The role of hydrogen sulphide in the cardiovascular system emphasis on haemorrhagic shock

Resuscitated, haemorrhagic-shocked rats had increased plasma concentrations of TNF-α, IL-1β, IL-6, IL-10, ALT, urea, creatinine, amylase, lactate and nitrate/nitrite, increased liver and

ROLE OF HYDROGEN SULPHIDE IN THE

CARDIOVASCULAR SYSTEM:

EMPHASIS ON HAEMORRHAGIC SHOCK

MOK YING-YUAN, PAMELA

(B.Sc(Hons.), NUS)

A THESIS SUMBITTED FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

2007

I

Acknowledgements

I would like to express my sincere appreciation and gratitude to my supervisor, Professor Philip K Moore for his invaluable guidance, constant encouragement and support throughout these 4 years of my PhD studies Without his mentorship, I would not have finished the project

I would like to thank my laboratory officer Yoke Ping for her efficiency and expedience in purchasing laboratory supplies, and my co-workers Farhana, Yusuf, Li Ling and Baskar for their invaluable help and technical assistance, and most importantly, their friendship

Many thanks to Associate Professor Madhav Bhatia and his laboratory staff and students,

as well as Associate Professor Zhu Yi-Zhun and his laboratory staff and students, for their technical assistance and the loan of laboratory equipment

My thanks to my family Without their understanding, support and forbearance throughout

my studies, I would not have achieved anything

Last but not least, I wish to thank the Agency for Science, Technology and Research for their award of a post-graduate scholarship, without which I would not have been able to pursue a PhD

II

CONTENTS

Acknowledgements I Contents II Summary VI List of figures VIII List of publications X Abbreviations XI

1 Introduction 1

1.1 Haemorrhagic shock 1

1.1.1 Compensatory phase 2

1.1.2 Decompensatory phase 3

1.1.3 Overview of hypoxic responses leading to inflammation 8

1.1.4 Inflammatory mediators of haemorrhagic shock 10

1.1.5 Reperfusion injury – role of ROS/RNS 22

1.2 Hydrogen sulphide 24

1.2.1 Biosynthesis of H 2 S 24

1.2.2 Regulation of H 2 S biosynthesis 25

1.2.3 Metabolism of H 2 S 26

1.2.4 Mechanisms of action of H 2 S 27

1.2.5 H 2 S and the cardiovascular system 28

1.2.6 H 2 S and inflammation 32

1.2.7 H 2 S and the nervous system 35

1.2.8 H 2 S and the gastrointestinal system 36

1.2.9 H 2 S and diabetes mellitus 38

1.3 Premise for the current thesis 39

2 Drugs, chemicals, materials & equipment 40

2.1 Drugs and Chemicals 40

III

2.2 Materials 41

2.3 Equipment 41

3 Methodology 44

3.1 Experiments measuring blood pressure 44

3.1.1 PowerLab calibration 44

3.1.2 Surgical cannulation 45

3.1.3 Haemorrhagic shock & resuscitation 47

3.1.4 Drug administration 48

3.1.5 Blood sampling & organ harvesting 48

3.2 Organ bath experiments 49

3.2.1 PowerLab calibration 49

3.2.2 Aortic ring preparation 50

3.2.3 Measurement of rat aorta contractility 51

3.3 Biochemical assays 52

3.3.1 Plasma cytokines 52

3.3.2 Plasma organ injury markers 54

3.3.3 Nitrate / nitrite 55

3.3.4 Myeloperoxidase activity 56

3.3.5 Inducible nitric oxide synthase protein 58

3.3.6 Hydrogen sulphide 61

3.3.7 Cystathionine γ-lyase mRNA 63

3.4 Statistical analysis and representation 65

4 The interaction of H 2 S with K ATP channels and NO in the vascular system 66

4.1 Introduction 66

4.2 Experimental design 67

4.2.1 Drug preparation 67

4.3 Results 68

4.3.1 Effect of NaHS on rat aorta contractility: Role of K ATP channels 68

4.3.2 Effect of NaHS / H 2 S on blood pressure & heart rate 73

4.3.3 Role of K ATP channels in the cardiovascular effects of NaHS / H 2 S 78

IV

4.3.4 Role of NO and prostaglandins in NaHS-induced vasorelaxation 79

4.3.5 NaHS effect on rat aorta contractility: Role of NO 85

4.3.6 Role of NO and K ATP channels in NaHS-induced increase in blood pressure 86

4.4 Discussion 92

4.4.1 NaHS-induced relaxation of isolated rat aorta 92

4.4.2 NaHS-induced contraction of isolated rat aorta 94

4.4.3 NaHS-induced fall in mean arterial pressure in the anaesthetized rat 96

4.4.4 NaHS-induced increase in mean arterial pressure in the anaesthetized rat 98

5 Changes in endogenous H 2 S production in haemorrhagic shock 101

5.1 Introduction 101

5.2 Results 102

5.2.1 Changes in H 2 S levels and biosynthesis in haemorrhagic shock 102

5.2.2 Effect of PAG & BCA on MAP & HR in haemorrhagic shock 106

5.2.3 Effect of PAG & BCA on H 2 S levels & biosynthesis in haemorrhagic shock 107

5.2.4 Effect of glibenclamide on MAP, HR & H 2 S levels in haemorrhagic shock 110

5.2.5 Effect of PAG and aminoguanidine on MAP and HR in haemorrhagic shock 111

5.2.6 Effect of PAG on organ damage & inflammation in haemorrhagic shock 111

5.3 Discussion 116

5.3.1 H 2 S formation in haemorrhagic shock 116

5.3.2 Use of CSE inhibitors in haemorrhagic shock 117

5.3.3 H 2 S and nitric oxide in haemorrhagic shock 118

5.3.4 Use of glibenclamide in haemorrhagic shock 119

5.3.5 H 2 S and inflammation in haemorrhagic shock 120

6 Implications of increased H 2 S production in haemorrhagic shock 123

6.1 Introduction 123

6.2 Experimental Design 124

6.3 Results 125

6.3.1 Effect of PAG on contractility of the isolated rat aorta 125

6.3.2 Effect of PAG on vasoconstrictor response 131

6.3.3 Effect of reinfusion of shed blood on MAP and HR of shocked rats 136

V

6.3.4 Effect of PAG on inflammatory mediators in haemorrhagic shock 136

6.3.5 Effect of PAG on plasma lactate & organ injury in haemorrhagic shock 143

6.4 Discussion 149

6.4.1 H 2 S & contractility of the isolated haemorrhagic-shocked rat aorta 149

6.4.2 H 2 S & cardiovascular responses in haemorrhagic shock 151

6.4.3 H 2 S & inflammation in haemorrhagic shock 153

6.4.4 H 2 S & haemorrhagic shock-induced organ injury 156

7 General discussion & summary 159

7.1 H2S and cardiovascular responses in haemorrhagic shock 159

7.2 H2S and inflammatory responses in haemorrhagic shock 160

7.3 Concluding remarks 162

8 References 163

VI

Summary

Haemorrhagic shock is a condition of reduced perfusion of organs resulting in inadequate delivery of oxygen and nutrients necessary for normal tissue and cellular function, due to an excessive loss of blood volume It is characterized by marked hypotension, vascular hyporesponsiveness to vasoconstrictors, and an inflammatory response Hydrogen sulphide (H2S) is a vasodilator and is endogenously produced in vascular tissues It has been implicated in cardiovascular disorders (e.g pulmonary hypertension) and is a mediator of inflammation Therefore, H2S may play a role in the aetiology of haemorrhagic shock

The cardiovascular effects of H2S were examined using sodium hydrogen sulphide (NaHS; H2S donor), henceforth referred to as H2S In accordance with published literature,

H2S relaxes phenylephrine pre-contracted aortic rings and causes a fall in mean arterial pressure (MAP) in anaesthetized rats However, the present study suggests that whilst H2S causes vasorelaxation by opening KATP channels in vitro, H2S causes a fall in MAP that is independent of KATP channels or the synthesis of nitric oxide (NO) and prostaglandin I2(PGI2) In addition, low concentrations/doses of H2S increases the tone of phenylephrine pre-contracted aortic rings, and causes a small but significant increase in MAP in anaesthetized rats This effect of H2S is possibly due to an interaction with NO (i.e NO

“quenching”), as H2S reverses acetylcholine and sodium nitroprusside (SNP)-induced relaxation of phenylephrine pre-contracted aortic rings In the healthy rat, H2S may not actively modulate vascular tone, as inhibition of H2S biosynthesis with D,L-propargylglycine (PAG) did not affect the MAP of anaesthetized rats

Haemorrhagic-shocked rats had increased plasma H2S levels and increased liver

H2S biosynthesis 1 h after blood withdrawal, both of which subsequently decreased to basal

VII

levels Increased liver H2S biosynthesis was likely due to upregulation of lyase (CSE) enzyme as haemorrhagic-shocked livers had increased liver CSE mRNA levels compared to non-shocked livers H2S partially mediates the hypotension in haemorrhagic shock, as PAG given both prophylactically and therapeutically could partially restore the lowered MAP Haemorrhagic-shocked aortic rings were hyporesponsive to phenylephrine

cystathionine-γ-as well cystathionine-γ-as to acetylcholine and SNP, and incubation with PAG further decrecystathionine-γ-ased the sensitivity to acetylcholine and SNP but did not affect the response to phenylephrine Thus,

H2S may increase vascular response to acetylcholine/NO in haemorrhagic shock Haemorrhagic shock was also associated with an increase in plasma concentrations of tumour necrosis factor-α (TNF-α), interleukin-1β (IL-1β), alanine aminotransferase (ALT) activity, urea and creatinine PAG-treated, haemorrhagic-shocked rats had lower plasma concentrations of IL-1β and ALT activity compared to saline/untreated rats

Resuscitated, haemorrhagic-shocked rats had increased plasma concentrations of TNF-α, IL-1β, IL-6, IL-10, ALT, urea, creatinine, amylase, lactate and nitrate/nitrite, increased liver and lung myeloperoxidase (MPO) activity, and lung inducible nitric oxide synthase (iNOS) protein Rats pre-treated with PAG had decreased plasma concentrations

of TNF-α, IL-6, ALT and nitrate/nitrite, reduced liver and lung MPO activity, and decreased lung iNOS protein compared to saline/untreated rats This suggest that H2S is pro-inflammatory in haemorrhagic shock and may contribute to haemorrhagic shock-associated multiple organ injury, thus inhibitors of H2S biosynthesis may be therapeutic

VIII

List of Figures

Figure 1.1 H2S production from L-cysteine 25

Figure 1.2 H2S metabolism 26

Figure 4.1 Effect of NaHS and cromakalim on pre-contracted isolated rat aortic rings 70

Figure 4.2 Effect of glibenclamide on pre-contracted isolated rat aortic rings 71

Figure 4.3 Effect of glibenclamide on NaHS and cromakalim-induced relaxation of pre-contracted isolated rat aortic rings 72

Figure 4.4 Representative traces showing drug-induced changes in arterial blood pressure 74

Figure 4.5 Representative traces showing drug-induced changes in heart rate 75

Figure 4.6 Effect of NaHS and H2S on MAP in the anaesthetized rat 76

Figure 4.7 Effect of NaHS and H2S on heart rate in the anaesthetized rat 77

Figure 4.8 Effect of glibenclamide on cromakalim-induced changes in MAP and heart rate 80

Figure 4.9 Effect of glibenclamide on NaHS-induced changes in MAP and heart rate 81

Figure 4.10 Effect of glibenclamide on H2S-induced changes in MAP and heart rate 82

Figure 4.11 Effect of glibenclamide on NaHS infusion-induced changes on MAP and

heart rate 83 Figure 4.12 Effect of glibenclamide on cromakalim-induced changes on MAP and heart rate .84

Figure 4.13 Effect of L-NAME and ODQ on NaHS-induced relaxation of pre-contracted isolated rat aortic rings Effect of NaHS on acetylcholine and SNP-induced relaxation of pre-contracted isolated rat aortic rings 88

Figure 4.14 Effect of glibenclamide on NaHS-induced increase in MAP and heart rate 89

Figure 4.15 Representative traces of arterial blood pressure changes caused by i.v injection of a mixture of SNP and NaHS 90

Figure 4.16 Effect of L-NAME pre-treatment on NaHS-induced increase in MAP and on heart rate 91 Figure 5.1 H2S biosynthesis in haemorrhagic shock 104

Figure 5.2 Liver CSE mRNA levels in haemorrhagic shock 105

IX

Figure 5.3 Effect of PAG and BCA on MAP and heart rate in haemorrhagic shock 108

Figure 5.4 Effect of PAG and BCA on plasma & tissue H2S 109

Figure 5.5 Effect of PAG, AG and both drugs on MAP and heart rate 113

Figure 5.6 Effect of PAG on inflammatory markers 114

Figure 5.7 Effect of PAG on organ injury markers .115

Figure 6.1 Effect of phenylephrine on pre-contracted isolated rat aortic rings 128

Figure 6.2 Effect of acetylcholine and SNP on pre-contracted isolated rat aortic rings 129

Figure 6.3 Effect of NaHS on pre-contracted isolated rat aortic rings 130

Figure 6.4 Effect of PAG on phenylephrine and angiotensin II-induced changes in MAP 133

Figure 6.5 Effect of PAG on phenylephrine and angiotensin II-induced increase in MAP 134

Figure 6.6 Effect of PAG on phenylephrine and angiotensin II-induced changes in heart rate 135

Figure 6.7 Effect of PAG on MAP in reinfused rats 137

Figure 6.8 Effect of PAG on heart rate in reinfused rats 138

Figure 6.9 Effect of PAG on plasma and organ NOX levels 140

Figure 6.10 Effect of PAG on plasma TNF-α and IL-1β 141

Figure 6.11 Effect of PAG on IL-6 and IL-10 142

Figure 6.12 Effect of PAG on plasma lactate levels 145

Figure 6.13 Effect of PAG on liver injury 146

Figure 6.14 Effect of PAG on kidney injury 147

Figure 6.15 Effect of PAG on lung and pancreatic injury 148

X

List of Publications

1 Mok, Y.Y., Atan, M.S., Yoke Ping, C., Zhong Jing, W., Bhatia, M., Moochhala, S., Moore, P.K., 2004 Role of hydrogen sulphide in haemorrhagic shock in the rat: protective effect of inhibitors of hydrogen sulphide biosynthesis Br J Pharmacol 143: 881-889

2 Mok, Y.Y., Moore, P.K Hydrogen sulphide (H2S) plays a role in haemorrhagic shock in the rat The 7th World Congress on Inflammation, Melbourne, Australia,

2005

3 Mok, Y.Y., Moore, P.K Role of H2S in the cardiovascular system and its possible interaction with NO The 15th World Congress of Pharmacology, Beijing, China,

2006

4 Mok, Y.Y., Moore, P.K Hydrogen sulphide is pro-inflammatory in haemorrhagic

shock Submitted to Inflammation Research

CINC Cytokine-induced neutrophil chemoattractant

eNOS Endothelial nitric oxide synthase

ICAM Intercellular cell adhesion molecule

iNOS Inducible nitric oxide synthase

L-NAME Nω-nitro-L-arginine methyl ester

XII

TNF-α Tumour necrosis factor-α

TRPV Transient receptor potential vanilloid

1.1 Haemorrhagic shock

Haemorrhagic shock is a condition of reduced perfusion of vital organs resulting in inadequate delivery of oxygen and nutrients necessary for normal tissue and cellular function, due to an excessive loss of blood volume This may be due to trauma, or spontaneous haemorrhage such as bleeding from the gastrointestinal tract (e.g non-steroidal anti-inflammatory drug (NSAID)-induced gastric injury) Blood loss leads to a

of arterioles and capacitance veins that leads to shunting of blood from the venous system

to the arterial system, an increase in cardiac pre-load and an increase in peripheral (systemic) resistance, all of which serves to increase the mean arterial pressure (Bond and Johnson, 1985; Tornebrandt et al, 1985)

In addition, the increase in sympathetic activity also causes an increase in angiotensin II levels via activation of the RAAS (Jakschik et al, 1974) Hypovolaemia also stimulates the release of renin (Zehr et al, 1980) Angiotensin II causes vasoconstriction directly by stimulating angiotensin II subtype 1 receptors on vascular smooth muscle (Nishimura et al, 1982; Zhang et al, 1994) and also indirectly through stimulation of the

Introduction

3

sympathetic nervous system (Clough et al, 1980; Hilgers et al, 1993; Seidelin et al, 1991) leading to an increase in both arterial and venous tone Angiotensin-stimulated release of aldosterone from the adrenal cortex (Aguilera, 1992; Balla et al, 1991) increases salt and water reabsorption from the kidney (Stumpe and Ochwadt, 1968), as well as increases thirst, both of which serve to increase blood volume and hence increase blood pressure An increase in sympathetic activity also causes the release of epinephrine from the adrenal medulla that in turn acts on β1-adrenergic receptors in the heart to reinforce the sympathetic-induced increase in heart rate and myocardial contractility (Van Loon et al, 1981) Sympathetic-induced release of vasopressin also occurs from the posterior pituitary gland, which directly causes vasoconstriction via activation of vasopressin receptors present on vascular smooth muscle cells (Caramelo et al, 1989; Fox et al, 1987; Gopalakrishnan et al, 1991; Hirsch et al, 1987; Penit et al, 1983) The decrease in tissue perfusion causes hypoxia and leads to anaerobic metabolism and subsequent acidosis Both hypoxia and acidosis trigger chemoreceptors that in turn further increase sympathetic activity (Chalmers et al, 1967; Halliwill et al, 2003; Korner and White, 1966; Rose et al, 1983)

The compensatory phase of haemorrhagic shock progresses into a state of vascular decompensation (decompensatory phase), which is characterized by a progressive vasodilatation and a continuous decrease in peripheral vascular resistance despite an increase in sympathetic activity This ultimately results in death and the transition to the decompensatory phase of haemorrhagic shock is associated with vascular hyporeactivity to

in haemorrhagic shock was more severe in the celiac artery and left femoral artery than in the superior mesenteric artery and left renal artery Gene expression levels of inducible nitric oxide synthase (iNOS), endothelial nitric oxide synthase (eNOS), interleukin-1β (IL-1β), IL-6, tumour necrosis factor-α (TNF-α), and endothelin-1 was increased in haemorrhagic shock and was upregulated to a greater extent in the liver and skeletal muscle compared to the ileum and left kidney (Liu and Dubick, 2005)

An increase in nitric oxide (NO) production in haemorrhagic shock due to an increase in level or activity of both eNOS and iNOS has been associated with the

hyporesponsiveness to norepinephrine and angiontensin II both in vitro and in the intact

animal (e.g Liu and Dubick, 2005; Pieber et al, 1999; Shirhan et al, 2004; Thiemermann et

al, 1993; Zingarelli et al, 1997) Furthermore, the upregulation of both eNOS and iNOS levels and activity is lipopolysaccharide (LPS)-independent and not associated with an elevation of endogenous LPS, suggesting that bacterial translocation across the gut is not a primary cause (Hierholzer and Billiar, 2001; Thiemermann et al, 1993) Activation of the endothelin-1 receptor on vascular smooth muscle cells also appears to partially mediate vacular hyporeactivity in haemorrhagic shock A recent study by Liu and Dubick (2005) showed that endothelin-1 mRNA levels was upregulated in organs such as the liver,

Introduction

5

intestine and skeletal muscle in haemorrhagic shock and was associated with the consequent vascular hyporesponsiveness to norepinephrine In particular, the authors demonstrated that administration of an endothelin-1 receptor antagonist (PD142893) could attenuate the hyporesponsiveness to phenylephrine in haemorrhagic shock in the celiac artery, left femoral artery, superior mesenteric artery and left renal artery In addition to a direct vasoconstrictor effect, angiotensin II could also attenuate the vascular hyporesponsiveness of isolated rat superior mesenteric artery to norepinephrine (Li et al, 2006b; Xu and Liu, 2005) Similarly, vasopressin administration could also increase the vascular response to norepinephrine in both the isolated mesenteric artery and in haemorrhagic-shocked rats (Yang et al, 2006a) These results suggest that vasoactive mediators may also influence the vascular hyporeactivity to norepinephrine It is also possible that inflammatory cytokines (TNF-α, IL-1) may play a role in the aetiology of vascular hyporeactivity produced in haemorrhagic shock Administration of anti-TNF-α antibodies to rats prior to haemorrhage could attenuate the vascular hyporeactivity to norepinephrine in aortic rings taken from these animals (Zingarelli et al, 1994) Incubation

of rat aortic rings with IL-1 also caused a decrease in response to norepinephrine and this effect was inhibited by Nω-nitro-L-arginine methyl ester (L-NAME), suggesting that inflammatory cytokines may mediate vascular hyporesponsiveness to norepinephrine by stimulating NO production (Robert et al, 1992)

Pieber and co-workers (1999) reported that the hyporeactivity to angiontensin II in the mesenteric circulation in haemorrhagic-shocked rats was exacerbated by an inhibitor of cyclooxygenase (COX) (indomethacin) and attenuated by a non-specific inhibitor of NOS (L-NAME), suggesting that endogenously produced prostanoids may delay or inhibit the vascular hyporeactivity while endogenous NO (possibly produced from inflammatory

Introduction

6

cells) are deleterious The authors’ results suggest that hyporeactivity to angiotensin II may

be mediated in part by NO as L-NAME restored vascular response to angiotensin II both systemically and in the mesenteric circulation In addition, the authors demonstrated that angiotensin II hyporesponsiveness in haemorrhagic-shocked rats does not involve the participation of KATP channels, as an inhibitor of KATP channels (glibenclamide) did not affect angiotensin II hyporesponsiveness

Emerging evidence indicates that the mechanisms involved in the decrease in response to norepinephrine in haemorrhagic shock despite an increase in sympathetic activity involve, at least in part, a decrease in calcium sensitivity of vascular smooth muscle cells, as well as a state of hyperpolarisation As mentioned above, angiotensin II and vasopressin both attenuate the vascular hyporeactivity to norepinephrine in rats subjected to haemorrhagic shock The primary mechanism of action of angiotensin II and vasopressin to attenuate norepinephrine hyporesponsiveness appears to be inhibition of a reduction in sensitivity to calcium in vascular smooth cells via a Rho-kinase signaling pathway (Li et al, 2006b; Xu and Liu, 2005; Yang et al, 2006a) In these studies, haemorrhagic shock caused a decrease in sensitivity to Ca2+-mediated contraction of the isolated rat superior mesenteric artery This decrease in Ca2+ sensitivity in haemorrhagic-shocked mesenteric artery was attenuated by angiotensin II and vasopressin treatment Stimulation of Rho-kinase enhanced, while Rho-kinase antagonists (Y-27632) inhibited, the attenuation of the calcium sensitivity reduction in isolated haemorrhagic-shocked rat mesenteric artery by angiotensin II and vasopressin respectively Xu and Liu (2005) also demonstrated the involvement of protein kinases C and G in the aetiology of Ca2+desensitization in isolated haemorrhagic-shocked rat mesenteric artery The authors showed that agonists of protein kinase C and antagonists of protein kinase G could inhibit the

Introduction

7

reduction in Ca2+ sensitivity in haemorrhagic-shocked rat mesenteric artery Furthermore, the authors also demonstrated that the attenuation of the reduction in Ca2+ sensitivity in haemorrhagic-shocked rat mesenteric artery by protein kinase C and Rho-kinase could be augmented by myosin light chain kinase inhibitors The findings of Xu and Liu (2005) suggest that biochemical modification or modulation of some components of the contractile apparatus may contribute to the decrease in calcium sensitivity of vascular smooth muscle cells in haemorrhagic shock Zhao and Zhao (2007) also demonstrated that part of the hyporesponsiveness to phenylephrine in haemorrhagic shock was due to a decrease in phenylephrine-induced Ca2+ accumulation in arteriolar smooth muscle cells The authors also showed that the decrease in phenylephrine-induced Ca2+ accumulation in haemorrhagic-shocked arteriolar smooth muscle cells was due to a reduction of Ca2+ influx through L-type Ca2+ channels

The vascular hyporesponsiveness to norepinephrine is also associated with

hyperpolarisation of vascular smooth muscle cells, at least in vitro Studies by Zhao and

co-workers (Zhao et al, 2000; Zhao and Zhao, 2007) implicate NO in associated vascular hyporeactivity to norepinephrine in the arterioles of haemorrhagic-shocked rats In these studies, arteriolar strips and arteriole smooth muscle cells taken from rats subjected to haemorrhagic shock demonstrated hyperpolarisation In particular, the reaction of NO with superoxide anion (O2-) to form peroxynitrite (ONOO-) appears to be a key step in NO-associated hyperpolarisation as treatment of haemorrhagic-shocked arteriolar strips with O2- scavengers reverses NO-associated hyperpolarisation The authors also demonstrated that ONOO--associated hyperpolarisation in haemorrhagic-shocked arteriolar strips was mediated by the opening of KATP and KCa channels as glibenclamide and charybdotoxin could reduce the hyperpolarisation caused by S-nitroso-N-

hyperpolarisation-Introduction

8

acetylpenicillamine (SNAP, NO donor) in haemorrhagic-shocked arteriolar strips The involvement of reactive oxygen species (ROS) in vascular hyporeactivity has also been demonstrated by Altavilla and co-workers (2001).In this study, administration of a vitamin

E analogue (IRFI-042) attenuated vascular hyporeactivity to phenylephrine in haemorrhagic-shocked aortas In another study, haemorrhagic-shocked rat aortas which demonstrated a vascular hyporesponsiveness to norepinephrine also showed an increase nitrotyrosine staining, an indicator of peroxynitrite formation (Zingarelli et al, 1997) In this study, treatment of rats with mercaptoethylguanidine, a dual inhibitor of inducible NOS and COX and a scavenger of ONOO-, attenuated the vascular hyporesponsiveness to norepinephrine in haemorrhagic shocked aortas and also reduced ONOO- formation in these vessels, as evidenced by a decrease in nitrotyrosine staining in these vessels (Zingarelli et al, 1997)

1.1.3 Overview of hypoxic responses leading to inflammation

Ischaemia refers to an insufficient supply of blood to a tissue leading to an insufficient supply of oxygen to meet the metabolic demands of the tissue In haemorrhagic shock, the low blood pressure is insufficient to perfuse organ tissues adequately, leading to ischaemia and subsequent hypoxia Hypoxia triggers a myriad of responses in hypoxic tissues and prolonged hypoxia eventually leads to hypoxia-induced cell death (apoptosis/necrosis), which results in tissue injury and initiation of an inflammatory response that is independent of reperfusion (Seth et al, 2004)

An insufficient supply of oxygen inhibits an oxygen-sensitive outward K+ current in cells and causes a hypoxia-induced depolarisation (Conforti et al, 2001; Lopez-Barneo, 1996) The depletion of ATP also leads to the failure of ion pumps (e.g NA+-K+-ATPase)

Introduction

9

and a consequent inability to maintain cell membrane potential, thus contributing to hypoxia-induced depolarisation (Ikenouchi et al, 1993) Failure of such ion pumps also disrupts cell osmolarity, leading to cell swelling and death Depolarisation of the cell activates voltage-gated Ca2+ channels, causing Ca2+ influx and a rise in intracellular Ca2+concentrations (Lopez-Barneo, 1996) The rise in Ca2+ concentration activates numerous

Ca2+-dependent enzymatic pathways and is central to the hypoxic response and also causes hypoxia-induced apoptosis/necrosis and hence inflammation (see review, by Seta et al, 2004)

Hypoxia also causes mitochondrial dysfunction and this leads to impairment of oxidative phosphorylation, which in turn leads to the generation of ROS (e.g Nakahara and Takeo, 1986; Niknahad et al, 1995; Vandeplassche et al, 1989; Young et al, 1983) Dysfunction of the mitochondria, which is the main energy supplier of the cell, further contributes to cellular dysfunction and may also promote apoptosis through the release of cytochrome c and activation of caspases (Araya et al, 1998; Chae et al, 2001; de Moissac et

al, 2000) The generation of NO and other reactive nitrogen species (RNS) and ROS contributes to hypoxia-induced inflammation Ultimately, cell death by apoptosis or necrosis occurs, and this leads to leukocyte infiltration and an inflammatory response Other responses to ischaemia include cytoskeletal changes For example, ATP-depleted endothelial cells have shortened F-actin filaments that increase endothelial permeability (Hinshaw et al, 1998)

One of the key molecules activated/upregulated in hypoxia is the hypoxia inducible factor-1 (HIF-1), which activates transcription of numerous genes and is central to cellular adaptation low oxygen levels (Ke and Costa, 2006) HIF-1 mediates hypoxia-induced inflammation by increasing transcription and activation of pro-inflammatory enzymes such

Introduction

10

as iNOS (Melillo et al, 1995; Palmer et al, 1998) and COX-2 (Hierholzer et al, 2001) In addition, HIF-1 also activates transcription of both pro-survival and pro-apoptic proteins and is involved in hypoxia-induced cell death (Ke and Costa, 2006; Hellwig-Burgel, 2005)

Another key transcription factor activated during hypoxia is nuclear factor κ B (NFκB) NFκB is a ubiquitous transcription factor present in the cytosol of cells and it is involved in immune and inflammatoryreactions In unstimulated cells, the binding of IκB

to NFκB inhibits NFκB-dependent gene transcription (e.g Baeuerle and Baltimore, 1989; Ito et al, 1994; Novak et al, 1991) Activation of NFκB-dependent gene transcription occurs through the phosphorylation and subsequent degradation of IκB, which enables NFκB to translocate to the nucleus, where it activates gene transcription (Cheshire and Baldwin, 1997; Steffan et al, 1995) NFκB is activated by signals associated with stress, injury and inflammation and upregulates the gene expression of cytokines, chemokines, cell adhesion molecules, growth factors and immunoreceptors (May and Ghosh, 1998) Upregulation of NFκB activity in haemorrhagic shock has been demonstrated in the various organs such as the liver (Altavilla et al, 2001), heart (Meldrum et al, 1997a), lung (Fan et

al, 1998) and kidney (Jiang et al, 1997)

1.1.4 Inflammatory mediators of haemorrhagic shock

Cytokines are a group of proteins and peptides that are important in cell-cell signaling They play critical roles in both innate and adaptive immune responses, and their effects may include the up-regulation and/or down-regulation of gene transcription, which

in turn affects the production of other cytokines, the number of surface receptors for other molecules, or causes feedback inhibition In addition, cytokines have "redundancy" as many cytokines share similar functions

Introduction

11

Cytokines

TNF-α Haemorrhagic shock has been associated with an increase in TNF-α

concentration in various organs such as the liver (Altavilla et al, 2001), heart (Meldrum et

al, 1997a and 1997b) and lung (Fan et al, 2000) The increase in TNF-α concentration has been attributed to an increase in oxidative stress-dependent activation of NFκB, at least in the liver (Altavilla et al, 2001) The authors demonstrated that administration of a free radical scavenger (IRFI-042) to haemorrhagic-shocked rats could inhibit the increase in NFκB binding activity, which in turn inhibited the rise in TNF-α concentration In the lung

at least, the production of TNF-α is partially dependent upon NO production as mice treated with a NO scavenger (NOX, a dithiocarbamate) demonstrated reduced levels of TNF-α mRNA expression (Hierholzer et al, 2002) Furthermore, Schwartz and co-workers (1995) demonstrated that the ROS source leading to the increase in TNF-α concentration in haemorrhagic shock was likely the ROS produced by xanthine oxidase as the increase in TNF-α concentration was inhibited by allopurinol (xanthine oxidase inhibitor) TNF-α production in haemorrhagic shock also promotes the production of pro-inflammatory IL-6 (see below) Treatment of mice with anti-TNF-α antibodies attenuates the production of IL-

6 following haemorrhage (Ertel et al, 1991a)

TNF-α mediates some pathology associated with haemorrhagic shock TNF-α may partially mediate the vascular hyporeactivity to norepinephrine in haemorrhagic shock (Chapter 1, Section 1.1.2) TNF-α has been shown to mediate the decrease in myocardial contractility seen in haemorrhagic shock (Meldrum et al, 1997b; Shahani et al, 2001) In these studies, haemorrhagic shock caused an increase in α1 adrenergic activity in the myocardium that led to an increase in NFκB activation and increased TNF-α concentrations The authors showed that this increase in TNF-α concentration in the

Introduction

12

myocardium after haemorrhagic shock could be inhibited by a TNF-α neutralizing antibody

or α1-adrenoceptor blockade, which also reversed the decrease in myocardial contractiliy TNF-α also mediates haemorrhagic shock-induced predisposition to acute respiratory distress syndrome by inducing the production of pro-coagulant molecules and reducing plasminogen activator levels, leading to lung fibrin disposition (Fan et al, 2000) Haemorrhagic shock is associated with subsequent depression of cell-mediated immunity due to a decreased major histocompatibility complex (MHC) class II antigen presentation (Ertel et al, 1990) This immunosuppression has been attributed, at least in part, to increased TNF-α production (Ertel et al, 1991a) The authors showed that pre-treatment of mice with anti-TNF-α antibodies prior to haemorrhagic shock could improve MHC class II antigen presentation in Kupffer cells after haemorrhage TNF-α may also contribute to increased neutrophil infiltration into organs following haemorrhagic shock and resuscitation by promoting leukocyte adhesion to endothelial cells (Maier et al, 2003)

IL-1β Haemorrhagic shock is associated with increased IL-1β production in

various organs such as the lung, liver and kidney (Ayala et al, 1992; Jiang et al, 1997; Suter

et al, 1992) The increase in IL-1β production following haemorrhagic shock and reperfusion appears to be dependent on ROS produced by xanthine oxidase as inhibition of xanthine oxidase with allopurinol decreased the increase in IL-1β production from pulmonary mononuclear cells (Schwartz et al, 1995) In the lung, the increase in IL-1β production appears to be mediated, at least in part, by an increase in α1 adrenergic activity,

as adrenergic blockade could inhibit IL-1β increase (Laffon et al, 1999) In addition, lung IL-1β production in haemorrhagic shock is also dependent upon the production of NO, as mice treated with a NO scavenger (NOX) demonstrated reduced levels of IL-1β mRNA expression (Hierholzer et al, 2002)

Introduction

13

IL-1β appears to be a factor in haemorrhagic shock-associated mortality (Pellicane

et al, 1993) The authors showed that administration of an IL-1 receptor antagonist (IL-1ra)

to haemorrhagic-shocked mice could improve the 5-day survival rate IL-1β may contribute

to the development of acute respiratory distress syndrome following haemorrhagic shock IL-1β mediates the decrease in alveolar fluid clearance observed following haemorrhagic shock as administration of an IL-1 receptor antagonist (IL-1ra) to haemorrhagic-shocked mice could reverse this effect (Laffon et al, 1999) It is also likely that IL-1β may enhance inflammatory responses (e.g cytokine production) by inducing the degradation of IκB and increasing the nuclear binding activity of NFκB (Parikh et al, 2000)

IL-6 Haemorrhagic shock is associated with an increase in IL-6 production in the

liver, peritoneum, kidney and lung (Ayala et al, 1992; Jiang et al, 1997; Zhu et al, 1994) The production of IL-6 in haemorrhagic shock appears to be dependent on TNF-α as mice pre-treated with anti-TNF-α antibody had decreased plasma concentrations of IL-6 compared to saline controls following haemorrhage (Ertel et al, 1991a) In the lung at least, the production of IL-6 is partially mediated by an NO-dependent mechanism as mice treated with a NO scavenger (NOX) demonstrated reduced levels of IL-6 mRNA expression (Hierholzer et al, 2002) In the intestinal epithelium, IL-6 increases the gene-expression of Bcl-2 (Rollwagen et al, 1998); in the lung and liver, IL-6 may activate NFκB (Toth et al, 2004)

The role of IL-6 in haemorrhagic is controversial Some studies suggest a protective effect of IL-6 on organ injury while others demonstrate a deleterious effect Increased IL-6 production may reduce systemic sepsis following haemorrhage (Rollwagen et al, 1996) The authors showed that feeding mice with crude or recombinant IL-6 could alleviate systemic bacteraemia following haemorrhagic shock and was associated with a restoration

Introduction

14

of intestinal mucosal integrity IL-6 appears to mediate this effect by increasing the gene expression of anti-apoptotic Bcl-2 and inhibiting apoptosis of intestinal epithelial cells (Rollwagen et al, 1998) However another study by Yang and co-workers (2003) demonstrated that IL-6-deficient mice subjected to haemorrhagic shock had decreased ileal mucosal permeability In the lung, Hierholzer and co-workers (1998b) showed that IL-6 might contribute to the development of acute respiratory distress syndrome following haemorrhagic shock by mediating neutrophil recruitment In this study, haemorrhagic shock followed by resuscitation was associated with an increase in IL-6 mRNA levels especially in the bronchial and alveolar cells In addition, the authors demonstrated that intratracheal instillation of IL-6 alone into healthy rats could result in neutrophil infiltration into lung interstitium and alveoli However in another study, infusion of IL-6 into rats subjected to haemorrhage resulted in a decrease in lung and liver inflammation and injury, and this was possibly due to a down-regulation of NFκB activity (Meng et al, 2000b) The protective effect of IL-6 was demonstrated in another study by Brundage and co-workers (2004) where infusion of recombinant IL-6 into haemorrhagic-shocked pigs could inhibit the increase in pro-inflammatory granulocyte colony-stimulating factor in the lung In

2004, Toth and co-workers showed that increase in liver IL-6 was associated with liver dysfunction and injury, and this effect on the liver could be attenuated by administration of anti-IL-6 antibodies A study by Meng and co-workers (2001) suggested that the deleterious effect of IL-6 in haemorrhagic shock predominates over its apparent protective effect The authors demonstrated that IL-6-deficient mice subjected to haemorrhagic shock had decreased lung and liver neutrophil infiltration, increased alveolar cross-sectional area and decreased liver necrosis, as well as decreased NFκB activity Another study suggests that IL-6 may also contribute to depressed cardiac contractility following haemorrhagic

In the same study, an increase in cardiac IL-6 in haemorrhagic shock was associated with

an upregulation of the leukocyte adhesion molecule intercellular cell adhesion molecule-1 (ICAM-1) and the chemokine cytokine-induced neutrophil chemoattractant (CINC), as well

as an increase in myeloperoxidase (MPO) activity Yang and co-workers (2006b) demonstrated that these effects could be reduced by anti-IL-6 antibody administration into haemorrhagic-shocked rats

IL-10 Haemorrhagic shock is associated with an increase in IL-10 concentration

(Ayala et al, 1994) In this study, IL-10 concentration appears to be correlated with IL-6 concentrations, as a reduction in IL-6 levels is accompanied by a concomitant decrease in IL-10 concentration In addition, the authors showed that the increase in IL-10 concentration in haemorrhagic shock was dependent upon IL-4 production

In haemorrhagic shock, IL-10 appears to down-regulate the pro-inflammatory cytokines IL-6 and TNF-α Karakozis and co-workers (2000) demonstrated that administration of IL-10 to haemorrhagic-shocked rats could inhibit haemorrhagic shock-induced increase in TNF-α levels In another study, Yokoyama and co-workers (2004) showed that haemorrhagic shock was associated with increase IL-10 in the liver The authors demonstrated that the production of TNF-α and IL-6 by Kupffer cells taken from haemorrhagic-shocked rats could be increased by incubating the cells with anti-IL-10 antibody Ayala and co-workers (1994) also demonstrated that increased levels of IL-10 in

al, 1989; Richard et al, 1991) NO activates guanylyl cyclase in vascular smooth muscle cells, which leads to an increase in cGMP and decrease in cytosolic Ca2+ concentrations, thus causing vascular smooth muscle relaxation and vasodilatation NO is a key molecule

in maintaining vascular smooth muscle relaxation and blood pressure (e.g Bassenge, 1991; Hendriks et al, 1993; Johnson and Freeman, 1992; Moncada et al, 1991; Perrella et al, 1991; Persson et al, 1990; Rees et al, 1989) Increased eNOS activity is associated with a reduction in blood pressure, plasma cholesterol and atherosclerotic lesions (van Haperen et

al, 2002) while the lack of eNOS activity is associated with hypertension (and its related diseases), possibly due to an imbalance between vasodilatory NO and vasoconstrictor endothelin (Quaschning et al, 2007; Shesely et al, 1996) However, excessive NO production from iNOS (and possibly eNOS as well) is generally associated with inflammation and plays a major role in the immune response In haemorrhagic shock, NO production is increased (Naziri et al, 1995; Smail et al, 1998) and NO plays a role in mediating the vascular hyporeactivity to vasoconstrictors during the vascular decompensation phase (Chapter 1, Section 1.1.2), and also mediates haemorrhagic shock-induced inflammatory responses However, whether NO production in haemorrhagic shock

is beneficial or deleterious is controversial

NO produced from iNOS appears to be be a primary mediator of the inflammatory response following haemorrhagic shock (Hierholzer et al, 1998a) The authors showed that inhibition of iNOS with N6-(iminoethyl)-L-lysine could inhibit the activation of NFκB as

Introduction

18

well as inhibit the increase in IL-6 and granulocyte-colony stimulating factor mRNA levels

in haemorrhagic-shocked lung and liver In addition, the authors showed that inhibition of iNOS with N6-(iminoethyl)-L-lysine also led to a reduction of lung and liver injury caused

by haemorrhagic shock In a subsequent study, Hierholzer and co-workers (2002) demonstrated that NO scavenging with NOX could decrease haemorrhagic shock-associated increase in ICAM-1 and neutrophil infiltration, as well as pulmonary oedema in the lung In addition, the authors showed that NOX administration to haemorrhagic-shocked rats could also reduce lung activation of NFκB and decrease levels of pro-inflammatory IL-6, TNF-α, and IL-1β In another study, Hua and Moochhala (1999) showed that both L-NAME and aminoguanidine increased the survival rate of haemorrhagic-shocked rats In addition, the authors demonstrated that L-NAME and aminoguanidine reduced macroscopic and microscopic organ injuries, and decreased PGE2and creatinine production in haemorrhagic-shocked animals Furthermore, these beneficial effects were reversed by L-arginine, suggesting the involvement of NO in the pathophysiology of hemorrhagic shock (Hua and Moochhala, 1999) The deleterious effects of excessive NO in haemorrhagic shock is also corroborated by Menezes and co-workers (1999), who showed that an NO scavenger (NOX) could decrease hepatic neutrophil infiltration and hepatic injury, and improve survival rates in haemorrhagic-shocked rats Excessive NO produced by iNOS (or eNOS) may mediate intestinal bacterial translocation in haemorrhagic shock as both L-NAME and aminoguanidine treatment could inhibit bacterial translocation and reduce organ damage (Hua and Moochhala, 2000) NO may also play a role in haemorrhagic shock-induced decrease in alveolar fluid clearance (Pittet et al, 2001) In this study, the haemorrhagic shock-induced, iNOS-dependent increase in the lung production of NO was associated with a failure of the alveolar

Introduction

19

epithelium to up-regulate vectorial fluid transport in response to β adrenergic agonists, and inhibition of iNOS restored the normal catecholamine-mediated up-regulation of alveolar liquid clearance Some of these deleterious effects of NO could be attributed to the formation of RNS (see below) For example, ONOO-has been shown to be formed early in haemorrhagic shock by eNOS and contributes to oxidative damage (Szabo et al, 1995) The upregulation of iNOS (and possibly eNOS) can lead to the generation of large amounts of

NO and mediates inflammation primarily through the generation of ROS/RNS (Chapter 1, Section 1.1.5)

Some published literature suggests a beneficial effect of NO in haemorrhagic shock Inhibition of NO production with L-NAME in haemorrhagic-shocked rats has been shown

to increase hepatic and gastric injury as well as decrease coronary blood flow leading to exacerbation of myocardial ischaemic injury (Adachi et al, 1998; Denizbasi et al, 2000; Harbrecht et al, 1995) Other studies using NO-donors have also demonstrated a beneficial effect of NO in haemorrhagic shock Administration of an NO donor, S-nitroso-N-acetylpenicillamine (SNAP), to rats subjected to haemorrhagic shock decreased neutrophil accumulation in the intestine and increased survival rate (Symington et al, 1992) Similar findings were made by Wallace and co-workers (1997) who demonstrated that administration of a NO-releasing derivative of aspirin to haemorrhagic-shocked rats could reduce leukocyte adherence to the endothelium and decrease neutrophil accumulation in the intestine, thus reducing intestinal injury NO has been suggested to restore the depressed cell-mediated immune response and ameliorate the inflammatory response after haemorrhage and resuscitation, thus promoting healing (Loehe et al, 2007) In this study, administration of L-arginine was found to restore the depressed organ blood flow and to reduce tissue injury following haemorrhagic shock, which was associated with restoration

Introduction

20

of the depressed cell-mediated immune responses and attenuation of the massive inflammatory response and improved wound healing In addition, the authors showed that the excessive infiltration of the liver with neutrophils in haemorrhage was decreased by L-arginine administration NO may mediate its anti-inflammatory effects by inhibiting the phosphorylation and degradation of IκB and thus preventing the activation of NFκB (Colasanti & Persichini, 2000; Katsuyama et al, 1998; Park et al, 1997) NO donors (e.g SNP) have been shown to inhibit TNF-α-induced activation of NFκB and may decrease the expression of endothelial cell adhesion molecules via downregulation of NFκB (Peng et al, 1995a and 1995b) NO may also reduce oxidative stress associated with ischaemia and reperfusion Kim and Kim (1998) demonstrated that L-arginine (NO precursor) could prevent the increase in lipid peroxide production and the decreases in glutathione contents

in gastric cells subjected to hypoxia and reoxygenation Furthermore, the increase in iNOS activity/expression is associated with an increase in gene expression levels of superoxide dismutase (Frank et al, 1999) This may be a cellular mechanism to reduce the formation of deleterious ONOO- which plays a role in mediating NO-induced cellular injury Thus it is likely that the deleterious effect of nitric oxide in haemorrhagic shock depends on its concentration in tissues Indeed, a study by Anaya-Prado and co-workers (2004) demonstrated that inhibiting excessive NO production by iNOS while supplying exogenous

NO in the form of SNP decreased the formation of pro-inflammatory cytokines such as TNF-α, IL-1β and IL-6 and reduced injury in haemorrhagic-shocked lung tissues

PGE2 / COX-2

Haemorrhagic shock is associated with an increase in plasma PGE2 (prostaglandin

E2) and upregulation of tissue COX-2 in rats (Ertel et al, 1991b; Hierholzer et al, 2001; Tsukada et al, 2000) PGE2 appears to play a role in haemorrhagic shock-induced

Introduction

21

immunosuppression Ertel and co-workers (1991b) suggested that this effect was mediated

by PGE2-induced decrease in antigen presentation by macrophages In this study, inhibition

of PGE2 production in haemorrhagic-shocked mice using ibuprofen (non-specific COX inhibitor) could increase antigen presentation and IL-1 synthesis by peritoneal macrophages In 1994, Ayala and co-workers showed that increased PGE2 in haemorrhagic-shocked mice was associated with increased anti-inflammatory IL-10 production In addition, the authors showed that PGE2 might mediate this effect through IL-4-dependent production of IL-10, as PGE2-induced production of IL-10 by haemorrhagic-shocked T-cells could be inhibited by anti-IL-4 antibody treatment PGs produced by COX-2 may play a role in ameliorating haemorrhagic shock-induced liver and intestinal injury as well (Tsukada et al, 2000) In this study, inhibition of COX-2 prostaglandin synthesis with NS398 in haemorrhagic-shocked rats was associated with an exacerbation of liver and intestinal injury However, some studies suggest that upregulation

of COX-2 (and PG production) in haemorrhagic shock may be detrimental instead Knoferl and co-workers (2001) demonstrated that PGs produced by COX-2 in haemorrhagic shock was associated with an increase in pro-inflammatory IL-6 production in Kupffer cells that could be inhibited with NS398 In 2005, Shirhan and co-workers (Md et al, 2005) also demonstrated that inhibition of COX-2 in haemorrhagic-shocked rats was associated with reduced organ injury as examined histologically Furthermore, the authors demonstrated that upregulation of COX-2 could also be inhibited by iNOS inhibition, suggesting that NO produced by iNOS could induce COX-2 upreulgation and subsequent PG production

Introduction

22

1.1.5 Reperfusion injury – role of ROS/RNS

Severe haemorrhage on its own leads to tissue hypoxia and dysfunction, and may ultimately result in death of the organism However, reperfusion of hypoxic tissues after prolonged ischaemia is associated with an exacerbation of organ injury known as ischaemia-reperfusion injury Initiation of ischaemic and inflammatory responses during the ischaemic phase contributes to this reperfusion injury Reperfusion exacerbates ischaemic injury firstly, by generating ROS/RNS due to the re-introduction of oxygen, and secondly, by facilitating the immune response through the circulation of pro-inflammatory mediators and enabling leukocytes to reach the sites of tissue damage An excessive accumulation of leukocytes may cause microvessel occlusion and exacerbate ischaemic damage Excessive neutrophil activation also leads to further generation of RNS/ROS and release of pro-inflammatory mediators, leading to multiple organ dysfunction (Li and Jackson, 2002)

Generation of ROS/RNS Hypoxia is associated with a down-regulation of

mitochondrial enzyme activity, including cytochrome c oxidase and manganese superoxide dismutase (Murphy et al, 1984) The absence of cytochrome c, the final electron acceptor, increases ROS (i.e O2-) formation by the proximal complexes and reduction of manganese superoxide dismutase also decreases mitochondrial antioxidant capacity, thus promoting the formation of ROS (Fridovich, 1978) There is also a general reduction of glutathione (GSH), superoxide dismutases and other antioxidant enzymes, which promotes ROS formation (Jackson et al, 1996; Kirshenbaum and Singal, 1992; Shlafer et al, 1987) Besides reacting with cellular components (e.g proteins, lipids, DNA), O2- formed during reoxygenation can also react with NO to form ONOO- (Beckman et al, 1990) In addition, during ischaemia, ATP is metabolized into hypoxanthine and xanthine dehydrogenase is

Introduction

23

converted to xanthine oxidase (Parks et al, 1988; McCord et al, 1985) Upon reoxygenation, a large amount of ROS is generated from the action of xanthine oxidase on hypoxanthine (McCord et al, 1985) Neutrophils migrated to the site of inflammation also contribute to the generation of ROS through the respiratory burst In addition to ROS, there

is also an increase in RNS formation, mainly in the form of NO and ONOO-, due to the increase in NO production by iNOS and eNOS (Li et al, 1999; Valdez et al, 2000; Van der Vliet et al, 1997)

ROS/RNS-mediated injury ROS/RNS generated during reoxygenation causes

oxidative damage by reacting with cellular components, as evidenced by increased lipid peroxidation, 3-nitrotyrosine, and nitrogen oxides (Tan et al, 1999) ONOO- is highly reactive and can oxidize sulfhydryl groups in proteins, leading to protein dysfunction and/or inactivation (Radi et al, 1991) ROS such as H2O2 can inhibit Na+-K+-ATPase pump activity, leading to loss of membrane potential and ionic balance and cell death (Kako et al, 1988) Reoxygenation can lead to cell death by necrosis or apoptosis, as both pathways can

be triggered by ROS/RNS (Bossenmeyer et al, 1998; Cai and Jones, 1998; Saikumar et al, 1998) In addition to apoptotic pathways, ROS/RNS can also participate in other signaling pathways Increase in oxidative stress has been associated with an increase in NFκB binding activity, which in turn activates NFκB-dependent inflammatory responses such as

an increase in IL-6 production (Kupatt et al, 1997) Oxidative stress-dependent activation

of NFκB can also promote apoptosis by suppressing the expression of anti-apoptotic proteins such as Bcl-2 (Matsushita et al, 2000) Reoxygenation has also been associated with an increase in cellular Ca2+ concentrations, which in turn leads to cellular injury through mitochondrial dysfunction and promotion of apoptosis via cytochrome c release (Delcamp et al, 1998; Griffiths et al, 1998) Other pathways also activated by

1.2 Hydrogen sulphide

Hydrogen sulphide (H2S) is a colourless gas with the characteristic smell of rotten eggs and has been traditionally viewed as a toxic gas and an environmental pollutant (Guidotti, 1996) H2S is a broad-spectrum toxicant, and at high concentrations, it exerts its toxic effect on various organs such as the brain, liver, lung and respiratory tract primarily through inhibition of the respiratory enzyme, cytochrome c oxidase, by the sulphide (S2-) ion (Dorman et al, 2002) However, in the last decade, it has become increasingly evident that H2S is synthesized endogenously in many non-vertebrate/vertebrate tissues and may have a physiological function (e.g Abe and Kimura, 1996; Julian et al, 2002; Olson et al, 2006)

H2S is synthesized in most tissues, with the highest rate of production found in the brain, liver, kidney and cardiovascular system (Doeller et al, 2005) H2S is produced

Introduction

25

endogenously from L-cysteine metabolism by two pyridoxal-5’-phosphate dependent enzymes, cystathionine-β-synthase (CBS, predominant H2S-synthesizing enzyme in the central nervous system) and cystathionine-γ-lyase (CSE, predominant H2S-synthesizing enzyme in the cardiovascular system) (Stipanuk and Beck, 1982; Swaroop et al, 1992) (Figure 1.1) Under physiological conditions, one-third of H2S is undissociated and two-thirds dissociate into H+ and HS- H2S can be detected in plasma and tissues at a concentration of about 50 µM, although the concentration in the brain is relatively higher (e.g Zhao et al, 2001; Warenycia et al, 1989)

There is little in the published literature regarding the regulation of the enzymes responsible for H2S biosynthesis CSE may be subject to regulation by redox mechanisms and has been shown to be inhibited by ROS (Diwakar and Ravindranath, 2007) CSE levels

in lung are down-regulated in rats with high pulmonary blood flow (Xiaohui et al, 2005) and it is possible that CSE levels and activity may be modulated by NO as CSE mRNA levels are decreased in animals treated with L-NAME (Zhong et al, 2003) In addition, CSE activity may be modulated by sex hormones and is increased during late gestation and

Figure 1.1 H 2 S production from the metabolism of L-cysteine

CSE, cystathionine-γ-lyase; CBS, cystathionine-β-synthase

Introduction

26

during lactation (Akahoshi et al, 2006, Barber et al, 1999) CBS can be activated by adenosyl-L-methionine (Stipanuk and Beck, 1982) Hormones may also regulate CSE and CBS activity, for example, glucagon has been demonstrated to increase CSE and CBS activity (Jacobs et al, 2001) However, it is possible that some these “regulatory” mechanisms are simply the result of altered L-cysteine levels in the various tissues

The figure below shows the known pathways and metabolites of H2S metabolism to date H2S is oxidized in the oxygen rich mitochondria into thiosulphate, which is in turn converted to sulphite and then to sulphate by thiosulphate:cyanide sulphurtransferase and sulphite oxidase respectively In the gut at least, H2S may also be methylated by thiol S-methyltransferase into dimethylsulphide (Furne et al, 2001) H2S may also directly interact with haemoglobin to form sulphaemoglobin (Nichol et al, 1968) However, it is not known

if any of these metabolites of H2S are bioactive, and if so, which of these pathways predominate

Figure 1.2 H 2 S metabolism SO, sulfite

oxidase; TSMT, thiol S-methyltransferase; TST, thiosulfate:cyanide sulfurtransferase

(rhodanese) Taken from Lowicka and

Beltowski, 2007

et al, 2005), the myocardium (Geng et al, 2004a), the carotid sinus (Xiao et al, 2006a), intestinal smooth muscle (Teague et al, 2002), some neurons of the nervous system (Distrutti etal, 2006; Kimura et al, 2006) and pancreatic beta cells (Yang et al, 2005)

H2S reacts readily with ROS/RNS and may mediate some of its effects through its interaction with various ROS Indeed, H2S can potentially act as an antioxidant H2S has been shown to react with O2- in neutrophils and in the myocardium (Geng et al, 2004a; Mitsuhashi et al, 2005), H2O2 in the myocardium (Geng et al, 2004a), scavenge peroxynitrite (ONOO-) (Whiteman et al, 2004) and hypochlorite (ClO-) (Whiteman et al, 2005) and also reacts with NO (Ali et al, 2006; Whiteman et al, 2006) In addition, H2S may stimulate cysteine transport into cells and increase GSH synthesis (Kimura and

Kimura, 2004) In vitro at least, H2S has been shown to protect proteins and lipids from ROS/RNS-mediated damage (Whiteman et al, 2004; Whiteman et al, 2005)

H2S may also regulate the activity and/or gene expression of eNOS and iNOS H2S may inhibit eNOS as the conversion of [3H]-arginine into [3H]-citrulline by recombinant eNOS could be inhibited by exogenous H2S (Kubo et al, 2007) A study by Hu and co-workers (2007) showed that exogenous application of NaHS and stimulation of endogenous H2S production attenuated LPS-stimulated NO production by decreasing LPS-stimulated increase in iNOS expression In addition, Oh and co-workers (2006) suggested that this inhibitory effect of H2S on iNOS was due to an upregulation of heme oxygenase-1