Whole body metabolism, muscle and mitochondrial function, and the role of uncoupling protein-3 in a mouse model of sepsis. Dr Parjam Seyed Zolfaghari A thesis submitted for the Degree of Doctor of Philosophy University College London Funded by the Medical Research Council 2012 1 DECLARATION I, Parjam Seyed Zolfaghari confirm that the work presented in this thesis is my own. Where information has been derived from other sources, I confirm that this has been indicated in the thesis. 2 ABSTRACT Sepsis, the exaggerated systemic inflammatory response to infection, often leads to multi-organ failure (MOF) and death. Skeletal muscle function is often profoundly affected, with patients requiring prolonged ventilatory support and rehabilitation. The pathophysiology underlying MOF and muscle failure in sepsis remains poorly understood. Recent evidence points to mitochondrial dysfunction and cellular energetic down-regulation, related in part to excess generation of reactive oxygen and nitrogen species (ROS). Substrate oxidation by the mitochondrial respiratory chain generates a proton gradient across the inner mitochondrial membrane that is coupled to ATP production. Mitochondrial uncoupling proteins (UCPs) may reduce mitochondrial ROS formation, though at the expense of oxidative phsophorylation. Notably, UCP3 is up-regulated in sepsis but its physiological role is unknown. I therefore hypothesized that muscle dysfunction in sepsis has a mitochondrial aetiology and increased UCP3 expression offers a protective role. I investigated this in a fluid-resuscitated murine model of faecal peritonitis/MOF (optimized and characterized during this project). I used Ucp3(-/-) mice to explore the role of this uncoupling protein in sepsis. I observed a profound hypothermic and hypometabolic response early in the course of severe sepsis. This was accompanied by respiratory muscle dysfunction and fatigue, as well as a decrease in mitochondrial proton-motive force (measured using novel live-cell 2- photon confocal imaging techniques in freshly isolated ex-vivo diaphragm muscles). By contrast, the mitochondrial proton-motive force was 3 unaffected in septic Ucp3(-/-) mice, but no difference was seen in whole- body metabolic response or respiratory muscle fatigue. Furthermore, proton leak and substrate utilization of mitochondria isolated from whole body skeletal muscle from wild-type septic mice were unchanged, though superoxide production was higher. These findings suggest that up- regulation of UCP3 in sepsis has no whole- body metabolic or functional consequence in skeletal muscle. The low global oxygen consumption and diaphragm mitochondrial proton-motive force suggest a reduced cellular metabolic demand in the septic mice. 4 ACKNOWLEDGMENTS I would like to extend my gratitude to my supervisors Professors Mervyn Singer (UCL), Michael Duchen (UCL) and Nancy Curtin (Imperial college), for their energy, guidance and advice throughout this project. Also to Drs Jane Carre, Nadeene Parker, Sean Davidson and Alex Dyson for their invaluable help with technical and theoretical aspects of the project as well as guidance in preparation of this manuscript. A special thanks also to Valerie Taylor and Raymond Stidwill for teaching and assistance with animal husbandry, surgery and in vivo experiments. This project was kindly funded by the Medical Research Council through a Clinical Research Training Fellowship. I would like to thank my wife, Dr Louise Barber, for help and support at home, and frequent trouble-shooting with science, computers and software! Also for bringing to life our daughter Ayla who has brought joy and perspective to our lives. Finally, I would like to dedicate this body of work to the memory of my father, Seyed Abbas Zolfaghari, who is missed greatly and forever in our hearts. 5 TABLE OF CONTENTS TITLE 1 DECLARATION 2 ABSTRACT 3 ACKNOWLEDGEMENTS 5 TABLE OF CONTENTS 6 LIST OF FIGURES 13 LIST OF TABLES 16 ABBREVIATIONS 17 Chapter 1 INTRODUCTION 20 1.1 Sepsis definition and epidemiology 20 1.2 The pathophysiology of sepsis and MOF 21 1.2.1 Tissue pO in septic multi-organ failure 22 2 1.2.2 Body metabolism and temperature in sepsis 24 1.3 Mitochondrial physiology and sepsis 26 1.3.1 Cellular and mitochondrial metabolism 26 1.3.1.1 The electron transport chain and ATP synthesis 28 1.3.1.2 Mitochondrial proton leak 31 1.3.1.3 Mitochondria and reactive oxygen species (ROS) 32 1.3.2 Nitric oxide, sepsis and mitochondrial physiology 34 1.3.3 Mitochondrial calcium accumulation 36 1.3.4 Mitochondrial dysfunction in sepsis 37 1.3.4.1 Electron transport chain dysfunction 38 1.3.4.2 Mitochondrial transmembrane potential (Δψ ) 39 m 1.3.4.3 ATP Synthase defect 39 1.3.4.4 Uncoupled respiration 40 1.4 The uncoupling proteins (UCPs) 42 1.4.1 Genetic manipulation of UCP-2 and -3 43 1.4.2 UCPs and ROS production 44 1.4.3 UCP-2 and -3 in sepsis 48 1.5 Skeletal muscle function in sepsis 49 6 1.5.1 Mechanism of septic skeletal muscle dysfunction 51 1.5.2 Inracellular Ca2+ in septic muscle dysfunction 52 1.5.3 Muscle fatigue in sepsis 54 1.6 Animal models of septic organ failure 56 1.7 Project aims 59 Chapter 2 MURINE MODEL OF SEPSIS 61 2.1 Introduction 61 2.2 Methods 62 2.2.1 Animal model 62 2.2.1.1 General anaesthesia (GA) for procedures 63 2.2.1.2 Original model of sepsis 63 2.2.1.3 New model of mouse faecal peritonitis 65 2.2.2 Standardisation of the septic insult 67 2.2.3 Cardiac output measured by echocardiography 68 2.2.4 Measurement of whole body metabolic rate 71 2.2.4.1 The effect of re-warming on metabolic rate in sepsis 72 2.3 Results 73 2.3.1 Survival in the original mouse model 73 2.3.2 New mouse model of sepsis 75 2.3.2.1 Serum biochemistry at 24 hours 78 2.3.2.2 Arterial blood gases at the 24h time-point 78 2.3.2.3 Blood glucose changes over 24 hours 81 2.3.2.4 Effect of sepsis on body temperature in mice 82 2.3.2.5 Weight changes following induction of sepsis 83 2.3.3 Cardiac output and peak aortic blood flow velocity 84 decrease in severe sepsis 2.3.3.1 Cardiovascular response to intravenous fluid 87 boluses in early severe sepsis is different to that seen in late severe sepsis 2.3.4 Oxygen consumption (VO ) and carbon dioxide 94 2 production (VCO ) in healthy fed and starved mice 2 2.3.4.1 Metabolic rate, temperature and weight in starvation 95 2.3.5 Metabolic rate is reduced in severe sepsis 98 7 2.3.5.1 Relationship between temperature and VO 102 2 2.3.5.2 Effect of re-warming on VO and cardiac output 103 2 in severe septic mice 2.4. Discussion 107 2.4.1 Animal model of sepsis and in-vivo findings 107 2.4.2 Cardiovascular response of mouse model of sepsis 113 2.5 Conclusion 116 Chapter 3 DIAPHRAGM MUSCLE FUNCTION 118 3.1 Introduction 118 3.2 Methods 120 3.2.1 Solutions and reagents 120 3.2.2 Mouse septic model 122 3.2.3 Tissue preparation 124 3.2.4 Equipment setup 125 3.2.5 Experimental protocol 127 3.2.5.1 Length/Tension relationship 127 3.2.5.2 Caffeine stimulated force generation 129 3.2.5.3 Length/Power relationship 130 3.2.5.4 Power output during repeated stimulation 133 3.2.5.5 Hypoxia experiments 133 3.2.5.6 Timing of experiments 134 3.2.6 Ca2+ imaging by fluorescence confocal microscopy 135 3.2.7 Statistical analysis 137 3.3 Results 138 3.3.1 Force/Power experiments 138 3.3.1.1 Demographic data 138 3.3.1.2 Maximal isometric force is reduced in sepsis 139 but not by starvation alone 3.3.1.3 Force generation by caffeine stimulation is 140 maintained in septic diaphragm muscle strips 3.3.1.4 Power output measurements using the work-loop 142 technique is lower in muscle strips from septic mice 8 3.3.2 Power output with repeated stimulation over 143 one-minute starts lower and decline more rapidly in septic muscle strips. 3.3.3 Hypoxia reduces power output with earlier fatigue 145 in repetitive work-loop cycles 3.3.4 Live cell imaging to study Ca2+ signal changes 149 in muscle cells 3.4 Discussion 152 3.4.1 Muscle force in starvation and sepsis 152 3.4.1.1 Effect of starvation on muscle force 153 3.4.1.2 Effect of sepsis on muscle force 154 3.4.2 Measuring intracellular Ca2+ 156 3.4.3 Reduced power output in septic diaphragms 157 3.4.4 Diaphragm muscle fatigue in sepsis 158 3.4.5 Limitations of this study 160 3.5 Conclusion 162 Chapter 4 MITOCHONDRIAL FUNCTION IN SEPTIC MICE 163 4.1 Introduction 163 4.2 Methods 165 4.2.1 Mouse septic model 165 4.2.2 Live-cell imaging 166 4.2.2.1 Measuring mitochondrial membrane potential (Δψ ) 166 m 4.2.2.2 Measuring mitochondrial redox state 169 4.2.2.3 Fluorophore excitation, microscope set-up and 172 experimental protocol to measure Δψ m and NADH autofluorescence 4.2.3 Permeabilized skinned muscle fibre respirometry 174 4.2.4 Modular kinetic analysis of mitochondrial function 181 in sepsis 4.2.4.1 Preparation of isolated mitochondria from 181 skeletal muscle 4.2.4.1.1 Solutions 181 9 4.2.4.1.2 Skeletal muscle mitochondria isolation procedure 182 4.2.4.2 Quality control of the isolated mitochondria 183 4.2.4.2.1 Respiratory control ratio 183 4.2.4.2.2 Citrate synthase (CS) activity 184 4.2.4.2.2.1 Stock solutions for measuring citrate synthase activity 186 4.2.4.3 Modular kinetic analysis experimental protocol 188 4.2.3.3.1 Titration protocols 192 4.2.3.3.1.1 Set-up protocol 193 4.2.3.3.1.2 Substrate oxidation kinetics 194 4.2.3.3.1.3 Phosphorylation kinetics 194 4.2.3.3.1.4 Proton leak kinetics 195 4.2.3.3.1.4 Statistical analysis 195 4.2.4 Mitochondrial ROS production 197 4.2.4.1 Measurement of Reactive Oxygen Species (ROS) 197 production by isolated mitochondria using Amplex Red fluorescence 4.2.5 Diaphragm muscle electron microscopy in sepsis 200 4.2.5.1 Reagents used for EM studies 201 4.3 Results 202 4.3.1 Skeletal muscle mitochondrial membrane potential 202 is lower in mice with severe sepsis 4.3.2 NADH autofluorescence in sham and septic 206 groups were equal 4.3.3 Permeabilized muscle fibre oxygen consumption 207 did not change with sepsis severity 4.3.4 Respiratory kinetics in isolated mitochondria 211 4.3.4.1 Respiratory control ratio of skeletal muscle 212 mitochondria were equal in all three groups of mice 4.3.4.2 Mitochondrial yield was equal in all groups of mice 213 4.3.4.3 Mitochondrial kinetic measurements 215 4.3.5 Reactive Oxygen Species production was higher 217 in starved sham and septic skeletal muscle mitochondria 4.3.6 Mitochondria in septic mouse diaphragms were 220 morphologically similar to sham fed. 10