 68% increase in mitoplasts) 388 (Fig. 4A). The observations suggest that while a BR oxidation system exists in both 389 organelles it differs in the kinetics of inducibility. 390 391 Role of CYP2A5 in BR oxidation in microsomes and mitoplasts. 392 It has been hypothesised that the initial step in BR oxidation is hydrogen abstraction 393 from the central methylene bridge of the tetrapyrrole, followed by loss of another hydrogen to 394 give BV (De Matteis et al., 2006). This has been shown to take place in the active site of 395 CYP2A6 enzyme (the human orthologue of CYP2A5) (Abu-Bakar et al., 2012). In a system 396 where ROS is accessible, random oxygen binding takes place subsequent to hydrogen 397 abstraction to form BR peroxyradical, which is then fragmented into putative dipyrroles with 398 m/z values of 301, 315, and 333 (Abu-Bakar et al., 2011; De Matteis et al., 2006). The BR 399 peroxyradical can also be formed by a superoxide anion attacking BR directly, followed by 400 fragmentations into the putative dipyrroles (Abu-Bakar et al., 2011; Abu-Bakar et al., 2012). 401 Accordingly, BV is the predominant product of the CYP2A5-catalysed BR oxidation, whilst

402 dipyrroles are major products of ROS-driven BR oxidation (non-enzymatic oxidation) (Abu- 403 Bakar et al., 2011; Abu-Bakar et al., 2012). 404 Given that both microsomes and mitochondria produce ROS through leakage of their 405 respective electron transfer systems (Cadenas and Davies, 2000; Zangar, 2004), the observed 406 BR disappearance activity — depicted in Fig. 4A — may be caused by both enzymatic and 407 non-enzymatic oxidations. To assess the contribution of the respective oxidation pathways in 408 the observed BR disappearance we added vitamin C — a ROS scavenger — to the incubation 409 mixtures to exclude random oxidation by ROS. 410 Vitamin C did not significantly inhibit BR degradation in control microsomes but 411 inhibited about 20% of the reaction in treated microsomes (Fig. 4B), indicating that random 412 oxidation by ROS is a minor event in microsomes. By contrast, ROS driven BR 413 disappearance seems to be the major, if not the only degradation pathway in the control 414 mitoplasts, as 100% of the BR disappearance was blocked by vitamin C (Fig. 4B). In the 415 treated mitoplasts, however, vitamin C blocked only 54% of BR degradation (Fig. 4B), 416 suggesting that in response to oxidative stress ROS-independent oxidation is as important as 417 the ROS-dependent oxidation. 418 To explore the contribution of CYP2A5 to BR oxidation in both organelles we first 419 screened BV and dipyrroles formations in incubations with either microsomes or mitoplasts. 420 In the second set of experiments we used anti-CYP2A5 antibody to inhibit CYP2A5- 421 catalysed BR oxidation.

422 The absence of BV formation in the chemical oxidation of BR by Fe-EDTA/ H2O2 423 system (Fig. 5A) confirms published observations that dipyrroles are the major products of 424 non-enzymatic BR oxidation (Abu-Bakar et al., 2011; Abu-Bakar et al., 2012; De Matteis et 425 al., 2006). The formation of BV and dipyrroles in incubations with microsomes or mitoplasts 426 (Fig. 5B) suggests BR oxidation in these organelles is driven by enzymatic and non- 427 enzymatic pathways. In control microsomes BV formation predominates that of the 428 dipyrroles (Fig. 5B). PYR treatment increased the formation of dipyrroles (Fig. 5B), 429 suggesting elevation in ROS-driven BR oxidation, which commensurate with ascorbic acid 430 inhibition depicted in Fig. 4B. In the case of mitoplasts, dipyrroles formation predominates 431 that of BV. The fold-increased of BV formation is 4 times that of dipyrroles in treated 432 mitoplasts (Fig. 5B). This observation suggests that in mitoplast the enzymatic oxidation is as 433 important as the ROS-mediated oxidation, which parallels with our earlier observation in 434 ascorbic acid inhibition study (Fig. 4B).

435 Fig. 6 shows that anti-CYP2A5 antibody blocks BR degradation in both organelles. It 436 inhibited microsomal BR oxidation by 70 - 72% (Fig. 6A). The amount of CYP2A5 antibody 437 needed to achieve the inhibition in treated microsomes was about 13-fold higher than in the 438 control microsomes, indicating that larger amounts of CYP2A5 protein responsible for BR 439 disappearance were present in the treated samples. This commensurate with the observed 440 protein levels and COH activity (Fig. 3 and Table 1). By contrast, we did not observe 441 significant inhibition in control mitoplasts. However, a maximum of 50% inhibition was 442 observed after PYR treatment (Fig. 6B), which parallels with the increased CYP2A5 activity 443 (Table 1). These results suggest that the other mechanism for BR degradation — in addition 444 to oxidation by ROS — is via CYP2A5-catalysed oxidation, as together these two metabolic 445 pathways would explain 100% of the BR degradation in each case. 446 To confirm that the enzymatic BR oxidation is supported by the respective mono- 447 oxygenase complexes in each organelle, anti-adrenodoxin reductase (AdxR) antibody was 448 added to the incubations with treated microsomes or treated mitoplasts. AdxR, is a key 449 component of the mitochondrial electron transfer chain (ETC) that provides electrons to the 450 mitochondrial CYP catalysed reactions and is absent in the microsomes. Fig. 6C shows that 451 the antibody did not inhibit microsomal BR disappearance, indicating that the microsomal 452 BR degradation is essentially driven by the microsomal ETC. This observation also 453 confirmed that the microsomal fractions are essentially free from mitoplast contamination. By 454 contrast, the anti-AdxR antibody inhibited 50% of BR degradation in mitoplasts (Fig. 6C), 455 which is in agreement with Fig. 4B & 5B and confirms that in treated mitoplasts 50% of the 456 oxidation is driven by mitochondrial CYP2A5. 457 458 Microsomal and mitochondrial CYP2A5 enzymes have similar affinity to BR. 459 We have previously established that both the microsomal and mitochondrial CYP2A5 460 enzymes have a high affinity towards coumarin, which can be used as a diagnostic substrate 461 for the enzyme in both organelles (Honkakoski et al., 1988). We also showed in this study 462 (Fig. 6C) and in earlier investigations that the enzyme has been modified to fit in the electron 463 transfer chains of the respective organelles (Honkakoski et al., 1988). To further characterise 464 the interaction of BR with mitochondrial CYP2A5 we inhibited the coumarin 7- 465 hydroxylation (COH) activity in treated mitoplasts with different concentrations of BR. 466 Fig. 7A shows that inhibition of mitochondrial coumarin 7-hydroxylation by BR is of a

467 mixed type that affected both the Vmax and Km, where the Vmax is reduced and the Km is

468 increased (Fig. 7B). The Km for coumarin 7-hydroxylation is 6.14±0.5 (Fig. 7C), which is 13

469 times more than the Km for microsomal coumarin 7-hydroxylation (Abu-Bakar et al., 2005), 470 suggesting that coumarin affinity to the mitochondrial CYP2A5 is less than the affinity to 471 microsomal CYP2A5. This aligns with previous observations that targeted xenobiotic- 472 metabolizing CYPs are modified to depend on mitochondrial ETC for a supply of electrons 473 for catalytic activity toward altered substrate specificity/affinity (reviewed in Kinonen et al., 474 1995).

475 The plot Km observed vs BR concentrations (Fig. 7C) yielded a linear graph, which

476 confirmed the competitiveness of BR inhibition to the reaction with Ki = 2.67 μM. This 477 indicates that BR affinity for mitochondrial CYP2A5 is as high as the reported affinity for 478 microsomal enzyme (ranging from 0. 7 to 2.2 μM), and comparable to coumarin affinity for 479 microsomal CYP2A5 (ranging from 0.5 to 2.0 μM) (Abu-Bakar et al., 2012; Abu-Bakar et 480 al., 2005; Draper et al., 1997; Kinonen et al., 1995). Importantly, BR affinity for 481 mitochondrial and microsomal CYP2A5 (0.5 to 2.7 μM) appears to be similar (unlike that of 482 coumarin) and optimal for maintaining the intracellular BR concentrations at sub-toxic levels 483 of <10 µM. 484 485 Effect of pyrazole on the release of cytochrome c from mitochondria 486 Cytochrome c — a component of the mitochondrial ETC — is often released from 487 mitochondria during the early stages of apoptosis when mitochondrial membranes were 488 severely damaged (Sanchez-Alcazar et al., 2000). Once in the cytosol, cytochrome c initiates 489 the process of apoptosis. Hence, analysis of this protein in mitoplasts and cytosol will give a 490 glimpse on whether or not PYR treatment caused severe damage to mitochondrial membrane. 491 Fig. 8 shows that PYR treatment did not cause the release of cytochrome c protein into the 492 cytosol. The cytochrome c is primarily concentrated in the mitoplasts, which is in agreement 493 with previous observations that the protein is tightly bound to the inner membrane by 494 cardiolipin, a phospholipid found primarily in the mitochondrial inner membrane (Cortese et 495 al., 1998; Gallet et al., 1997; Gorbenko, 1999; Rytomaa et al., 1995). 496 The observed lack of cytochrome c leakage suggests that despite a significant increase 497 (> 200%) in mitochondrial HMOX1 activity (Table 1) — the enzyme that produces BR — 498 mitochondrial BR levels were well below the concentration range (> 10 μM) that disrupts 499 mitochondrial membrane permeability (Rodrigues et al., 2002). The observation is also 500 aligned with the observed reduction of mitochondrial protein carbonylation after PYR 501 treatment (Fig. 2A). 502

503 Discussion 504 In the present study we have demonstrated that HMOX1, BVR and CYP2A5 — the 505 enzymes critical for maintaining hepatic intracellular BR levels optimal for its antioxidant 506 role — are targeted to mitochondria in response to PYR-induced oxidative stress. The 507 treatment regime evoked oxidative stress as indicated by nuclear translocation of cytoplasmic 508 Nrf2, as well as carbonylation of cellular proteins, particularly to endoplasmic reticulum 509 proteins, which aligns with previous findings (Cederbaum, 2013; Gilmore and Kirby, 2004; 510 Lu et al., 2008). 511 The CYP2A5, a BR oxidase, in particular is strongly upregulated in the mitochondria 512 and is also adapted to be part of the mitochondrial ETC. Interestingly in this condition, 513 microsomal UGT1A1, the sole enzyme that catalyses BR elimination via glucuronidation is 514 not targeted to mitochondria. By contrast, the activity of BVR, the enzyme that catalyses BV 515 reduction to BR is increased in the mitochondria. Collectively, these observations and the fact 516 that the major metabolite of the CYP2A5-catalysed BR oxidation is BV, indicate that 517 oxidative stress creates local conditions in mitochondria for increased capacity of the BR- 518 BV-BR loop, which does not favour BR elimination. It is plausible that the loop is associated 519 with heightened mitochondrial antioxidant capacity as indicated by reduction of 520 mitochondrial proteins carbonylation and lack of cytochrome c leakage into the cytoplasm. 521 It is essential to note the different purposes of the three different metabolic routes of 522 BR. Firstly, BR conversion to BV by the CYP2A5 does not aim at its elimination, instead to 523 reduce its toxicity and to keep it in the cells for later use as supported by previous 524 observations (Abu-Bakar et al., 2011; Abu-Bakar et al., 2012). Secondly, the purpose of 525 random BR oxidation by ROS to dipyrroles — which are excreted in the urine (Arthur et al., 526 2012; Kobayashi et al., 2003) — is to protect cells and mitochondria against oxidative 527 damage. Lastly, the well-known UGT1A1-catalysed glucuronidation pathway is important in 528 the clearance of excessive amounts of circulatory BR (McDonagh, 2010) and is not involved 529 in regulating minute intracellular levels of BR. The absence of this pathway in mitochondria 530 further emphasise the importance of BR oxidation in managing intracellular BR. It is thus 531 plausible that the random BR oxidation by ROS protects cellular structures against oxidative 532 stress, whilst the enzymatic BR oxidation protects cells against excessive intracellular BR — 533 which mechanism predominates would depend on the levels of intracellular ROS and BR. 534 In the case of mitochondria our observations suggest that under normal condition BR 535 oxidation by ROS predominates that of the enzymic oxidation, presumably to protect 536 mitochondria against ROS-mediated damage. Under oxidative stress conditions — when

537 local BR production is heightened due to increased HMOX1 activity — the ROS and 538 enzymic oxidation of BR is equally important. This indicates that in a risk situation of 539 structural damage, recruitment of microsomal HMOX1 and CYP2A5, and cytosolic BVR to 540 mitochondria elevates local cytoprotective capacity to confer protection against elevated ROS 541 and the potential toxic effects of excessive BR. This proposition is supported by our 542 observations that mitochondrial proteins carbonylation was significantly reduced and that 543 mitochondrial cytochrome c was not released to the cytosol in treated animals. 544 Cytochrome c is tightly anchored in the inner membrane by binding to cardiolipin 545 (Cortese et al., 1998; Gallet et al., 1997; Gorbenko, 1999; Rytomaa et al., 1995). Thus, it has 546 been proposed that the release of this protein into the cytosol involves its detachment from 547 the inner mitochondrial membrane by disruption of the cardiolipin–cytochrome c complex, 548 which is followed by permeabilisation of the outer membrane and the release of cytochrome c 549 into the cytoplasm (Ott et al., 2002). In a study with rat liver mitochondria, Ott et al (2002) 550 demonstrated that permeabilization of the outer membrane or disruption of the cardiolipin– 551 cytochrome c complex, alone, does not sufficiently trigger the release of cytochrome c. 552 Additionally, the observations that cardiolipin confers fluidity and stability of the inner 553 membrane (Lesnefsky et al., 2001; Ostrander et al., 2001; Paradies et al., 2000) and that 554 oxidation of cardiolipin by mitochondrial ROS solubilises cytochrome c (Kagan et al., 2005; 555 Ott et al., 2002; Shidoji et al., 1999), suggest that damage to the inner membrane is required 556 for the release of cytochrome c. Accordingly, our results indicate that the inner membrane has 557 not been severely damaged to cause the release of cytochrome c to the cytoplasm. It is thus 558 plausible that induced recruitment of key BR regulatory enzymes to mitochondria occurred 559 prior to severe membrane damage. This proposition is supported by the observed elevated 560 nucleo-Nrf2 levels, which indicates initiation of a downstream cytoprotective mechanism. 561 Indeed, Nrf2 has recently been implicated as a new therapeutic target for the treatment of 562 liver fibrosis (review in Yang et al., 2013). 563 Importantly, enzymic bilirubin-oxidizing activity has been described earlier in the liver 564 and brain mitochondria, and has been suggested as the major metabolic pathway for 565 mitochondrial BR but the enzyme(s) involved has not be identified (Cardenas-Vazquez et al., 566 1986). Our present findings confirmed the earlier observations and identified CYP2A5 as the 567 mitochondrial BR oxidase. 568 We surmise that our preliminary data clearly demonstrate that the recruitment of key 569 BR regulatory enzymes into mitochondria is an adaptive response to PYR-induced oxidative 570 stress, potentially to protect mitochondria against severe oxidative damage. The placement of

571 a local system to regulate mitochondrial BR homeostasis is perhaps to protect the organelles 572 against ROS generated by mitochondrial respiration and from exposure to toxic chemicals. 573 Our future studies aim at further understanding the role of mitochondrial BR regulating 574 enzyme system in cytoprotection and in mitochondria associated diseases. 575 576 Conflict of interest statement 577 The authors declare that they have no conflict of interest in this work. 578 579 Acknowledgements 580 The authors would like to thank Dr. Mahmoud Shahin for his committed participation 581 in animal care and treatment, Mr. Geoff Eaglesham for his assistance with HPLC-MS/MS 582 analysis of the bilirubin oxidative metabolites, and Dr. Sergej Usanov, National Academic of 583 Sciences, Republic of Belarus, for the provision of the anti-adrenodoxin reductase antibody 584 This paper is partly based on Siti Nur Fadzilah Muhsain’s (SNFM) doctoral research funded 585 by the University of Queensland and the Universiti Teknologi Mara, Malaysia. The National 586 Research Centre for Environmental Toxicology is a partnership between Queensland Health 587 and the University of Queensland. 588 589 References 590 Abu-Bakar, A., Arthur, D.M., Aganovic, S., Ng, J.C., Lang, M.A., 2011. Inducible bilirubin 591 oxidase: A novel function for the mouse cytochrome P450 2A5. Toxicol. Appl. 592 Pharmacol. 257, 14-22. 593 Abu-Bakar, A., Arthur, D.M., Wikman, A.S., Rahnasto, M., Juvonen, R.O., Vepsalainen, J., 594 Raunio, H., Ng, J.C., Lang, M.A., 2012. Metabolism of bilirubin by human cytochrome 595 P450 2A6. Toxicol. Appl. Pharmacol. 261, 50-58. 596 Abu-Bakar, A., Hakkola, J., Juvonen, R., Rahnasto-Rilla, M., Raunio, H., Lang, M.A., 2013. 597 Function and regulation of the Cyp2a5/CYP2A6 genes in response to toxic insults in the 598 liver. Curr. Drug Metab. 14, 137-150. 599 Abu-Bakar, A., Moore, M.R., Lang, M.A., 2005. Evidence for induced microsomal bilirubin 600 degradation by cytochrome P450 2A5. Biochem. Pharmacol. 70, 1527-1535. 601 Abu-Bakar, A., Satarug, S., Marks, G.C., Lang, M.A., Moore, M.R., 2004. Acute cadmium 602 chloride administration induces hepatic and renal CYP2A5 mRNA, protein and activity 603 in the mouse: Involvement of transcription factor NRF2. Toxicol. Letts. 148, 199-210. 604 Addya, S., Anandatheerthavarada, H.K., Biswas, G., Bhagwat, S.V., Mullick, J., Avadhani, 605 N.G., 1997. Targeting of NH2-terminal-processed microsomal protein to mitochondria: 606 a novel pathway for the biogenesis of hepatic mitochondria P450MT2. J. Cell Biol. 607 139, 589-599. 608 Anandatheerthavarada, H.K., Biswas, G., Mullick, J., Sepuri, N.B., Otvos, L., Pain, D., 609 Avadhani, N.G., 1999. Dual targeting of cytochrome P4502B1 to endoplasmic 610 reticulum and mitochondria involves a novel signal activation by cyclic AMP- 611 dependent phosphorylation at ser128. EMBO J. 18, 5494-5504.

612 Anandatheerthavarada, H.K., Sepuri, N.B.V., Avadhani, N.G., 2009. Mitochondrial targeting 613 of cytochrome P450 proteins containing NH2-terminal chimeric signals involves an 614 unusual TOM20/TOM22 bypass mechanism. J. Biol. Chem. 284, 17352-17363. 615 Arthur, D.M., Ng, J.C., Lang, M.A., Abu-Bakar, A., 2012. Urinary excretion of bilirubin 616 oxidative metabolites in arsenite-treated mice. J. Toxicol. Sci. 37, 655-661. 617 Asikainen, A., Tarhanen, J., Poso, A., Pasanen, M., Alhava, E., Juvonen, R.O., 2003. 618 Predictive value of comparative molecular field analysis modelling of naphthalene 619 inhibition of human CYP2A6 and mouse CYP2A5 enzymes. Toxicol. In Vitro. 17, 449- 620 455. 621 Bae, S.H., Sung, S.H., Lee, H.E., Kang, H.T., Lee, S.K., Oh, S.Y., Woo, H.A., Kil, I.S., 622 Rhee, S.G., 2012. Peroxiredoxin III and sulfiredoxin together protect mice from 623 pyrazole-induced oxidative liver injury. Antioxid. Redox Signal. 17, 1351-1361. 624 Bansal, S., Biswas, G., Avadhani, N.G., 2014. Mitochondria-targeted heme oxygenase-1 625 induces oxidative stress and mitochondrial dysfunction in macrophages, kidney 626 fibroblasts and in chronic alcohol hepatotoxicity. Redox Biol. 2, 273-283. 627 Brodersen, R. & Bartels, P., 1969. Enzymatic oxidation of bilirubin. Eur. J. Biochem. 10, 628 468-473 629 Cadenas, E., Davies, K.J., 2000. Mitochondrial free radical generation, oxidative stress, and 630 aging. Free Radic. Biol. Med. 29, 222-230. 631 Cardenas-Vazquez, R., Yokosuka, O., Billing, B.H., 1986. Enzymic oxidation of 632 unconjugated bilirubin by rat liver. Biochem. J .236, 625-633. 633 Cederbaum, A.I., 2013. Nrf2 and antioxidant defense against CYP2E1 toxicity. Suncell. 634 Biochem. 67, 105-30. 635 Chernogolov, A.A., Usanov, S.A., 1997. Immunochemical characterization of steroid 636 hydroxylases of adrenocortical mitochondria. 1. Antibodies as inhibitors of cytochrome 637 P450scc (CYP11A1) and P450(11 beta) (CYP11B1) activities. Biochem. Biokhim. 62, 638 1375-1384. 639 Converso, D.P., Taille, C., Carreras, M.C., Jaitovich, A., Poderoso, J.J., Boczkowski, J., 640 2006. HO-1 is located in liver mitochondria and modulates mitochondrial heme content 641 and metabolism. FASEB J. 20, 1236-1238. 642 Cortese, J.D., Voglino, A.L., Hackenbrock, C.R., 1998. Multiple conformations of 643 physiological membrane-bound cytochrome c. Biochemistry. 37, 6402-6409. 644 Dalle-Donne, I., Rossi, R., Giustarini, D., Milzani, A., Colombo, R., 2003. Protein carbonyl 645 groups as biomarkers of oxidative stress. Clin. Chim. Acta. 329, 23-38. 646 De Matteis, F., Lord, G.A., Kee Lim, C., Pons, N., 2006. Bilirubin degradation by uncoupled 647 cytochrome P450. Comparison with a chemical oxidation system and characterization 648 of the products by high-performance liquid chromatography/electrospray ionization 649 mass spectrometry. Rapid Commun. Mass Spectrom. 20, 1209-1217. 650 Draper, A.J., Madan, A., Parkinson, A., 1997. Inhibition of coumarin 7-hydroxylase activity 651 in human liver microsomes. Arch. Biochem. Biophys. 341, 47-61. 652 Fagan, J.M., Sleczka, B.G., Sohar, I., 1999. Quantitation of oxidative damage to tissue 653 proteins. Int. J. Biochem. Cell Biol. 31, 751-757. 654 Gallet, P.F., Petit, J.M., Maftah, A., Zachowski, A., Julien, R., 1997. Asymmetrical 655 distribution of cardiolipin in yeast inner mitochondrial membrane triggered by carbon 656 catabolite repression. Biochem. J. 324, 627-634. 657 Genter, M.B., Clay, C.D., Dalton, T.P., Dong, H., Nebert, D.W., Shertzer, H.G., 2006. 658 Comparison of mouse hepatic mitochondrial versus microsomal cytochrome P450 659 following TCDD treatment. Biochem. Biophys. Res. Commun. 342, 1375-1381.

660 Gilmore, W.J., Kirby, G.M., 2004. Endoplasmic reticulum stress due to altered cellular redox 661 status positively regulates murine hepatic CYP2A5 expression. J. Pharmacol. Exp. 662 Ther. 308, 600-608. 663 Gorbenko, G.P., 1999. Structure of cytochrome c complexes with phospholipids as revealed 664 by resonance energy transfer. Biochim. Biophys. Acta. 1420, 1-13. 665 Guzov, V.M., Zel'ko, I.N., Chernogolov, A.A., Usanov, S.A., 1993. Interrelationship of 666 microsomal and mitochondrial systems for hydroxylating steroids from cattle adrenal 667 cortex: effect of highly-purified components of the mitochondrial monoxygenase 668 system on hydroxylation of steroids by microsomes. Biokhimiia. 58, 1761-1770. 669 Honkakoski, P., Kojo, A., Raunio, H., Pasanen, M., Juvonen, R., Lang, M.A., 1988. Hepatic 670 mitochondrial coumarin 7-hydroxylase: Comparison with the microsomal enzyme. 671 Arch. Biochem. Biophys. 267, 558-567. 672 Honkakoski, P., Lang, M.A., 1989. Mouse liver phenobarbital-inducible P450 system: 673 Purification, characterization, and differential inducibility of four cytochrome P450 674 isozymes from D2 mouse. Arch. Biochem. Biophys. 273, 42-57. 675 Honkakoski, P., Maenpaa, J., Leikola, J., Pasanen, M., Juvonen, R., Lang, M.A., Pelkonen, 676 O., Raunio, H., 1993. Cytochrome P4502A-mediated coumarin 7-hydroxylation and 677 testosterone hydroxylation in mouse and rat lung. Pharmacol. Toxicol. 72, 107-112. 678 Jansen, T., Hortmann, M., Oelze, M., Opitz, B., Steven, S., Schell, R., Knorr, M., Karbach, 679 S., Schuhmacher, S., Wenzel, P., Munzel, T., Daiber, A., 2010. Conversion of 680 biliverdin to bilirubin by biliverdin reductase contributes to endothelial cell protection 681 by heme oxygenase-1: evidence for direct and indirect antioxidant actions of bilirubin. 682 J. Mol. Cell Cardiol. 49, 186-195. 683 Johnson, G.D., 1953. Correlation of color and constitution. I. 2,4-dinitrophenylhydrozones. J. 684 Am. Chem. Soc. 75, 2720-2723. 685 Juvonen, R.O., Shkumatov, V.M., Lang, M.A., 1988. Purification and characterization of a 686 liver microsomal cytochrome P-450 isoenzyme with a high affinity and metabolic 687 capacity for coumarin from pyrazole-treated D2 mice. Eur. J. Biochem. 171, 205-211. 688 Kagan, V.E., Tyurin, V.A., Jiang, J., Tyurina, Y.Y., Ritov, V.B., Amoscato, A.A., Osipov, 689 A.N., Belikova, N.A., Kapralov, A.A., Kini, V., Vlasova, I.I., Zhao, Q., Zou, M., Di, P., 690 Svistunenko, D.A., Kurnikov, I.V., Borisenko, G.G., 2005. Cytochrome c acts as a 691 cardiolipin oxygenase required for release of proapoptotic factors. Nat. Chem. Biol. 1, 692 223-232. 693 Kaipainen, P., Nebert, D.W., Lang, M.A., 1984. Purification and characterization of a 694 microsomal cytochrome P-450 with high activity of coumarin 7-hydroxylase from 695 mouse liver. Eur. J. Biochem. 144, 425-431. 696 Kinonen, T., Pasanen, M., Gynther, J., Poso, A., Jarvinen, T., Alhava, E., Juvonen, R., 1995. 697 Competitive inhibition of coumarin 7-hydroxylation by pilocarpine and its interaction 698 with mouse CYP2As and human CYP2A6. Br. J. Pharmacol. 116, 2625-2630. 699 Kobayashi, A., Takahashi, T., Sugai, S., Miyakawa, Y., Iwatsuka, H., Yamaguchi, T., 2003. 700 Urinary excretion of oxidative metabolites of bilirubin in fenofibrate-treated rats. J 701 Toxicol. Sci. 28, 71-75. 702 Lesnefsky, E.J., Slabe, T.J., Stoll, M.S., Minkler, P.E., Hoppel, C.L., 2001. Myocardial 703 ischemia selectively depletes cardiolipin in rabbit heart subsarcolemmal mitochondria. 704 American J Physiol. 280, H2770-2778. 705 Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with 706 the Folin phenol reagent. J. Biol. Chem. 193, 265-275. 707 Lu, Y., Gong, P., Cederbaum, A.I., 2008. Pyrazole induced oxidative liver injury independent 708 of CYP2E1/2A5 induction due to Nrf2 deficiency. Toxicol. 252, 9-16.

709 McDonagh, A.F., 2010. Controversies in bilirubin biochemistry and their clinical relevance. 710 Semin. Fetal Neonatal Med. 15, 141-147. 711 Ostrander, D.B., Zhang, M., Mileykovskaya, E., Rho, M., Dowhan, W., 2001. Lack of 712 mitochondrial anionic phospholipids causes an inhibition of translation of protein 713 components of the electron transport chain. A yeast genetic model system for the study 714 of anionic phospholipid function in mitochondria. J. Biol. Chem. 276, 25262-25272. 715 Ott, M., Robertson, J.D., Gogvadze, V., Zhivotovsky, B., Orrenius, S., 2002. Cytochrome c 716 release from mitochondria proceeds by a two-step process. PNAS. 99, 1259-1263. 717 Paradies, G., Petrosillo, G., Pistolese, M., Ruggiero, F.M., 2000. The effect of reactive 718 oxygen species generated from the mitochondrial electron transport chain on the 719 cytochrome c oxidase activity and on the cardiolipin content in bovine heart 720 submitochondrial particles. FEBS Letts. 466, 323-326. 721 Radominska-Pandya, A., Ouzzine, M., Fournel-Gigleux, S., Magdalou, J., 2005. Structure of 722 UDP-glucuronosyltransferases in membranes. Methods Enzymol. 400, 116-147. 723 Reznick, A.Z., Packer, L., 1994. Oxidative damage to proteins: Spectrophotometric method 724 for carbonyl assay. Methods Enzymol. 233, 357-363. 725 Rodrigues, C.M., Sola, S., Brito, M.A., Brites, D., Moura, J.J., 2002. Bilirubin directly 726 disrupts membrane lipid polarity and fluidity, protein order, and redox status in rat 727 mitochondria. J. Hepatol. 36, 335-341. 728 Ryter, S.W., Kvam, E., Tyrrell, R.M., 2000. Heme oxygenase activity. Current methods and 729 applications. Methods Mol. Biol. 99, 369-391. 730 Rytomaa, M., Kinnunen, P.K., 1995. Reversibility of the binding of cytochrome c to 731 liposomes. Implications for lipid-protein interactions. J. Biol. Chem. 270, 3197-3202. 732 Sanchez-Alcazar, J.A., Ault, J.G., Khodjakov, A., Schneider, E., 2000. Increased 733 mitochondrial cytochrome c levels and mitochondrial hyperpolarization precede 734 camptothecin-induced apoptosis in Jurkat cells. Cell Death Diff. 7, 1090-1100. 735 Shidoji, Y., Hayashi, K., Komura, S., Ohishi, N., Yagi, K., 1999. Loss of molecular 736 interaction between cytochrome c and cardiolipin due to lipid peroxidation. Biochem. 737 Biophys. Res. Comm. 264, 343-347. 738 Srivastava, P., Pandey, V.C., 1996. Mitochondrial heme oxygenase of Mastomys coucha. Int 739 J. Biochem. Cell Biol. 28, 1071-1077. 740 Tenhunen, R., Ross, M.E., Marver, H.S., Schmid, R., 1970. Reduced nicotinamide-adenine 741 dinucleotide phosphate dependent biliverdin reductase: partial purification and 742 characterization. Biochemistry 9, 298-303. 743 Tukey, R.H., and Strassburg, C.P., 2000. Human UDP-glucuronosyltransferases: metabolism, 744 expression, and disease. Annu. Rev. Pharmacol. Toxicol. 40, 581-616. 745 Usanov, S.A., Chernogolov, A.A., Honkakoski, P., Lang, M., Passanen, M., Raunio, H., 746 Pelnonen, O., 1990. Cholesterol-hydroxylating cytochrome P-450 from bovine adrenal 747 cortex mitochondria and human placenta: immunochemical properties and structural 748 characteristics. Biokhimiia 55, 865-877. 749 Usanov, S.A., Honkakoski, P., Lang, M.A., Pasanen, M., Pelkonen, O., Raunio, H., 1989. 750 Comparison of the immunochemical properties of human placental and bovine adrenal 751 cholesterol side-chain cleavage enzyme complex. Biochim. Biophys. Acta. 998, 189- 752 195. 753 Usanov, S.A., Turko, I.V., Akhrem, A.A., Chashchin, V.L., 1984. Immunochemical 754 characteristics of cholesterol-hydroxylating cytochrome P-450 from adrenal cortex 755 mitochondria. Biokhimiia 49, 1810-1818. 756 Wang, T.Y., 1967. The Isolation and purification of mammalian cell nuclei. Methods 757 Enzymol. 12, 417-421.

758 Yang, J.J., Tao, H., Huang, C., Li, J., 2013. Nuclear erythroid 2-related factor 2: a novel 759 therapeutic target for liver fibrosis. Food Chem. Toxicol. 59, 421-427. 760 Zangar, R.C., Davydov, D.R., Verma, S. , 2004. Mechanisms that regulate production of 761 reactive oxygen species by cytochrome P450. Toxicol. Appl. Pharmacol. 199, 316-331.

Table 1

762 Table 1: Activities of coumarin 7-hydroxylase, haem oxygenase and biliverdin reductase in 763 the livers of control and PYR-treated mice. 764 Enzyme Cytosol Microsomes Mitoplasts Activity CTR PYR CTR PYR CTR PYR COH# ND ND 743 ± 99 7329 ± 500* 10 ± 1 622 ± 53* HMOX## ND ND 260 ± 23 964 ± 63* 83 ± 10 295 ± 30* BVR## 30 ± 2 36 ± 10 ND ND 24 ± 5 45 ± 1* 765 766 CTR = control; PYR = pyrazole. #Coumarin 7-hydroxylase activity; pmol 767 umbelliferone/min/mg protein; ##Haem oxygenase (HMOX) and biliverdin reductase (BVR) 768 activities; pmol BR/min/mg protein. CTR, control; PYR, pyrazole. The values are of 3 769 separate pools each consisting of 2 animals. Values were normalized against percentage of 770 microsomal contaminations. *Mean difference is significant from control group at p < 0.05 771 (Student t-test).

Figure Captions

772 Figure legends

773 Figure 1: Estimation of the purity of microsomes and mitoplasts. Mice (N = 12) were 774 administered 200 mg pyrazole/kg/day for three consecutive days or normal saline. Twenty- 775 four hours after the final dose, liver mitoplasts and microsomal fractions were prepared. (A) 776 After SDS-PAGE electrophoresis, gels were blotted on PVDF membranes, which were then 777 probed with antibodies specific for either endoplasmic reticulum (NPR) or mitoplasts 778 (COXIV) marker proteins. Each blot represents one of three blots (each blot showed the same 779 pattern of induction). The amounts of protein (µg) loaded on the gels are indicated. Positive 780 control (+) = NPR and COXIV peptides. (B) The activities of NPR and COXIV in mitoplasts 781 and microsomes from livers of PYR-treated and control mice. CTR, control; PYR, pyrazole. 782 The values are of 3 separate pools each consisting of 2 animals and were normalized against 783 percentage of microsomal contaminations. *Mean difference is significant from control group 784 at p < 0.05. 785 786 Figure 2: Effects of pyrazole treatment on biomarkers of oxidative stress. Mice (N = 12) 787 were administered 200 mg pyrazole/kg/day for three consecutive days or normal saline. 788 Twenty-four hours after the final dose, liver cytosol, microsomes, mitoplasts, and nuclei were 789 isolated. (A) Protein carbonyl contents in cytosol, microsomes and mitoplasts of control and 790 treated livers. (B) Nrf2 antibody-immunoreactive protein levels in cytosol and nuclei. 791 Electrophoresis was performed with 20 µg cytosolic and nuclei protein. The proteins were 792 blotted and probed with anti-Nrf2 antibody (1:500 dilutions). Shown are representative of 793 three blots (each blot showed the same pattern of induction). Nuclear extracts from HeLa

794 cells treated with 50 μM CdCl2 for 4, whole cell lysates from HeLa cells, and β-Actin peptide 795 were used as positive control (+) for Nrf2, Lamin B1 and β-actin, respectively. β-Actin and 796 Lamin B1 served as loading control for cytosolic and nuclei proteins, respectively. (C) 797 Densitometry analysis of the blots probed with the various antibodies. Relative density = 798 density of signals from protein of interest relative to density of loading control signals. CTR, 799 control; PYR, pyrazole. The values are means of 3 separate pools each consisting of 2 800 animals and were normalized against percentage of microsomal contaminations.*Mean 801 difference is significant from control group at p < 0.05. #Mean difference is significant from 802 cytosol and microsomes control groups at p < 0.05 803

804 Figure 3: Effects of pyrazole treatment on induction of BR regulatory enzymes. Mice (N = 805 12) received doses of PYR (200 mg/kg) on three consecutive days. Twenty-four hours after 806 the final dose, liver cytosolic, microsomal and mitoplasts fractions were prepared. (A) 807 CYP2A5, HMOX1, BVR and UGT1A1 antibody-immunoreactive protein levels in cytosol, 808 microsomes and mitoplasts. The amount of protein loaded on the gel for each subcellular 809 fraction was 25 μg. Each blot represents one of three blots (each blot showed the same pattern 810 of induction). Positive controls (+) = CYP2A5 recombinant protein and peptides of HMOX1, 811 UGT1A1 and BVR. (B) Densitometry analysis of the blots probed with the various 812 antibodies. Relative density = density of signals from protein of interest relative to density of 813 loading control signals. CTR, control; PYR, pyrazole. The values are of 3 separate pools each 814 consisting of 2 animals and were normalized against percentage of microsomal 815 contaminations. *Mean difference is significant from control group at p < 0.05. 816 817 Figure 4: Effect of pyrazole treatment on microsomal and mitochondrial BR disappearance 818 rates. (A) Liver microsomes and mitoplasts (10 µg protein) were incubated in reaction 819 mixtures containing 1.5 mM NADPH. The reaction was initiated by addition of 20 µM BR 820 and absorbance at 440 nm was recorded for 20 cycles at 1 min/cycle. The rate of BR 821 disappearance is expressed as pmol BR disappearing/min/mg protein, using a εmM=41.15 822 cm−1, which was obtained experimentally under the conditions of the assay. All experiments 823 were done in the dark and 2 mM (final concentration) of EDTA was added in all incubation 824 mixtures. (B) Inhibition of microsomal and mitochondrial BR disappearance rate by 200 mM 825 vitamin C. CTR, control; PYR, pyrazole. The values are of 3 separate pools each consisting 826 of 2 animals and were normalized against percentage of microsomal contaminations. *Mean 827 difference is significant from control group at p < 0.05 (Student t-test). 828 829 Figure 5: Summary of the oxidative products of bilirubin formed in the chemical and

830 enzymic systems. (A) Chemical oxidation of BR by the Fe-EDTA/H2O2 system produced the 831 putative dipyrroles. (B) Bilirubin incubations with microsomes or mitoplasts produced both 832 biliverdin and the putative dipyrroles. *Significant different from control group, p<0.05 833 (Student t-test). 834 835 Figure 6: Effects of anti-CYP2A5 and anti-AdxR antibodies on bilirubin degradation activity 836 in microsomes and mitoplasts. (A) Inhibition of bilirubin degradation activity in control and 837 treated microsomes in the presence of anti-CYP2A5 antibody (▪) and in the presence of serum

838 IgG (▫). (B) Inhibition of bilirubin degradation activity in control and treated mitoplasts in the 839 presence of anti-CYP2A5 antibody (●) and in the presence of serum IgG (○). The antibody 840 was added in increasing amount so as to achieve the antibody to microsomal/mitochondrial 841 protein concentration ratios indicated. (C) Treated microsomes or mitoplasts were incubated 842 with BR in the presence of absence of anti-AdxR antibody. The antibody was added in 843 increasing amount so as to achieve the antibody to microsomal/mitochondrial protein 844 concentration ratios indicated. CTR = control; PYR = pyrazole; AdxR, adrenodoxin 845 reductase. The values are of 3 separate pools each consisting of 2 animals. 846 847 Figure 7: Effect of bilirubin on 7-hydroxylation of coumarin in mitochondria. (A) 848 Mitoplasts were incubated with various concentrations of coumarin in the presence of 849 increasing amounts of BR. The rates of coumarin 7-hydroxylation were determined by 850 fluorescence spectroscopy. The formation of 7-hydroxycoumarin by mitochondrial CYP2A5

851 followed simple Michaelis–Menten kinetics. (B) BR increased the Km and reduced the Vmax

852 of coumarin 7-hydroxylation. (C) Km observed vs BR concentrations plot indicates that BR is a

853 competitive inhibitor with Ki = 2.67 μM. 854 855 Figure 8: Effects of pyrazole on cytochrome c protein expression. (A) Western blot analysis 856 of cytosolic and mitoplasts proteins from livers of control and pyrazole-treated DBA/2J mice 857 probed with anti-cytochrome c antibody. Cytochrome c peptide was used as a positive control 858 (+). (B) Densitometric analysis of cytochrome c blots. The means ± SD are of three separate 859 pools each consisting of two animals.

Figure 1 Click here to download high resolution image Figure 2 Click here to download high resolution image Figure 3 Click here to download high resolution image Figure 4 Click here to download high resolution image Figure 5 Click here to download high resolution image Figure 6 Click here to download high resolution image Figure 7 Click here to download high resolution image Figure 8 Click here to download high resolution image *Conflict of Interest Statement Abu-Bakar Click here to download Conflict of Interest Statement: COI_YTAAP_ABakar.doc *Conflict of Interest Statement Lang Click here to download Conflict of Interest Statement: COI_YTAAP_Lang.doc *Conflict of Interest Statement Muhsain Click here to download Conflict of Interest Statement: COI_Muhsain.pdf