REDOX STATE OF ERYTHROCYTE PEROXIREDOXIN 2 DURING OXIDATIVE STRESS AND ITS EFFECT ON MEMBRANE BINDING
Simone Birgit Bayer
A thesis submitted for the degree of Doctor of Philosophy in Pathology at the University of Otago, Christchurch, New Zealand
February 2015
ii
Abstract
Peroxiredoxin 2 (Prx2) is the third most abundant protein in erythrocytes. It is a thiol-specific antioxidant protein that is able to reduce hydroperoxides. Prx2 is a physiologically vital protein, since mice lacking Prx2 develop severe haemolytic anaemia. Much is known about Prx2 structure and activity, and its redox state has been linked to several functions in other cells. However, if and how the redox state of erythrocyte Prx2 is modified during inflammation or blood storage has not been established. Further, Prx2 membrane binding in erythrocytes has been observed frequently, but it is not clear if Prx2 redox state affects membrane binding. The aims of this thesis were to investigate how inflammation, neutrophil activation and blood storage affect Prx2 redox state, and how Prx2 membrane binding is influenced by its redox state.
When oxidised, Prx2 is converted to a disulfide-linked homodimer. Non-reducing SDS-PAGE and Western blotting was exploited to measure the redox state of Prx2. Prx2 was oxidised when exposed to hydrogen peroxide (H2O2). Neutrophils generate H2O2 during inflammation, so activated neutrophils were able to oxidise Prx2 in neighbouring erythrocytes. Prx2 oxidation by neutrophil-derived H2O2 was also observed in a mouse model of endotoxaemia, raising the possibility of Prx2 being a useful real-time marker for oxidative stress in inflammatory diseases.
For blood transfusions, erythrocytes are stored at 4 °C for up to six weeks. Under standard storage conditions, Prx2 remained mainly reduced during the first three weeks, after which oxidised Prx2 accumulated progressively. While supplementation with dihydrolipoic acid (DHLA) showed some promise in prevention of Prx2 oxidation, storage of erythrocytes in an alternative buffer similar to a recently FDA approved solution with high pH and some buffering capacity delayed the onset on Prx2 oxidation successfully. When the antioxidant capacity of stored erythrocytes was tested with additional H2O2, it was revealed that while Prx2 recycling was impaired, it was not absent. This is likely due to a heterogeneous population of erythrocytes, where older cells may lose their ability to recycle oxidised Prx2. Prx2 could therefore be a useful biomarker for oxidation in stored erythrocytes.
Prx2 binds to the erythrocyte membrane, which was previously suggested to depend on Prx2 redox state. In whole erythrocytes, Prx2 membrane binding increased with calcium, and iii
decreased with H2O2, while the redox state of Prx2 was only affected by the latter.
Furthermore, when both calcium and H2O2 were added, no Prx2 was lost from the membrane, indicating that two different mechanisms are involved in Prx2 membrane binding. After purification of erythrocyte Prx2, isolation of erythrocyte membranes, and incubation of the membranes with hyperoxidised, reduced and oxidised Prx2, it was established that membrane binding was independent of Prx2 redox state. This indicates that H2O2 acts on another component of erythrocytes to inhibit binding. When oxyhaemoglobin and haemichromes, which are denatured haemoglobin, were added, the haemichromes interfered with Prx2 binding to the membrane. Haemoglobin had no effect. This is consistent with the hypothesis that Prx2 protects the membrane from oxidative damage arising from haemoglobin oxidation, but is displaced during erythrocyte senescence.
iv
Acknowledgements
I want to thank Prof. Christine Winterbourn and Prof. Mark Hampton for their support and trust in my abilities, and the chance to do this project in this extraordinary group. I hope I was able to live up to the expectations, even though the course of this PhD had many ups and downs, and not all of them due to the quakes. I am indebted to their nearly endless patience and guidance. I am also forever grateful for the help that Prof. Margreet Vissers and the University of Otago gave me during the hardest and darkest time of my life. And nothing of this would have been possible without the University of Otago doctoral scholarship and the extension of the scholarship after the earthquakes. I also want to thank Prof. Roland Stocker and Dr. Ghassan Maghzal from the Centre for Vascular Research, School of Medical Sciences (Pathology) and Bosch Institute, The University of Sydney, and the Vascular Biology Division from The Victor Chang Cardiac Research Institute in Darlinghurst in New South Wales, Australia, for the conduction of the mouse experiments in Chapter 3.
Working with the centre of free radical research has been amazing, and the people are outstanding. I felt very welcome and a part of something bigger. Thank you to Dr. Alexander Peskin for the short and precise emails and answers to my many questions; you are a fountain of knowledge. I am grateful to Dr. Louise Paton and Dr. Nina Dickerhof for performing the mass spectrometry analyses in this thesis. I also want to acknowledge the amazing help I got from Dr. Nick Magon for the Prx2 purification; I hope it was not a complete waste of your time. I need to give my thanks to Dr. Melissa Stacey, who was my mentor in my first months, and to Steph Moran and her family who were real friends from the start and even gave me a roof over my head. I have been blessed by all the blood donors and –takers, especially by Dr. Rufus Turner, Tessa Mocatta, Stephanie Moran, and Irada Khalikova.
I am also grateful to Canterbury Scientific Ltd., who provided me and many others from my group a place to work after the earthquakes shut us out of our main building.
To my “peer reviewers” Dr. Nick Magon, Kate Vicks, Emma Spencer, Dr. Rebecca Poynton, and Dr. Stephanie Bozonet, I want to say:” You rock!”
And finally, I have to thank my beloved husband Manuel Clavel. You were my mountain and my lifeboat; I would have given up without you. v
Publications arising from this thesis
PUBLISHED MANUSCRIPTS
Bayer, S.B., Maghzal, G., Stocker, R., Hampton, M.B., Winterbourn, C.C., Neutrophil- mediated oxidation of erythrocyte peroxiredoxin 2 as a potential marker of oxidative stress in inflammation. FASEB J, 2013. 27(8): p. 3315-22.
MANUSCRIPT PRE-PUBLISHED ONLINE
Bayer, S.B., Hampton, M.B., Winterbourn, C.C., Accumulation of Oxidized Peroxiredoxin 2 in Erythrocytes. Transfusion, 2015.
vi
Contents
Abstract ...... ii
Acknowledgements ...... iv
Publications arising from this thesis ...... v
Contents ...... vi
List of Tables ...... xii
List of Figures ...... xiii
Abbreviations ...... xvii
1 Introduction ...... 1
1.1 Oxidative stress ...... 2 1.1.1 Reactive oxygen species (ROS) and the definition of oxidative stress ...... 2 1.1.2 Overview of reactive oxygen species ...... 2 1.1.3 General consequences of oxidative stress ...... 3 1.2 Endogenous source of oxidative stress in erythrocytes ...... 3 1.2.1 The function of haemoglobin ...... 3 1.2.2 Haemoglobin autoxidation ...... 4 1.2.3 Haemoglobin denaturation ...... 5 1.2.4 Special cases ...... 7 1.2.4.1 Haemoglobinopathies ...... 7 1.2.4.2 Drug induced oxidative stress ...... 8 1.2.4.3 Erythrocyte storage ...... 8 1.3 Exogenous source of oxidative stress in erythrocytes ...... 13 1.3.1 Erythrocytes as scavengers ...... 13 1.3.2 Neutrophils ...... 13 1.4 Antioxidant mechanisms in the erythrocyte ...... 15 1.4.1 Non-enzymatic systems ...... 15 1.4.2 Enzymatic systems ...... 16 1.4.2.1 Superoxide dismutase ...... 16 1.4.2.2 Catalase ...... 16 vii
1.4.2.3 The glutathione/ glutathione peroxidase/ glutathione reductase system ...... 16 1.4.2.4 NADPH ...... 18 1.5 Peroxiredoxins ...... 19 1.5.1 The catalytic cycle of peroxiredoxins ...... 19 1.5.2 The thioredoxin/ thioredoxin reductase system ...... 21 1.5.3 Hyperoxidation of peroxiredoxins and its repair by retroreduction ...... 22 1.5.4 Higher molecular weight structures of peroxiredoxin ...... 24 1.5.5 Proposed functions of peroxiredoxins ...... 25 1.6 Peroxiredoxin 2 in erythrocytes ...... 28 1.6.1 History of peroxiredoxin 2 discovery ...... 28 1.6.2 The reactivity of peroxiredoxin 2 in comparison with other antioxidant systems ...... 28 1.6.3 Peroxiredoxin 2 susceptibility to oxidation and hyperoxidation ...... 29 1.6.4 Haematological phenotypes of peroxiredoxin knock out mouse models ...... 30 1.6.5 Binding of peroxiredoxin 2 to erythrocyte membranes ...... 31 1.7 Aims ...... 33
2 Methods and Materials ...... 35
2.1 Erythrocyte peroxiredoxin 2 oxidation state during inflammation and sepsis ..... 35 2.1.1 Erythrocyte preparation ...... 35 2.1.2 N-Ethylmaleimide-blocking ...... 35 2.1.3 Effect of fever-like temperatures on erythrocyte peroxiredoxin 2 oxidation ...... 35 2.1.4 Neutrophil preparation ...... 36 2.1.5 Co-incubation of isolated erythrocytes and neutrophils ...... 36 2.1.5.1 Incubation of erythrocytes with activated neutrophils at different ratios ...... 36 2.1.5.2 Incubation of erythrocytes and activated neutrophils over time ...... 37 2.1.5.3 Activation of neutrophils in full blood ...... 37 2.1.5.4 Effect of inhibitors and scavengers on erythrocyte peroxiredoxin 2 oxidation by activated neutrophils ...... 37 2.1.5.5 Comparison of normal and MPO-deficient neutrophils ...... 37 2.1.5.6 Incubation of erythrocytes with LPS-stimulated neutrophils ...... 37 2.1.5.7 Cytochrome c assay for superoxide production by LPS-stimulated neutrophils .. 38 2.1.5.8 Incubation of erythrocytes with Staphylococcus aureus-stimulated neutrophils . 38 2.1.6 Measurement of oxidised erythrocyte peroxiredoxin 2 in a mouse model for endotoxaemia, in collaboration with University of Sydney, Australia ...... 38 2.1.7 SDS-PAGE and immunoblotting ...... 39 2.1.8 Visualisation of bands and calculation of the percentages ...... 40 viii
2.2 Erythrocyte peroxiredoxin 2 redox state during blood storage ...... 42 2.2.1 Preparation of erythrocytes for blood storage ...... 42 2.2.2 Measurement of erythrocyte peroxiredoxin 2 oxidation state during storage ...... 42 2.2.3 Reversal of peroxiredoxin 2 oxidation in stored erythrocytes ...... 42 2.2.4 Reversal of peroxiredoxin 2 oxidation in stored erythrocytes with Rejuvesol™, dithiothreitol, α-lipoic acid, dihydrolipoic acid and N-acetylcysteine ...... 42 2.2.5 Prevention of peroxiredoxin 2 oxidation during storage of erythrocytes ...... 43 2.2.6 Peroxiredoxin 2 oxidation in stored erythrocytes after additional stress with hydrogen peroxide ...... 43 2.2.7 Separation of erythrocytes by density and measurement of peroxiredoxin 2 oxidation ...... 43 2.2.8 Spectral measurement of methaemoglobin and haemichrome content ...... 44 2.2.9 Visualisation of vesicle proteins by in-gel fluorescence staining ...... 45 2.2.10 GSH-quantification of stored erythrocytes with and without NAC ...... 45 2.2.10.1 Quantification of total low molecular weight thiols by DTNB-assay ...... 45 2.2.10.2 Quantification of GSH in stored erythrocytes by stable isotope tandem mass spectroscopy ...... 46 2.3 Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state ...... 46 2.3.1 Membrane binding in whole cells ...... 46 2.3.2 Purification of erythrocyte peroxiredoxin 2 ...... 47 2.3.2.1 Purification of erythrocyte peroxiredoxin 2 via anion-exchange chromatography ...... 47 2.3.2.2 Determination of purified peroxiredoxin 2 via SDS-PAGE and immunoblotting ...... 48 2.3.2.3 Determination of purified peroxiredoxin 2 via electrospray ionisation–mass spectrometry ...... 50 2.3.3 Binding of purified peroxiredoxin 2 to isolated membranes ...... 51 2.3.3.1 Preparation of erythrocyte membranes ...... 51 2.3.3.2 Preparation of inside-out erythrocyte membrane vesicles ...... 52 2.3.3.3 Acetylcholine esterase assay to determine sideness of IOM ...... 52 2.3.3.4 Visualisation of IOM by membrane-labelling ...... 53 2.3.3.5 Preparation of reduced, oxidised and hyperoxidised peroxiredoxin 2 ...... 53 2.3.3.6 Binding of purified peroxiredoxin 2 to inside-out membrane vesicles ...... 54 2.3.3.7 Binding of purified peroxiredoxin 2 to Hb free, peroxiredoxin 2 free ghosts ...... 54 2.3.3.8 Additional experiments to achieve insight into mechanism of peroxiredoxin 2 binding ...... 55 2.3.3.9 Binding of peroxiredoxin 2 to isolated erythrocyte ghosts in the presence of haemichromes ...... 55 ix
2.3.3.10 Crosslinking of purified peroxiredoxin 2 and isolated erythrocyte membrane with glutardialdehyde ...... 55 2.3.4 SDS-PAGE, immunoblotting and Coomassie-staining of membrane proteins .... 55 2.4 Statistics ...... 56 2.5 Materials ...... 57 2.5.1 Erythrocyte peroxiredoxin 2 oxidation state during inflammation and sepsis ..... 57 2.5.2 Erythrocyte peroxiredoxin 2 redox state during simulated blood storage ...... 58 2.5.3 Changes in membrane-bound peroxiredoxin 2 ...... 59 2.5.4 Instruments and equipment ...... 60 2.5.5 Common recipes for buffers ...... 61
3 Erythrocyte peroxiredoxin2 oxidation state during inflammation ...... 63
3.1 Introduction ...... 63 3.2 Experimental approach ...... 63 3.3 Results ...... 65 3.3.1 Effect of fever-like temperatures on erythrocyte peroxiredoxin 2 oxidation ...... 65 3.3.2 Effect of acidity on erythrocyte peroxiredoxin 2 oxidation ...... 65 3.3.3 Effect of PMA-stimulated neutrophils on erythrocyte peroxiredoxin 2 oxidation ...... 66 3.3.3.1 Effects of inhibitors and scavengers on erythrocyte peroxiredoxin 2 oxidation .. 69 3.3.3.2 Comparison of blood from a healthy patient and an MPO-deficient patient ...... 70 3.3.4 Effect of LPS-stimulated neutrophils on erythrocyte peroxiredoxin 2 oxidation 71 3.3.5 Effect of neutrophils stimulated with Staphylococcus aureus on erythrocyte peroxiredoxin 2 oxidation ...... 73 3.3.6 Effect of LPS on peroxiredoxin 2 oxidation in murine erythrocytes in vivo ...... 74 3.4 Discussion ...... 75 3.4.1 Neutrophil activation and peroxiredoxin 2 oxidation ...... 75 3.4.2 Reversal of erythrocyte peroxiredoxin 2 oxidation during inflammation ...... 75 3.4.3 Role of neutrophil MPO in erythrocyte peroxiredoxin 2 oxidation ...... 76 3.4.4 Erythrocyte peroxiredoxin 2 as a scavenger of extracellular hydrogen peroxide 76 3.4.5 Effect of temperature and pH on peroxiredoxin 2 oxidation within erythrocytes 77 3.5 Summary ...... 77
4 Changes in peroxiredoxin 2 redox state during storage of erythrocytes...... 78
4.1 Introduction ...... 78 4.2 Experimental approach ...... 78 4.3 Results ...... 80 x
4.3.1 Storage of erythrocytes and peroxiredoxin 2 oxidation ...... 80 4.3.2 Prevention or delay of peroxiredoxin 2 oxidation during storage of erythrocytes ...... 82 4.3.3 Reversal of peroxiredoxin 2 oxidation in stored erythrocytes ...... 84 4.3.4 Differences in peroxiredoxin 2 oxidation, hyperoxidation and haemoglobin derivatives in stored erythrocytes separated by density ...... 87 4.4 Discussion ...... 90 4.4.1 Peroxiredoxin 2 oxidation and GSH levels during erythrocyte storage ...... 90 4.4.2 Hypotheses for the impaired reduction of erythrocyte peroxiredoxin 2 after storage ...... 91 4.4.3 Possible differences in erythrocyte populations ...... 92 4.4.4 Rejuvenation, small molecular weight antioxidants and the antioxidant quality of stored blood ...... 93 4.4.5 Storage buffers and quality of stored erythrocytes ...... 93 4.4.6 Peroxiredoxin 2 and vesiculation ...... 94 4.4.7 Peroxiredoxin 2 redox state as a biomarker ...... 94 4.5 Summary ...... 95
5 Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state ...... 96
5.1 Introduction ...... 96 5.2 Experimental approach ...... 96 5.3 Results ...... 97 5.3.1 Membrane binding of peroxiredoxin 2 in whole cells ...... 97 5.3.2 Binding of purified peroxiredoxin 2 to isolated membranes ...... 101 5.3.2.1 Binding of purified Prx2 to inside-out membrane vesicles ...... 101 5.3.2.2 Binding of purified peroxiredoxin 2 to haemoglobin-free, peroxiredoxin 2-free ghosts ...... 103 5.3.2.3 Additional experiments to achive insight into mechanism of peroxiredoxin 2 binding ...... 106 5.3.2.4 Binding of peroxiredoxin 2 to isolated erythrocyte ghosts in the presence of haemichromes ...... 107 5.3.2.5 Peroxiredoxin 2 crosslinking to membrane proteins ...... 109 5.4 Discussion ...... 109 5.4.1 Effect of calcium on membrane binding of Prx2 ...... 109
5.4.2 Redox state of peroxiredoxin 2 and effect of H2O2 ...... 110 5.4.3 Effect of haemichromes on membrane binding of peroxiredoxin 2 ...... 110 5.4.4 Specificity of peroxiredoxin 2 binding and possible binding partners ...... 111 5.5 Summary ...... 111 xi
6 General discussion ...... 113
6.1 Summary of major findings ...... 113 6.2 Implications and future work ...... 115 6.2.1 Peroxiredoxin 2 as a biomarker for oxidative stress ...... 115 6.2.2 Improvement of the quality of blood storage ...... 117 6.2.3 Peroxiredoxin 2 membrane binding, decameric structure and calcium ...... 118 6.2.4 Peroxiredoxin 2: holdase and thiol protector? ...... 118 6.2.5 Peroxiredoxin 2 and Gardos channel activation ...... 120 6.2.6 Peroxiredoxin 2: a major player in erythrocyte senescence?...... 122 6.3 General conclusions ...... 124
7 References ...... 125
8 Appendix ...... 165
8.1 Calculations used in Chapter 4 ...... 165 8.2 Calculations used in Chapter 5 ...... 165
xii
List of Tables
Chapter 1 Table 1.1 Selection of biologically important reactive species...... 3 Table 1.2 Properties of mammalian peroxiredoxins...... 19
Chapter 2
Table 2.1 Standard recipes for SDS-PAGE...... 40 Table 2.2 Purification of human peroxiredoxin 2 from erythrocytes...... 49 Table 2.3 Materials used in Section 2.1...... 57 Table 2.4 Materials used in Section 2.2...... 58 Table 2.5 Materials used in Section 2.3...... 59 Table 2.6 Instruments and equipment used in Sections 2.1, 2.2 and 2.3...... 60 Table 2.7 Common recipes for buffers used in Sections 2.1, 2.2 and 2.3...... 61
Chapter 4
Table 4.1 Content of OxyHb, metHb and haemichromes within separated populations of stored erythrocytes...... 89
Chapter 5
Table 5.1 Concentrations of oxyHb, metHb and haemichromes before and after incubation with APH...... 108
xiii
List of Figures
Chapter 1 Figure 1.1 Erythrocytes...... 1 Figure 1.2 Formation of reversible and irreversible haemichromes...... 6 Figure 1.3 The sequence of events leading to Heinz body formation...... 7 Figure 1.4 Erythrocyte morphology during refrigerated storage...... 10 Figure 1.5 Overview of the changes involved in the storage lesion...... 11 Figure 1.6 NOX2 assembly on the membrane...... 14 Figure 1.7 The production of HOCl by MPO...... 15 Figure 1.8 Structure of glutathione...... 17 Figure 1.9 The catalytic cycle of typical 2-Cys peroxiredoxins...... 20 Figure 1.10 The thioredoxin cycle...... 21 Figure 1.11 Hyperoxidation of typical 2-Cys peroxiredoxins...... 23 Figure 1.12 Simplified mechanism of peroxiredoxin hyperoxidation repair by sulfiredoxin...... 23 Figure 1.13 Higher molecular weight structures of peroxiredoxin 2...... 25 Figure 1.14 Possible roles for hyperoxidised and oxidised peroxiredoxin...... 27 Figure 1.15 Crystal structures of dimeric and decameric peroxiredoxin 2...... 28
Chapter 2
Figure 2.1 Differences in calculated ratios due to uneven peroxiredoxin 2 antibody affinity...... 41 Figure 2.2 Calculation of the percentage of oxidised erythrocyte peroxiredoxin 2. .... 41 Figure 2.3 Purification of peroxiredoxin 2 from human erythrocytes by ion exchange chromatography...... 48 Figure 2.4 Electrophoretic and immunological detection of peroxiredoxin 2 in the purified preparation...... 49 Figure 2.5 Electrospray ionisation–mass spectrometry analysis of purified peroxiredoxin 2...... 51 Figure 2.6 Preparation of hyperoxidised, oxidised and reduced peroxiredoxin 2...... 54
xiv
Chapter 3
Figure 3.1 The necessity of alkylation of free thiol groups for evaluation of peroxiredoxin 2 redox status...... 64 Figure 3.2 Effect of temperature on erythrocyte peroxiredoxin 2 oxidation...... 65 Figure 3.3 Effect of pH on erythrocyte peroxiredoxin 2 oxidation...... 66 Figure 3.4 Effect of PMA-stimulated neutrophils on erythrocyte peroxiredoxin 2 oxidation...... 67 Figure 3.5 Erythrocyte peroxiredoxin 2 content compared to neutrophil peroxiredoxin 2 content...... 68 Figure 3.6 Oxidation of erythrocyte peroxiredoxin 2 by PMA-stimulated neutrophils over time...... 68 Figure 3.7 Effect of PMA on erythrocyte peroxiredoxin 2 oxidation in blood...... 69 Figure 3.8 Structure of diphenylene iodonium (DPI)...... 69 Figure 3.9 Effect of inhibitors and scavengers on peroxiredoxin 2 oxidation...... 70 Figure 3.10 Comparison of erythrocyte Prx2 oxidation by MPO-deficient and non- MPO-deficient neutrophils...... 71 Figure 3.11 Effect of neutrophils stimulated with LPS on erythrocyte peroxiredoxin 2 oxidation...... 72 Figure 3.12 Superoxide-induced reduction of cytochrome c by LPS-stimulated neutrophils...... 72 Figure 3.13 Effect of neutrophils stimulated with S. aureus on erythrocyte peroxiredoxin 2 oxidation...... 73 Figure 3.14 Effect of LPS on peroxiredoxin 2 oxidation in murine erythrocytes in vivo over time...... 74
Chapter 4
Figure 4.1 Oxidation of erythrocyte peroxiredoxin 2 during storage at 4 °C...... 80 Figure 4.2 Representative Western blot of peroxiredoxin 2 in stored erythrocytes under reducing and non-reducing conditions...... 81 Figure 4.3 Hyperoxidation of peroxiredoxins during storage at 4 °C...... 81 Figure 4.4 Effect of α-lipoic acid (ALA), N-acetyl cysteine (NAC) or dihydrolipoic acid (DHLA) on oxidation of erythrocyte peroxiredoxin 2 during storage...... 82 xv
Figure 4.5 Effect of NAC, glycine and glutamine (NGQ) supplementation on peroxiredoxin 2 oxidation in stored erythrocytes...... 83 Figure 4.6 Effect of supplementation with NAC, glycine and glutamine (NGQ) on free GSH and low molecular weight thiols (LMW) in erythrocytes during storage...... 83 Figure 4.7 Comparison of peroxiredoxin 2 oxidation in erythrocytes stored in SAGM or EAS-76v6...... 84 Figure 4.8 Effect of glucose (A) and DTT (B) on erythrocyte peroxiredoxin 2 oxidation after storage...... 85 Figure 4.9 Effect of Rejuvesol™ (A), NAC, α-lipoic acid (ALA) (B) or dihydrolipoic acid (DHLA) (C) on peroxiredoxin 2 oxidation in erythrocytes after storage...... 86 Figure 4.10 Peroxiredoxin 2 oxidation in stored erythrocytes after challenge with hydrogen peroxide...... 87 Figure 4.11 Peroxiredoxin 2 oxidation in stored erythrocytes after inhibition of autoxidation with carbon monoxide and challenge with hydrogen peroxide...... 87 Figure 4.12 Comparison of the separation profiles of fresh erythrocytes and erythrocytes stored for 42 days...... 88 Figure 4.13 Prx2 oxidation (A) and hyperoxidation (B) in separated fractions of erythrocytes stored for 42 days...... 89 Figure 4.14 Content of erythrocyte vesicles after 42 days of storage...... 90
Chapter 5
Figure 5.1 Effects of calcium on (A) membrane binding and (B) redox state of peroxiredoxin 2...... 98 Figure 5.2 Effects of hydrogen peroxide on (A) membrane binding and (B) redox state of peroxiredoxin 2...... 99 Figure 5.3 Observed oxidative damage on membrane proteins by hydrogen peroxide...... 100 Figure 5.4 Effect of hydrogen peroxide on the absorption spectra of haemoglobin. . 100 Figure 5.5 Effect of hydrogen peroxide and calcium on membrane binding of peroxiredoxin 2...... 101 xvi
Figure 5.6 Micrograph showing lipid-containing membranes encapsulating beads. . 102 Figure 5.7 Binding of reduced (A) and oxidised (B) peroxiredoxin 2 to IOM...... 103 Figure 5.8 Oxidised and hyperoxidised peroxiredoxin 2 on a Western blot after non-reducing SDS-PAGE...... 104 Figure 5.9 Effect of redox state on membrane binding of peroxiredoxin 2...... 105 Figure 5.10 Competition between peroxiredoxin 2 and C-terminal peptide of peroxiredoxin 2 binding to erythrocyte ghosts...... 106 Figure 5.11 Binding of recombinant peroxiredoxin 3 to erythrocyte ghosts...... 106 Figure 5.12 Effect of calcium on peroxiredoxin 2 binding to erythrocyte ghosts...... 107 Figure 5.13 Competition of peroxiredoxin2 and haemichromes binding to erythrocyte ghosts...... 108 Figure 5.14 Crosslinking of membrane proteins and oxidised peroxiredoxin 2 with glutardialdehyde...... 109 Chapter 6
Figure 6.1 The balance of oxidation and reduction of peroxiredoxin 2 in erythrocytes...... 116 Figure 6.2 Hypothesis of oxidised peroxiredoxin 2 as holdase...... 120 Figure 6.3 Schematic diagram of the possible role of peroxiredoxin 2 in erythrocyte senescence...... 123
xvii
Abbreviations
ABAH 4-Aminobenzoic acid hydrazide ADSOL Additive solution ALA α-Lipoic acid ANOVA Analysis of variance APH Acetylphenyl hydrazine APS Ammonium persulfate AS-7 Additive solution 7 ASK1 Apoptosis signalling kinase 1 ATP Adenosine triphosphate CO Carbon monoxide Cp Peroxidatic cysteine Cr Resolving cysteine Cys Cysteine DHLA Dihydrolipoic acid DMSO Dimethyl sulfoxide DPG 2,3-bisphosphoglycerate DPI Diphenyleneiodonium chloride DTNB Dithionitrobenzoic acid DTT Dithiothreitol EAS76v6 Experimental additive solution 76 version 6 ECL™ Enhanced chemiluminescence reagent EDTA Ethylenediaminetetraacetic acid EGTA Ethylene glycol tetraacetic acid Fe Iron G6PD Glucose-6-phosphate dehydrogenase Gpx1 Glutathione peroxidase 1 GR Glutathione reductase GSH Reduced glutathione GSSH Glutathione disulfide xviii
H2O2 Hydrogen peroxide Hb Haemoglobin Hbs Haemoglobins HBSS Hank’s balanced salt solution HCL Hydrochloric acid HOCl Hypochlorous acid hsK1 Human small conductance potassium channel 1 IOM Inside-out membrane vesicles IP Isotonic Percoll solution kDa Kilo Dalton KO Knock-out LPS Lipopolysaccharide metHb Methaemoglobin MPO Myeloperoxidase m/z Mass to charge NAC N-acetylcysteine NEM N-ethylmaleimide NETs Neutrophil extracellular traps NGQ N-acetylcysteine, glycine, glutamine NKEF-B Natural killer enhancing factor B NOX NADPH-oxidase Orp1 Oxidant receptor protein 1 OxyHb Oxygenated haemoglobin PBS Phosphate buffered saline PKC Protein kinase C PMA Phorbol myristate acetate PMS Phenazine methosulfate PPP Pentose phosphate pathway PRP Protector protein Prxs Peroxiredoxins Prx2 Peroxiredoxin2 PS Phosphatidylserine xix
PVDF Polyvinylidene difluoride ROS Reactive oxygen species RS Reactive species SAGM Sodium adenine glucose mannitol SB Sample buffer SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis SE Standard error SH Sulfhydryl group SOD Superoxide dismutase SOH Sulfenic acid
SO2H Sulfinic acid
SO3H Sulfonic acid Srx Sulfiredoxin TBST Tris buffered saline with TWEEN® 20 TEMED Tetramethylethylenediamine TSA-II Thiol-specific antioxidant protein II Chapter 1: Introduction 1
1 Introduction
Peroxiredoxin 2 (Prx2) is a member of a family of antioxidant proteins that are able to break down peroxides rapidly. Erythrocyte Prx2, which is present at a high concentration, has a number of thought-provoking properties including the ability to bind to the membrane, and being unusually resistant to inactivation in the presence of excess peroxide. Prx2 has been shown to be essential for normal erythrocyte survival. Mice lacking Prx2 show high levels of oxidatively denatured haemoglobin and suffer from severe haemolytic anemia. Erythrocyte Prx2 efficiently decomposes hydrogen peroxide, either produced as a result of haemoglobin autoxidation or when added exogenously, a role traditionally thought to be filled by glutathione peroxidase and catalase. This suggests that Prx2 is involved in the defence against oxidative stress, and that it is important for preservation of the erythrocyte lifespan. However, it is not yet understood if Prx2 is able to protect erythrocytes from pathological oxidative stress during inflammation and disease, or if it is able to prolong erythrocyte lifespan, such as in blood storage. It is also unclear if the Prx2 redox state influences membrane binding. This thesis focuses on the redox state of erythrocyte Prx2 during the oxidative stresses associated with inflammation and blood storage, and aims to determine whether Prx2 redox state influences membrane binding.
Figure 1.1 Erythrocytes. Courtesy of the National Science Foundation, USA. This image was taken with a scanning electron micrograph, and uses false colour. Chapter 1: Introduction 2
Erythrocytes (Figure 1.1) are cells shaped like biconcave discs that make up nearly half of the blood volume. They exist mainly to distribute oxygen throughout the body. To fulfil this role, they lack a nucleus and other intracellular organelles, allowing them to pass through the smallest capillaries. Erythrocytes have an average life span of 120-150 days [1-3]. During this time they maintain their shape, keep ionic and pH equilibrium, and protect haemoglobin and membranes from damage. Moreover, erythrocytes are capable of protecting other cells from oxidative damage by scavenging oxidants [4, 5]. To achieve this, erythrocytes rely solely on their metabolic activity and antioxidant equipment.
1.1 Oxidative stress
1.1.1 Reactive oxygen species (ROS) and the definition of oxidative stress
The transfer of electrons onto oxygen produces a variety of different oxygen intermediates: .- including the superoxide anion radical, O2 , hydrogen peroxide, H2O2, and the hydroxyl radical, HO.. These reactive oxygen species (ROS) are able to oxidatively damage proteins, lipids and nucleotides [6-8]. Antioxidative protective systems exist to limit oxidative damage and the resulting functional impairment. The antioxidative systems involve radical scavengers, chelators which prevent ROS generation, antioxidant enzymes, and also repair systems. If the balance between oxidants and antioxidants is askew and veered towards oxidant generation, it is commonly referred to as “oxidative stress” [9]. However, since it is known that reactive species take part in redox signalling, there has been an approach to re- define the term oxidative stress as “a disruption of redox signalling and control” [10].
1.1.2 Overview of reactive oxygen species
The term ROS typically includes not only oxygen radicals, but also non-radical derivatives of oxygen, such as H2O2 [8]. Additionally, some reactive oxidants do not contain any oxygen at all. The term reactive species (RS) includes all reactive bromine, chlorine, nitrogen, oxygen and sulfur derivatives [8]. A selection of RS is shown in Table 1.1.
Chapter 1: Introduction 3
Table 1.1 Selection of biologically important reactive species.
Reactive species (RS) Radicals Non-radicals Reactive oxygen species (ROS) .- Superoxide O2 Hydrogen peroxide H2O2 Hydroxyl OH. Hypochlorous acid HOCl Hydroperoxyl HOO. Organic peroxides ROOH . Peroxyl ROO Ozone O3 Reactive chlorine species Atomic chlorine Cl. Hypochlorous acid HOCl
Chloramines R2NCl, RNCl2
Chlorine gas Cl2 Reactive nitrogen species Nitric oxide NO. Peroxynitrite ONOO- . Nitrogen dioxide NO2 Nitrous acid HNO2
1.1.3 General consequences of oxidative stress
Oxidative stress caused by increased ROS production or impairment of the antioxidant systems can lead to a range of results including cell injury, senescence or even cell death, but also increased proliferation and adaption. Injury is a broad term, which covers any transient change in cell homeostasis [11-13]. Senescence, the permanent cell cycle arrest, can be induced by high levels of oxidative stress [14]. Apoptosis, necrosis and necroptosis can also be the result of oxidative stress [15-19], and mild to moderate oxidative stress has been linked to proliferation [20] and adaption [21, 22]. In erythrocytes, oxidative stress impairs oxygen transport [23] as well as membrane deformability [24] and integrity by oxidation of membrane lipids [25-28].
1.2 Endogenous source of oxidative stress in erythrocytes
1.2.1 The function of haemoglobin
Haemoglobin (Hb) is the major protein in erythrocytes, constituting approximately 97% of the dry weight [29]. The main function of Hb is the transport of oxygen from the lungs to the tissues, where the oxygen is released and used in the mitochondria of cells to produce energy. Additionally, Hb is involved in the transport of other gases including carbon dioxide and nitric oxide [30, 31].
Hb is a metalloprotein composed of four globular subunits, known as globins. Each globin subunit is connected with a non-protein haem-group. The haem-group consists of ferrous iron (Fe2+) inside a porphyrin ring. The iron is bound to the nitrogen atoms via a coordinate Chapter 1: Introduction 4
covalent bond, which enables binding of oxygen in a reversible way by sharing an electron with Fe2+ [30]. This allows the iron to stably transition between its Fe2+ and Fe3+ state due to electron delocalisation while oxygen is bound. Additionally, the globin pocket is hydrophobic, maintaining the iron in its ferrous state after oxygen release. The iron is covalently bound to the nitrogen atoms in the porphyrin ring, and covalently stabilised by a histidine residue of the Hb (see Figure 1.2 for structure). In the deoxygenated state, the histidine residue pulls the iron out of the plane of the porphyrin, which leads to a “tight” conformation of the Hb tetramer. When oxygen binds to the iron in the haem group, the iron takes up a planar conformation [32]. This has a widespread effect on Hb conformation: the binding of oxygen to one of the four subunits relaxes the conformation of the tetramer, which increases the affinity of other subunits for oxygen [32, 33], while also increasing Hb affinity for the erythrocyte membrane [24, 34]. The change in affinity by binding of oxygen is called cooperativity. Other effects on Hb are of allosteric nature, namely the Haldane and the Bohr effects, which also regulate the uptake and release of oxygen. The Haldane effect explains how deoxygenation of Hb increases its affinity for carbon dioxide [35], while the Bohr effect describes how the decrease of pH and the binding of protons reduces its oxygen-binding affinity [36]. 2,3-Bisphosphoglycerate (DPG), the isomer of the glycolytic intermediate 1,3- bisphosphoglycerate, binds preferentially to deoxygenated Hb (deoxyHb). The allosteric binding of DPG to deoxyHb lowers its affinity for oxygen further and promotes the release of oxygen in tissues where oxygen is needed most [37, 38]. Additionally, this interaction limits glycolysis within erythrocytes and stimulates the pentose phosphate pathway (PPP), which generates NADPH for biological processes [39] (see Section 1.4.2.4 for NADPH).
1.2.2 Haemoglobin autoxidation
The protection of the ferrous iron in the hydrophobic globin pocket of Hb is not complete. Spontaneous alterations of the conformation of the Hb pocket allow small anions or water to enter [40-42]. As a consequence, the oxygen is displaced, and removes an electron from the 3+ .- iron, forming Fe and superoxide (O2 ) [40, 43-45]:
2+ 3+ .- haem-Fe -O2 → haem-Fe + O2
This Fe3+ product is called methaemoglobin (metHb) [46, 47]. MetHb is unable to bind oxygen, which limits the cell’s ability to exchange O2 and CO2. Roughly 3-4% of Hb is oxidised to metHb daily [3, 43, 45], and acidic pH and intermediate oxygen pressures increase the rate of autoxidation [44, 48]. Autoxidation would severely impair the erythrocyte ability Chapter 1: Introduction 5
for oxygen transport over time. To counteract metHb production and limiting superoxide flux, erythrocytes contain metHb reductase. MetHb reductase reduces metHb back to ferrous Hb, through the utilisation of NADH. There is also a NADPH-dependent metHb reductase enzyme [3], but most metHb reductase activity in erythrocytes is NADH-dependent.
Superoxide is not very reactive with most biological molecules [49], but it is very short lived due to spontaneous or enzyme catalysed dismutation to H2O2 [49, 50].
.- .- + O2 + O2 + 2H → O2 + H2O2
H2O2 is a small molecule, which could allow it to diffuse easily through cell membranes.
There is also evidence that H2O2 diffusion might be facilitated by pore-forming proteins called aquaporins [51, 52]. H2O2 is a strong oxidant but has a high activation energy for most reactions, thereby increasing its lifetime in biological systems [53, 54]. This property enables it to act as a redox-signal molecule in various pathways [55, 56]. H2O2 readily reacts with Hb and metHb, leading to ferryl-Hb (Fe+4-Hb) or the oxoferryl-Hb-radical [45], which are strong oxidants themselves [57, 58]. Furthermore, H2O2 is able to recycle these species, which leads to even more superoxide production as well as Hb degradation in the case of reaction with Fe+4-Hb [59]:
4+ 3+ .- Fe -O + H2O2 → Fe +O2 + H2O
The reaction of H2O2 with oxygenated Hb (oxyHb) also gives rise to fluorescent Hb- degradation products [60]. These products have been observed in vivo and during blood storage [45]. Additionally, H2O2 releases iron from oxyHb [60]. The free iron is capable of exacerbating H2O2-induced oxidative stress via the Fenton reaction [61]:
2+ 3+ - . Fe + H2O2 → Fe + OH +OH
1.2.3 Haemoglobin denaturation
Haemichromes are denatured Hb [62]. The first step in haemichrome formation seems to be the dissociation of the Hb tetramer [63, 64], which leaves unstable Hbs inherently prone to Hb denaturation. There are two distinct forms of haemichromes, the reversible, and the irreversible. When the sixth coordination position of the haem-iron is occupied by OH-, or the distal histidine (His63) (Figure 1.2, middle), which is not involved in stabilising the iron, a reversible haemichrome is formed [65]. The reversible haemichromes are transient and can change back to metHb. If this position is occupied by a nitrogenous base, or protonated His63 Chapter 1: Introduction 6
(Figure 1.2, right), the haemichrome is irreversible [65]. Moreover, binding of an exogenous ligand, for example acetylphenylhydrazine or phenylhydrazine, also allows for haemichrome formation [43]. While all haemichromes are unstable, the irreversible haemichromes are especially prone to precipitation [66] and dissociation from the haem [62, 67]. The precipitate forms small, coccoid inclusion bodies (Heinz bodies), which represent the end product of Hb oxidation (see Figure 1.3).
Figure 1.2 Formation of reversible and irreversible haemichromes. Illustration of selected partial structures in the process of haemichrome formation. For a detailed description see text.
Haemichromes have a high and very specific affinity to the cytoplasmic domain of Band3, the membrane protein responsible for exchange of chloride and bicarbonate [68-70]. Two Band3 dimers bind five haemichrome dimers [69, 70]. OxyHb as well as deoxyHb also bind to Band3, but the interaction of Band3 and deoxyHb is favoured [71-73]. While binding of deoxyHb to Band3 is physiologically relevant for modulation of intracellular glycolysis [39], binding of haemichromes and Heinz bodies is known to damage Band3 and phospholipids, to change the erythrocyte structure, and to lead to vesiculation and removal of the erythrocyte [74-77] (see Section 1.2.4.3 and Figure 1.5). Haemichromes are argued to directly damage proteins and lipids in presence of H2O2 [75, 78] and to produce hydroxyl radicals by Fenton chemistry [79]. Haemichromes can also induce damage by releasing haemin (oxidised haem). Haemin is known to oxidise membrane protein thiols [80, 81]. Membrane-bound haemichromes are observed in blood storage, sickle cell anemia, β-thalassemia, glucose-6- phosphate dehydrogenase (G6PD) deficiency, and malaria [70, 75, 77, 82-87]. Chapter 1: Introduction 7
Figure 1.3 The sequence of events leading to Heinz body formation. The sequence of events during the oxidation of Hb results in its intracellular denaturation and precipitation as Heinz bodies. Most external and internal factors increase autoxidation rates and, therefore, increase metHb production.
1.2.4 Special cases
1.2.4.1 Haemoglobinopathies
There are a number of known single gene mutations that can affect the structure of Hb. The most common haemoglobinopathies are sickle cell disease, and unstable Hbs. Sickle cell disease is a hereditary blood disorder in which erythrocytes change their biconcave shape to a rigid, sickle shape. A mutation causes the β-chain of Hb to aggregate upon low oxygen tension. The Hb aggregates lead to the shape change, and the sickled erythrocytes obstruct capillaries. Since sickled erythrocytes have a limited life span, the disease results in haemolytic anaemia [88]. It is important to note here that the term haemolytic anemia is misleading, because the erythrocytes do not lyse, but rather get removed rapidly and prematurely due to markers of erythrocyte damage [89].
Unstable Hbs is an umbrella term for hereditary defects in Hb structure which present in patients as congenital haemolytic anaemias [90]. Several of the unstable Hbs have mutations which cause substitutions of hydrophilic amino acids to hydrophobic amino acids in the haem pocket of Hb. These substitutions facilitate autoxidation of the unstable Hbs, resulting in Hb precipitation and Heinz body formation upon heating, low oxygen pressure or drug induced oxidative stress [67, 90] (see Sections 1.2.2, and 1.2.3). Chapter 1: Introduction 8
1.2.4.2 Drug induced oxidative stress
Other sources for increased oxidative stress in erythrocytes are pharmaceuticals and xenobiotics, for example vicine, divicine, convicine, and isoramil found in fava beans [91-97]. Drugs and chemicals like aniline dyes, benzocaine (a local anaesthetic), nitrates, hydrazine derivatives and sulfons like dapsone are known to induce metHb anaemia and Heinz body haemolytic anaemia [98-101]. Some compounds such as pyrimidines deplete the free glutathione pool [94], whereas others form radicals by reacting directly with Hb or metHb [43, 102], thus accelerating metHb formation up to 1000-fold [103]. Drug induced oxidative stress is also a problem in G6PD deficiency. G6PD is an important enzyme in the PPP. In G6PD deficiency, the PPP is inhibited and NADPH production is therefore restrained. This severely limits the recycling of reducing systems including glutathione and peroxiredoxins during acute oxidative stress [94, 104, 105] (see Sections 1.4.2 and 1.5). In some individuals with G6PD deficiency, the consumption of fava beans consequently leads to an acute haemolytic crisis caused by oxidative stress called favism [106].
1.2.4.3 Erythrocyte storage
Another case where haemichromes are observed regularly within erythrocytes is during the storage of blood for transfusion [85]. When erythrocytes are stored at low temperatures and normal oxygen pressure in either SAGM (saline, adenine, glucose, mannitol) (see Table 2.7 for details), CPDA-1 (citrate, phosphate, dextrose, adenine) or AS-1 (additive solution, Adsol) for transfusion, the cells undergo biochemical and structural changes over time, generally referred to as the “storage lesion” [28, 107-109] (see Figure 1.5 for an overview). These changes include, but are not limited to, increased oxygen affinity [110-114], depletion of ATP [112, 115, 116], and modified membrane lipids and proteins [115, 117-119], some of which can be reversed [113, 120-122].
It is commonly acknowledged that oxidative stress contributes to the storage lesion [28, 114, 115, 123]. The observed oxidative damage during storage consists largely of membrane protein oxidation, and loss or oxidation of antioxidants like glutathione (GSH) [80, 118, 119, 124]. Some of the oxidative damages may be limited by addition of antioxidants like ascorbate, GSH, and N-acetylcysteine (NAC), a precursor amino acid for GSH [125-127]. However, similar changes also occur during erythrocyte senescence [24, 128, 129], which makes it difficult to distinguish between damage induced by oxidative stress during storage and damage induced by erythrocyte senescence ex vivo. Chapter 1: Introduction 9
Another problem is the lack of a single culprit for the observed oxidative damages [108, 130, 131]. The discussed origins of the observed oxidative stress ranges from high oxygen pressure during storage [122, 132, 133], decrease of the already acidic pH of storage buffer as well as depletion of endogenous antioxidants [134-136] and nitric oxide [137]. For example, erythrocytes rely solely on the glycolytic pathway to fulfil their energy needs [39]. Under normal conditions, 92% of glucose is consumed by this pathway [138]. To gain energy, one glucose molecule is broken down to two molecules of lactate and two protons. The free protons cause the environment of stored erythrocytes to become acidic over time [139-141], dropping the pH from 7.32-7.4 to below 7 in four weeks [142] (Figure 1.5, 2). The lower pH inhibits hexokinase and phosphofructokinase [32], two of the first three enzymes of the glycolysis pathway (Figure 1.5, 2). In consequence, this inhibition not only slows down the production of ATP and NADH, but also limits the production of NADPH by limiting substrates for the PPP (Figure 1.5, 3). Additionally, a low pH increases the breakdown of DPG [141, 143], an important allosteric effector of Hb for oxygen release, to 3- phosphoglycerate. 3-Phosphoglycerate is rapidly metabolised in glycolysis during storage [28]. It has been reported to drop from 12-15 µmol/ g Hb in fresh cells to 0.46-1.3 µmol after 28 days [142]. This means that after a short increase in ATP concentration due to the complete breakdown of DPG during the first two weeks of storage [144], ATP drops steadily [139, 145]. In the absence of ATP, the ATP-dependent pumps, including the calcium pump, slow down [146, 147], leading to an increase in intracellular calcium [139] (Figure 1.5, 1). Another ATP-dependent pump is the Na+/K+-ATPase, whose activity is restricted at low temperatures [28]. This means that slowly leaking potassium, as well as the increased calcium dependent loss of potassium, is not replaced [139]. The high concentration of extracellular potassium in stored blood has been known to cause severe arrhythmias and sudden death in patients since 1960 [148].
The most obvious structural change of erythrocytes is the change of shape. The lack of ATP, the increased intracellular calcium and the low pH cause the erythrocytes to transform progressively from a biconcave disk to an echinocyte with thorny projections, and then to a spheroechinocyte (see Figure 1.4) [109, 141, 145]. Spheroechinocytes are shaped more like balls and are unable to adapt their shape to squeeze through the microcirculation, which increases erythrocyte clearance [149-152]. Chapter 1: Introduction 10
Figure 1.4 Erythrocyte morphology during refrigerated storage. The biconcave erythrocyte changes progressively from a smooth discoid morphology to a more spherical form [109]. (1) Erythrocytes. (2a) Echinocytes stage 2. (2b) Echinocyte stage 3. (2c) Spheroechinocyte.
The shape change is caused by a loss of membrane surface [153], which includes both phospholipids and some structural membrane proteins. The loss of membrane is caused by budding of vesicles from the erythrocytes [154] (Figure 1.5, 5). These vesicles are rich in oxidised lipids and Hb [155], contain membrane proteins but lack spectrin [156-160], generally have phosphatidylserine (PS) exposed [161] (a signal for removal by macrophages), and can be problematic once transfused [85, 162, 163]. Yet, extreme vesiculation is only observed in a very small population of stored erythrocytes [28, 154], which includes the younger population [164, 165]. However, as long as the erythrocyte has not lost too much of its membrane, echinocytes are reversible by rejuvenation [120]. Rejuvenation occurs when the stored erythrocyte is incubated in warmed, fresh and nutrient rich buffer, which results in increasing ATP concentration, normalisation of calcium and potassium gradients and restoration of DPG [121, 122]. Rejuvenation solutions are FDA-approved for use between three days after erythrocyte collection and three days after expiration of the blood packs [166]. Another structural change affects Band3, a large anion transporter responsible for the exchange of bicarbonate and chloride, and anchor protein within the membrane (Figure 1.5, 7), which is reorganised to large oligomers as early as day 12 of storage [129, 156, 167] (Figure 1.5, 4). Band3 is lost from the erythrocyte membrane by vesiculation after three weeks or later. Band3 aggregation is found to promote removal of erythrocytes, fulfilling the role of an antigen for recognition by the immune system [2, 168-170]. Chapter 1: Introduction 11
Figure 1.5 Overview of the changes involved in the storage lesion. In clockwise order: (1) Cation homeostasis is affected by low temperatures and depletion of ATP and DPG. (2) Glycolysis produces ATP and lactate, which lowers the intra- and extracellular pH. The low pH inhibits glycolytic enzymes. (3) Autoxidation drives ROS production, which promotes the PPP and impairs GSH homeostasis. Inhibition of glycolysis impairs metHb recycling, leading to an increase in haemichromes. (4) Haemichromes bind to Band3. Oxidation of Band3 thiols by ROS lead to Band3 clusters, a clearance signal for the immune system. (5) Band3 clustering, together with increased calcium influx and PS exposure, impairs membrane deformability, increases osmotic fragility and leads to blebbing. (6) Protein oxidation is somewhat curbed by antioxidants and chaperone molecules. Yet, storage leads to redox modification to proteins and lipids (MDA, carbonylation, glycation, oxidation). Additionally, storage affects degradation of proteins and lipids. (7) Band3 is responsible for the modulation of pH, glycolysis, and indirectly influences gas exchange by Hb. Fragmentation of the cytosolic domain by ROS, calpain and caspases, as well as phosphorylation by kinases displaces glycolytic enzymes and structural proteins (Ankyrin (ANK), Band4.1, Band4.2). This impairs membrane stability and aids blebbing. (8) Increased calcium influx activates the Gardos channel, which increases cell dehydration and influences blebbing. (9) The increased calcium influx also activates proteolytic enzymes (caspases, calpain) and various kinases which change the phosphorylation status of many proteins. Figure reproduced and modified from [171].
Chapter 1: Introduction 12
Inhibition of glycolysis also decreases available NADH, which results in reduced activity of the enzyme metHb reductase, leading to the slow accumulation of metHb [28, 172]. MetHb accumulation is further elevated by a decrease in the pH of the buffer, which increases the autoxidation rate of Hb [173]. Enhanced autoxidation results in oxidative stress as it generates superoxide [3] (Figure 1.5, 3, also Section 1.2.2). This oxidative stress may be exacerbated if the unstable metHb loses its iron, making it available for Fenton reactions. Fenton reactions may drive the observed oxidation of proteins and membrane lipids [108, 115, 130, 174], together with the attachment of haemichromes to membranes [124, 175]. However, damage to membranes by haemichromes has been reported under both anaerobic and aerobic conditions [24, 60, 77, 85, 176, 177], and metHb does not decrease in erythrocytes stored under anaerobic conditions [178]. Additionally, it is more and more acknowledged that free Hb and free iron are not really free, but packed in membrane vesicles [155, 156].
While glycolysis slows down during storage, the increased oxidative stress results in a shift of erythrocyte metabolism towards the PPP [138, 139, 179]. This leads to an increase of NADPH available for antioxidant systems [139] (see Section 1.4.2.4). While recycling of oxidised glutathione (GSSG) to reduced glutathione (GSH) is dependent on NADPH and the GSH reductase, the increase of NADPH seems to have no effect on GSSG reduction [139, 180] (see Section 1.4.2.3). The decreasing activity of GSH reductase during storage may even be partially responsible for the accumulation of NADPH during storage and the decrease in GSH, since amassed GSSG is not recycled [139, 181]. Additionally, after 21 days in storage, the ATP dependent glutathione synthesis is inhibited [134], resulting in a dramatic loss of GSH [115, 126, 161]. However, not only GSH decreases, but also GSSG is lost during storage, which could be caused by degradation of GSH since cysteine, glutamate and glycine have been shown to accumulate [131, 139, 171, 182, 183]. The loss of glutathione and the lack of recycling and de novo synthesis could affect the other antioxidant systems.
One way to limit the effects of the storage lesion is to prevent the changes from arising in the first place. The removal of leukocytes from the stored blood was one of the ways to prevent damage. Leukocytes contain digestive enzymes which are released when they break down during storage. These digestive enzymes damage the erythrocyte membrane, for example by generating lysophospholipids and removal of terminal sugars [28, 184]. The practice of leukocyte reduction has resulted in reduced haemolysis and less inflammatory markers [162, 185, 186]. Other attempts focus on the improvement of the storage buffers. Traditionally, the storage buffers are acidic because the glucose within the buffers would caramelise at neutral Chapter 1: Introduction 13
or basic pH during autoclaving [187]. However, it has become clearer that the pH, sodium chloride and mannitol content, and buffering capacity all have a different effect on erythrocyte metabolism [141]. The invention of new storage buffers has shown considerable advantages [136, 140, 188-190], and some of these new buffers are FDA approved [191].
1.3 Exogenous source of oxidative stress in erythrocytes
1.3.1 Erythrocytes as scavengers
It has been acknowledged for years that erythrocytes are able to act as free radical scavengers or a “sink” for ROS generated in the vasculature [4, 5, 23, 192-195]. In many of these studies activated neutrophils have been proposed as source for ROS, mainly for H2O2 [4, 5, 192-200]. Further, it has been shown that perfusion of rat hearts with autologous erythrocytes is beneficial in ischaemia-reperfusion [201-203]. In ischaemia-reperfusion, mitochondria- derived ROS are acknowledged to participate in damaging the heart tissue [204, 205]. Additionally, it has been observed that erythrocytes are able to modulate T-lymphocyte proliferation and survival by scavenging ROS which arise after lymphocyte activation [206, 207].
1.3.2 Neutrophils
Neutrophils are the first line of cellular defence in infection and are a hallmark of acute inflammation. They have three methods for microbial killing: phagocytosis of microbes and digestion inside the phagosome [208], release of antimicrobial substances from their granules (degranulation), and release of neutrophil extracellular traps (NETs) [209].
Neutrophils carry three different kinds of granules: primary or azurophil granules, secondary or specific granules, and tertiary granules. The primary granules contain myeloperoxidase (MPO) and proteinases like elastase and lysozyme. The secondary granules hold lactoferrin, collagenase and more lysozyme, and the tertiary granules carry a variety of digestive enzymes [210, 211]. This underlines that the neutrophil main function is to digest pathogens using the contents of their granules [212], whether it is inside phagosomes or outside the cell. Upon activation by phagocytosis, neutrophils undergo an oxidative burst: they generate large quantities of superoxide and H2O2 while consuming oxygen [213-216]. The oxidative burst is caused by the NADPH oxidase NOX2 [217-219], which is located at the phagosomal membrane. NOX2 consists of 6 subunits, which assemble at the membrane of the phagosome after phosphorylation of subunits by protein kinase C (PKC) (see Figure 1.6). The assembly Chapter 1: Introduction 14
of the NADPH oxidase can be irreversibly blocked with the flavoprotein inhibitor diphenylene iodonium (DPI), or triggered by the PKC activator phorbol 12-myristate 13- acetate (PMA) [211]. Assembly can also be activated by binding of lipopolysaccharide (LPS), an endotoxin found on the outer membrane of gram negative bacteria, together with LPS- binding protein, found in serum, to CD14 on the neutrophil surface [220-225]. NOX2 oxidises
NADPH, taking one electron to reduce oxygen to superoxide. H2O2 is formed by dismutation of superoxide which occurs spontaneously in phagosomes.
Figure 1.6 NOX2 assembly on the membrane. The messenger molecule DAG (diacylglycerol), located on the membrane, or PMA activates PKC, which phosphorylates the NOX2 subunits p67 and p47. The phosphorylated subunits translocate to the membrane and form a complex with the GTPase RAC, p40 and the transmembrane subunits gp91 and p22 of the active NOX2. NOX2 oxidises NADPH to NADP+ and reduces oxygen to superoxide, which dismutates to H2O2. Phox denotes phagocytic oxidase.
Additionally, neutrophils are able to produce NETs to kill microbes [226, 227]. These NETs are released from neutrophils and consist of a fibrous network made from DNA and histones, in which microbes become trapped. Antimicrobial proteins and enzymes like MPO, elastase and calprotectin from granules are attached to the DNA scaffolding [228]. MPO is thought to be the main contributor to the antimicrobial activity of NETs [229] and to killing inside the phagosome [208, 230]. It primarily utilises H2O2 to produce HOCl [208, 218, 230, 231] (see Figure 1.7). HOCl reacts with amines to generate chloramines, and oxidises sulfhydryl groups [232]. HOCl and chloramines could oxidise erythrocyte proteins, for example Prx2 [233-235]. To prevent damaging oxidation, erythrocytes have sophisticated antioxidant mechanisms. Chapter 1: Introduction 15
Figure 1.7 The production of HOCl by MPO. Ferric MPO reacts with H2O2 to form Compound I. Compound I oxidises halogens and thiocyanate to their respective hypohalous acids. Simplified and adapted from [236].
1.4 Antioxidant mechanisms in the erythrocyte
Erythrocytes do not have a nucleus, and they have to maintain their structure and function for 120-150 days without the capacity to synthesize new proteins that may become oxidatively damaged. Also, the membranes of erythrocytes contain high amounts of polyunsaturated fatty acids, which are prone to radical chain reactions [3]. Additionally, free catalytic iron from haemoglobin denaturation may be present in the erythrocyte cytoplasm, available for Fenton reactions (see Sections 1.2.2 and 1.2.3). Moreover, erythrocytes face constant physical stress since they deform to fit through the capillary net, which increases cation leakage [237]. For these reasons, erythrocytes feature enzymatic and non-enzymatic antioxidant defence systems.
1.4.1 Non-enzymatic systems
The diet-derived and low molecular weight radical scavengers ascorbate, tocopherols and carotenoids can act as sinks for a variety of ROS. Superoxide, H2O2 and other radicals can be removed by ascorbate, even though it is not very efficient [238-240]. α-Tocopherol, a powerful intercepting radical chain-breaking compound in the membrane, scavenges peroxyl radicals [241]. During this process, a relatively stable tocopheryl-radical is formed, which is usually resolved by ascorbate. While all diet-derived antioxidant levels depend on intake, ascorbate in erythrocytes is approximately 60% of the ascorbate plasma level [242], carotenoids are selectively enriched [243, 244], and tocopherol levels seem to be dependent on plasma lipoproteins [243, 245]. Chapter 1: Introduction 16
1.4.2 Enzymatic systems
1.4.2.1 Superoxide dismutase
Superoxide is converted to H2O2 by superoxide dismutase (SOD). The SOD in erythrocytes contains copper (Cu) and zinc (Zn) ions in the active site. The Cu- ion is oxidised and reduced .- in turn to catalyse the dismutation of O2 [246], whereas the Zn-ion stabilises the enzyme:
2+ .- + SOD-Cu + O2 → SOD-Cu + O2
+ .- + 2+ SOD-Cu + O2 + 2H → SOD-Cu + H2O2
.- .- + Net reaction: O2 + O2 + 2H → O2 + H2O2
.- 5 -1 -1 Although O2 is able to dismutate spontaneously, the rate is low (km of 5 x 10 M s at pH 7) and pH dependent [49]. Conversely, SOD accelerates this reaction to at least 1.5 x 109 M-1s-1 and is almost pH independent [247]. The SOD concentration in erythrocytes is approximately 0.73 ± 0.10 µg/mg protein [248], with an activity of 2352 U/g Hb [249].
1.4.2.2 Catalase
Catalase is a large enzyme with four homologous 60 kDa subunits. It converts H2O2 to water and oxygen. Each subunit contains a Fe3+-heme group in the active site [247], used to reduce their substrate:
3+ Catalase-Fe + H2O2 → compound I + H2O 3+ Compound I + H2O2 → Catalase-Fe + H2O + O2
Net reaction: 2 H2O2 → 2 H2O + O2
7 -1 -1 Catalase is one of the fastest enzymes known, with a rate constant of kH2O2 ~10 M s [250,
251]. One molecule of catalase is able to decompose millions of H2O2 molecules per second, and its mechanism makes it almost unsaturable at high concentrations of H2O2 [252]. Catalase was previously thought to be the main enzyme required for the removal of high concentrations of H2O2 [253-255].
1.4.2.3 The glutathione/ glutathione peroxidase/ glutathione reductase system
The other enzyme that has been considered as being the main enzyme for H2O2 removal in erythrocytes is glutathione peroxidase 1 (Gpx1) [256-258]. Gpx1 consists of four subunits that each contain an active site selenocysteine [259], which catalyses the following reaction with hydrogen peroxide as substrate: Chapter 1: Introduction 17
- + Gpx-Se + H2O2 + H → Gpx-SeOH + H2O
Gpx-SeOH + GSH → GSH-Gpx-SeOH → H2O + Gpx-Se-SG Gpx-Se-SG + GSH → GSH-Gpx-Se-SG → Gpx-SeH-GSSG → Gpx-Se- + H+ + GSSG
Net reaction: 2 H2O2 +2 GSH → GSSH + 2 H2O
In summary, GPx1 reduces hydroperoxides (ROOH), H2O2 or lipid peroxides with two GSH molecules as hydrogen donors [256], which results in an oxidised, disulfide-linked GSSG. The generated GSSG is recycled back to GSH by glutathione reductase (GR). Under stress conditions, Gpx1 is able to protect a wide range of tissues from oxidative damage [260]. However, knockout (KO) models of Gpx1 show no phenotype [261], until stressed with redox-cycling herbicides like paraquat, a compound that produces high amounts of superoxide [262, 263]. This indicates Gpx1 may not play such a defined role in the removal of basal
H2O2 levels, even though its rate constant for the reaction with H2O2 in humans (formation of Gpx-SeOH) is 4.1×107 M-1s-1 [264].
GSH is a small peptide consisting of the three amino acids glycine, cysteine and glutamate (Figure 1.8) and has an unusual gamma-peptide bond between glutamate and cysteine. The GSH concentration in erythrocytes is approximately 2 mM [254] and is important to reduce the low level of oxidative stress in cells [265]. GSH is defined as an antioxidant because its thiol group is able to serve as an electron donor, allowing reduction of important thiols on other proteins. The oxidised form of GSH is a disulfide linked dimer, referred to as GSSG. Oxidative stress is generally accompanied by an increase in GSSG, and is expressed as the GSH/GSSG ratio. The GSH/GSSG ratio is lower in cells experiencing high levels of oxidative stress [266].
Figure 1.8 Structure of glutathione.
Chapter 1: Introduction 18
Recycling of oxidised GSSG to reduced GSH is achieved by GR in a NADPH-dependent reaction [267]:
GSSG + NADPH + H+ → 2 GSH + NADP+
Additionally, GSH is involved in other processes. It is able to act like an antioxidant itself, reacting directly with a variety of ROS [267, 268]. GSH is also involved in recycling of ascorbate in erythrocytes [267, 269, 270]. All animal cells are able to synthesise GSH [268], even erythrocytes. The synthesis of GSH involves two enzymes, γ-glutamylcysteine synthetase, an enzyme which catalyses the rate limiting first step, and glutathione synthetase required for the second step [271]:
glutamate + cysteine + ATP → γ-glutamyl-cysteine + ADP + Pi
γ-glutamyl-cysteine + glycine + ATP → GSH + ADP + Pi
1.4.2.4 NADPH
NADPH is a necessary co-factor and co-substrate in most antioxidant systems. It provides the needed reducing equivalents for metHb reductase activity (see Section 1.2.2), for recycling of antioxidants like GSH and could even represent a repair or protection pathway, at least for catalase [272]. The PPP is responsible for maintaining NADPH levels. During the normal metabolism of erythrocytes within the blood stream, 8% of glucose is metabolised through the PPP, and 92% through glycolysis [138]. This changes during oxidative stress, which is able to increase the metabolism of glucose by the PPP to 90% [138, 179, 273].
For decades, there has been an ongoing debate about which of the two enzymes, catalase or
GPx, is the most significant for removal of H2O2 [253, 254, 256-258, 274] within erythrocytes. It was generally agreed that catalase is more efficient to remove high concentrations of H2O2 since it practically cannot be saturated, and that Gpx is better suited for removal of low concentrations of H2O2 [255, 275, 276]. Things changed when Johnson et al. showed that peroxiredoxin 2 was necessary in their model of peroxide metabolism within the erythrocyte [255], and when Low et al. demonstrated that peroxiredoxin 2 (Prx2) reacted faster with H2O2 than previously thought [233]. This showed that the significance of Prx2 had been underestimated for many years.
Prx2, a member of the highly conserved Prx family, is the main focus of this thesis. For this reason, a general introduction to Prxs is necessary before their possible functions in erythrocytes can be explored. Chapter 1: Introduction 19
1.5 Peroxiredoxins
The Prxs are a family of thiol-specific antioxidant proteins, which are able to reduce H2O2, organic hydroperoxides, peroxynitrite, and even chloramines and hypochlorous acid [234, 235, 277-282]. Prxs occur ubiquitously in cells of every known life form, even in archaea [277, 283-285]. They are located not only in the cytosol, but are also found in mitochondria, nuclei, endoplasmic reticulum, and bound to the inner surface of membranes [286]. Prx isoforms can make up to 0.1-1% of soluble protein in various cultured cells and rat tissues [287-289]. Most cell types contain more than one isoform of Prx. Erythrocytes, for example, contain Prx1, Prx2 and Prx6 (a 1-Cys Prx).
However, the other two isoforms are not expressed as highly as Prx2 [55, 290, 291]. Prxs are obligate homodimers [277, 278]. There are six members of the Prx family in mammals, grouped into three main classes: the typical 2-Cys Prx, the atypical 2-Cys Prx and the 1-Cys Prx [277] (see Table 1.2). These three classes differ on the number of cysteinyl residues which are involved in catalysis [292] and the catalytic mechanisms.
Table 1.2 Properties of mammalian peroxiredoxins. The molecular masses are for the monomers of human Prx, the nomenclature lists human and other mammalian isoforms. Note the separation of exactly 121 residues between the two cysteines involved in catalysis in the typical 2-Cys Prxs. Details selected and reproduced from [277].
Prx1 Prx2 Prx3 Prx4 Prx5 Prx6
Class Typical 2- Typical 2-Cys Typical 2-Cys Typical 2-Cys Atypical 2-Cys 1-Cys Cys Previous HBP23 Torin AOP-1 AOE372 AOPP AOP-2 nomenclature PAG Calpromotin MER5 Trank AOEB166 ORF-6 NKEF-A NKEF-B SP22 PMP20 MSP23 Molecular 22.1 21.8 27.7 (cleaved 30.5 22.0 (cleaved 24.9 mass (kDA) to 21) to 17) Peroxidatic/ Cys52/ Cys51/ Cys108/ Cys124/ Cys100/ Cys47/ No resolving Cys173 Cys172 Cys229 Cys245 Cys204 resolving Cys Cys residues Subcellular Cytosol, Cytosol, Mitochondria Endoplasmic Mitochondria, Cytosol location nucleus membrane reticulum, peroxisome, extracellular cytosol, space nucleus
1.5.1 The catalytic cycle of peroxiredoxins
Prx monomers are orientated in a head-to-tail manner [293]. One monomer uses its peroxidatic cysteine (Cp) to attack the peroxide and is oxidised to a sulfenic acid (Cp-SOH) [294]. This appears to be common to Prxs [295-297]. In typical 2-Cys Prxs, the new sulfenic acid group then reacts with the resolving cysteine (Cr) on another monomer and forms an intermolecular disulfide bond. This bond is predicted to be reduced by thioredoxin (Trx) and Chapter 1: Introduction 20
thioredoxin reductase (TrxR) in mammalian cells (see Section 1.5.2), or another thiol-based protein [233, 277, 278, 293, 297-300] (Figure 1.9).
Figure 1.9 The catalytic cycle of typical 2-Cys peroxiredoxins. The functional unit of a 2-Cys Prx is a head-to-tail homodimer. The peroxidatic cysteine of each monomer is oxidised (e.g. by H2O2) to a sulfenic acid. The resolving cysteine of the opposite monomer reacts with the sulfenic acid to form an intermolecular disulfide that is reduced back by the thioredoxin/ thioredoxin reductase/ NADPH system. Adapted from [301].
However, glutaredoxin and glutathione are inadequate reductants [287], while in vitro dithiotreitol (DTT) is able to substitute for Trx [302, 303]. In atypical 2-Cys Prxs, the intermolecular disulfides rearrange into intramolecular disulfide bridges [304], which also depend on Trx for regeneration [288]. Since 1-Cys Prxs lack the Cr, the sulfenic acid is reduced by GSH and π-glutathione-transferase (π-GST) in a two-step mechanism [305-307]:
Prx6-SOH + GSH → Prx6-SSG + H2O (catalysed by π-GST) Prx6-SSG + GSH → Prx6-SH + GSSG
The early enzymatic studies on Prxs deemed them catalytically inefficient compared to GPx 8 -1 -1 7 -1 -1 (kH2O2 ~10 M s ) and catalase (kH2O2 > 10 M s ), because they exhibited a rate of kH2O2 ~104-105 M-1s-1 [278, 308-312]. However, these measurements were limited by the Trx/TrxR- coupled oxidation of NADPH [313, 314]. More recent studies, based on a competitive kinetic approach, showed a reactivity which is roughly two orders of magnitude higher: kH2O2 ~1.3 x 7 -1 -1 7 -1 -1 10 M s for human Prx2 [235], and kH2O2 ~1-4 x 10 M s for bacterial and yeast Prxs [314,
315]. This demonstrates that Prx are effective scavengers of H2O2, comparable to catalase, even though they are limited by their recycling rates. This is especially true at low intracellular concentrations or activities of Trx and TrxR, like in human erythrocytes [233], Chapter 1: Introduction 21
which may lead to accumulation of the oxidised form of Prx2 under conditions which may overwhelm antioxidant systems.
1.5.2 The thioredoxin/ thioredoxin reductase system
Trx is a 12 kDa polypeptide with two sulfhydryl (-SH) groups in its active site that can form an intramolecular disulfide bond when oxidised [316]. This allows Trx to covalently bind to other oxidised proteins with thiol groups and form a mixed disulfide. Trx then reduces the other protein by becoming oxidised. Oxidised Trx is recycled by thioredoxin reductase (TrxR) [316, 317]. TrxR is a homodimeric (55 kDa subunits) and NADPH dependent enzyme. It bears a selenocysteine residue in its active centre, which modulates its activity [318, 319]. When TrxR is oxidised, in contains two disulfide bonds, one between a N-terminal dithiol motif, and the other between the active site selenocysteine and its adjoining cysteine [320, 321]. Both disulfide bridges are reduced by NADPH each redox cycle via a flavin-adenine dinucleotide prosthetic group. A simplified mechanism of the occurring electron transfer is shown in Figure 1.10.
It has recently been discovered that Trx also contains additional non-active site thiols that can be oxidised as well to an intramolecular disulfide [322]. This oxidation prevents reduction of Trx by TrxR, and also its ability to inhibit apoptosis signal-regulating kinase 1 (ASK1) [323, 324]. The two-disulfide forms can be re-activated by reduction through GSH, glutathione reductase, and glutaredoxin1 [322].
Figure 1.10 The thioredoxin cycle. Protein disulfides are reduced by Trx, which is oxidised to a disulfide. In return, the oxidised Trx is reduced via TrxR, which utilises NADPH from the PPP as reducing equivalents. For clarity, only one subunit of TrxR is shown, and the N-terminal dithiol motif is not included. Adapted from [325].
Chapter 1: Introduction 22
1.5.3 Hyperoxidation of peroxiredoxins and its repair by retroreduction
Eukaryotic Prxs are more susceptible to hyperoxidation than prokaryotic Prxs [326]. The Cp
Prx sulfenic acid can be oxidised to a sulfinic acid (-SO2H), or even a sulfonic residue (Cp-
SO3H) [327-330], (see Figure 1.11). Once the Cp becomes oxidised, the protein has to unfold locally to allow disulfide formation, otherwise the opposing Cp and Cr of the two monomers would be too far apart to interact (13 Å separation in Prx2) [293]. In prokaryotes, unfolding is faster and therefore prokaryotic Prxs are less susceptible for hyperoxidation [277]. After comparison of crystal structures and sequences of 2-Cys Prx isoforms in all three redox states, an explanation for the susceptibility of hyperoxidation of eukaryotic Prxs was found [326]. Eukaryotic 2-Cys Prxs have two distinct sequence motifs, the GGLG motif located in the middle of the protein, and the YF motif found at the C-terminus of the protein [326]. These motifs are absent in prokaryote Prxs [293]. In Prx2, both motifs are not directly part of the active site Cp, but surround it. The highly conserved active site that surrounds the Cp serves as a solvent-accessible pocket [277]. In the reduced state, the GGLG-motif “protects” the active site and keeps the resolving Cr buried [326]. Movement of the C-terminal arm brings the two Cys residues close enough to allow disulfide formation [293]. However, the structural rearrangements slow down disulfide bond formation [326] and H2O2 is able to hyperoxidise the peroxidatic cysteine [327, 329, 330] (Figure 1.11), inactivating the peroxidase activity. Thus, a mutation of the C-terminus of these sensitive Prxs would render the Prxs more resistant to hyperoxidation. This resistance was reported using mutated TpxY, a 2-Cys Prx in yeast [331], and in mutated 2-Cys Prxs of Schistosoma mansoni [332]. In Prx1, only 0.072% of the active enzyme is hyperoxidised under steady state H2O2 levels (Glucose oxidase, H2O2 < 1 µM) per redox cycle [329]. Under these conditions, the hyperoxidised form is allowed to accumulate. With concentrations of H2O2 in excess to Prx, substantial amounts of hyperoxidation are also achievable without cycling [333]. Chapter 1: Introduction 23
Figure 1.11 Hyperoxidation of typical 2-Cys peroxiredoxins. The eukaryotic 2- Cys Prxs are able to undergo hyperoxidation [334]. If hydrogen peroxide attacks the already formed sulfenic acid before it is able to form a disulfide, a sulfinic or even sulfonic acid will form. While the sulfenic acid residue can be resolved in an ATP-dependent process by sulfiredoxins [335], no such pathway is known for the sulfonic acid. Adapted from [301]
Hyperoxidation of Prx cannot be reversed by reducing agents such as GSH or Trx. However, reversal of hyperoxidation was observed during experiments with cultured cells involving metabolic labelling, and the first biological sulfinic acid reductases, the sulfiredoxins, were unmasked [336]. Sulfiredoxin (Srx) is a small 13 kDa protein which was first discovered in yeast [335], but it has been detected in plants and mammals as well [337, 338], for example in erythrocytes [339]. The reactivation of hyperoxidised Prx by Srx is specific to 2-Cys Prxs only, hyperoxidised forms of Prx5 and Prx6 cannot be recycled [340]. It has been shown that Srx uses ATP and reductants like Trx and GSH during a four-step reaction cascade [341, 342] -1 (Figure 1.12). However, with a kcat of 0.18 min , Srx is too slow to be an efficient enzyme [296, 337].
Figure 1.12 Simplified mechanism of peroxiredoxin hyperoxidation repair by sulfiredoxin. As soon as the Srx-ATP complex binds to Prx, ATP is hydrolysed [343]. Dephosphorylation and subsequent reduction steps are mediated by low molecular weight thiols (RSH) such as GSH and Trx. Only one Prx monomer is shown for clarity. Chapter 1: Introduction 24
1.5.4 Higher molecular weight structures of peroxiredoxin
While the catalytic unit of Prx is a homodimer, many isoforms have been observed to oligomerise. Studies on purified proteins, most importantly crystallographic studies, have revealed a variety of quaternary structures [277, 299, 312, 344]. Human Prx1 and Prx2, as well as bacterial AhpC, form decamers, or rather a pentamer of dimers [293, 299, 345, 346], while Prx3 forms a dodecamer [347]. High ionic strength was reported to increase [348] as well as to decrease [349] oligomerisation. The conditions that seem to aid the assembly of the decamers include low pH [350], high calcium concentrations [351, 352], phosphorylation [353], protein concentration [354] and a reduced or hyperoxidised redox status of the peroxidatic centre [293, 299, 355, 356]. These observations have firmly linked redox status and decamer formation [277]. It is apparent at this point that the reduced and hyperoxidised forms prefer the decameric formation [357]. While disulfide formation favours dimers, oxidised Prxs exist in an equilibrium between decameric and dimeric forms, depending on protein concentration [282, 293, 299, 314, 348, 350, 358]. It is thought that the decameric structure stabilises the active site and thereby increases the Prx effectiveness as a peroxidase, while the large conformational changes required for disulfide formation destabilises the dimer:dimer interface of the decamer, leading to its dissociation [293, 299, 314].
The oligomerisation to higher molecular weight structures has been linked to chaperone activity, after it was discovered that exposure to heat or high H2O2 induced oligomerisation of yeast Prx into spherical structures, preventing protein denaturation or cell death, and abolishing peroxidase activity [359]. Heat and high concentrations of H2O2 had a similar effect on Prx2 in HeLa cells [360], and even the usually hyperoxidation resistant AhpC was coaxed into comparable changes [361]. Since hyperoxidation occurred in all those studies, it was proposed that hyperoxidation could be a physical and functional switch for Prx [362]. However, phosphorylation also induces oligomerisation of Prx1 [353], which does not affect its redox state. Additionally, even though the redox status of the peroxidatic cysteine appears to be important for the stability of the oligomer, the chaperone activity does not seem to be dependent on hyperoxidation [357]. Moreover, it was observed during sedimentation experiments that oxidised AhpC also presents as a mixture of oligomeric structures varying from dimers to decamers, depending on the protein concentration [299]. Additionally, Prx1 seems to have a higher chaperone activity than Prx2, which is predicted to be the result of an extra cysteine which can stabilise oligomeric structures through disulfide bonds [294]. This Chapter 1: Introduction 25
suggests that it might not be hyperoxidation by itself that activates the chaperone, but rather the tendency to form stable decamers in general [363].
Interestingly, some Prxs show even higher molecular weight structures than decamers and dodecamers. Prx3 was reported to form two dodecameric rings, linked into each other [347], and just recently Prx3 dodecamers have been used to form long fibres for future nanotechnological implications [364]. Before Prx2 had been recognised as a member of the typical 2-Cys Prxs, it was called torin [345], which was based on the round decameric shapes that were observed on erythrocyte membranes via transmission electron microscopy (TEM) [358, 365-367] (Figure 1.13, A, upper third; also Figure 1.15). It was found that the decameric rings were able to stack and form long high molecular weight fibrils in vitro [345] (Figure 1.13, A, lower half). Under specific conditions, Prx2 was even shown to be able to form ball- like shapes like the 12 decamer dodecahedron [368] (Figure 1.13, B). Prx1 was observed to form dense ball-like shapes [359], and two-ring chaperones with a clear loss of peroxidase activity [356].
Figure 1.13 Higher molecular weight structures of peroxiredoxin 2. (A) Transmission electron microscopy showing decameric rings (upper third) and 2-D arrays of Prx2 in vitro, scale bar indicates 100 nm [345] (B) 3D-reconstruction of the Prx2 dodecahedron, based on TEM data [368].
1.5.5 Proposed functions of peroxiredoxins
The importance of Prxs within the antioxidant network is impressive. Prx2 is able to protect Hb and erythrocyte membranes from external oxidants and oxidation related aggregation [365, 369]. It may even be capable of removing hydroperoxides directly at the membrane [286], and seems to be the sole protein protecting erythrocytes from oxidation-induced haemolysis in mice [301, 370] (see Section 1.6.4 for details on Prx2 KO mice). Mice lacking Prx1 show haemolytic anaemia, but to a less severe degree than Prx2 KO mice, and they are predisposed to various cancers and cardiovascular problems [290, 371]. In diabetic men, Prx1 Chapter 1: Introduction 26
is upregulated in erythrocytes [372], and mice lacking Prx6 are more prone to ischaemia- reperfusion injury [373, 374]. Chronic inflammation and ROS are involved in the pathogenesis of all these conditions [375], which underlines the antioxidant role of Prxs.
However, the evolutionary conservation of the motifs leading to hyperoxidation of eukaryotic Prxs, and therefore less efficient peroxidase activity points towards an alternate biological role for Prxs. The “floodgate model” was the first hypothesis for a role of Prxs in cell signalling (Figure 1.14, A). In the flood gate model, Prxs take the role as redox sensors [326]. In a resting cell, the high amount of Prx present [287] is capable of detoxifying H2O2 [376]. Given
H2O2 can act as a second messenger in receptor-mediated cell signalling and other pathways [377, 378], this might limit or even completely inhibit signals [55, 326, 379]. Increased generation of endogenous H2O2 would lead to hyperoxidation and inactivation of Prx and to accumulation of the dimer [327]. This would increase H2O2 levels in the local environment of inactivated Prx. H2O2 could then act as a second messenger by oxidising other target proteins [326, 377, 380, 381].
An alternate role for Prxs in cell signalling could be as a sensor protein for H2O2-mediated signals. Within a cell, various thiol-containing proteins compete with each other to react with
H2O2 [382]. This means that to react with a low flux of H2O2, the protein needs to have a high specificity and reactivity towards H2O2, and to either be localised at the source of the H2O2 or highly abundant [383, 384]. The Prxs fulfil these criteria by having both a high reactivity (~107 M-1s-1) and abundance (0.2-1% soluble protein) [277, 302, 385]. Once they are oxidised, the Prxs could relay the “signal” via oxidation of specific thiol-protein targets [386- 388]. This behaviour can also be interpreted as disulfide exchange with other downstream sensor proteins (Figure 1.14, B) [334]. Disulfide exchange has been observed in yeasts, where the disulfide is handed from Prx to Trx, which releases the Trx-bound apoptosis signalling kinase (ASK1) [389], or the activation of Yap1, a transcription factor, by the Prx oxidant receptor protein (Orp1) [390].
As a signalling protein, Prxs seem to be involved in the regulation of various biological processes. For example, Prxs have been proposed to be involved in circadian oscillation in both erythrocytes [391, 392] and adrenal cortex cells [393]. Prx3 apparently regulates apoptotic signalling in mitochondria [394] as well as differentiation of erythrocytes [395]. Prx6, as the only mammalian 1-Cys Prx, has been shown to have calcium independent phospholipase A2 activity [396], involved with the production and recycling of surfactant Chapter 1: Introduction 27
[397]. It has also been found to interact with the NADPH-oxidase of neutrophils, regulating its activity [398].
Alternatives for the physiological role of Prxs also include the mentioned chaperone activity (Figure 1.14, C) (see Section 1.5.4). It was observed that the formation of stable high molecular weight oligomers of yeast Prx by hyperoxidation is able to prevent the aggregation of protein during heat shock [359]. In mammalian cells, the chaperone activity of Prxs seems to be able to prevent apoptosis [360].
Figure 1.14 Possible roles for hyperoxidised and oxidised peroxiredoxin. (A) The floodgate model is a loss of function model, where the inactivation of Prx by oxidation/ hyperoxidation allows the signal molecule H2O2 to oxidise its main targets. (B) Disulfide exchange is a model, in which the oxidation of Prx allows downstream transfer of disulfide bridges. Hyperoxidation would inhibit said transfer, which allows tight regulation of the activity of target proteins, with or without intermediate transduction proteins like Trx. (C) The chaperone switch is a gain of function model, where the role of peroxiredoxin changes from antioxidant to protection from denaturation. Adapted from [334] and [382].
Chapter 1: Introduction 28
1.6 Peroxiredoxin 2 in erythrocytes
Prx2, a typical 2-Cys Prx, is the third most abundant protein in the erythrocyte with an intracellular concentration of either 240 µM [399], 410 µM [248], or 5.6 mg/ ml of packed red blood cells [400], depending on the assay used. Erythrocytes also contain Prx1 and Prx6, but in comparison the Prx2 concentration is 80 times the concentration of Prx1, and 200 times more than Prx6 [248].
1.6.1 History of peroxiredoxin 2 discovery
In 1968, Prx2 was discovered by Harris et al. as one of two new proteins released from fragmented, Hb-free erythrocyte ghosts [358]. By electron microscopy, one of the observed proteins appeared as ring structure [358], the other, a stacked four ring structure, was later determined to be the 20S proteasome [401, 402]. Harris also suggested at the time that the ring structures were formed from ten smaller subunits [358, 365, 366] and were not an integral part of the erythrocyte membrane, but a cytosolic protein [366]. Because of the ring shape, it was named torin [403]. The antioxidant properties of Prx2 were determined 20 years after its discovery [302, 400], and its structure was finally solved in 2000 by Schroder et al. [293] (Figure 1.15).
Figure 1.15 Crystal structures of dimeric and decameric peroxiredoxin 2. (A) Ribbon diagram showing reduced Prx2 dimer. (B) Surface diagram showing reduced Prx2 decamer. Monomers alternately coloured in teal or dark purple. Structure rendered on PyMol 0.99 using data deposited in the Protein Data Bank [293], accession number 1QMV.
1.6.2 The reactivity of peroxiredoxin 2 in comparison with other antioxidant systems
Before the discovery of Prx2, it was believed that catalase and Gpx embodied the main defence against H2O2 in erythrocytes. The debate about which of the two was more significant spanned decades [253, 254, 256-258, 274]. It was generally agreed that catalase is more efficient to remove high concentrations of H2O2 and Gpx is better suited for removal of low Chapter 1: Introduction 29
concentrations of H2O2 [255, 275, 276]. Peskin et al. and Low et al. demonstrated that Prx2 reacted faster with H2O2 than previously thought [233, 235]. In these studies, based on a competitive kinetic approach, a rate constant of ~1.3 x 107 M-1s-1 for human Prx2 was determined [235]. Therefore, Prx2 is able to compete with catalase, which has a kH2O2 of 3 x 107 M-1s-1 in humans [250]. Additionally, Johnson et al. showed that Prx2 was necessary in their model of peroxide metabolism [255].
1.6.3 Peroxiredoxin 2 susceptibility to oxidation and hyperoxidation
Prx2 in erythrocytes is mainly reduced [233]. However, it is highly sensitive to oxidation by 6 H2O2. At 5 x 10 cells/ml, 5 µM H2O2 is enough to cause significant accumulation of oxidised Prx2 dimer, despite other antioxidant pathways being present [233]. The accumulation of oxidised Prx2 dimer is easily detected via non-reducing SDS-PAGE [233]. Prx2 oxidation can also occur during cell lysis, due to trace amounts of H2O2 in lysis buffers [235].
Recycling of oxidised Prx2 in erythrocytes is slow, which could be because TrxR is only present a low concentrations [233]. When TrxR was inhibited, oxidised Prx2 accumulated, and was prevented by carbon monoxide, implying that Hb autoxidation is a major source of
H2O2 in the erythrocyte. The low TrxR activity is sufficient to keep Prx2 mainly reduced under normal conditions. Increased oxidative stress in inflammation or blood storage, however, may lead to accumulation of the oxidised dimer.
Even though Prx2 has been shown to become hyperoxidised in various other cell types [329, 330, 404], in erythrocytes, it shows unusual resistance [233, 248]. High concentrations of hydrogen peroxide are necessary to induce hyperoxidation in erythrocytes [333]. Only very recently, hyperoxidation has been observed in erythrocytes under physiological conditions [392]. The lower amount of active TrxR may be a reason for this behaviour [233], since active Prx2 cycling facilitates hyperoxidation [329]. Other possibilities include N-terminal de- methionylation and acetylation or carbamylation of Prx2 [293, 405]. Recombinant mutants lacking the N-terminal modification showed increased susceptibility to irreversible hyperoxidation (-SO3H), a less ordered structure and a lower thermostability [405]. This was interpreted as a change in secondary structure, caused by the N-terminal modification, leading to a higher thermostability of Prx2. Another possibility could be a structural limitation of access to the active site due to structural changes induced by binding to other proteins [406, 407]. The high presence of catalase is also not a factor, since inhibition of catalase only modestly increases the sensitivity to hyperoxidation [233]. Chapter 1: Introduction 30
1.6.4 Haematological phenotypes of peroxiredoxin knock out mouse models
The physiological importance and interplay of the erythrocyte antioxidants catalase, GPx1 and Prx2 has been studied in various murine knockout models. Catalase-deficient mice showed no haematological abnormalities [275]. This was also the case in humans with congenital acatalasaemia [408-410], a condition characterised by very low levels of catalase. This implies that, under normal conditions, catalase is not a crucial enzyme for the erythrocyte.
Gpx1 KO mice also presented a haematological profile similar to wild type mice, and showed equal resistance to H2O2 induced haemolysis [261, 411]. While Gpx1 deficiency has been seen occasionally with haemolytic disorders in humans, a clear association has so far been elusive [412, 413]. Overall, this implies a redundancy for both Gpx1 and catalase in the antioxidant network of erythrocytes absent of oxidative stress.
In comparison, the Prx2 knockout mouse shows a variety of health problems [370]. The mice have severe haemolytic anaemia, a lowered Hb content, a lowered haematocrit, splenomegaly, and Heinz bodies within their erythrocytes. While the lowered Hb content and haematocrit are results of haemolysis, the higher reticulocyte count indicates erythropoetic compensation for the haemolysis, and the splenomegaly is usually caused by an abnormal high clearance of erythrocytes. The occurrence of Heinz bodies in Prx2 KO mice, as early as five weeks, is an indication of increased oxidative stress, which causes the destruction of Hb. The basal levels of metHb in the KO mice were the same as in wild type mice, but there was increased formation of metHb on addition of H2O2. Additionally, the erythrocytes also had a subtle increase in membrane thiol oxidation, which would trigger increased erythrocyte clearance. Other enzymes known for susceptibility to haemolytic anaemias, G6PD and pyruvate kinase [414], were unaffected, as were catalase and GPx1. Since catalase and GPx1 were unable to metabolically compensate for the lack of Prx2 in these mice, it points to Prx2 having an essential function in erythrocytes under normal metabolic conditions.
There have also been Prx1, Prx3 and Prx6 knockout mice studies. Prx6 and Prx3 KO mice have a normal haematological phenotype [415, 416]. This allows for the assumption that Prx6 and Prx3 are not of vital importance in the erythrocyte, even though Prx3 might be of importance during erythropoiesis [395]. Prx1 KO mice, however, develop malignant tumours and Heinz body anaemia. The Heinz body anaemia develops after nine months, which is a much later onset than in Prx2 KO mice [290]. It was proposed that the delayed onset of Chapter 1: Introduction 31
anaemia indicates a metabolic compensation which could not be sustained with age [290]. While Prx1 shares 91% homology to Prx2, is also a cytosolic protein which has been observed bound to membranes [417], and has the same catalytic mechanism, they differ in concentration within the erythrocyte [248]. Prx2 is 80 times more abundant [248], potentially explaining why Prx2 could compensate for the loss of Prx1, but not vice versa.
1.6.5 Binding of peroxiredoxin 2 to erythrocyte membranes
Another possible explanation for the inability of Prx1 to compensate for Prx2 is that Prx2 could have additional protective properties. Prx2 was originally referred to by a variety of other names, all of which stemmed from the various functions, conditions and forms it has been associated with: natural killer enhancing factor B (NKEF-B) [287, 380, 400, 418], thiol- specific antioxidant protein II (TSA-II) in yeast [292, 419], thioredoxin peroxidase B [287, 300, 420], and PRP for protector protein [297]. The most prominent names for Prx2 were torin, Band8 and calpromotin. Torin was the name Harris gave Prx2 when he discovered it bound to erythrocyte ghosts in 1968 [358]. Only a few years later, Palek et al. [421, 422] and Riggs et al. [423] noticed a protein appearing on the SDS-PAGE of sickle cell and ATP- depleted erythrocyte membranes, which they named Band8. Band8 was clearly a cytosolic protein, a homodimer, which was enriched on erythrocyte membranes with high intracellular calcium concentrations. Mizukami et al. [424] found that the cytosolic protein was bound loosely to membranes of fresh erythrocytes at neutral pH, and was washed away over time with a more basic buffer, indicating an ionic interaction. In the same paper it was also mentioned that the monomeric protein forms dimers via disulfide bonds, and polymers depending on quaternary structure, protein concentration, calcium concentration and the presence of NaCl [424].
Plishker et al. named this protein calpromotin, on the basis that it is needed for the calcium- dependent efflux of potassium [425, 426]. While calcium-dependent potassium efflux historically was initiated by energy depletion via alkylation [427], Plishker found that Gardos channel activity was inhibited upon alkylation of the protein, which points to the involvement of reactive thiol groups [399, 426]. The channel responsible for the potassium efflux is called the Gardos channel [428]. It is a calcium dependent potassium channel, also known as human small conductance potassium channel (hsK1). The “Gardos effect“ describes the loss of potassium ions in erythrocytes following ATP depletion, which depends on an increase of intracellular calcium and activation of the mentioned channel [428-430]. The activation of the Chapter 1: Introduction 32
Gardos channel by increased intracellular calcium leads to loss of potassium and subsequently to cell dehydration [430]. Cell dehydration makes cells more dense, which can be observed in diseases like sickle cell anemia, in which active sickling is associated with an increased concentration of dense cells [431, 432]. Additionally in dense sickle cells, Prx2 concentration on the membrane is increased [433, 434]. Calcium-dependent cell shrinkage and possible Gardos channel involvement has also been observed in eryptosis [435] and vesiculation [436].
It has been a striking question ever since as to why and how Prx2 binds to the erythrocyte membrane. Though Prx2 is the third most abundant protein in the cytosol [400], as mentioned by Harris et al. [358], membrane bound Prx2 at most makes up only 5% of the protein of the intact erythrocyte membrane. This corresponds to 0.05 % of the cell’s Prx2 content (approx. 3 µg/ml packed cells) [358, 399, 437]. One reason for the binding of Prx2 to the membrane might lie in its ability to reduce lipid hydroperoxides [286] and therefore to protect membranes from damage by membrane-bound haemichromes [301] (see Section 1.2.3). One proposition was that Prx2 binds directly to the phospholipid layer with its C-terminus [286], another that Prx2 binds to integral membrane proteins like Band3 [406, 438] and stomatin [439]. However, binding to Band3 was questioned since in one work, it bound instead to an unknown protein with the size of approximately 40 kDa [439]. Interestingly, cleaved versions of the cytosolic domain of Band3 are often observed on gels in the 41-43 kDa region [437, 440-442]. Calpain, a membrane-recruited, calcium-activated protease shown to digest Prx2 as well as Band3 or spectrin [355, 443, 444], might then be responsible for the removal of Prx2 by C-terminal truncation [286, 355]. Some researchers have found that oxidative stress increases Prx2 binding [445-447], while others have observed a decrease in Prx2 binding [406, 438, 448]. Depending on the used storage solution, Prx2 is also removed or enriched in the membranes during blood storage [85]. However, protein cleavage would neither explain the displacement of Prx2 from the membrane by haemichromes [406, 438, 448], nor why inhibition of calpain decreases Gardos channel activity as well as membrane-bound Prx2 [434]. The membrane binding structure of Prx2 has been reported to be oxidised dimers as well as oligomers [345, 351, 365], which is not exclusive since oligomerisation seems to be more dependent on Prx2 concentration than on redox state [282, 293, 299, 351, 358, 449]. However, it is generally assumed that the oligomeric structure is only stabilised when Prx2 is reduced [357] or hyperoxidised [293, 355]. The same antagonistic outcome was also observed when proteins were alkylated with iodoacetic acid; while alkylation reduced Prx2 binding [351], it did not inhibit it [425]. Chapter 1: Introduction 33
Since observations regarding Gardos channel activity, membrane binding and the redox status of Prx2 are counterintuitive, the role of Prx2 redox state in membrane binding warrants further research.
1.7 Aims
Ever since the identification of the Prx family as antioxidant proteins, many studies have been conducted to understand their behaviour structurally, to characterise their antioxidant activity, their involvement in redox signal transduction, and chaperone function. However, the role of Prx2 in erythrocytes has not been clearly elucidated, even though it is present in high amounts and shows unexpected properties not seen in other cells. While Prx2 redox state has been observed and linked to several functions in other cells, it has not been determined if the redox state of erythrocyte Prx2 is altered in disease or blood storage, and if it affects cell function. Additionally, Prx2 membrane binding in erythrocytes has been consistently observed, but the question of how it actually binds to the membrane and if membrane binding is affected by Prx2 redox state has not been answered. The overall aim of this thesis was to determine what affects Prx2 redox state in erythrocytes, and how these changes in redox state influence membrane binding.
The specific aims were to:
Assess the effects of disease on Prx2 redox state (Chapter 3):
- Examine the redox behaviour of Prx in erythrocytes exposed to temperatures and pH fluctuations commonly observed in disease
- Investigate the effect of neutrophil activation on Prx2 redox state in neighbouring erythrocytes
- Determine the main Prx2 oxidising species during neutrophil activation
- Explore the physiological relevance of the in vitro observations
Investigate Prx2 redox state during blood storage (Chapter 4):
- Evaluate Prx2 redox changes during blood storage, and investigate possible reasons for the observed changes
- Develop a method to prevent, delay or reduce the observed Prx2 oxidation during or after blood storage Chapter 1: Introduction 34
Examine how Prx2 membrane binding is influenced by the redox state of Prx2 (Chapter 5):
- Investigate the effect of intracellular calcium and H2O2 induced oxidation on membrane binding and redox state of Prx2 in erythrocytes
- Determine the effect of Prx2 redox state on membrane binding with purified Prx2 and isolated erythrocyte ghosts
Chapter 2: Materials and Methods 35
2 Methods and Materials
2.1 Erythrocyte peroxiredoxin 2 oxidation state during inflammation and sepsis
2.1.1 Erythrocyte preparation
Blood was obtained for all experiments from 30 healthy human volunteers (ages 18-65, no smokers) and one individual who is MPO-deficient (less than 5% MPO in neutrophils [450]) with informed consent. The study was approved by the New Zealand Southern A Regional Ethics Committee. Blood was collected into EDTA or heparinised tubes and used the day of collection. The blood was centrifuged at 3,800 g for five minutes with no brake. Plasma and buffy coats were removed and cells were washed three times with two volumes of cold phosphate-buffered saline (PBS, see Table 2.7). Packed cells were diluted 1:10 with PBS and cell counts were performed using a haemocytometer. Erythrocyte suspensions were kept on ice throughout the preparation.
2.1.2 N-Ethylmaleimide-blocking
To preserve the redox state of Prx2 [233], it is necessary to alkylate the cysteine thiols with N-ethylmaleimide (NEM) before lysis. After each treatment, cells were pelleted at 3,800 g for one minute, the supernatant was removed, NEM (100 mM) in PBS was added to a volume of 100 µl, and the cells were incubated for 20 minutes at 37 °C. The cells were either lysed in 400 µl of 1:1 solution of 100 mM NEM in PBS and 1x non-reducing sample buffer (SB, see Table 2.7) or in 400 µl of SB and frozen at -20 °C.
2.1.3 Effect of fever-like temperatures on erythrocyte peroxiredoxin 2 oxidation
Erythrocytes were diluted to 5 x 107/ml (approximately 0.1% cell suspension) in Hank’s balanced salt solution (HBSS, see Table 2.7), and either incubated at 37 °C, 39 °C and 41 °C, pH 7.4, or at 37 °C in HBSS where the pH was adjusted to 7.1 and 6.8 with hydrochloric acid (HCl). After one hour incubation, the cells were NEM-blocked and prepared for non-reducing SDS-PAGE. Chapter 2: Materials and Methods 36
2.1.4 Neutrophil preparation
Neutrophils were isolated using Ficoll/ Hypaque centrifugation, dextran sedimentation, and hypotonic lysis [451] as follows: between 30 to 50 ml of blood was collected into tubes with 100 µl sodium heparin. For each 10 ml of blood, a 50 ml tube was prepared with 12.5 ml Ficoll. Each 10 ml of blood was diluted with 27.5 ml PBS, layered carefully on top of the prepared Ficoll, and then centrifuged at 1,000 g for 20 minutes with no brakes. The supernatant and the interface layers were removed and discarded until approximately one cm above the erythrocyte fraction. This fraction was then diluted with 5% dextran to a volume of 50 ml, gently mixed, and left for 30-45 minutes at room temperature to allow sedimentation of the erythrocytes. The upper, neutrophil-enriched layer was then transferred into another tube, topped up to 50 ml with PBS, and centrifuged for five minutes at 1,000 g. The supernatant was removed and the cell pellet was resuspended in 10 ml PBS. Twenty ml H2O was added and gently mixed for two minutes to lyse the remaining erythrocytes. The lysis was stopped by returning the solution to isotonicity with 10 ml of 2.7% NaCl and 10 ml PBS. The neutrophils were centrifuged at 500 g for five minutes, washed one more time with PBS, resuspended in five ml HBSS and counted using a haemocytometer and trypan blue for exclusion of dead cells.
2.1.5 Co-incubation of isolated erythrocytes and neutrophils
2.1.5.1 Incubation of erythrocytes with activated neutrophils at different ratios
Erythrocytes and neutrophils were prepared as outlined in Sections 2.1.1 and 2.1.4. Erythrocytes (2 x 107/ml) in HBSS were added to an equal volume of neutrophil suspension, which was adjusted to give ratios between 1 neutrophil to 100 erythrocytes and 1 neutrophil to 2,000 erythrocytes. Neutrophils were stimulated with 60 ng/ml phorbol myristate acetate (PMA), which was prepared from a stock solution of 100 µg/ml PMA in dimethyl sulfoxide (DMSO) by dilution of 1:10 with PBS, and the samples were incubated for 10 minutes at 37 °C. After treatment, the samples were centrifuged for 10 seconds at 4,000 g, the supernatant removed, and the cell pellet treated with NEM as stated above. After additional centrifugation for 10 seconds and removal of the NEM buffer, erythrocytes were lysed with 100 µl of 100 mM NEM in H2O, added to 400 µl non-reducing SB, and frozen at -20 °C.
Chapter 2: Materials and Methods 37
2.1.5.2 Incubation of erythrocytes and activated neutrophils over time
Erythrocytes (2 x 107/ml) in 250 µl HBSS were added to an equal volume of neutrophil suspension, which was adjusted to a ratio of 1 neutrophil to 500 erythrocytes. Neutrophils were stimulated with 60 ng/ml PMA, and samples were incubated for up to 60 minutes at 37 °C. The treatment was suspended at different time points by placing the sample tubes in wet ice, followed by centrifugation at 4,000 g for 10 seconds and removal of the supernatant. The cell pellet was treated with NEM and prepared for SDS-PAGE as stated in Section 2.1.5.1.
2.1.5.3 Activation of neutrophils in full blood
Whole blood (12 µl) was treated with 2 µg/ml PMA (3 µl of 1:10 dilution of PMA stock in PBS) for 10 minutes at 37 °C, immediately blocked with NEM, and prepared for SDS-PAGE as outlined in Section 2.1.5.1.
2.1.5.4 Effect of inhibitors and scavengers on erythrocyte peroxiredoxin 2 oxidation by activated neutrophils
Erythrocytes (2 x 107/ml) in HBSS were added to an equal volume of neutrophil suspension that was adjusted to a ratio of 1 neutrophil to 500 erythrocytes. Neutrophils were stimulated at 37 °C with 60 ng/ml PMA, and samples were incubated with inhibitors or oxidant scavengers at 10 µM (diphenyleneiodonium chloride; DPI), 100 µg/ml (catalase, from bovine liver, activity >10,000 units/mg protein), 20 µg/ml (SOD), 10 mM (L-methionine) and 100 µM (4- aminobenzoic acid hydrazide; ABAH) for up to 60 minutes. Samples were then centrifuged for 10 seconds at 4,000 g, the supernatant removed, the cell pellet treated with NEM and prepared for SDS-PAGE as stated in Section 2.1.5.1.
2.1.5.5 Comparison of normal and MPO-deficient neutrophils
Erythrocytes (2 x 107/ml) in HBSS were added to an equal volume of neutrophil suspension, which was adjusted to a ratio of 1 neutrophil to 500 erythrocytes. Neutrophils were stimulated at 37 °C with 60 ng/ml PMA, and samples were incubated with 10 µM DPI and 100 µg/ml catalase for 10 minutes at 37 °C. Cells were centrifuged, treated with NEM and prepared for SDS-PAGE as in Section 2.1.5.1.
2.1.5.6 Incubation of erythrocytes with LPS-stimulated neutrophils
Erythrocytes (2 x 107/ml) in HBSS were added to an equal volume of neutrophil suspension that was adjusted to a ratio of 1 neutrophil to 500 erythrocytes. HBSS containing 10 µg/ml lipopolysaccharide (LPS) and 10% human serum was added, and incubated for up to 60 Chapter 2: Materials and Methods 38
minutes. At the given time points, the cells were centrifuged for 10 seconds at 4,000 g, the buffer removed, the cell pellet treated with NEM and prepared for SDS-PAGE as mentioned in Section 2.1.5.1.
2.1.5.7 Cytochrome c assay for superoxide production by LPS-stimulated neutrophils
Neutrophils (5 x 105/ml) were co-incubated at 37 °C with 107/ml erythrocytes in HBSS containing 10% human serum, catalase (20 µg/ml), cytochrome c (40 µM) and LPS (10 µg/ml) in the presence and absence of 20 µg/ml SOD. Samples were centrifuged for 15 seconds at intervals up to 45 minutes and the absorbance of the supernatant was measured at 550 nm.
2.1.5.8 Incubation of erythrocytes with Staphylococcus aureus-stimulated neutrophils
Preparation of erythrocytes and neutrophils was performed as stated in Sections 2.1.1 and 2.1.4. Erythrocytes (2 x 107/ml) in HBSS were added to an equal volume of neutrophil suspension that was adjusted to a ratio of 1 neutrophil to 500 erythrocytes. S. aureus ATCC 27217 were grown, washed and opsonised as previously described [451, 452]. Briefly, single colonies were relocated from agar plates into sterile trypticase soy broth (3 g/100 ml) and grown at 37 °C under agitation overnight. The bacteria were pelleted at 1,000 g and washed three times with PBS before resuspension in PBS. The concentration of the bacteria was determined by spectrophotometric measurement at 550 nm, and the measured optical density was related to 0.09, which corresponds to 108 cells/ml. Bacteria were opsonised with 10% human serum in HBSS for 20 minutes at 37 °C. Bacterial cell suspension containing 4x105 S. aureus and 10% human serum was then added to erythrocytes and neutrophils, and incubated for up to 60 minutes at 37 °C. This is deemed to be the appropriate ratio of bacteria to neutrophil to give maximal stimulation [230]. At the given time points, cells were centrifuged for 10 seconds at 4,000 g, the buffer removed, the cell pellet treated with NEM and prepared for SDS-PAGE as mentioned in Section 2.1.5.1.
2.1.6 Measurement of oxidised erythrocyte peroxiredoxin 2 in a mouse model for endotoxaemia, in collaboration with University of Sydney, Australia
C57BL/6J mice (7-11 weeks old) were sourced from Animal Resources Centre (Perth, Australia), while p47phox-/- mice [453] and NOX2-/- mice [454] (12-13 weeks old) were both on a C57BL/6J background and were sourced from Monash University (Melbourne, Australia). All protocols were approved by the Animal Ethics Committee of the University of Sydney. The main experiment was kindly conducted by Dr. Ghassan Maghzal in Australia. Chapter 2: Materials and Methods 39
Endotoxaemia was induced by intra-peritoneal injection of 7.5 mg LPS (Escherichia coli serotype O111:B4, dissolved in 200 µl saline) per kg body weight, as described previously [455]. Mice were then given a standard chow diet and water ad libitum for up to 24 hours. At selected time points, mice were anaesthetised by placing them in chambers that allow for controlled delivery of isoflurane and scavenging of waste gases. Once anaesthetised, mice were connected to the anaesthetic circuit via a nose cone T piece to allow continuous anaesthesia. Blood (250 µl) was collected by cardiac puncture under anaesthesia into
K2EDTA microtainer tubes containing 250 µl of NEM (100 mM). Tubes were mixed gently and left to incubate at room temperature for one hour. Five µl of blood/NEM solution was removed and added to 500 µl of non-reducing gel loading buffer. Samples were then shipped to Christchurch, NZ, stored at -20 ºC upon arrival until SDS- PAGE and immunoblotting analysis for Prx2 oxidation. Zero time samples were obtained from mice treated as above without LPS injection.
2.1.7 SDS-PAGE and immunoblotting
Erythrocyte samples were diluted 1:50 (5-12 µg protein) and separated by 12% SDS- polyacrylamide gel electrophoresis (SDS-PAGE, see Table 2.1) under non-reducing conditions using top tank buffer and lower tank buffer (see Table 2.7). Electrophoresis was completed after 45 minutes at 200 V. Separated proteins were transferred onto a polyvinylidene difluoride (PVDF) membrane by wet western blotting with western blot buffer (see Table 2.7) for 50 minutes at 90 V. The membrane was blocked for one hour in 5% milk in Tris-buffered saline with TWEEN® 20 (TBST, see Table 2.7), 15 mM sodium azide
(NaN3), and 2% H2O2 to inhibit the pseudoperoxidase activity of haemoglobin. After blocking, the membrane was immunostained with a polyclonal rabbit anti-human Prx2 antibody (diluted 1:10,000 in 2% milk in TBST) over night, washed three times with TBST, and a horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (diluted 1:10,000 in 2% milk in TBST) was applied for one hour as previously stated [233].
To test for hyperoxidation of Prx2, erythrocyte samples were either diluted 3:4 with reducing SB (5% β-mercaptoethanol), or protein concentration was estimated and 15-20 µg protein was loaded onto the gel before separation and blotted as above. The blots were blocked with 5% milk in TBST, 15 mM NaN3, and 2% H2O2, as above. For immunostaining, a polyclonal rabbit anti-human Prx-SO2/3 antibody in 5% milk (dilution 1:2,500) and a horseradish Chapter 2: Materials and Methods 40
peroxidase-conjugated goat anti-rabbit secondary antibody were used. The bands were visualised and quantified as mentioned above.
Table 2.1 Standard recipes for SDS-PAGE.
Chemicals 12% Resolving gel Stacking gel 40% acrylamide/bis solution 37.5:1 3 ml 0.5 ml
H2O 4.9 ml 3 ml Tris-HCl, 2 M, pH 8.8 2 ml - Tris-HCl, 0.5 M, pH 6.8 - 1.25 ml
Sodium dodecyl sulfate (20%) in H2O (SDS) 50 µl 25 µl Ammonium persulfate (10%) in H2O (APS) 50 µl 25 µl Tetramethylethylenediamine (TEMED) 15 µl 8 µl
2.1.8 Visualisation of bands and calculation of the percentages
Western blots were incubated with the enhanced chemiluminescence reagent (ECL™) for 30 seconds, and the bands were visualised with the Chemidoc XRS gel documentation system for 60 seconds. In my experience, the Prx2 antibody has a greater affinity to the monomer. That means that calculated percentages depend on the protein amount loaded, on the time the ECL™ is left on the blot for development, and on the exposure time for the picture. This dependence may lead to overestimation of the dimer band signal, because the monomer band signal is saturated due to overexposure or even burned through the membrane. However, it could also lead to an underestimation of the dimer band signal if it is not completely visible (see Figure 2.1). To minimise this common error and to enable confidence in the gained results, haemolysate samples were processed and diluted the same way to be comparable. Additionally, the same volume was loaded onto gels. Finally, the development time with ECL™ and the exposure time for the camera were also kept identical. ImageJ software (National Institutes of Health, USA) was used to quantify the intensities of entire lanes. The background was removed automatically by the program. To limit variation in quantification due to differences in protein amounts, the percentage of oxidised dimer in relation to reduced monomer and oxidised dimer in each lane were calculated as presented in Figure 2.2. This way, the possible trends were depicted, instead of quantification of absolute numbers. Chapter 2: Materials and Methods 41
Figure 2.1 Differences in calculated ratios due to uneven peroxiredoxin 2 antibody affinity. (A), (B) Western blots of Prx2 haemolysate. Erythrocytes were lysed with and without blocking of thiols with NEM, resulting in either completely reduced or completely oxidised Prx2. The two samples were mixed at the given percentages, separated by SDS-PAGE under non-reducing conditions and visualised as described in Section 2.1.8. (A) Example of underestimation of oxidised dimer. (B) Example of overestimation of oxidised dimer. (C) Comparison between measured oxidation and expected oxidation of Prx2. Diagram represents average of A and B, error bars are ± standard deviation.
Figure 2.2 Calculation of the percentage of oxidised erythrocyte peroxiredoxin 2. After alkylation of free thiols, separation of oxidised Prx2 dimer and reduced Prx2 monomer by SDS-PAGE and visualisation of the bands by immunoblotting, the band intensities are measured with ImageJ and the percentage is calculated using the equation shown. The resulting percentages are illustrated as bar diagrams. Chapter 2: Materials and Methods 42
2.2 Erythrocyte peroxiredoxin 2 redox state during blood storage
2.2.1 Preparation of erythrocytes for blood storage
Blood was provided for all experiments during this research by 30 healthy human volunteers (ages 18-65, no smokers) with informed consent. The study was approved by the New Zealand Southern A Regional Ethics Committee. Blood was drawn into sterile EDTA- containing tubes and left at room temperature for two hours. The tubes were mixed, centrifuged at 3,800 g for 10 minutes without braking, and plasma and buffy coat were removed entirely under sterile conditions. Packed cells (1 ml) were added to 400 µl sterile filtered SAGM buffer, to 1 ml EAS76v6, or to 400 µl sterile filtered AS-7 [188, 191, 456] (Table 2.7) and stored at 4 °C in 15 ml Falcon tubes to simulate blood banking conditions. The cells remained oxygenated throughout.
2.2.2 Measurement of erythrocyte peroxiredoxin 2 oxidation state during storage
Samples (14 µl) were taken weekly up to 42 days, and added to 100 µl of 100 mM NEM in PBS to block free thiols and preserve the Prx2 redox state [233]. Following incubation at room temperature for 25 minutes, the cells were lysed in 400 µl of 1x SB and frozen at -20 °C for SDS-PAGE (Section 2.1.7)
2.2.3 Reversal of peroxiredoxin 2 oxidation in stored erythrocytes
Samples of the stored erythrocytes (70 µl) were centrifuged at 3,800 g for 30 seconds, the pelleted cells resuspended in HBSS containing 10 µg/ml catalase, and incubated at 37 °C for up to two hours. At each time point, the samples were pelleted by centrifugation at 3,800 g for 10 seconds and 10 µl of packed cells were removed. All samples were blocked with NEM and processed for non-reducing SDS-PAGE as stated in Section 2.2.2.
2.2.4 Reversal of peroxiredoxin 2 oxidation in stored erythrocytes with Rejuvesol™, dithiothreitol, α-lipoic acid, dihydrolipoic acid and N-acetylcysteine
Rejuvesol™ was prepared according to the published protocol [121] (Table 2.7). Samples (100 µl) of the stored erythrocytes were taken every 7 days between 21 and 42 days. A ratio of one part Rejuvesol™ was added to eight parts stored erythrocytes in SAGM. Other samples (100 µl) of the stored erythrocytes were centrifuged for 30 seconds at 3,800 g, the supernatant discarded, and the cells were resuspended in PBS containing 1 mM dithiothreitol (DTT), 5 mM N-acetylcysteine (NAC, non-neutralised), 250 µM α-lipoic acid (ALA, 0.33% ethanol) or Chapter 2: Materials and Methods 43
250 µM dihydrolipoic acid (DHLA, 0.005% DMSO) and incubated at 37 °C under gentle agitation for up to two hours. After incubation, the samples were centrifuged at 3,800 g for 30 seconds, and 10 µl of packed cells were blocked with NEM and processed for SDS-PAGE.
2.2.5 Prevention of peroxiredoxin 2 oxidation during storage of erythrocytes
Erythrocytes were stored in SAGM buffer in the presence of 70 µM ALA (0.36 % ethanol), 1.4 mM NAC (non-neutralised), or a mixture of 2.5 mM NAC, 2.5 mM glycine, and 2.5 mM glutamine (neutralised with NaOH) (NGQ) [172]. Samples (14 µl) were taken weekly, treated with NEM, and separated by SDS-PAGE.
2.2.6 Peroxiredoxin 2 oxidation in stored erythrocytes after additional stress with hydrogen peroxide
Samples (100 µl) of the stored erythrocytes were taken weekly up to 42 days. Aliquots were centrifuged at 3,800 g for 30 seconds, the pelleted cells were resuspended in 1 ml of 100 µM
H2O2 in PBS, and incubated at 37 °C under gentle agitation for 10 minutes. The H2O2 -1 -1 concentration in the stock solution was determined by A240 measurements (ε = 43.6 M cm ) prior to use [457]. The erythrocytes were re-pelleted, resuspended in 500 µl HBSS with 10 µg/ml catalase, and incubated at 37 °C under gentle agitation for up to two hours. All samples were NEM-blocked and separated by SDS-PAGE.
In another experiment, samples were stored in SAGM for 42 days , then treated with or without carbon monoxide (CO), to inhibit Hb autoxidation. CO was bubbled through the stored erythrocytes-buffer mix and the mix was incubated on ice under CO-rich atmosphere for one hour to ensure tight binding to Hb. The erythrocytes were then centrifuged at 3,800 g for 30 seconds, and the pelleted cells resuspended in 10 ml PBS containing 100 µM H2O2. After 10 minutes of incubation at 37 ºC, the erythrocytes were re-pelleted, resuspended in 500 µl HBSS with 10 µg/ml catalase, and incubated at 37 °C under gentle agitation for up to two hours. All samples were NEM-blocked and separated by SDS-PAGE.
2.2.7 Separation of erythrocytes by density and measurement of peroxiredoxin 2 oxidation
When erythrocytes age, they become increasingly dense [74, 458]. This change in density is theorised to be caused by an alteration of the water to haemoglobin ratio. The alteration occurs due to intracellular water loss during storage [459]. Using this difference in density, erythrocytes can be separated by “age” using density gradients, for example with polyvinyl- Chapter 2: Materials and Methods 44
pyrrolidone-coated colloidal silica matrix (Percoll) [460-462]. To achieve this separation, stored erythrocytes were loaded onto a discontinuous gradient of Percoll, loosely based on the protocol described by Alderman [463]: first, 250 µl of stored erythrocytes were diluted 1:1 with PBS containing 10 µg/ml catalase to prevent oxidation of Prx2. A Percoll calculator (GE
Healthcare) was used to determine the necessary amounts of Percoll, 10x PBS and H2O to prepare isotonic solutions (IP) with specific density values of 1.086 to 1.116 g/ml. All solutions contained 10 µg/ml catalase to protect the free cysteines of Prx2 from oxidation during separation. A 15 ml falcon tube was prepared by adding 2.5 ml of IP1.116 to each tube and carefully overlayering with 2.5 ml volumes of IP1.110, IP1.100, IP1.090, and IP1.080. Erythrocytes were carefully layered on top, and then centrifuged at 500 g for 60 minutes. The supernatant, which does not enter the top Percoll layer, was carefully aspirated for collection of erythrocyte vesicles. The fractions with a density lower than 1.1 g/ml were carefully aspirated as well and pooled (F1), the same as the fractions with a higher density (F2) [464]. The two pooled fractions were diluted 1:2 with PBS and washed one time by centrifugation to pellet the cells, while the supernatant was centrifuged for 10 minutes at 12,000 g to pellet the vesicles. 10 µl of the packed cells were incubated in 100 µl NEM (100 mM) for 25 minutes at room temperature, lysed in 400 µl 1x SB and stored at -20 °C until determination of protein content and separation by SDS-PAGE. The erythrocyte vesicles were washed one more time, incubated in 100 µl NEM (100 mM) for 25 minutes at room temperature, lysed in 200 µl Dodge buffer and stored at -20 °C until determination of protein content and separation by SDS-PAGE.
2.2.8 Spectral measurement of methaemoglobin and haemichrome content
Separated erythrocyte fractions from Section 2.2.7 were lysed with water and analysed spectrophotometrically for their Hb, metHb, haemichrome and choleglobin content [465]. The necessary absorptions are A560, A577, A630 and A700. The measured absorbances were used to calculate the haemichrome content as follows:
- Subtract A700 – 0.005 (choleglobin) from all absorbances
- [haemichrome (µM)] = -133A577 – 144A630 + 233A560
- [oxyHb (µM)] = 119A577 – 39A630 – 89A560
- [metHb (µM)] = 28A577 + 307A630 – 233A560 - [oxyHb (%)] = ([oxyHb (µM)] / ([oxyHb (µM)] + [haemichrome (µM)] + [metHb (µM)])) * 100 Chapter 2: Materials and Methods 45
- [haemichrome (%)] = ([haemichrome (µM)] / ([oxyHb (µM)] + [haemichrome (µM)]+ [metHb (µM)])) * 100 - [metHb (%)] = ([metHb (µM)]/ ([oxyHb (µM)] + [haemichrome (µM)] + [metHb (µM)])) * 100
2.2.9 Visualisation of vesicle proteins by in-gel fluorescence staining
The vesicle proteins were diluted 1:1 with 1x SB and loaded onto a 12% acrylamide gel, which was prepared as described in Section 2.1.7, but with the addition of 50 µl 2,2,2 trichloroethanol before setting. This allows for detection of protein by UV-induced crosslinking of the compound to the tryptophan residues in proteins [466]. The gels were irradiated with UV for up to ten minutes and pictures of the fluorescence were taken using the UVITech system.
2.2.10 GSH-quantification of stored erythrocytes with and without NAC
2.2.10.1 Quantification of total low molecular weight thiols by DTNB-assay
Erythrocytes were stored in SAGM with or without 2.5 mM NGQ as described in Section 2.2.5. Once a week, 28 µl were removed, centrifuged for 30 seconds and the packed cells were used to determine the intracellular free low molecular weight thiols content, which in erythrocytes consist almost entirely of GSH plus NAC if added, according to Beutler [467]:
20 µl packed erythrocytes were lysed in 180 µl H2O and proteins were precipitated with 300 µl of precipitation solution (5.37 mM EDTA, 5.13 M NaCl, and 0.17 M orthophosphoric acid). After five minutes incubation at room temperature, the samples were centrifuged at 12,000 g for 10 minutes. Two hundred µl of the clear supernatant was mixed with 890 µl of
0.3 M disodium phosphate (Na2HPO4) and the absorption (A1) was measured spectrophotometrically at 412 nm against a blank of 0.3 M Na2HPO4. After addition of 10 µl dithionitrobenzoic acid (DTNB; 500 µM) and incubation in the dark for 10 minutes, the absorption was measured again (A2). The intracellular thiol content was then calculated using the following formula [467]:
Chapter 2: Materials and Methods 46
2.2.10.2 Quantification of GSH in stored erythrocytes by stable isotope tandem mass spectroscopy
GSH was analysed (as its NEM derivative) by liquid chromatography tandem mass spectrometry (LC-MS/MS) as adapted from Harwood et al. [468]. Twenty µl packed erythrocytes were treated with 50 µl of 100 mM NEM for 20 minutes, then after water lysis, proteins were precipitated as in Section 2.2.10.1 and supernatants were analysed. Briefly, samples were diluted with water containing 0.25% formic acid (v/v) and isotopically labelled internal GSH-NEM standard was added. Analytes were separated on a Hypercarb column operated at 60 °C using an Ultimate 3000 RS system with elution solvents as described [468]. The HPLC was coupled inline to an electrospray ionisation source of an Applied Biosystems 4000 QTrap mass spectrometer. Quantification of GSH-NEM was by selective reaction monitoring in positive ion mode. Settings for the target analytes were m/z 433→304 (parent ion→fragment ion) for GSH-NEM and m/z 436→307 for the isotopically labelled internal standard. The analysis was kindly performed by Dr. Nina Dickerhof.
2.3 Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state
2.3.1 Membrane binding in whole cells
Erythrocytes were washed and isolated as detailed in Section 2.1.1. To test for the calcium dependence of Prx2 membrane binding, one ml packed erythrocytes were diluted in 19 parts PBS containing either 0.1 mM ethylene glycol tetraacetic acid (EGTA) or calcium chloride
(CaCl2) concentrations between 5 µM and 1 mM. After the addition of 65 nM ionomycin in PBS (20% DMSO), the cells were incubated under gentle agitation for 25 minutes at room temperature. The erythrocytes were pelleted at 3,800 g for 5 minutes, the supernatant was removed, and the cells were washed in PBS containing 0.1 mM EGTA.
To measure H2O2 dependence, the packed erythrocytes were diluted in 9 volumes of PBS with
10 mM NaN3 and incubated at room temperature for 10 minutes to inhibit endogenous catalase. Ten ml PBS containing H2O2 concentrations between 25-200 µM were prepared and added into the erythrocyte suspension while vortexing. H2O2 concentration of the stock solution was determined prior to use as in Section 2.2.6. The samples were incubated at 37 °C for 10 minutes before they were centrifuged at 3,800 g for 5 minutes at 4 °C, the supernatants removed, and washed with PBS and 10 µg/ml catalase. Chapter 2: Materials and Methods 47
To assess how calcium and H2O2 together affect membrane binding, 1 ml packed erythrocytes were pretreated with 9 ml PBS containing 10 mM NaN3, 65 nM ionomycin (20% DMSO), and 200 µM CaCl2 for five minutes at room temperature before adding 10 ml PBS with 0-100
µM H2O2 while vortexing. The samples were incubated at room temperature under gentle agitation for 25 minutes before centrifugation for 5 minutes at 3,800 g at 4 °C. After removal of the supernatant, all cells were washed with 10 ml PBS including 10 µg/ml catalase and 0.1 mM EGTA.
To isolate the membranes, the packed cells were lysed with 35 ml ice-cold Dodge buffer [469] (5 mM phosphate buffer, pH 8; see Table 2.7), and 1 mM phenylmethanesulfonyl fluoride (PMSF), and centrifuged at 15,000 g for 20 minutes at 4 °C. For spectral analysis of haemoglobin in H2O2 treated samples, 100 µl lysate was diluted with 1 ml Dodge buffer and measured at 500-700 nm. The membranes were washed twice with Dodge buffer until they were free of Hb, and solubilised at a ratio of 1:1 with 4% SDS in 50 mM Tris-buffer, pH 7.6, and 0.5 mM EDTA.
2.3.2 Purification of erythrocyte peroxiredoxin 2
2.3.2.1 Purification of erythrocyte peroxiredoxin 2 via anion-exchange chromatography
Based on the methods of Lim [302] and Peskin [235] et al. Prx2 was purified from erythrocytes of healthy volunteers. After washing the heparinized blood 3 times with ice-cold PBS, pH 7.4, typically five ml packed erythrocytes were lysed with column buffer (Table 2.7). The lysate was centrifuged for 15 minutes at 15,000 g, at 4 °C. The supernatant was carefully aspirated and the membrane fraction was discarded. A 10 cm x 1 cm diethylaminoethyl (DEAE) sepharose column was equilibrated with 100 ml column buffer, and the supernatant (typically 50 ml) was loaded with the use of a Bio-Logic Duo-flow pump at a flow rate of two ml/min. The column was then washed with 100 ml lysis buffer to remove Hb, which does not bind to the column matrix under these conditions. The red colour representing Hb was observed to move through the column into the wash until the column matrix was white again. At the same time, the UV-trace settled onto a new base line, and no further peaks were observed at 280 nm. Prx2 did not bind as expected onto the column at pH 7.4, even though the IP of Prx2 is 5.66, which has been reported before [406]. The bound Prx2 was eluted with a linear 0-500 mM NaCl gradient in column buffer (elution buffer), and was put on hold every time the UV-detector showed absorbance higher than 0.100, to provide better separation (Figure 2.3). The fractions were collected with the Bio-logic Bio-frac Chapter 2: Materials and Methods 48
fraction collector, with three ml per fraction. Usually, Prx2 elution began at 160 mM NaCl in column buffer. At the end of the run, the column was washed with 2 M NaCl in column buffer to remove any other tightly bound proteins.
2.3.2.2 Determination of purified peroxiredoxin 2 via SDS-PAGE and immunoblotting
To locate Prx2 and to ensure exclusion of Prx1, 5 µl of each protein fraction was dot blotted onto nitrocellulose and immunostained as in Section 2.1.7 against Prx2 and Prx1 (Figure 2.4, A). The fractions containing Prx1 (F 8 + 9, Figure 2.4, B) were discarded. The fractions containing Prx2 were run on a 12% SDS-PAGE under reducing conditions and stained overnight with Coomassie Brilliant Blue stain G250. The gel was destained with water and visually assessed for Prx2 presence, concentration and purity (Figure 2.4, C). The molecular mass of human Prx2 with its N-terminal methionine still attached is 21.9 kDa (Swiss-Prot accession number P32119). When the fractions which contained Prx2, as proven by immunostain, were run on SDS-PAGE, a major band migrated slightly above the 20 kDa marker (Figure 2.4, C).
Figure 2.3 Purification of peroxiredoxin 2 from human erythrocytes by ion exchange chromatography. A representative elution trace derived from the ion exchange chromatography of erythrocyte haemolysate. 40 ml of lysate was loaded (this did not exceed the binding capacity of the column). Hb did not bind to the DEAE sepharose and eluted promptly during loading of the column (as represented by the large peak at the start of chromatography), while Prx2 was retained. A 0–500 mM NaCl gradient (light blue line) was used to elute Prx2 from the column. The detailed methodology is described in Section 2.3.2.1. Chapter 2: Materials and Methods 49
Figure 2.4 Electrophoretic and immunological detection of peroxiredoxin 2 in the purified preparation. (A), (B) Representative dot blots of fractions derived from DEAE-sepharose chromatography. (A) Immunostained for Prx2. (B) Stained for Prx1. (C) Representative gel after reducing SDS-PAGE of fractions containing most Prx2. Fractions 8 + 9 were excluded due to contamination with Prx1. Bands were stained with Coomassie blue. Values of the molecular mass marker (kDa) are indicated on the left.
Fractions containing more than 80% Prx2, as determined by visual assessment, were pooled (Fractions 7-18 in Figure 2.4). The pooled fractions were concentrated by centrifugation using Amicon Ultra-15 centrifugal filter concentrators with a molecular mass cut-off of 10 kDa. The concentrate was dialysed into 50 mM Tris-buffer, pH 7.6, containing 0.5 mM EDTA, to remove most of the NaCl. The protein content was estimated via DC™-assay as in Section 2.3.4. The yield of Prx2 from several purification preparations is displayed in Table 2.2. Due to the very low yield, all preparations were pooled and protein concentration was determined by DC™-assay as in Section 2.3.4 to be approximately 0.41 mg/ml, at a volume of 10 ml.
Table 2.2 Purification of human peroxiredoxin 2 from erythrocytes. All preparations were prepared independently and erythrocytes were from different donors.
Prep Prep Prep A B C Starting volume of packed erythrocytes (ml) 5 5 5 Volume of purified Prx2 (µl) 1000 1200 8000 Prx2 concentration (mg/ml) 2.9 1.09 0.18 Effective amount of Prx2 assuming 80% purity (mg) 2.32 1.04 1.26 Yield of Prx2 (mg/ml packed erythrocytes) 0.46 0.21 0.25 Yield of erythrocyte Prx2 assuming a cytoplasmic concentration of 8.3 % 3.8% 4.5% 5.6 mg/ml packed erythrocytes [399] Chapter 2: Materials and Methods 50
2.3.2.3 Determination of purified peroxiredoxin 2 via electrospray ionisation–mass spectrometry
Electrospray ionisation–mass spectrometry (ESI–MS) was performed with an LCQ DECA plus XP ion trap instrument under positive ionisation mode. Purified Prx2 was desalted by passage through a spin column equilibrated with water, then diluted 1:1 with a solution containing 50% acetonitrile and 0.1% (v/v) formic acid. The sample was then directly infused using a syringe at a flow rate of 5 μl/min. Nitrogen was used as the nebulising and drying gas. A full scan of product ions from a mass/charge (m/z) ratio of 100–2,000 was performed, data were collected for one minute and deconvolution was carried out using BioworksBrowser 3.1. The analysis was kindly performed by Dr. Lou Paton.
While the protein mass for reduced Prx2, as given by Swiss-prot, is 21.892 kDa, the experimental value usually obtained from ESI–MS of human erythrocyte Prx2 in our laboratory is 21.805 kDa. This is consistent with the observation that in human erythrocytes, Prx2 is lacking its N-terminal methionine, and is either alkylated [405] or carbamylated instead (E. Schröder, pers. comment, F. Low, unpublished results). The relative abundance of ions having an m/z ratio of 100 to 2,000 showed the typical Gaussian distribution of multiply charged peaks. Deconvolution of the m/z spectrum yielded a major peak with a mass of 43,606 Da. Since no reducing agent was present, this relates to oxidised Prx2-dimer, with a mass of 2 x 21.805 Da (Figure 2.5). It also confirms that no major higher molecular weight contaminant is present. Chapter 2: Materials and Methods 51
Figure 2.5 Electrospray ionisation–mass spectrometry analysis of purified peroxiredoxin 2. (A) Mass-to-charge spectrum (m/z 100–2,000) derived from ESI–MS analysis of purified Prx2 showing the distribution of the product ions (B) Deconvolution of A, showing the resulting molecular mass profile. (C) Zoom into the molecular mass profile between 43 kDa and 45 KDa. 43.606 kDa was identified as oxidised Prx2-dimer. Analysis was performed by Dr. Louise Paton.
2.3.3 Binding of purified peroxiredoxin 2 to isolated membranes
2.3.3.1 Preparation of erythrocyte membranes
To prepare the cell membranes (ghosts), erythrocytes were washed three times with cold PBS, with subsequent removal of the buffy coat. To lyse the cells, 1 ml of packed erythrocytes was added to 49 ml of ice-cold Dodge buffer and 0.5 mM EDTA. The membranes were pelleted by centrifugation at 15,000 g for 15 minutes, at 4 °C. To remove any traces of Hb or other non-covalently bound proteins, the ghosts were washed six times with Dodge buffer and 0.5 Chapter 2: Materials and Methods 52
mM EDTA, and the membrane pellet was passed through a 22-gauge needle between the washes to further assist lysis. The ghosts were washed three times with 0.9 % NaCl to remove bound Prx2, and washed one more time with Dodge buffer to remove the NaCl. Ghosts were resuspended in Dodge buffer II and the protein content was measured as in Section 2.3.4.
2.3.3.2 Preparation of inside-out erythrocyte membrane vesicles
To prepare inside-out membrane vesicles (IOMs) with a high yield and little contamination of rightside-out materials, and without loss of important structural membrane proteins such as spectrins, the method of Kuross et al. [77], a modified version of Jacobson et al. [470], was employed. The method uses positively charged coated beads which interact strongly with the negatively charged outside of erythrocyte membranes. Five ml methacrylate copolymer beads with quaternary amine functional groups were washed three times with five ml of 0.2 M NaCl and further four times with five ml of 5 mM Dodge buffer II, pH 7.4 (Table 2.7). The beads were resuspended 1:1 in Dodge buffer II. Erythrocyte membranes were prepared as described in Section 2.3.3.1. To cover the beads with membranes, the beads were gently mixed for five minutes with the ghosts in Dodge buffer II at 2 mg ghost protein per ml beads. Unbound material was removed by centrifugation at 100 g, the unbound protein content was measured, and the membrane-covered beads were further washed three times with Dodge buffer II, at a centrifugation speed of 100 g for five minutes, at 4 °C. One ml beads were resuspended with five ml Dodge buffer II, and sonicated for seven seconds using the Omni-ruptor 4000 at a power output of 20%. The beads were washed eight more times with 10 ml Dodge buffer II, at, diluted 1:1 after the final washing step, and stored at -80 °C until use.
2.3.3.3 Acetylcholine esterase assay to determine sideness of IOM
To determine the yield of IOMs, an assay for acetylcholine esterase accessibility was used [471]. This assay measures the catalytic activity of acetylcholine esterase, which is located on the outside of the erythrocyte membrane, and should be inaccessible on IOMs. Briefly, 5 µl of IOM slurry was added to a semi-micro cuvette containing either 95 µl Dodge buffer or 95 µl Dodge buffer with 0.2% Triton X-100 to disrupt the membranes. After one minute, 600 µl of sodium phosphate buffer (100 mM, pH 7.5) and 50 µl DTNB (10 mM, in 100 mM phosphate buffer, pH 7, with 3 mg NaHCO3 per 8 mg DTNB) were added and starting measurements A1 and B1 were taken at 412 nm against a blank of 100 mM phosphate buffer. To initiate the reaction, 50 µl of acetylthiocholine chloride (12.5 mM) was added to both cuvettes, mixed well, and incubated in the dark for 10 minutes before measuring A2 and B2. The percentage of IOMs is calculated as followed: Chapter 2: Materials and Methods 53
The average yield of two preparations of IOMs was 92%.
2.3.3.4 Visualisation of IOM by membrane-labelling
To visualise the erythrocyte membrane-covered beads, the membranes were labelled with Vybrant™ DiO cell-labelling solution. Vybrant™ DiO is lipophilic, has an absorbance maximum at 484 nm and a fluorescent emission maximum of 501 nm when detected with an Omega XF23 filter. Briefly, 2 µl control beads and beads treated to generate IOMs were suspended in 1 ml Dodge buffer, and 5 µl of labelling solution was added. After ten minutes in the dark at 37 °C, the beads were washed three times with Dodge buffer, transferred onto microscopy slides, and brightfield and fluorescence photographs were taken using the AxioCam MRm microscope.
2.3.3.5 Preparation of reduced, oxidised and hyperoxidised peroxiredoxin 2
For the IOM-experiments, purified, oxidised Prx2 was exchanged into Tris-buffer (50 mM Tris, pH 7.5, 20 mM NaCl) with microcentrifuge spin columns for size exclusion chromatography. The spin columns were filled with 1 ml of BioGel P-6DG resin. The resin filled tubes were washed five times with 1 ml MilliQ-water, and five times with 1 ml of Tris- buffer before 70 µl of Prx2 solution was applied to the spin columns which were centrifuged for four minutes. To reduce the Prx2, the eluate was treated with 10 mM DTT in Tris-buffer for 30 minutes, and DTT was either removed with microcentrifuge spin columns equilibrated with Tris-buffer containing 1 µg/ml catalase.
For the experiments with purified ghosts, purified, oxidised Prx2 was exchanged into Dodge buffer II with microcentrifuge spin columns as described above. To reduce the Prx2, the eluate was treated with 10 mM DTT for 30 minutes and used without removal of DTT. To hyperoxidise the Prx2, the purified protein solution was incubated with 10 mM DTT for 30 minutes, 1 mM H2O2 was added and incubated at room temperature for further 30 minutes. The protein was then exchanged into Dodge buffer II using the spin column (see Figure 2.6). After exchange into Dodge buffer II, the protein concentration was estimated by measuring the absorption at 280 nm with the Nanodrop spectrophotometer. Protein redox state was determined via non-reducing SDS-PAGE and immunoblotting. Chapter 2: Materials and Methods 54
Figure 2.6 Preparation of hyperoxidised, oxidised and reduced peroxiredoxin 2. See text for details.
2.3.3.6 Binding of purified peroxiredoxin 2 to inside-out membrane vesicles
Prx2-free erythrocyte ghosts were prepared as in Section 2.3.3.1, IOMs were prepared as in Section 2.3.3.2, and oxidised, hyperoxidised, and reduced Prx2 was prepared as in Section 2.3.3.5. Oxidised and reduced Prx2 (DTT removed) were added to 50 µl IOMs at concentrations between 0 and 15 µg/ml, and incubated at room temperature for 15 minutes. The IOMs were washed three times with Tris-buffer, and proteins and membranes were eluted from the beads by adding elution buffer (70% glycerol, 5% SDS, 0.1 mM EDTA, 100 µg/ml PMSF) and heating the samples for five minutes at 80 ºC. The eluted proteins were stored at - 20 °C for electrophoresis.
2.3.3.7 Binding of purified peroxiredoxin 2 to Hb free, peroxiredoxin 2 free ghosts
Erythrocyte membranes were prepared as in Section 2.3.3.1. Ghosts containing 100 µg of membrane protein were mixed with 0-4.6 µM oxidised, hyperoxidised and reduced Prx2 in 100 µl Dodge buffer II (a total of 0-10 µg per sample). The ghosts were incubated at room temperature for 15 minutes, then pelleted at 15.000 g for 10 minutes at 4 °C. The supernatant was removed, and the membranes were washed three more times with Dodge buffer II. For the reduced Prx2, the Dodge buffer II was supplemented with 200 µM DTT to guarantee a reducing environment. The membranes were stored at -20 °C until electrophoresis. Chapter 2: Materials and Methods 55
2.3.3.8 Additional experiments to achieve insight into mechanism of peroxiredoxin 2 binding
Binding of recombinant human Prx3 (kindly provided by Dr. R. Poynton, from cDNA pCMV/XL5 and expressed in E.coli) to isolated erythrocyte membranes was conducted under oxidised and reduced conditions as in Section 2.3.3.7. In another experiment, ghosts containing 100 µg membrane protein were preincubated with 15 µM C-terminal Prx2 peptide for 15 minutes before addition of 0-4.6 µM Prx2 to see if the C-terminal peptide was interfering with binding. Ghosts containing 100 µg membrane protein were incubated with 0- 4.6 µM Prx2 under reducing conditions as in Section 2.3.3.7, with addition of 1 mM calcium chloride to see whether calcium still increased binding of Prx2 in this limited system.
2.3.3.9 Binding of peroxiredoxin 2 to isolated erythrocyte ghosts in the presence of haemichromes
Stored, Hb-rich supernatant, which was left over after lysis of erythrocytes with Dodge buffer and centrifugation at 15.000 g for 15 minutes as in Section 2.3.3.1, was dialysed over night at 4 °C against Dodge buffer II to remove GSH and other low molecular weight constituents. One ml of lysate was mixed with 2 mM 1-acetyl 2-phenylhydrazine (APH) and incubated for two hours at 37 °C. The lysate was spun down at 15,000 g for ten minutes to remove any precipitated protein, and the supernatant was analysed spectrophotometrically for its Hb, metHb, haemichrome and choleglobin content as in Section 2.2.8. Ghosts containing 100 µg membrane protein were incubated with 0-4.6 µM Prx2 and 15 µM haemichromes for 15 minutes to test for differences in binding behaviour.
2.3.3.10 Crosslinking of purified peroxiredoxin 2 and isolated erythrocyte membrane with glutardialdehyde
Purified Prx2 at concentrations of 0-4.6 µM were incubated with ghosts containing 100 µg of membrane protein in Dodge buffer II and 5 µl glutardialdehyde (pentane-1,5-dial, 2.3 %), in a total volume of 100 µl, for 25 minutes at room temperature. The reaction was terminated by addition of 10 µl Tris-HCL (1 M, pH 8); the membranes were centrifuged and washed as in Section 2.3.3.7, stored at -20 °C and resolved by SDS-PAGE.
2.3.4 SDS-PAGE, immunoblotting and Coomassie-staining of membrane proteins
To prepare samples for electrophoresis, the protein content was estimated using the DC™ (detergent compatible) assay, which is based on the Lowry method [472, 473]. The estimation was performed as a microplate assay, with a standard curve of bovine serum albumin (BSA). The absorbance was measured at 750 nm using the Varioskan Flash plate reader (Thermo Chapter 2: Materials and Methods 56
Scientific Inc., Waltham, MA, USA). Each sample was diluted with 1x SB containing 0.7 M β-mercaptoethanol for reducing SDS-PAGE. Six µg proteins were loaded per well and separated by 12% SDS-PAGE under reducing conditions as in Section 2.1.7. For the experiments conducted in Sections 2.3.3.8 and 2.3.3.9, the membranes were diluted to the same volume that would represent a similar membrane protein concentration of 6 µg per well, without testing for a changed protein content as above.
After SDS-PAGE, the gel was cut just above the 50 kDa marker and the lower part (<75 kDa) was transferred by wet western blocking and immunostained as detailed in Section 2.1.7. The upper part (>75 kDa) of the gel was stained for one hour in Coomassie brilliant blue R 250, destained in destain buffer (Table 2.7) for one hour, and the destaining process was completed overnight with water. Bands were visualised with ECL™ and the UVITech system. The high molecular weight bands on the gel were photographed with the Chemidoc XRS gel documentation system. ImageJ software was used to quantify the band intensities, which were used to calculate the bands as fractions relative to each other while normalising for protein loading by either using the band intensities of spectrin and Band3, or an added running control of 10 ng purified Prx2 (see Appendix 8.2 for detailed calculations).
2.4 Statistics
Data are expressed as means ± standard error of the mean (SEM) of n independent experiments using cells from different individuals. Statistical analyses were performed using SigmaPlot 11 (Systat Software, San Jose, CA, USA). Unless otherwise stated, paired t-tests were used to compare differences between two groups. To compare differences among more than two groups, one way repeated measures analysis of variance (ANOVA) or two way ANOVA was used. Significance was determined with the Holm-Sidak multiple comparison method. Differences in data were taken as statistically significant when p <0.05.
Chapter 2: Materials and Methods 57
2.5 Materials
All materials are listed in order of their first appearance in the methods section.
2.5.1 Erythrocyte peroxiredoxin 2 oxidation state during inflammation and sepsis
Table 2.3 Materials used in Section 2.1.
Material Supplier Location Erythrocyte preparation Sodium heparin Pfitzer New York, NY, USA Sodium chloride (NaCl) Biolab Ltd Scoresby, Australia Potassium chloride (KCl) Biolab Ltd Scoresby, Australia Monopotassium dihydrogen BDH Laboratory Supplies Poole, England phosphate (KH2PO4) Disodium phosphate (Na2HPO4) BDH Laboratory Supplies Poole, England NEM blocking N-ethylmaleimide (NEM) Sigma-Aldrich St Louis, MO, USA Tris(hydroxymethyl)amino- Roche Diagnostics Mannheim, Germany methane (TRIS) Sodium dodecyl sulfate (SDS) Sigma-Aldrich St Louis, MO, USA Bromophenol blue BDH Laboratory Supplies Poole, England Glycerol BDH Laboratory Supplies Poole, England Erythrocyte peroxiredoxin 2 oxidation at fever-like temperatures or at acidosis like pH Glucose Scharlau Barcelona, Spain Magnesium chloride (MgCl) Merck New York, NY, USA
Calcium chloride (CaCl2) BDH Laboratory Supplies Poole, England Hydrochloric acid (HCL) Pauling Industries Auckland, NZ Neutrophil preparation Ficoll GE Healthcare Uppsala, Sweden Dextran Sigma-Aldrich St Louis, MO, USA Trypan blue Sigma-Aldrich St Louis, MO, USA Co-incubation of isolated erythrocytes and activated neutrophils Phorbol myristate acetate Sigma-Aldrich St Louis, MO, USA (PMA) Lipopolysaccharide (e.coli Sigma-Aldrich St Louis, MO, USA serotype 026:b6) (LPS) Catalase (bovine liver) Sigma-Aldrich St Louis, MO, USA Superoxide dismutase (SOD) Sigma-Aldrich St Louis, MO, USA Cytochrome c (equine heart) Sigma-Aldrich St Louis, MO, USA S.aureus ATCC 27217 New Zealand Communicable Porirua, New Zealand Disease Centre Diphenyleneiodonium chloride Sigma-Aldrich St Louis, MO, USA (DPI) Superoxide dismutase (SOD) Sigma-Aldrich St Louis, MO, USA L-methionine Sigma-Aldrich St Louis, MO, USA 4-aminobenzoic acid hydrazide Fluka Chemicals Buchs, Germany (ABAH) Dimethyl sulfoxide (DMSO) Sigma-Aldrich St Louis, MO. USA Trypticase soy broth BDH Laboratory Supplies Poole, UK Measurement of oxidised erythrocyte peroxiredoxin 2 in a mouse model for endotoxaemia
K2EDTA microtainer tube BD USA C57BL/6J mice Animal Resources Centre Perth, Australia p47phox-/- mice Monash University Melbourne, Australia NOX2-/- mice Monash University Melbourne, Australia Chapter 2: Materials and Methods 58
Lipopolysaccharide (e.coli Sigma-Aldrich St Louis, MO, USA serotype O111:B4) (LPS) SDS-PAGE and Immunoblotting BenchMark® prestained protein Molecular Probes/ Invitrogen Auckland, New Zealand ladder 40% acrylamide/bis solution Bio-Rad Laboratories Hercules, CA, USA 37.5:1 Ammonium persulfate (APS) Sigma-Aldrich St Louis, MO, USA Tetramethylethylenediamine Sigma-Aldrich St Louis, MO, USA (TEMED) Enhanced chemiluminescence Amersham Biosciences Buckinghamshire, England (ECL™) western blotting system
TWEEN® 20 20 Sigma-Aldrich St Louis, MO, USA Polyclonal rabbit anti-human Sigma-Aldrich St Louis, MO, USA Prx2 antibody Polyclonal rabbit anti-human Abcam plc Cambridge, UK Prx-SO2/3 Horseradish peroxidase- Sigma-Aldrich St Louis, MO, USA conjugated goat anti-rabbit antibody (GARP) Polyclonal rabbit anti-human Sigma-Aldrich St Louis, MO, USA Prx1 antibody Methanol BDH Laboratory Supplies Poole, UK Glycine Sigma-Aldrich St Louis, MO, USA
Sodium azide (NaN3) Fisons Scientific Apparatus Loughborough, UK
Hydrogen peroxide (H2O2) Labserv, Pronalys, Thermo Waltham, MA, USA Fischer β-mercaptoethanol Sigma-Aldrich St Louis, MO, USA
2.5.2 Erythrocyte peroxiredoxin 2 redox state during simulated blood storage
Table 2.4 Materials used in Section 2.2.
Material Supplier Location Preparation of erythrocytes for simulated blood storage Adenine Sigma-Aldrich St Louis, MO, USA Mannitol BDH Laboratory Supplies Poole, England Sodium bicarbonate Sigma-Aldrich St Louis, MO, USA Reversal of peroxiredoxin 2 oxidation in stored erythrocytes with Rejuvesol™, dithiothreitol, α- lipoic acid, dihydrolipoic acid and N-acetylcysteine Pyruvic acid Sigma-Aldrich St Louis, MO, USA Inosine Sigma-Aldrich St Louis, MO, USA Monosodium phosphate BDH Laboratory Supplies Poole, UK (NaH2PO4) Dithiothreitol (DTT) Sigma-Aldrich St Louis, MO, USA α-lipoic acid Sigma-Aldrich St Louis, MO, USA N-acetyl cysteine Sigma-Aldrich St Louis, MO, USA Glutamate Sigma-Aldrich St Louis, MO, USA Glycine Sigma-Aldrich St Louis, MO, USA Dihydrolipoic acid Sigma-Aldrich St Louis, MO, USA Peroxiredoxin 2 oxidation in stored erythrocytes after additional stress with hydrogen peroxide Carbon monoxide BOC Gases NZ limited Christchurch, NZ Separation of erythrocytes by density and measurement of peroxiredoxin 2 oxidation Chapter 2: Materials and Methods 59
Polyvinyl-pyrrolidone-coated GE healthcare Uppsala, Sweden colloidal silica matrix (Percoll) Visualisation of vesicle proteins by in-gel fluorescence staining 2,2,2 trichloroethanol Sigma-Aldrich St Louis, MO, USA GSH-assay of stored erythrocytes with and without NAC Ethylenediaminetetraacetic acid Sigma-Aldrich St Louis, MO, USA (EDTA) Orthophosphoric acid BDH Laboratory Supplies Poole, England Dithionitrobenzoic acid (DTNB) Sigma-Aldrich St Louis, MO, USA Formic acid Merck New York, NY, USA
2.5.3 Changes in membrane-bound peroxiredoxin 2
Table 2.5 Materials used in Section 2.3.
Material Supplier Location
Calcium and H2O2 dependence of peroxiredoxin 2 membrane binding Ethylene glycol tetraacetic acid Sigma-Aldrich St Louis, MO, USA (EGTA) Phenylmethanesulfonyl fluoride Sigma-Aldrich St Louis, MO, USA (PMSF) Ionomycin Sigma-Aldrich St Louis, MO, USA SDS-PAGE, immunoblotting and coomassie staining of membrane proteins DC™ protein assay Bio-Rad Laboratories Hercules, CA, USA Bovine serum albumin (BSA) Gibco/ Thermo-Fisher Waltham, MA, USA Coomassie brilliant blue R250 Sigma-Aldrich St Louis, MO, USA Acetic acid Ajax Finechem/ Thermo-Fisher Waltham, MA, USA Ethanol Merck New York, NY, USA Purification of erythrocyte peroxiredoxin 2 Diethylaminoethyl (DEAE) GE healthcare Uppsala, Sweden sepharose fast flow column Coomassie brilliant blue G250 Sigma-Aldrich St Louis, MO, USA Ammonium sulfate Merck New York, NY, USA Acetonitrile Thermo-Fisher Waltham, MA, USA Preparation and characterisation of inside-out erythrocyte membranes Macro-Prep® High Q Support Bio-Rad Laboratories Hercules, CA, USA Acetylthiocholine chloride Sigma-Aldrich St Louis, MO, USA Sodium hydrogencarbonate BDH Laboratory Supplies Poole, England Triton X-100 Bio-Rad Laboratories Hercules, CA, USA Vybrant™ DiO Molecular probes Inc. Eugene, OR, USA Binding of purified peroxiredoxin 2 to isolated erythrocyte ghosts Micro-Bio-Spin Chromatography Bio-Rad Laboratories Hercules, CA, USA columns Glutardialdehyde BDH Laboratory Supplies Poole, England 1-Acetyl 2-phenylhydrazine Sigma-Aldrich St Louis, MO, USA C-terminal peptide of Prx2 (AA Sigma-Aldrich St Louis, MO, USA 184-198) BioGel P-6DG Bio-Rad Laboratories Hercules, CA, USA cDNA Prx3 pCMV-XL5 Origene Rockville, MD, USA
Chapter 2: Materials and Methods 60
2.5.4 Instruments and equipment
Table 2.6 Instruments and equipment used in Sections 2.1, 2.2 and 2.3.
Instruments/Equipment Supplier Location Improved haemocytometer Brand GmbH Wertheim, Germany EDTA tubes Belliver Industrial Estate Plymouth, UK AccuBlock digital drybath Labnet Edison, NY, USA Mini-PROTEAN® tetra cell Bio-Rad Laboratories Hercules, CA, USA system Mini-Trans-Blot Cell Bio-Rad Laboratories Hercules, CA, USA Power pack basic 300 Bio-Rad Laboratories Hercules, CA, USA Hybond- polyvinylidene Amersham Biosciences Buckinghamshire, England difluoride (PVDF) membrane Chemidoc XRS gel Bio-Rad Laboratories Hercules, CA, USA documentation system Thermomixer compact Eppendorff Hamburg, Germany Nano photometer Implen Westlake Village, CA, USA 4000Q-trap mass spectrometer Applied Biosystems Concord, Canada Hypercarb column (150 x 2.1 Thermo scientific San Jose, CA, US mM) Ultimate 3000 RS system Dionex Sunnyvale, Ca, US 8453 UV-Vis spectrophotometer Agilent Technologies Loveland, Colorado, USA Varioskan Flash Thermo Scientific Inc./ Thermo- Waltham, MA, USA Fisher UVItech system UVItech systems Cambridge, UK Bio-Logic Duo-flow pump Bio-Rad Laboratories Hercules, CA, USA Bio-logic Bio-frac fraction Bio-Rad Laboratories Hercules, CA, USA collector Amicon Ultra-15 centrifugal filter Millipore Bedford, MA, USA concentrators LCQ DECA XPplus ion trap ThermoFinnigan San Jose, CA, USA instrument BioworksBrowser 3.1 ThermoFinnigan San Jose, CA, USA Omni-Ruptor 4000 sonicator Omni International Kennesaw, GA, USA AxioCam MRm Carl Zeiss Microscopy New York, USA Nanodrop 3.1.2 Thermo Scientific Inc./ Thermo- Waltham, MA, USA Fisher Micro-cuvette Implen Westlake Village, CA, USA
Chapter 2: Materials and Methods 61
2.5.5 Common recipes for buffers
Table 2.7 Common recipes for buffers used in Sections 2.1, 2.2 and 2.3.
Common recipes for buffers Phosphate buffered saline, PBS Prepared from a 10x stock solution:
26.8 mM KCl, 1.36 M NaCl, 80.3 mM Na2HPO4,
14.7 mM KH2PO4, 2 l H2O, pH 7.4 4x Sample buffer for SDS-PAGE 26.5 mM Tris (pH 6.8), 373 µM Bromophenol
blue, 2.7 M Glycerol, 4% SDS, in 16 ml H2O Hank’s balanced salt solution, HBSS Prepared from 1x PBS and the following 100x stock solutions: 0.5 M Glucose, 0.05 M MgCl, 0.1
M CaCl2, each in 10 ml H20 Top Tank buffer (between the gels) Prepared from a 10x stock solution, 250 mM Tris,
190 mM Glycine, 35 mM SDS in 2 l H2O
Lower tank buffer 82.6 mM Tris in 2 L H20
Western blot running buffer: Prepared from a 10x stock solution: 250 mM Tris, 190 mM Glycine, and addition of 10% Methanol to the 1x buffer Tris buffered saline with TWEEN® 20 20, TBST Prepared from 10x stock solution of TBS: 200 mM Tris, 1.36 M NaCl, pH 7.6, addition of 2 ml/L
TWEEN® 20 20 (25%) to the 1x buffer SAGM buffer (Suru International Pvt. Ltd.) 45 mM Glucose, 150 mM NaCl, 1.25 mM
Adenine, 29 mM Mannitol in H2O EAS76v6 [188] 30 mM NaCl, 30 mM Sodium bicarbonate, 9 mM
Na2HPO4, 2 mM Adenine, 50 mM Glucose, and 30 mM Mannitol (pH 8.4) AS-7 [191] 80 mM Glucose, 55 mM Mannitol, 26 mM Sodium
bicarbonate, 12 mM Na2HPO4, 2 mM Adenine (pH 8.5) Rejuvesol™ [121] 100 mM Pyruvic acid, 99.9 mM Inosine, 5 mM
Adenine, 70.4 mM Na2HPO4, and 29 mM
NaH2PO4 in H2O Dodge buffer Prepared from a 100x stock solution (500mM):
493 mM NaH2PO4, 7 mM Na2HPO4, in H2O, pH 8 Dodge buffer II Prepared from a 100x stock solution (500mM):
493 mM NaH2PO4, 7 mM Na2HPO4, in H2O, pH 7.4 Coomassie destaining solution 450 ml Ethanol (industrial grade), 100 ml Acetic
acid (glacial), 450 ml H2O Chapter 2: Materials and Methods 62
Coomassie brilliant blue R250 staining solution 3 mM brilliant blue R250, 228 ml Ethanol (industrial grade), 45 ml Acetic acid (glacial), 228
ml H2O Coomassie brilliant blue G250 staining solution 100 g Ammonium sulfate, 11.8 ml Ortho-
phosphoric acid (85%), solubilised in 800 ml H2O and 1 g brilliant blue G250, solubilised in 50 ml
H2O Solutions were slowly mixed together, filled up to
1 liter with H2O. Before use, 30% ethanol was added to 70%
staining solution, destained with H2O Column buffer 50 mM Tris, 1 mM DTT, 0.5 mM EDTA, pH 7.6 Elution buffer 50 mM Tris, 1 mM DTT, 0.5 mM EDTA, 2 M NaCl, pH 7.6
Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 63
3 Erythrocyte peroxiredoxin2 oxidation state during inflammation
Results of this chapter have been published in:
Bayer, S.B., et al., Neutrophil-mediated oxidation of erythrocyte peroxiredoxin 2 as a potential marker of oxidative stress in inflammation. FASEB J, 2013. 27(8): p. 3315-22.
3.1 Introduction
Although normally there is little oxidised Prx2 in freshly isolated erythrocytes, a number of disease-related processes associated with oxidative stress, both internal and external to the erythrocyte, could give rise to its accumulation. One possibility is that high temperature or acidosis could increase autoxidation of haemoglobin (see Section 1.2.2) or oxidant generation from other metabolic processes. Also, the activation of neutrophils and/or monocytes during infection or inflammation could result in oxidative stress (see Section 1.3.2).
3.2 Experimental approach
The purpose of this work was to assess the effects of disease on Prx2 redox state, by (1) examining how temperatures and pH change the redox state of Prx2 in erythrocytes, (2) measuring Prx2 oxidation in erythrocytes after activation of neutrophils, and (3) to determine which of the neutrophil-derived reactive oxygen species (ROS) is causing the observed oxidative changes. Fresh, isolated human erythrocytes were incubated at different pH levels and temperatures in the presence of catalase to remove residual H2O2 from the buffer (Section 2.1.3), or in the presence of freshly isolated, activated neutrophils (see Section 2.1.4 for details on neutrophil isolation). Prx2 oxidation state was determined by SDS-PAGE under non-reducing conditions and Western blotting for Prx2. Oxidised Prx2 forms dimers, which have a higher molecular weight than the reduced Prx2 monomers, and can be easily distinguished on a Western blot. To observe the Prx2 redox-status in its native form, it is important to alkylate with NEM, since residual H2O2 in buffers will otherwise oxidise the reduced Prx2 during lysis (Figure 3.1) (see Section 2.1.2 on NEM-alkylation). The neutrophils were activated with phorbol myristate acetate (PMA), an activator of the protein kinase C (PKC, see Figure 1.6) (Section 2.1.5.1). Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 64
Figure 3.1 The necessity of alkylation of free thiol groups for evaluation of peroxiredoxin 2 redox status. Reduced Prx2 is highly susceptible to oxidation. Any
contact with residual H2O2 will lead to oxidation and dimerisation of Prx2. Alkylation of the free thiol groups with NEM before lysis of the cells traps the reduced Prx2 in its native, monomeric state, which can be easily distinguished from the dimeric, oxidised form by non-reducing SDS-PAGE and Western blotting.
The neutrophils were also stimulated with soluble lipopolysaccharide and the phagocytic stimulus Staphylococcus aureus. Additionally, experiments were undertaken in blood without isolating individual cells, thereby mimicking physiological conditions (Sections 2.1.5.3, 2.1.5.6, and 2.1.5.8).
To characterise the nature of the neutrophil-derived ROS responsible for Prx2 oxidation, neutrophils were incubated with the NADPH-oxidase (NOX) inhibitor diphenyleneiodonium chloride (DPI), the myeloperoxidase (MPO) inhibitor 4-aminobenzoic acid hydrazide (ABAH), the HOCl scavenger methionine, as well as the antioxidants catalase and SOD (see Sections 1.3.2, 1.4.2.1, and 1.4.2.2 on background, and Section 2.1.5.4 for detailed methods). An additional experiment was conducted with neutrophils from an MPO-deficient individual.
Finally, a mouse model of LPS-induced endotoxaemia was examined to determine if Prx2 oxidation occurred in vivo.
Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 65
3.3 Results
3.3.1 Effect of fever-like temperatures on erythrocyte peroxiredoxin 2 oxidation
Fever is a sign of the inflammatory response. Heat stress promotes Hb autoxidation, since increasing temperatures decrease the affinity of Hb for oxygen [474]. This leads to an increased flux of superoxide and H2O2. It is therefore plausible that erythrocyte Prx2 may become oxidised with increased temperatures. For this reason erythrocytes were incubated for 60 minutes in HBSS, to maintain physiological metabolism, at 37 °C, 39 °C and 41 °C. After 60 minutes at 37 °C, only 2% of the intracellular Prx2 existed as a dimer in control cells, and 98% was reduced (n = 5). There was no significant effect of temperature on Prx2 oxidation in the temperature range between 37 °C and 41 °C. The 32 kDA band between the monomer and the dimer represents a globin dimer, which shows pseudoperoxidase activity and reacts with the substrate used for visualisation [302].
Figure 3.2 Effect of temperature on erythrocyte peroxiredoxin 2 oxidation. Erythrocytes (5 x 107/ml) were incubated in HBSS at the indicated temperatures for 60 minutes (see Section 2.1.3), followed by NEM treatment as stated in Section 2.1.2. Results are the mean and SE of five independent experiments, with a representative Western blot shown. Note that in this and subsequent figures, molecular weight markers line up the monomer and dimer bands at the expected 22 kDa and 44 kDa positions. The percentage of oxidised Prx2 was calculated as described in Section 2.1.8. There was no statistical significance in comparison with cells incubated at 37 °C (repeated measures ANOVA, with Holm-Sidak multiple comparison method).
3.3.2 Effect of acidity on erythrocyte peroxiredoxin 2 oxidation
Another pathophysiological condition that could potentially increase oxidant generation and cause erythrocyte Prx2 oxidation is low pH. Protonation of Hb decreases oxygen affinity, the so-called Bohr effect [36, 474], and thus increases Hb autoxidation and the flux of ROS [173] (see Sections 1.2.1 and 1.2.2 on Bohr effect and Hb autoxidation). Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 66
Lowering pH to values achievable during severe metabolic acidosis [475-478] did, indeed, increase the extent of Prx2 oxidation (Figure 3.3). After 60 minutes at 37°C the highest amount of disulfide-linked dimer was seen in erythrocytes incubated at pH 6.8, but there was also significantly more oxidation at pH 7.1 than pH 7.4 (n = 5). The Western blot in Figure 3.3 shows the highest observed Prx2 oxidation, while the average of the five independent experiments lay around 15%.
Figure 3.3 Effect of pH on erythrocyte peroxiredoxin 2 oxidation. Erythrocytes (5 x 107/ml) were treated in HBSS at the indicated pH for 60 minutes, followed by NEM treatment as in Section 2.1.2. Results are the mean and SE of five independent experiments, with a representative Western blot shown. *p≤0.05 in comparison with cells incubated at pH 7.4 (repeated measures ANOVA, with Holm-Sidak multiple comparison method).
3.3.3 Effect of PMA-stimulated neutrophils on erythrocyte peroxiredoxin 2 oxidation
Neutrophils are an important source of ROS in pathophysiological conditions [208] and are potentially able to oxidise erythrocyte lipids and proteins [4, 198]. To test whether neutrophil ROS oxidise erythrocyte Prx2, neutrophils and erythrocytes were isolated from the same healthy donor and incubated for ten minutes at 37 °C at different ratios, with PMA added to stimulate the neutrophil oxidative burst. As observed previously [233], Prx2 in erythrocytes without neutrophils was predominantly in its reduced form, with 6 ± 2% (n = 4) detected as the disulfide-linked dimer by non-reducing SDS-PAGE and western blotting (Figure 3.4, left lane). Stimulated neutrophils caused a significant increase in Prx2 oxidation (Figure 3.4). Even at the low ratio of 1 neutrophil to 2,000 erythrocytes, 23 ± 4% of the Prx2 was dimeric. There was no significant oxidation of Prx2 when incubated with non-stimulated neutrophils (5.2 ± 0.3%, n=3, data not shown). Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 67
Figure 3.4 Effect of PMA-stimulated neutrophils on erythrocyte peroxiredoxin 2 oxidation. Erythrocytes (2 x 107/ml) were incubated with neutrophils at different ratios, with 60 ng/ml PMA, in HBSS for ten minutes before the redox state of Prx2 was trapped with NEM as described in Sections 2.1.5.1 and 2.1.2. Under these conditions in our laboratory, neutrophils routinely generate 6-8 nmol/min/106 cells of superoxide when stimulated with PMA. Results are the mean and SE of four independent experiments, with a representative Western blot shown. The left hand lane is from the same gel and the adjacent lane was removed. *p≤0.05 in comparison with untreated erythrocytes (repeated measures ANOVA, with Holm-Sidak multiple comparison method).
To establish that the Prx2 being measured was of erythrocyte rather than neutrophil origin, an extract from 104 neutrophils (20 times more than in the sample with the highest number of neutrophils in Figure 3.4) was subjected to Western blotting with anti-Prx2 antibodies (Figure 3.5). No signal was detected with this number of neutrophils, indicating that it was solely the erythrocyte Prx2 that was being visualised. Assuming that erythrocytes contain roughly 240 µM Prx2 [301, 399], and that PMA-stimulated neutrophils generate approximately 3 6 nmol/min H2O2 per 10 cells [479], we made a broad estimate of the efficiency of Prx2 oxidation. With 107 erythrocytes (0.2 nmol Prx2) at a ratio of 500:1 in Figure 3.4, the neutrophils should generate ~0.6 nmol H2O2, implying that a substantial fraction of this reacted with the erythrocyte Prx2.
When monitored over time, at a ratio of 500 erythrocytes to 1 neutrophil, Prx2 oxidation occurred rapidly. Dimerisation was maximal at 5-15 minutes then gradually declined over the next 40 minutes (Figure 3.6). The rate of superoxide production by PMA-stimulated neutrophils decreases over time, dependent on factors like availability of NADPH [479]. Therefore, the decrease in levels of oxidised Prx2 may be due to the rate of Prx2 dimer reduction by the thioredoxin pathway gradually exceeding the rate of Prx2 oxidation. Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 68
Figure 3.5 Erythrocyte peroxiredoxin 2 content compared to neutrophil peroxiredoxin 2 content. Western blot following non-reducing SDS-PAGE. Cells were lysed in 1x sample buffer after NEM treatment. Western blot is representative of two independent experiments.
Figure 3.6 Oxidation of erythrocyte peroxiredoxin 2 by PMA-stimulated neutrophils over time. Time-course for Prx2 oxidation was conducted under similar conditions as in Figure 3.4 with an erythrocyte to neutrophil ratio of 500:1. Controls are untreated erythrocytes, incubated for 10 minutes. Results are the mean and SE of four independent experiments. *p≤0.05 in comparison with untreated control erythrocytes (repeated measures ANOVA, with Holm-Sidak multiple comparison method).
To test for the observed oxidation of erythrocyte Prx2 in whole blood as opposed to isolated cells, freshly drawn blood was incubated with PMA. After 10 minutes of incubation, a small but significant increase in the proportion of dimer was visible (Figure 3.7). Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 69
Figure 3.7 Effect of PMA on erythrocyte peroxiredoxin 2 oxidation in blood. Blood (15 µl) was treated with 2 µg/ml PMA for ten minutes and NEM-blocked as described in Sections 2.1.5.3 and 2.1.2. Results are the mean and SE of six independent experiments, with a representative Western blot shown. *p=0.03 in comparison with the untreated control (paired t-test).
3.3.3.1 Effects of inhibitors and scavengers on erythrocyte peroxiredoxin 2 oxidation
To determine the main source of the ROS which oxidises Prx2 in erythrocytes, I tested the flavoprotein inhibitor DPI (Figure 3.8). DPI is converted into a radical by addition of an electron from flavoproteins like NADPH oxidase, forming an adduct with the protein, and inhibiting the enzyme in the process [480]. Oxidation of Prx2 in erythrocytes exposed to PMA-stimulated neutrophils was completely inhibited by DPI. While DPI is known to inhibit flavoproteins in general, its major effect on neutrophils is to prevent the production of superoxide (and hence H2O2) by the NADPH oxidase. This observation implies a requirement for oxidant generation by NADPH oxidase.
Figure 3.8 Structure of diphenylene iodonium (DPI).
To test which oxidant(s) were responsible for Prx2 oxidation, the following were added to erythrocytes and PMA-stimulated neutrophils in HBSS : SOD to dismutate superoxide, 4- aminobenzoic acid hydrazide (ABAH) to inhibit MPO [481], L-methionine to scavenge HOCl
[450] and catalase to decompose H2O2. Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 70
Catalase gave significant (50%) inhibition of dimer formation (27 ± 5% vs. 55 ± 3% with PMA alone, n = 9), but the other inhibitors or scavengers had no significant effect (Figure 3.9). These results indicate that while HOCl or other MPO-products make little contribution to Prx2 oxidation, the main cause of Prx2 oxidation is H2O2.
Figure 3.9 Effect of inhibitors and scavengers on peroxiredoxin 2 oxidation. Erythrocytes (2 x 107/ml) were incubated with isolated neutrophils at a ratio of 500:1 and 60 ng/ml PMA in HBSS for ten minutes as in Figure 3.4. Where indicated, 100 µM ABAH, 20 μg/ml SOD, 10 mM methionine, 10 µM DPI or 100 μg/ml catalase was added (Section 2.1.5.4). Results are the mean and SE of four (DPI), five (SOD, ABAH, methionine) and nine (catalase) independent experiments. *p≤0.05 in comparison with untreated control erythrocytes (repeated measures ANOVA, with Holm-Sidak multiple comparison method).
3.3.3.2 Comparison of blood from a healthy patient and an MPO-deficient patient
To further substantiate that neutrophil-produced H2O2 is responsible for Prx2 oxidation, I compared neutrophils from an otherwise healthy individual with MPO-deficiency to those of a normal donor. MPO-deficiency is a genetic disorder where neutrophils show either very low MPO-content or barely any MPO-activity [450, 482]. The neutrophils from this individual contain <5% of the MPO content of control cells [450], and while they show a prolonged respiratory burst, they do not produce HOCl [483]. Most patients show no signs of an immune deficiency [484], however it is considered a risk for development of severe infections and chronic inflammation [485].
The Prx2 in erythrocytes exposed to MPO-deficient neutrophils (obtained from the same individual on two occasions and treated with PMA under the same conditions as in Section 3.3.3.1, Figure 3.9) increased from 4% dimer at baseline to 72 ± 2% (n = 2). Both DPI (6 ± 1% dimer) and catalase (19 ± 0.5% dimer) inhibited oxidation. These values were not significantly different from those obtained with MPO-replete neutrophils in the same experiment (Figure 3.10). Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 71
Figure 3.10 Comparison of erythrocyte Prx2 oxidation by MPO-deficient and non-MPO-deficient neutrophils. Erythrocytes from an MPO-deficient and a normal donor (2 x 107/ml) were incubated with isolated neutrophils at a ratio of 1:500 with 60 ng/ml PMA, 10 µM DPI, and 100 μg/ml catalase in HBSS for 10 minutes and treated with NEM as described in Sections 2.1.5.5 and 2.1.2. Results are the mean and SE of two independent experiments.
3.3.4 Effect of LPS-stimulated neutrophils on erythrocyte peroxiredoxin 2 oxidation
To determine the plausibility of neutrophil-induced erythrocyte Prx2 oxidation in vivo, a more physiological stimulus for the neutrophil immune response needed to be tested. For this reason, Escherichia coli lipopolysaccharide (LPS) was added to the isolated cells, together with human serum to provide necessary LPS-binding protein. LPS, a supreme instigator of the immune response to gram-negative bacteria, activates neutrophils by binding to the TLR-4 and CD14-receptor [220, 222] and up-regulates the assembly of NOX2 [486]. Activation of neutrophils is increased and accelerated in the presence of LPS binding protein (LBP), which is found in serum [487]. Furthermore, monocytes and macrophages recognize erythrocytes with bound LPS and LBP, but not LPS or LBP alone [223], so by implication the neutrophil response may be greater in the presence of erythrocytes. Prx2 oxidation was observed following incubation of erythrocytes with neutrophils and LPS, with the highest oxidation seen after 30 minutes (44 ± 16% vs. 2 ± 1% in control, n = 3) (Figure 3.11). Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 72
Figure 3.11 Effect of neutrophils stimulated with LPS on erythrocyte peroxiredoxin 2 oxidation. Erythrocytes (2 x 107/ml) were incubated with neutrophils at a ratio of 500:1 after stimulation with 10 µM LPS in HBSS (10% serum) for up to 60 minutes and treated with NEM as stated in Sections 2.1.5.6 and 2.1.2. Results are the mean and SE of three independent experiments, with a representative Western blot shown. *p≤0.05 in comparison with untreated control erythrocytes (repeated measures ANOVA, with Holm- Sidak multiple comparison method).
To establish that neutrophils produce superoxide under these conditions, a cytochrome c assay was carried out, with erythrocytes plus serum present as in Figure 3.11. SOD-inhibitable reduction of cytochrome c was observed up until around 30 minutes, at an approximate rate of 2 µM/min/106 cells (n = 2, Figure 3.12). This is likely to be an underestimation of superoxide production since addition of serum to the assay system partially reversed the extent of cytochrome c reduction, which might have been due to interference by human serum albumin [488].
Figure 3.12 Superoxide-induced reduction of cytochrome c by LPS-stimulated neutrophils. Erythrocytes (2 x 107/ml) were incubated with neutrophils at a ratio of 500:1 for up to 60 minutes as in Figure 3.11, spun down, and the supernatant was used to determine cytochrome c reduction as described in Section 2.1.5.7. Results are the mean and SE of two independent experiments.
Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 73
3.3.5 Effect of neutrophils stimulated with Staphylococcus aureus on erythrocyte peroxiredoxin 2 oxidation
PMA activation is very different to phagocytosis, where ROS production is limited to the phagosome [489]. To determine if ROS production during phagocytosis is able to oxidise Prx2 in neighbouring erythrocytes, neutrophils were co-incubated with Staphylococcus aureus. S. aureus is a gram-positive, catalase-positive bacterium, and a known inducer of the NADPH-oxidase inside the neutrophil phagosome [230].
Serum-opsonised S. aureus (20 bacteria per neutrophil) were added to a suspension of erythrocytes and neutrophils at a ratio of 1 neutrophil to 500 erythrocytes. This resulted in a time-dependent increase in the amount of oxidised Prx2 to a maximum of 32 ± 5% (n = 3) after 30 minutes (Figure 3.13). It is important to mention that this is likely not a linear reflection of the amount of ROS produced, since serum was present as in Section 3.3.4, and moreover, S.aureus is a catalase-positive bacterium which implicates that some ROS could have been scavenged instead of reaching reduced Prx2.
Figure 3.13 Effect of neutrophils stimulated with S. aureus on erythrocyte peroxiredoxin 2 oxidation. Erythrocytes (2 x 107/ml) were incubated with neutrophils at a ratio of 500:1 after stimulation with 4 x 105 S. aureus in HBSS (10% serum) for up to 60 minutes and treated with NEM as stated in Sections 2.1.5.8 and 2.1.2. Results are the mean and SE of three independent experiments, with a representative Western blot shown. *p≤0.05 in comparison with untreated control erythrocytes (repeated measures ANOVA, with Holm-Sidak multiple comparison method).
Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 74
3.3.6 Effect of LPS on peroxiredoxin 2 oxidation in murine erythrocytes in vivo
To investigate whether oxidation of Prx2 occurs in an inflammatory situation in vivo, mice were injected with LPS and their erythrocytes examined during the development of endotoxaemia. The Prx2 was almost entirely reduced at time zero and there was little change in the oxidation state for up to 6 hours (Figure 3.14, A & B). However, Prx2 oxidation was significantly increased at 10 hours, reaching a maximum of 16 ± 4% (n = 7) before returning to near baseline (n = 3) after 24 hours.
The role of neutrophil oxidant activity in this model was established using NOX2-/- mice. These mice lack the gp91 subunit of NOX2, and their neutrophils are not able to produce superoxide [490]. Erythrocytes from NOX2-/- mice showed no Prx2 oxidation at any time point examined following LPS challenge (Figure 3.14, C). Similarly, p47phox-/- mice, which also lack NOX2 activity, showed no Prx2 oxidation when treated with LPS (the Western blot looked identical to Figure 3.14, C).
Figure 3.14 Effect of LPS on peroxiredoxin 2 oxidation in murine erythrocytes in vivo over time. (A) Western blot following non-reducing SDS-PAGE; wild type (WT) C57BL/6J mice were injected with LPS and blood samples taken at baseline and at indicated intervals. NEM was immediately added to block subsequent Prx2 oxidation and cells were treated as described in Section 2.1.6. (B) Results are means and SE of 6 analyses (t=0), 5 analyses (2, 4 and 6 h), 7 analyses (6 and 10 h) and 3 analyses (24 h). Mice were sacrificed at the time of sampling so each analysis represents an individual mouse. *p≤0.05 in comparison with 2, 4, 6 and 24 hours (repeated measures ANOVA, with Holm-Sidak multiple comparison method). (C) As in A, but with NOX2-/- mice. Identical blots were obtained in two additional independent experiments. Experiments were performed by Dr. Ghassan Maghzal, University of Sydney, Australia. Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 75
3.4 Discussion
3.4.1 Neutrophil activation and peroxiredoxin 2 oxidation
Prx2 is highly susceptible to a variety of oxidants and is readily oxidised in erythrocytes treated with H2O2 and chloramines [4, 233, 234]. To test whether Prx2 becomes oxidised under stresses that could occur in pathological situations, I exposed erythrocytes to conditions mimicking fever, acidosis, bacterial infection and sepsis. I observed that Prx2 in erythrocytes is highly sensitive to oxidation by stimulated neutrophils. With PMA-stimulated neutrophils, Prx2 oxidation was time- and concentration-dependent and was observed at ratios of neutrophils to erythrocytes similar to those found in circulating blood (typically ~1:1000). Prx2 oxidation was additionally detected in whole blood when PMA was added to stimulate the neutrophils. In whole blood, erythrocytes are much more concentrated and other immune system cells, serum proteins, and antioxidants are present, which can influence the observed Prx2 oxidation. This suggests that even with other physiological influences present, neutrophil derived ROS oxidise Prx2 in neighbouring erythrocytes. Erythrocyte Prx2 also underwent oxidation with neutrophils stimulated by ingestion of bacteria or LPS. The extent of oxidation, however, was less than with PMA, which is consistent with the lower amount of superoxide produced by neutrophils exposed to LPS, and the generation of oxidants within the phagosome [197, 221, 491]. Based on these results, it is feasible that neutrophils could induce Prx2 oxidation in vivo if their activation is sufficiently high.
3.4.2 Reversal of erythrocyte peroxiredoxin 2 oxidation during inflammation
I also found that the oxidation of erythrocyte Prx2 was reversible. However, Prx2 is known to be recycled slowly in erythrocytes due to low levels of TrxR [233]. The extent of dimer accumulation reflects the relative rates of oxidation and reduction, and reversal over time can be explained by the gradual decline in NADPH oxidase activity following stimulation of the neutrophils, be it by loss of NADPH [479] or neutrophil cell death due to damage instigated by oxidants on the neutrophils themselves [492]. This balance between formation and removal means that the level of Prx2 dimer is a real-time rather than a cumulative measure of the oxidative stress experienced within erythrocytes.
Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 76
3.4.3 Role of neutrophil MPO in erythrocyte peroxiredoxin 2 oxidation
Neutrophils produce H2O2 and the MPO-derived oxidants, HOCl and chloramines [218]. As Prx2 is susceptible to all three, I tested different inhibitors and scavengers as well as cells from an MPO-deficient individual. Only the NADPH-oxidase inhibitor DPI and catalase inhibited erythrocyte Prx2 oxidation. Activation of MPO-deficient neutrophils resulted in similar levels of Prx2 oxidation to those seen in control cells, and inhibition of MPO or scavenging of HOCl had no significant effect. Therefore, I conclude that MPO-derived oxidants played only a minor role, and that the majority of the observed Prx2 oxidation was due to H2O2 derived from the NADPH-oxidase produced superoxide.
3.4.4 Erythrocyte peroxiredoxin 2 as a scavenger of extracellular hydrogen peroxide
It is possible that erythrocyte Prx2 could provide a protective mechanism against oxidants generated in the blood stream. Erythrocytes have been observed to protect against oxidative injury during hyperoxia or cardiac reperfusion [4, 194, 202, 203, 493]. This was attributed to catalase or reduced glutathione. However, glutathione was implicated on the basis of inhibition by thiol-blocking agents [194], which would have rendered Prx2 inactive. Therefore, based on previous findings that Prx2 is oxidised in erythrocytes exposed to low levels of H2O2 despite the presence of active catalase [233], my results support a role for Prx2 in protection from oxidative damage in the vasculature. While the slow turnover rate of oxidised protein could limit the oxidant removal by erythrocyte Prx2, its sheer abundance (200 µM in cells that constitute almost half of the approximately 5 l of total blood volume) and rapid transit through an inflammatory site provides an enormous antioxidant capacity. Additionally, neutrophils are known to be able to cause substantial collateral damage during their inflammatory response once they are activated [494-496]. While neutrophils migrate through tissue towards the focus of damage, instead of taking the shortest path, neutrophils stay inside blood vessels and capillaries as long as possible [497]. This could be to prevent damage of tissue and to ensure prolonged contact with the vast antioxidant capacity of erythrocytes. This supports the concept of erythrocyte Prx2 as protector against oxidative damage in the vasculature.
To examine whether these mechanisms of erythrocyte Prx2 oxidation operate in vivo, I studied a mouse model of relatively mild endotoxaemia [455]. I observed a transient increase in Prx2 oxidation six to ten hours after injection with LPS. This could reflect development followed by recession of the neutrophil phase of the response, and is consistent with Chapter 3: Erythrocyte peroxiredoxin2 oxidation state during inflammation 77
observations that neutrophils are activated during the initial stages of sepsis [498]. The absence of any changes in Prx2 oxidation status in mice lacking either gp91 or p47phox of the NADPH-oxidase, strongly suggests that NOX2 activity is responsible for the Prx2 oxidation. NOX2 is present not only in neutrophils but also in monocytes/macrophages and other cell types [490], and even erythrocytes contain various NOX enzymes [499]. Any of these cells could have contributed to the Prx2 oxidation observed in the mice. However, the known activation of neutrophils during endotoxaemia, plus the high capacity of their NADPH- oxidase to produce ROS, makes it likely that neutrophils were involved. Regardless of the source, my results indicate that there is sufficient oxidant generation in the vasculature, during the early stages of endotoxaemia, to cause systemic oxidation of erythrocyte Prx2. With a Prx2 concentration in the erythrocyte of ~240 µM [399], this is a significant oxidative load, which has been demonstrated previously in models of septic shock [500, 501]. Furthermore, it has been shown that Prx2 alone was able to protect mice from LPS-induced lethal shock [502], which underlines the importance of Prx2 for protection against oxidants.
3.4.5 Effect of temperature and pH on peroxiredoxin 2 oxidation within erythrocytes
I also investigated elevated temperature and low pH as pathological conditions for causing Prx2 oxidation in erythrocytes. My results indicate that pyrexia is unlikely to contribute, but that acidosis could increase Prx2 oxidation. The pH effect is most likely to be due to the increased oxidation of Hb, which is the major endogenous contributor to Prx2 oxidation [233, 301]. The lack of an effect of fever-like temperatures on the redox state of Prx2 could lie in the fact that temperature has a minor effect on oxygen affinity than protons [474].
3.5 Summary
In this chapter, I have shown that erythrocytes readily scavenge H2O2 generated by neutrophils which in turn leads to Prx2 oxidation within the erythrocyte. This oxidation occurs at neutrophil:erythrocyte ratios seen in whole blood and could be more extensive where neutrophils accumulate at inflammatory sites. Evidence of the in vivo relevance of this is my observation of transient Prx2 oxidation in a mouse model of endotoxaemia. As well as demonstrating the ability of erythrocyte Prx2 to scavenge oxidants generated both internally and externally, my results raise the possibility that accumulation of dimeric erythrocyte Prx2 could be an early indicator of sepsis and a useful real-time marker of oxidative stress.
Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 78
4 Changes in peroxiredoxin 2 redox state during storage of erythrocytes
Results of this chapter have been accepted for print in:
S. B. Bayer et al., Accumulation of Oxidized Peroxiredoxin 2 in Erythrocytes, Transfusion, 2015
4.1 Introduction
Transfusion of stored erythrocytes is an essential part of medicine. However, transfusion- related adverse effects, including multiple organ failure or myocardial infarction, are commonly reported and often related to the age of the stored erythrocytes [114, 503, 504]. While the mean storage age of blood packs for transfusion is approximately 19 days [505], erythrocytes can be stored for up to 42 days [506, 507], and more than a third of transfused blood packs are older than three weeks [505]. Adverse effects are thought to be caused by the progression of metabolic and structural changes [115, 129, 133, 508] known as “the storage lesion” [137] (see Section 1.2.4.3 for details on the storage lesion). A major contributor to the formation of the storage lesion is thought to be oxidative stress during storage [115, 124, 126, 127, 174], but a clear pathology has so far been elusive [108]. Based on the sensitivity of Prx2 to oxidation in erythrocytes, I reasoned that it could become oxidised in stored blood. This accumulation of oxidised Prx2 would consequently compromise antioxidant defences as well as provide a sensitive marker of oxidative stress.
4.2 Experimental approach
My objectives were to evaluate if Prx2 redox state changes occur during blood storage, and what factors were important in promoting oxidation. Additionally, I aimed to find a way to prevent, delay or reduce the observed Prx2 oxidation during or after blood storage. To achieve this, I simulated hypothermic blood storage by storing fresh erythrocytes in a commonly used storage buffer, SAGM (sodium, adenine, glucose, mannitol), in the same proportions used in blood banking (see Section 2.2.1 and Table 2.7). The erythrocytes were stored at 4°C for up to six weeks. In most experiments, samples were taken weekly under sterile conditions to determine the percentage of oxidised Prx2 as described in Section 2.2.2. Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 79
To test different methods to reverse Prx2 oxidation after storage, I tried to restart erythrocyte metabolism with glucose (Section 2.2.3), and the rejuvenation solution Rejuvesol™ (Section 2.2.4). As a different approach, reducing equivalents were added in the form of DTT, α-lipoic acid (ALA), dihydrolipoic acid (DHLA) or N-acetylcysteine (NAC). I also attempted to prevent Prx2 oxidation by including these compounds during the storage period (Section 2.2.5).
Additionally, I was curious to determine whether manipulation of the glutathione content by adding NAC in coordination with glycine and glutamine would have a preventative effect on Prx2. The effects of amino acid supplementation on low molecular weight thiols, which in erythrocytes consists almost entirely of GSH (plus NAC if added), were quantified spectrophotometrically at 412 nm with DTNB [509] (Section 2.2.10.1). GSH was also analysed (as its NEM derivative) by stable isotope tandem mass spectroscopy (Section 2.2.10.2). Further, I used the experimental buffer EAS76v6, and in a preliminary experiment, the FDA-approved AS-3, which can be used to prolong storage time.
To challenge the reduction capacity of erythrocytes, stored cells were stressed with H2O2, and reversal was monitored (Section 2.2.6). In an attempt to find out what affected Prx2 redox state and recycling during storage, a preliminary experiment was conducted with addition of carbon monoxide (CO) after storage to determine the contribution of autoxidation to Prx2 oxidation after storage.
To determine differences in the level of oxidised Prx2 in erythrocytes of different “ages”, the stored cells were fractionated by their densities via discontinuous Percoll gradient centrifugation (Section 2.2.7), and metHb and haemichrome concentrations as well as Prx2 redox state was assessed (Section 2.2.8). Additionally, supernatant from stored erythrocytes was used to collect vesicles and their protein and Prx2 content was visualised (Section 2.2.9).
Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 80
4.3 Results
4.3.1 Storage of erythrocytes and peroxiredoxin 2 oxidation
Samples of erythrocytes, stored in SAGM buffer under simulated blood bank conditions [507], were taken weekly and the oxidation state of Prx2 was analysed by SDS-PAGE under non-reducing conditions. As expected from the results in Chapter 3 and previous studies [233, 510], Prx2 from freshly isolated erythrocytes was present almost entirely in reduced, monomeric form (4 ± 1% dimer, n = 16). Oxidation levels of Prx2 in erythrocytes remained low until 21 days of storage (7 ± 1%), then rose progressively to reach 43 ± 6% after 42 days (Figure 4.1). When run on reducing gels, all the samples ran as a single monomeric band, consistent with the dimer being an interchain disulfide (Figure 4.2).
It is of importance to note that there is a high donor-specific variability in blood storage in general [131, 511, 512]. This is noticeable in the high variability of Prx2 oxidation, especially at the end of storage at day 42. For this reason, only the general trend was analysed, and samples from the same donated blood were treated and compared in each individual experiment. The Western blot in Figure 4.1 shows the highest Prx2 oxidation observed. The average of all 16 independent experiments was consistently lower.
Figure 4.1 Oxidation of erythrocyte peroxiredoxin 2 during storage at 4 °C. Haemolysates after storage at 4 °C in sterile SAGM buffer for up to 42 days. Samples were NEM-treated and immunoblotted as described in Section 2.2.2. A representative Western blot is shown. Results are the mean and SE of 16 independent experiments, with the exception of days 7 and 14, which were 3 and 6 experiments, respectively. *p≤0.05 in comparison with day zero (repeated measures ANOVA, with Holm-Sidak multiple comparison method). Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 81
Figure 4.2 Representative Western blot of peroxiredoxin 2 in stored erythrocytes under reducing and non-reducing conditions. Haemolysates after storage at 4 °C in sterile SAGM buffer as in Figure 4.1. Representative Western blot after SDS-PAGE under reducing and non-reducing conditions as stated in Section 2.1.7.
To test for hyperoxidation, reducing gels were immunoblotted with an antibody against Prx-
SO2/3. To gain a sufficient signal on the blots, 30 times more protein was necessary for detection than for detection of Prx2. The blots revealed low amounts of Prx hyperoxidation at day zero, which increased until day 42 (n = 3, Figure 4.3). Although the antibody is unable to distinguish between co-eluting Prx1 and Prx2, it is more likely that Prx2 was responsible for the signal as it is present in the erythrocyte at a 80 times higher concentration than Prx1[248]. Additionally, Prx2 hyperoxidation within the erythrocyte has been observed only recently [392]. However, the increase in hyperoxidised Prx in the stored cells probably represents only a small proportion, since hyperoxidised Prx2 cannot form once the fully oxidised dimer has accumulated.
Figure 4.3 Hyperoxidation of peroxiredoxins during storage at 4 °C. Same haemolysates as in Figure 4.1, but immunoblotted against Prx-SO2/3 after SDS-PAGE under reducing conditions. Results are the mean and SE of three independent experiments, relative intensity was calculated as stated in Appendix 8.1. *p≤0.05 in comparison with day zero (repeated measures ANOVA, with Holm-Sidak multiple comparison method). Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 82
4.3.2 Prevention or delay of peroxiredoxin 2 oxidation during storage of erythrocytes
ALA, DHLA, and the acidic NAC are small molecules which are known to affect thiol antioxidant networks. ALA is reduced by glutathione and thioredoxin reductases [513, 514] to its active form DHLA, which is able to scavenge some oxidants directly, as well as reduce protein disulfides [515]. NAC, a commonly used antioxidant, is a known precursor for glutathione that may also interact with protein disulfide networks [516, 517]. To test if ALA or NAC could delay Prx2 oxidation, each was added to the erythrocytes at the start of the storage period. Prx2 oxidation in erythrocytes was measured at days 21, 28, 35 and 42. There was no statistical difference with either treatment (n = 3), even though Prx2 oxidation with ALA and NAC appeared to be higher in comparison with the control (Figure 4.4). If NAC acted as a precursor of GSH, its effect could be limited by the availability of the other precursors, glycine and glutamine. However, supplementation with all three amino acids (NAC, glycine, glutamine (NGQ)) also gave no protection (n = 3, Figure 4.5). In contrast, when DHLA was added at day 21, subsequent oxidation of Prx2 was significantly decreased at day 42 (Figure 4.4). Of note is also that with addition of NGQ, which was pH neutralised, Prx2 shows less oxidation on day 21 when compared to the addition of unneutralised NAC.
Figure 4.4 Effect of α-lipoic acid (ALA), N-acetyl cysteine (NAC) or dihydrolipoic acid (DHLA) on oxidation of erythrocyte peroxiredoxin 2 during storage. Cells were lysed after storage in SAGM buffer containing either 250 µM ALA, or 5 mM NAC from day zero on, or 250 µM DHLA after day 21 (Section 2.2.5). Results are the mean and SE of three independent experiments. Controls are untreated stored erythrocytes. *p≤0.05 in comparison with untreated control (repeated measures ANOVA, with Holm-Sidak multiple comparison method).
To determine whether provision with GSH precursors during storage has any effect on GSH levels, the stored erythrocytes were tested for low molecular weight thiols via the DTNB assay [467, 509], and for GSH levels by stable isotope tandem mass spectroscopy [468]. Consistent with other studies [125, 131, 139, 172], erythrocyte GSH (which accounted for all Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 83
the low molecular weight thiols) dropped to about half during 42 days of storage, with most of the decline occurring after day 21 (n = 6, Figure 4.6). Supplementation with NGQ initially increased the total low molecular weight thiol content but not the level of GSH. By day 14 all the low molecular weight thiols could be accounted for by GSH and the loss appeared to be slightly slower. This also means that supplementation with NGQ had no effect on the intracellular concentration of free GSH.
Figure 4.5 Effect of NAC, glycine and glutamine (NGQ) supplementation on peroxiredoxin 2 oxidation in stored erythrocytes. Erythrocytes were stored at 4 °C in sterile SAGM buffer with or without NGQ supplementation for 42 days, blocked with NEM and separated with non-reducing SDS-PAGE as described in Section 2.2.2. Results are the mean and SE of three independent experiments. There was no statistical significance between NGQ and untreated controls (repeated measures ANOVA, with Holm-Sidak multiple comparison method).
Figure 4.6 Effect of supplementation with NAC, glycine and glutamine (NGQ) on free GSH and low molecular weight thiols (LMW) in erythrocytes during storage. Stored, packed erythrocytes were lysed with water and insoluble proteins were removed. The concentration of free small molecular weight thiols in the supernatant was determined using the DTNB assay. For GSH, the thiols were blocked with NEM before lysis and the supernatant was analysed by stable isotope dilution liquid chromatography tandem mass spectrometry assay as described in Section 2.2.10. Results are the mean and SE of three (NGQ) to six (controls) individual experiments. There was no statistical significance between NGQ and controls (repeated measures ANOVA, with Holm-Sidak multiple comparison method). GSH analysis was performed by Dr. Nina Dickerhoff. Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 84
As many of the observed changes in erythrocyte metabolism during storage seem to be driven by a drop in pH [122, 140, 141], I investigated Prx2 oxidation with an experimental additive solution (EAS-76v6) formulated by Hess et al. [188]. This solution, and a modified version AS-7 (additive solution 7), have a higher initial pH and greater buffering capacity than SAGM, since they contain bicarbonate (See Table 2.7 for details). These buffers have been shown to ameliorate the storage lesion and enable cold storage for up to 12 weeks [188, 191, 518]. On this basis AS-7 has recently been approved by the FDA for erythrocyte storage [191]. When erythrocytes were stored in EAS-76v6, the extent of Prx2 oxidation remained low throughout the 42 day testing period (Figure 4.7). A preliminary experiment with erythrocytes stored in AS-7 showed an average Prx2 oxidation of 13 ± 3% on day 42 (n = 3), which lies between the observed percentages of Prx2 of erythrocytes stored in SAGM (39 ± 14%, n = 4) or in EAS-76v6 (7 ± 2%, n = 3) on the same day.
Figure 4.7 Comparison of peroxiredoxin 2 oxidation in erythrocytes stored in SAGM or EAS-76v6. Erythrocytes were stored in either sterile SAGM, or in EAS-76v6 for up to 42 days. Results are the mean and SE of three (EAS-76v6) to four (SAGM) independent experiments. *p≤0.05 in comparison with SAGM sample at the same time point (repeated measures ANOVA, with Holm-Sidak multiple comparison method)
4.3.3 Reversal of peroxiredoxin 2 oxidation in stored erythrocytes
H2O2-induced Prx2 oxidation in fresh erythrocytes can be reversed metabolically if the cells are incubated in medium containing glucose [233]. To test for reversibility in stored erythrocytes, the SAGM buffer was removed and the cells were resuspended and incubated in
HBSS containing catalase (to prevent artifactual Prx2 oxidation by H2O2 in the buffer [233]). Incubation with glucose for two hours did not decrease the oxidation levels of Prx2 in stored samples collected at any time point between 21 to 42 days (n = 3, Figure 4.8, A). Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 85
To confirm that the dimer band in the stored erythrocytes was reducible in intact cells, we incubated 42 day samples in PBS containing 1 mM DTT. As predicted, DTT reduced the disulfide bonds of oxidised Prx2 back to a similar basal level as on day zero (n = 3, Figure 4.8, B).
Figure 4.8 Effect of glucose (A) and DTT (B) on erythrocyte peroxiredoxin 2 oxidation after storage. (A) Erythrocytes were stored as in Figure 4.1. Samples were incubated in fresh HBSS with 10 µg/ml catalase for two hours before NEM-treatment. *p≤0.05 in comparison with no incubation (repeated measures ANOVA, with Holm-Sidak multiple comparison method). (B) Samples were stored as in A, and incubated with 1 mM DTT for one hour. Results are the mean and SE of three independent experiments. *p≤0.05 in comparison with no incubation (repeated measures ANOVA, with Holm-Sidak multiple comparison method).
Rejuvesol™, a rejuvenation solution that contains pyruvate, inosine, adenine and phosphate, has been shown to reverse loss of ATP and 2,3-DPG within one hour after addition to stored erythrocytes [121, 122]. Hence, it was reasoned that Rejuvesol™ might be able to restart erythrocyte metabolism after storage and generate sufficient reducing equivalents to recycle the already oxidised Prx2. However, when Rejuvesol™, prepared according to Meyer et al. [121], was added to stored erythrocytes, there was no decrease in oxidation levels of Prx2 after incubation for two hours (n = 3, Figure 4.9, A).
I tested whether Prx2 reduction in stored erythrocytes could be achieved by addition of ALA, NAC or DHLA after storage. The SAGM buffer was removed and the cells were resuspended in PBS with either NAC, ALA or DHLA. No decrease in Prx2 oxidation was observable with addition of either NAC or ALA (n = 3, Figure 4.9, B). In fact, addition of NAC to the stored erythrocytes actually increased Prx2 oxidation, similar to what was observed in Figure 4.4 when NAC was present within the storage buffer. In contrast, addition of DHLA was able to completely reverse oxidised Prx2 levels back to basal levels (n = 3, Figure 4.9, C). Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 86
Figure 4.9 Effect of Rejuvesol™ (A), NAC, α-lipoic acid (ALA) (B) or dihydrolipoic acid (DHLA) (C) on peroxiredoxin 2 oxidation in erythrocytes after storage. Erythrocytes were stored as in Figure 4.1. (A) Samples were incubated with 24 µl Rejuvesol™ (1:8 Rejuvesol™:erythrocytes) at 37 °C for up to two hours. Results are the mean and SE of three independent experiments. p≤0.05 between zero hours and two hours treatment groups (repeated measures two way ANOVA, with Holm-Sidak multiple comparison method). (B) Samples were incubated with 5 mM NAC or 250 µM ALA (0.33% ethanol) at 37 °C for one hour. Controls were not incubated. Results are the mean and SE of three independent experiments. p≤0.05 between control and NAC, and between ALA and NAC. (Repeated measures two-way ANOVA, with Holm-Sidak multiple comparison method). (C) Samples were treated with 250 µM DHLA (0.005% DMSO) as in B. Results are the mean and SE of three independent experiments *p≤0.013 in comparison with control (paired t-test).
To see if the increased oxidation of Prx2 during storage is due to an impairment in reductive capacity, the cells were stressed by addition of 100 µM H2O2, then resuspended in HBSS containing catalase. With fresh erythrocytes, there was an initial increase in disulfide due to reaction with H2O2, then reversed back to the basal level within one hour, as reported by Low et al. [233] (Figure 4.10). With the stored erythrocytes, there was also a reduction of the disulfide bonds, but only back to initial levels observed prior to H2O2 treatment. Thus, the cells were able to reduce newly formed but not pre-existing dimers.
One possible explanation for the irreversible level of basal Prx2 oxidation is that it is caused by an increased basal level of Hb autoxidation. For this reason, erythrocytes were stored for six weeks in SAGM buffer, Hb autoxidation was inhibited with CO after storage, and cells were again challenged with H2O2 in a preliminary experiment. Inhibition of Hb autoxidation with CO had seemingly no effect on Prx2 redox state (Figure 4.11). This suggests that either Hb autoxidation is not the reason behind the limited reduction of Prx2 oxidation, or that CO saturation was not reached with the amount of CO used.
Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 87
Figure 4.10 Peroxiredoxin 2 oxidation in stored erythrocytes after challenge with hydrogen peroxide. Erythrocytes were stored as in Figure 4.1. Samples were incubated in fresh PBS containing 100 µM H2O2 for ten minutes at 37 ºC, then incubated in fresh HBSS with 10 µg/ml catalase for two hours at 37 ºC after removal of buffer. Results are the mean and SE of three independent experiments. p≤0.05 between before peroxide and after peroxide. (Repeated measures two-way ANOVA, with Holm-Sidak multiple comparison method).
Figure 4.11 Peroxiredoxin 2 oxidation in stored erythrocytes after inhibition of autoxidation with carbon monoxide and challenge with hydrogen peroxide.. Erythrocytes were stored as in Figure 4.1. Samples were pretreated with CO for one hour before incubation in fresh PBS with 100 µM H2O2 for ten minutes at 37 ºC. After removal of buffer, erythrocytes were incubated in fresh HBSS with 10 µg/ml catalase for one hour at 37 ºC. Results are the mean and SE of three independent experiments. There was no statistical significance (Repeated measures ANOVA, with Holm-Sidak multiple comparison method).
4.3.4 Differences in peroxiredoxin 2 oxidation, hyperoxidation and haemoglobin derivatives in stored erythrocytes separated by density
An alternate explanation is that Prx2 oxidation during storage could be limited to only a fraction of the erythrocyte population, explicitly the oldest cells, that have lost the ability to reduce their Prx2. These cells could have been closer to the end of their 100-150 day lifespan before they were stored. In this aged population, Prx2 could be completely oxidised, while the Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 88
rest of the erythrocytes could contain less oxidised Prx2. As mentioned in Section 2.2.7, the increased density of “aged” cells can be used to separate cells using density gradients [460, 462, 519]. For this reason, an attempt was made to separate stored erythrocytes into younger and older populations according to density. The separated erythrocyte populations were pooled into two distinct fractions. The two fractions were screened for their Prx2 redox states, metHb and haemichrome content. The vesicles from the supernatant were also tested for Prx2 presence.
The separation of the stored erythrocytes proved to be problematic, as their distribution was very different in comparison to fresh blood (Figure 4.12). While the fresh blood sample separated into seven distinct fractions, with the majority of erythrocytes being split between the top at 1.080 g/ml and the middle at 1.100 g/ml, the stored erythrocytes separated only into two fractions, with an estimate of less than 10% being denser than 1.110 g/ml and the rest staying at 1.080 g/ml. It is questionable how this relates to the age of the populations. Nevertheless, I examined Prx2 in the top and bottom fractions.
Figure 4.12 Comparison of the separation profiles of fresh erythrocytes and erythrocytes stored for 42 days. Fresh erythrocytes and erythrocytes stored for 42 days in sterile SAGM buffer at 4 °C were separated by centrifugation on a discontinous Percoll gradient. The density ranged from 1.116 g/ml to 1.080 g/ml. Percoll solutions contained PBS for isotonicity and 10 µg/ml catalase to prevent artefactual oxidation of Prx2 during centrifugation.
The oxidation and hyperoxidation state in both fractions were assessed by Western blotting. The denser, lower fraction F2 showed slightly less Prx2 oxidation (71 ± 14%, n = 6) than the upper fraction F1 (88 ± 7%), but the difference was not statistically significant (Figure 4.13, A). It is of note that the observed oxidation within the samples was extremely high in comparison to the other experiments, even before separation of the fractions. However, the hyperoxidation of Prx is also lower in the denser fraction F2, with 77 ± 8% less hyperoxidised Prx than the upper fraction F1 (Figure 4.13, B). Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 89
Figure 4.13 Prx2 oxidation (A) and hyperoxidation (B) in separated fractions of erythrocytes stored for 42 days. Erythrocytes were stored for 42 days at 4 °C in sterile SAGM buffer, separated by density. Samples were NEM-treated and immunoblotted, with a representative western blot shown. Results are the mean and SE of six independent experiments (A) Oxidised Prx2, 0.6 µg protein loaded. Intermittent bands were cut. There was no statistical significance (paired t-test). (B) hyperoxidised Prx after reducing SDS- PAGE, 15-20 µg protein loaded and immunoblotted against Prx-SO2/3. *p≤0.001 in comparison to F1 (paired t-test).
Interestingly, when the fractions were measured spectrophotometrically, the percentage of metHb content within F2 was significantly higher (n = 3, Table 4.1). Even though the differences in haemichrome percentage within each fraction was not statistically significant, the fraction with the lowest metHb and the highest hyperoxidation of Prx, F1, was also the fraction with the most haemichromes.
Table 4.1 Content of OxyHb, metHb and haemichromes within separated populations of stored erythrocytes. Spectral analysis of lysed erythrocytes after separation by density as described in Section 2.2.8 and 2.2.7 and calculation of OxyHb, metHb and haemichrome concentrations. Numbers are the mean and SE of three individual samples. *p=0.023 between F1 and F2 (t-test).
Fraction F1 F2 OxyHb 57.4 ± 18.8% 98.1 ± 0.8% MetHb 0.1 ± 0.1% * 1.1 ± 0.3% * Haemichromes 42.5 ± 33.7% 0.8 ± 0.6%
Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 90
The vesicles isolated via high-speed centrifugation from the supernatant on top of the Percoll gradient were tested for Prx2 content using the Prx2 antibody. Other proteins were assessed visually by In-Gel fluorescence-staining of the SDS-PAGE gel before blotting as detailed in Section 2.2.9 (Figure 4.14). The fluorescence-stain of marker proteins does not align with the usual Coomassie-stain of marker proteins. The vesicles contained detectable staining at positions corresponding to the membrane proteins Spectrin and Band3, as well as a variety of other proteins. Most protein visible was monomeric Hb. The vesicles also contained Prx2, which intensities related visibly with the intensities of Hb in the vesicles.
Figure 4.14 Content of erythrocyte vesicles after 42 days of storage. Erythrocytes were stored for 42 days at 4 °C in sterile SAGM buffer, separated by density and vesicles were isolated from the cell-free supernatant. The vesicles of six individual samples were lysed in reducing sample buffer and 6 µg proteins were separated by SDS- PAGE. (A) In-gel fluorescence stain of vesicle protein after reducing SDS-PAGE. (B) Same as in A, Western-blotted and immunostained against Prx2.
4.4 Discussion
4.4.1 Peroxiredoxin 2 oxidation and GSH levels during erythrocyte storage
I have shown that Prx2 remains mainly reduced in erythrocytes stored in SAGM at 4 °C for 21 days, after which it becomes oxidised and the disulfide progressively accumulates. Additionally, I observed an increase in the amount of the poorly reducible hyperoxidised form of Prx2, although this was not evident until a later time (after 28 days).
Prx2 oxidation during blood storage has recently been observed by others [520]. However, with our more frequent sampling, we could show that there is little change during the first three weeks, and that the rise becomes exponential towards the end of the six week storage period. This is of relevance as this three week period has been associated with changes Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 91
collectively known as “storage lesion” and with increased risk for transfusion related side effects [114, 503, 504]. As well as ATP loss [131, 139], these changes include loss of many membrane proteins during the first three weeks [156], increased binding of Prx2, GPx and catalase to the membrane [85, 139, 521], and increases in various other oxidative biomarkers [124, 127, 174]. A decrease in GSH level after three weeks storage has been reported [139, 171, 182, 183]. In one study this was accounted for by a rise in GSSG [139], but others have observed a decrease in total glutathione [171, 182, 183] as well as a decrease in rate of ATP- dependent de novo synthesis of GSH, [134], even with sufficient metabolites being present [131]. Our observed loss of GSH, mainly during the second three weeks, is consistent with these results. Supplementation with NGQ did not increase GSH initially, probably because synthesis is under feed-back control [134]. Another possibility is a stoppage within the S- adenosyl-methionine cycle, which impairs both GSH synthesis and methylation [131]. These results suggests light retardation of GSH loss, but any effect may have been limited by the low membrane permeability of glutamine [134] and the low level of ATP.
4.4.2 Hypotheses for the impaired reduction of erythrocyte peroxiredoxin 2 after storage
Although Prx2 oxidation during storage was reversible with DTT, there was no reversal when the cells were incubated in the presence of glucose. This contrasts with the reduction of Prx2 seen with fresh cells and implies a defect in recycling. Especially interesting is that when more Prx2 was oxidised by adding H2O2, this was reversed on incubation but there was no effect on basal oxidation levels. While this could just reflect a slower reduction rate, as has been proposed [520], it would not explain why the residual Prx2 dimers could not be recycled on extended incubation. Therefore, the oxidation state of Prx2 in cells could represent a new balance between oxidation and reduction. One possibility is that the rate of oxidation becomes relatively higher in the stored cells. It is known that the extra- and intracellular pH drops during erythrocyte storage [131, 141, 191, 518], and that this slows glycolysis, and decreases ATP [141, 522]. A low pH also increases the rate of Hb autoxidation [173], which is a major source of hydrogen peroxide in the erythrocyte [233]. However, incubation in HBSS should counteract the pH effect on Hb, anaerobically stored erythrocytes showed the same metHb levels than aerobically stored [178], and furthermore, there is only a small difference in pH after storage in SAGM and the newly introduced AS-7 [191]. Additionally, inhibition of Hb autoxidation with CO before addition of H2O2 and incubation in fresh HBSS did not increase Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 92
the reduction of Prx2. This leads to the conclusion that autoxidation has only a negligible effect on Prx2 recycling within stored erythrocytes.
The partial reduction of Prx2 in the stored erythrocytes could be explained by a heterogeneous population of cells, including those recently released from the bone marrow and those approaching senescence. Younger cells may retain their ability to recycle oxidised Prx2, whereas the older cells may lose this reductive capacity. NADPH levels are well maintained in stored erythrocytes [139], so this is unlikely to be the limiting factor. However, several other changes could be involved, such as loss of thioredoxin reductase activity in the older cells, or inhibition of thioredoxin itself [322]. A similar differential between young and older cells could also apply to other metabolites, and may explain why many erythrocytes are removed in 24 hours after transfusion [523, 524]. It is, however, difficult to distinguish selective loss of one component from a small population of cells [525] from a partial uniform loss from all cells. In one study, the most significant losses in metabolites were present within the erythrocyte population that represented only 1-2% of the erythrocytes in circulation [525]. Additionally, stored cells give anomalous profiles on the density gradients used to enrich young and old populations of fresh erythrocytes. The reason for the observed anomalous profiles could lie in the higher activity of the Gardos channel in younger erythrocyte population, which leads to a more rapid dehydration of younger cells [526]. Additionally, the increase in extracellular potassium during storage inhibits Gardos channel activation in older erythrocytes and leads to more homogenous cell densities within stored erythrocytes.
4.4.3 Possible differences in erythrocyte populations
I have shown that in lower density fractions of stored erythrocytes, Prx hyperoxidation and Prx2 oxidation might be higher than in fractions of higher density. Additionally, the lower density population in this work shows more haemichrome content, even though the difference is not statistically significant. The majority of erythrocytes are observed in the lower density fraction after 42 days of storage. While separation of fresh blood by density with Percoll gradients correlates with erythrocyte senescence in fresh cells [460, 462, 463, 519], this might not be true for stored erythrocytes. Membrane loss due to blebbing also diminishes cell volume and changes the density, since haemoglobin and other oxidised proteins are packed into those vesicles [175]. It has been observed that as early as 21 days into storage, erythrocyte populations would separate into less fractions than fresh cells, with a lack of clear boundaries between the populations [527]. It is therefore hard to interpret the findings without Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 93
certainty of the metabolical identity and age of the two separated fractions. Additionally, the observed Prx2 oxidation within these particular samples has been extremely high. Even though not statistically significant, the fraction with the highest amount of oxidised Prx2 appeared to have the higher haemichrome content and the more hyperoxidised Prx, which could indicate that in this fraction, the antioxidant network might have been more severely impaired during storage than in the other fraction.
4.4.4 Rejuvenation, small molecular weight antioxidants and the antioxidant quality of stored blood
A major focus of my work was on whether Prx2 oxidation could be the prevented or reversed to potentially improve blood quality for transfusions. Although Rejuvesol™ contains compounds necessary for restoring energy metabolism, like precursors for ATP [121], it lacks antioxidants or reducing agents, and had no effect on Prx2 when added to the stored erythrocytes. The thiol compounds, ALA and NAC were also ineffective when added during or after storage. Indeed, NAC increased Prx2 oxidation when added after storage. This is in opposition to the findings with NGQ. The most reasonable explanation lies in the acidic nature of NAC. When added during storage without neutralisation, the erythrocytes were able to buffer for the additional decrease in pH, which probably led to a faster inhibition of glycolysis, resulting in an earlier onset of the storage lesion and Prx2 oxidation. Added after storage to a lower concentration of erythrocytes, the decreased pH could have increased Hb autoxidation, which would explain the significant rise in Prx2 oxidation. The NGQ solution, however, had been neutralised before addition to the storage buffer. ALA is known to deplete GSH for turnover into DHLA [528], which could explain why ALA slightly increased Prx2 oxidation during storage. However, the reduced form of ALA, DHLA, was able to reduce oxidised Prx2 after storage, as well as to prevent it from accumulating. This is likely to be due to direct provision of reducing equivalents for the reduction of thioredoxin [529, 530]. Whether DHLA has an application in blood storage warrants further consideration.
4.4.5 Storage buffers and quality of stored erythrocytes
I also showed marked protection against Prx2 oxidation when erythrocytes were stored in an alternative solution with a higher pH and some buffering capacity (EAS76v6, AS-7) [188, 191]. These buffers have been designed to slow down metabolism, spare ATP and retain an appropriate ATP/DPG balance [141, 188, 191]. Their slightly higher pH could also decrease Hb oxidation [173] and hydrogen peroxide production. Storage lesion in general is decreased Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 94
and storage time can be extended without increasing negative outcomes. Preservation of erythrocyte reductive mechanisms and improved antioxidant capacity of Prx2 could be contributors to this improvement. Further studies are needed to determine if Prx2 protection from oxidation extends up to 12 weeks in EAS76v6 and how Prx2 oxidation is affected when erythrocytes are stored for up to 8 weeks in AS-7.
4.4.6 Peroxiredoxin 2 and vesiculation
Another interesting observation was that Prx2 is included in erythrocyte microvesicles after storage. However, if this is because of the increased binding of Prx2 to the erythrocyte membrane [521], or an attempt to rid the erythrocyte of oxidised and hyperoxidised Prx2 and other oxidised proteins [124, 154, 175, 531], is unclear. It could also be likely that vesicles contain a small amount of cytoplasm, which consists mostly of Hb and Prx2. However, blebbing is a regulated and specific process [154], which occurs more in young erythrocytes [164, 165], but is increased during blood storage [532, 533]. The majority of the observed free Hb caused by haemolysis during blood storage is Hb entrapped in vesicles [129, 155]. Additionally, the protein profile of micro- and nanovesicles is quite different [85, 156, 534]. One study showed that after calcium-induced blebbing, both vesicles contain Hb, but the nanovesicles do not contain Prx2 [534], which hints at the possibility of a highly selective mechanism.
4.4.7 Peroxiredoxin 2 redox state as a biomarker
Our results suggest that increased oxidation of Prx2 is a good biomarker of oxidative stress during storage of erythrocytes. We also propose that oxidised Prx2 could be used as an indicator for the quality of stored erythrocytes and the duration of the storage. Membrane- bound Prx2 has been proposed by Rinalducci et al. [464, 521] as a biomarker for blood storage, for example in blood doping. It has been shown that Prx2 binds increasingly to membranes under oxidative stress or in cases of oxidatively damaged membranes [24, 433, 445, 446, 535-537]. It has also been observed that Prx2 increases on the membrane during storage of blood [85, 156, 464, 521], and decreases when stored under anaerobic conditions [447, 521]. However, they concluded that measuring membrane binding is too intricate and time consuming to be practical. In contrast, the oxidation state of Prx2 is easily detectable by Western blotting, as elevated levels of oxidised Prx2 in a fresh blood sample can be caused by storage or potentially severe inflammation with neutrophil involvement [510]. Chapter 4: Changes in peroxiredoxin 2 redox state during storage of erythrocytes 95
4.5 Summary
In this chapter, I have shown that Prx2 oxidation increases progressively during simulated blood storage in SAGM, after it remained predominantly reduced for the first three weeks. Prx hyperoxidation increases during storage as well, but at a later time. In contrast to fresh cells, oxidation was not reversed by incubation with glucose. Storage of erythrocytes in a high pH, low chloride, and high phosphate/ bicarbonate buffer (EAS76v6 or AS-7) largely prevented accumulation of oxidized Prx for at least six weeks, and dihydrolipoic acid, but not Rejuvesol, N-acetylcysteine or α-lipoic acid, was able to reverse or protect against Prx2 oxidation. Further Prx2 oxidation occurred when hydrogen peroxide was added. However this was reversible, suggesting that the reductive capacity was compromised in some but not in all cells. Stored erythrocytes showed only two distinct fractions when separated by density with Percoll gradient centrifugation after 42 days of storage. The fraction with the lowest density showed a slight increase in Prx2 oxidation level, more Prx hyperoxidation, and more haemichrome formation in comparison with the higher density fraction. It is also of note that some Prx2 is expelled from the stored erythrocyte and packed into microvesicles. My results raise the possibility that accumulation of dimeric erythrocyte Prx2 could be a useful marker of the antioxidant quality of stored erythrocytes.
Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 96
5 Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state
5.1 Introduction
The flexible erythrocyte plasma membrane allows the erythrocyte to move through the vasculature to provide the necessary oxygen, all the while keeping its bi-concave shape. It also regulates intracellular cation concentrations, provides isolation of cell content against the exterior, and interacts with the environment through its membrane receptors. Additionally, the membrane provides signals for erythrocyte clearance. According to genome sequences, approximately 30% of the cellular proteins are integral membrane proteins. However, there are proteins that are associated with the membrane, but are not an integral part. They usually interact with integral proteins either by covalent binding, for example with intermolecular disulfide bridges, or by association via hydrophobic/hydrophilic patches on the tertiary or quaternary structures. One of those proteins is Prx2. Even though Prx2 is a cytosolic protein, 0.05% of the cell’s Prx2 content is associated with the membrane (approximately 3 µg/ml packed cells) [358, 399, 437]. While it is clear that Prx2 does not bind to the membrane by disulfide linkage, the main mechanism of how, why and where it binds has so far been controversial and elusive. Since membrane binding of Prx2 has been linked with its redox status, as well as its oligomeric status, I developed the hypothesis that Prx2 preferentially bound to the erythrocyte membrane in its oligomeric form, therefore reduced and hyperoxidised Prx2 would favour binding, while oxidised Prx2 would be destabilised and detach.
5.2 Experimental approach
The intention of this chapter was to determine if and how the membrane binding of Prx2 depends on its redox state, and to verify the effects of H2O2 and intracellular calcium on Prx2 membrane binding in erythrocytes. I aimed to use both intact erythrocytes and isolated membranes exposed to purified Prx2.
To accomplish this, I started by incubating fresh erythrocytes with H2O2, calcium, or both, and determined how the proportion of membrane-bound Prx2 was influenced by each treatment. Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 97
To assemble a simple system to measure Prx2 membrane binding outside of whole cells, I purified Prx2 from erythrocytes and isolated Hb-free erythrocyte ghosts by hypotonic lysis. Ghosts are defined as the intact erythrocyte membrane after haemolysis, which is devoid of its cytosolic proteins but has retained its sub-membranous cytoskeleton [469, 538, 539]. The cytoskeleton consists mostly of spectrin [540]. For Prx2 binding experiments, I evaluated the application of positively charged micro-beads for creation of inside-out membrane vesicles. Since this method showed a major flaw, I decided to conduct the binding experiments with isolated Hb-free erythrocyte ghosts instead. As long as the ghosts are kept in a hypotonic solution, the ruptured membranes will not reseal to form right-side-out vesicles [471]. Rather, the ghosts will stay accessible for proteins to interact with both the inside and the outside of the membranes. Reduced, oxidised and hyperoxidised Prx2 were prepared and incubated with ghosts to assess the effect of Prx2 redox state on Prx2 membrane binding behaviour. To gain insight into possible binding partners, Prx2 and membrane proteins were crosslinked with glutardialdehyde. To gauge a deeper understanding on what else might affect the membrane binding behaviour of Prx2, Hb was denatured into haemichromes and added to the simplified system, as well as calcium to test if Prx2 membrane binding still increased without cytosolic proteins present. Prx2 C-terminal peptide was used in the purified system to assess a possible competition, recombinant Prx3 was tested for the specificity of the membrane-Prx2 interaction.
5.3 Results
5.3.1 Membrane binding of peroxiredoxin 2 in whole cells
Fresh, isolated erythrocytes were incubated with increasing amounts of calcium, and ionomycin as an ionophore to allow entry into the cells, and the amount of membrane-bound Prx2 was assessed. The intensities of spectrin and Band3, the two largest integral membrane proteins, were quantified and used to normalise the Prx2 intensities. It is not feasible to measure the oxidation state of Prx2 bound to the membrane in this system because Prx2 oxidation occurs during lysis [233]. To trap Prx2 in its redox state, it is necessary to alkylate the protein before lysis. However, NEM alkylates all accessible protein thiols in erythrocytes, which influences Prx2 membrane binding [351, 425].
Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 98
The membrane-bound fraction of Prx2 rose with increasing amounts of intracellular calcium up to threefold at a calcium concentration of 1 mM in comparison with EGTA-treated controls (n = 3, Figure 5.1, A). EGTA-treated controls had a slightly higher amount of Prx2 bound than untreated controls (not shown). At the same time, the redox state of Prx2 within the haemolysate after NEM treatment did not change (Figure 5.1, B).
Figure 5.1 Effects of calcium on (A) membrane binding and (B) redox state of peroxiredoxin 2. (A) Erythrocytes were incubated with increasing amounts of calcium and ionophore, the cells were lysed and the ghosts were isolated and separated via SDS- PAGE under reducing conditions. Results are the mean and SE of three independent experiments, with a representative Coomassie-stained gel (top, Spectrin and Band3) and a Western blot from the same gel immunostained against Prx2 shown. Relative units relates to the difference in relative band intensities in comparison to the mean of the EGTA control, and intensities of Spectrin and Band3 on the Coomassie-stained gel were used for normalisation (see Appendix 8.2 for calculations used). *p≤0.05 in comparison with EGTA (repeated measures ANOVA, with Holm-Sidak multiple comparison method). (B) Redox state of Prx2 in haemolysate depending on calcium. The samples were the same as in A, but alkylated with NEM before lysis with sample buffer as in Sections 2.1.2 and 2.3.1.
To observe the effect of H2O2 on Prx2 membrane binding, erythrocytes were pretreated with sodium azide to inhibit catalase activity and increasing amounts of H2O2 added. The amount of membrane-bound Prx2 steadily decreased with increasing H2O2 concentrations. The loss of
Prx2 from the membrane reached 70% for a H2O2 concentration of 75 µM (n = 3, Figure 5.2, A). Prx2 was completely oxidised at this concentration (Figure 5.2, B). While it looks like
Prx2 binding is increasing after addition of 100 µM H2O2, further experiments with 200 µM Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 99
and 500 µM did not confirm this. At higher concentrations of H2O2, oxidative changes and crosslinking of other proteins to spectrin was regularly observed as blurring of the spectrin band, increase of protein bands with a higher molecular mass than spectrin, loss of β-actin and an increase of low molecular weight proteins on the Coomassie gels (Figure 5.3). This could be responsible for the apparent increase in Prx2 binding, since it would affect the normalisation with Band3 and Spectrin that was used in absence of a loading control. There were no obvious changes to the membrane up to 75 µM H2O2.
Figure 5.2 Effects of hydrogen peroxide on (A) membrane binding and (B) redox state of peroxiredoxin 2. (A) Erythrocytes were preincubated with 10 µM sodium azide and treated with increasing amounts of hydrogen peroxide. The cells were lysed and the ghosts were isolated and immunoblotted as in Figure 5.1. Results are the mean and SE of three independent experiments, with a representative Coomassie-stained gel and Western blot from the same gel shown. Spectrin and Band3 were used for normalisation as in Figure 5.1 *p≤0.05 in comparison with EGTA (repeated measures ANOVA, with Holm-Sidak multiple comparison method). (B) Redox state of Prx2 in haemolysate depending on hydrogen peroxide. The samples were prepared as in A, but alkylated with NEM before lysis as in Figure 5.1.
A one-off spectral analysis of H2O2-treated erythrocyte lysate revealed progressive oxidation to metHb (Figure 5.4, A, C). Low amounts of haemichromes (< 2%) were present at 75 µM
H2O2, which increased to 34% at 500 µM H2O2.
Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 100
Figure 5.3 Observed oxidative damage on membrane proteins by hydrogen peroxide. Erythrocytes were treated as in Figure 5.2. (A) Coomassie-stained acrylamide gel of membrane proteins after SDS-PAGE under reducing conditions. (B) In-gel fluorescence stained acrylamide gel after SDS-PAGE with trichloroethanol, as described in Section 2.2.9. The arrows indicate observed changes and damage induced by H2O2, such as non-reducible crosslinking with spectrin and increase of low molecular peptides.
Figure 5.4 Effect of hydrogen peroxide on the absorption spectra of haemoglobin. Erythrocytes were preincubated with 10 µM sodium azide and treated with increasing amounts of H2O2. The cells were lysed and the absorbance of the lysate was analysed spectrophotometrically between 500 and 700 nm. (A) Picture of haemolysates treated with increasing amounts of H2O2, from left to right. (B) Relative absorption spectra of Hb, with the double peaks of oxyHb clearly visible. (C) Changes in oxyHb, metHb and haemichrome content, calculated from B as described in Section 2.2.8. Results are from only one experiment. Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 101
To see whether calcium is able to compensate for the observed loss of Prx2 from the membrane, ghosts were incubated with increasing amounts of H2O2 in the presence of calcium and ionophore. There was no significant loss of membrane-bound Prx2 in the H2O2 treated cells in presence of calcium (n = 3, Figure 5.5), indicating that the calcium effect encorporates a different mechanism than the effect of H2O2 on Prx2 membrane binding. It is not affected by the H2O2 induced loss of Prx2 and is able to compensate for it.
Figure 5.5 Effect of hydrogen peroxide and calcium on membrane binding of peroxiredoxin 2. Erythrocytes were preincubated with 10 µM sodium azide and treated with increasing amounts of hydrogen peroxide in the presence of ionomycin and 100 µM CaCl2. The cells were lysed and the ghosts were isolated and immunoblotted as in Figure 5.1. Results are the mean and SE of three independent experiments, with a representative Coomassie-stained gel and Western blot from the same gel shown. Spectrin and Band2 were used for normalisation as in Figure 5.1. There was no statistical difference between the samples (repeated measures ANOVA, with Holm-Sidak multiple comparison method). For comfortable comparison, the same data as in Figure 5.2 is displayed as well.
5.3.2 Binding of purified peroxiredoxin 2 to isolated membranes
5.3.2.1 Binding of purified Prx2 to inside-out membrane vesicles
To study the interaction of purified Prx2 with isolated erythrocyte membranes, a few methods were considered. The ideal method involved inside-out membrane vesicles, since Prx2 binds to the inner membrane leaflet [358, 366]. However, some methods that create inside-out vesicles result in a loss of spectrin [471, 541, 542], or a high concentration of right-side-out vesicles [471]. For these reasons, the use of charged micro-beads for fast creation of inside- out membrane vesicles with a high yield and a minimal loss of membrane proteins was chosen [77, 470].
Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 102
Hb-free and Prx2-free inside-out membrane vesicles (IOM) were prepared as described by Kuross et al. [77]. To determine the “sideness” of the vesicles, acetylcholine esterase, an enzyme located on the outside of the erythrocyte membrane, was assayed [471]. For IOM, there should be no acetylcholine esterase activity. In two preparations, the acetylcholine eserase activity measured was 10% and 6% of the total determined in the presence of detergent to disrupt the vesicles.
To see whether the beads were completely covered by erythrocyte membrane, the IOM were stained with DiO, a lipophilic fluorescent tracer that penetrates membranes. Under the microscope, the beads appeared to be completely covered in erythrocyte membrane; however some membrane particles were free of beads (Figure 5.6, D).
Figure 5.6 Micrograph showing lipid-containing membranes encapsulating beads. Methacrylate copolymer beads (40 µm) were labelled with Vybrant™ DiO as described in Section 2.3.3.4. Photographs taken with AxioCam MRm. (A) Uncoated control beads, brightfield. (B) Same beads as in A, taken with FITC-filter, fluorescence emission at 501 nm. (C) Bead coated with erythrocyte membrane, photographed as in A. (D) Same bead as in C, photographed as in B.
To test for Prx2 binding, purified Prx2 was reduced with DTT, and the reduced and oxidised Prx2 preparations were added to IOM. Packed erythrocytes contain 5.6 mg/ml Prx2 [400], with a minimum of 3 µg/ml Prx2 bound to the membrane, which can be increased about threefold with calcium. Lysis of erythrocytes generates approximately 2 mg/ml membrane proteins. Therefore, 1.5 µg Prx2 should bind to each mg membrane protein. The IOM had a concentration of 1 mg/ml, so I chose a range of 0-15 µg/ml Prx2 to add to the beads to ensure saturation. Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 103
While the first results looked promising (Figure 5.7), there was more Prx2 binding to the IOM than expected even after three washes, and no difference in binding could be determined between reduced and oxidised Prx2. While the Western blot for membrane binding of reduced Prx2 showed binding with complex patterns when immunoblotted against Prx2 (Figure 5.7, A), the western blot for binding of oxidised Prx2 showed only two bands (Figure 5.7, B), even with increased concentrations of β-mercaptoethanol in both blots. Binding of Prx2 to the beads was saturated above 7.5 µg/ml Prx2 (Figure 5.7, C). Additionally, it was also observed that Prx2 would bind directly to uncoated micro-beads used to prepare IOM (Figure 5.7, D). This was not surprising, since the beads are strong anion exchangers used for protein purification. Since incomplete coating of the beads with erythrocyte ghosts could not be excluded, further experiments with IOM were dismissed.
Figure 5.7 Binding of reduced (A) and oxidised (B) peroxiredoxin 2 to IOM. (A) Reduced Prx2 was added to IOM and eluted from the beads as described in Section 2.3.3.6 and Western blotted after SDS-PAGE under reducing conditions as described in Section 2.3.4. (B) as in A, but oxidised Prx2 was added to IOM. (C) Western blot of unbound Prx2 present in the supernatant after incubation of Prx2 with IOM as in B. (D) Prx2 eluted from membrane-covered and uncovered beads. Western blots A, B and C are representative of two independent experiments, and D is representative of one experiment.
5.3.2.2 Binding of purified peroxiredoxin 2 to haemoglobin-free, peroxiredoxin 2-free ghosts
At this point, I decided not to be concerned about exposing only internal surfaces to the Prx2, and used ghosts prepared in a standard manner. Incubation of ghosts and Prx2 in hypotonic buffer with a pH below 8 should allow proteins to permeate the membranes [471] and membrane binding of Prx2 [350, 424]. Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 104
Hb-free and Prx2-free erythrocyte ghosts were prepared by hypotonic lysis as in Section 2.3.3.1, and resuspended in Dodge buffer II (5 mM phosphate buffer, pH 7.4, see Table 2.7). Purified, oxidised Prx2 was exchanged into Dodge buffer II using spin columns. Reduced Prx2, prepared by addition of DTT, and hyperoxidised Prx2, prepared by addition of DTT and
H2O2, were also tested. The Prx2 redox state of oxidised and hyperoxidised Prx2 was confirmed via non-reducing SDS-PAGE. Since Prx2 was not alkylated, completely hyperoxidised Prx2 is unable to form dimers and runs as a monomer on the Western blot (Figure 5.8). Because of the high sensitivity of reduced Prx2 to oxidation, it was necessary for DTT to be present in all buffers used after reduction of the protein with DTT.
Figure 5.8 Oxidised and hyperoxidised peroxiredoxin 2 on a Western blot after non-reducing SDS-PAGE. Purified, oxidised Prx2 was reduced with DTT and hyperoxidised by H2O2 as described in Section 2.3.3.5. The oxidised Prx2 was run side by side with the hyperoxidised Prx2 after buffer exchange by spin column centrifugation on a non-reducing SDS-PAGE and immunoblotted against Prx2. Most hyperoxidised Prx2 runs as a monomer, with only a small proportion of dimer visible. The band below the monomer band represents a small proportion of sulfonic acid Prx2.
Using the same calculations as in Section 5.3.2.1, around 150 ng purified Prx2 should bind to 100 µg membrane protein. To ensure saturation, 100 µg membrane protein was incubated with 0-10 µg purified Prx2 in all three redox states. Oxidised Prx2 bound to the erythocyte ghosts in a concentration-dependent manner (Figure 5.9). Hyperoxidised Prx2 also bound to erythrocyte ghosts in a dose-dependent manner, to a similar extend to that observed for the disulfide. The same dose-dependent binding was observed with reduced Prx2 in presence of DTT. When the membrane binding of Prx2 in all three redox states are compared, it appears that oxidised Prx2 binds slightly less to the membrane, while reduced and hyperoxidised Prx bind slightly better. However, there is no statistical significance between the redox state of the purified Prx2 and its binding behavior. The estimated amount of Prx2 that bound was approximately 1-2% of the concentration added, and reflected the calculated membrane binding capacity of 1.5 µg/mg membrane protein.
Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 105
Figure 5.9 Effect of redox state on membrane binding of peroxiredoxin 2. Erythrocyte ghosts were incubated with increasing amounts of Prx2, washed and immunoblotted as described in Sections 2.3.3.7 and 2.3.4. (A) Binding of oxidised Prx2 to erythrocyte ghosts. (B) Binding of hyperoxidised Prx2 to erythrocyte ghosts. (C) Binding of reduced Prx2 to erythrocyte ghosts. Results are the mean and SE of three independent experiments, with a representative Western blot and Coomassie-stained gel shown. Spectrin and Band3 served as a loading control, Prx2 concentrations on the membrane were calculated using the density of a Prx2 standard, as explained in Section 2.3.4. There is no statistical difference in binding between the three redox states (repeated measures ANOVA, with Holm-Sidak multiple comparison method). Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 106
5.3.2.3 Additional experiments to achive insight into mechanism of peroxiredoxin 2 binding
Since Cha et al. found that the C-terminus of Prx2 was involved in membrane binding [286], I investigated if membrane binding is specific to the Prx2 C-terminus by incubating the erythrocyte ghosts with purified Prx2 and letting it compete with excess C-terminal peptide. The C-terminal peptide is a synthetic peptide corresponding to amino acid residues 184-198 of Prx2. C-terminal peptide in excess did not interfere with Prx2 membrane binding (Figure 5.10). Additionally, BSA in excess did not compete with Prx2 for membrane binding either (not shown).
Figure 5.10 Competition between peroxiredoxin 2 and C-terminal peptide of peroxiredoxin 2 binding to erythrocyte ghosts. Erythrocyte ghosts were incubated with increasing amounts of Prx2 in the presence of 15 µM Prx2 C-terminal peptide, washed and immunoblotted as described in Figure 5.9. Western blot represents one independent experiment.
Having established that Prx2 membrane binding is independent of redox state, the question was if the binding is specific to Prx2. Recombinant Prx3, a mitochondrial Prx which forms a dodecamer when reduced or hyperoxidised, was added to erythrocyte ghosts under oxidising or reducing conditions. Prx3 also bound to the membrane in a dose dependent manner, and binding increased slightly but not significantly under reducing conditions (Figure 5.11)
Figure 5.11 Binding of recombinant peroxiredoxin 3 to erythrocyte ghosts. Erythrocyte ghosts were incubated with increasing amounts of recombinant Prx3 with or without 200 µM DTT, washed and immunoblotted as described in Section 2.3.4. Western blot represents three independent experiments. There was no statistical significance (repeated measures ANOVA, with Holm-Sidak multiple comparison method).
Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 107
To test whether the effect of calcium on binding of Prx2 to the membrane is dependent on another cytosolic protein, Prx2 was incubated with erythrocyte ghosts in the presence of 1 mM CaCl2. The presence of calcium significantly increased Prx2 binding (Figure 5.12), which implies that the calcium directly regulates the binding of Prx2 to the erythrocyte membrane.
Figure 5.12 Effect of calcium on peroxiredoxin 2 binding to erythrocyte ghosts. Erythrocyte ghosts were incubated with increasing amounts of Prx2 in the presence of 1 mM CaCl2, washed and immunoblotted as in Figure 5.9, with a representative Western blot and Coomassie-stained gel shown. Spectrin and Band3 served as loading controls. Results are the mean and SE of three independent experiments, relative units relates to the difference in relative band intensities in comparison to the mean of the 10 µg Prx2 control (see Appendix 8.2 for details on the calculation). p≤0.05 between controls and CaCl2 treated groups (repeated measures two way ANOVA, with Holm-Sidak multiple comparison method).
5.3.2.4 Binding of peroxiredoxin 2 to isolated erythrocyte ghosts in the presence of haemichromes
Haemichromes are denatured Hb [63] (see Section 1.2.3 for details). Haemichrome formation can occur spontaneously by autoxidation of Hb and further oxidation of metHb, or by binding of an exogenous ligand like acetylphenylhydrazine (APH) to Hb [43]. It was reported that haemichromes not only bind to erythrocyte membranes [170, 543, 544], but compete with Prx2 for binding sites [406, 438]. To test whether haemichromes do affect Prx2 binding in a pure system, haemichromes were produced by incubation of haemolysate with APH (Table 5.1).
Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 108
Table 5.1 Concentrations of oxyHb, metHb and haemichromes before and after incubation with APH. Concentrations were calculated as described in Section 2.2.8, representing two individual experiments.
OxyHb metHb Haemichrome Lysate before APH 548 µM 103 µM 29 µM Lysate after APH 66 µM 367 µM 205 µM
The resulting haemichromes were added to erythrocyte ghosts and increasing concentrations of Prx2, with Hb as a control (<2 µM haemichromes), and with or without DTT. Prx2 membrane binding did not change with incubation of Hb, while binding was decreased with addition of haemichromes (Figure 5.13). The inhibiting effect was the same in the presence of DTT (not shown). Another important observation was that after centrifugation of haemichrome-treated ghosts, the pellet appeared denser, smaller and of slight brown colour in comparison to Hb-treated ghosts, where Hb was completely removed after two washes, and untreated controls.
Figure 5.13 Competition of peroxiredoxin2 and haemichromes binding to erythrocyte ghosts. Erythrocyte ghosts were incubated with 15 µM haemichromes, the same volume of fresh lysate for Hb (<2 µM haemichromes) and increasing concentrations of oxidised Prx2 as in Section 2.3.3.9. For SDS-PAGE, same volumes of samples were loaded, representing 6 µg membrane proteins before addition of Prx2 or haemichromes. Spectrin and Band3 served as loading controls. Relative units relates to the difference in relative band intensities in comparison to the mean of the 10 µg Prx2 control (see Appendix 8.2 for detailed calculation). Results are the mean and SE of three independent experiments, with a representative Coomassie-stained gel and Western blot of the same gel shown. p≤0.05 between control and haemichrome and between Hb and haemichrome (repeated measures two way ANOVA, with Holm-Sidak multiple comparison method). Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 109
5.3.2.5 Peroxiredoxin 2 crosslinking to membrane proteins
As haemichromes bind specifically to Band3, I attempted to determine whether Prx2 binds to specific membrane proteins. In a preliminary experiment, Prx2 was crosslinked to its interaction partners with glutardialdehyde. In-gel fluorescence stained acrylamide gel shows Spectrin and Band3, but no Prx2 is visible. The Western blot immunostained against Prx2 shows, in addition to the Prx2 monomer and dimer, three more bands of potential interest (Figure 5.14). The high density band above 250 kDa could represent the Prx2 decamer, which has a molecular weight of 218 kDa, however since no similar experiment has been conducted with Prx2 and glutardialdehyde alone, this is only speculation.
Figure 5.14 Crosslinking of membrane proteins and oxidised peroxiredoxin 2 with glutardialdehyde. Erythrocyte ghosts were incubated with and without 5 µg oxidised Prx2 and crosslinked with glutardialdehyde as described in Section 2.3.3.10. (A) In-gel fluorescence stained acrylamide gel with trichloroethanol as described in Section 2.2.9. (B) Western blot of the same gel as in A, but immunostained against Prx2. Figure is representative of two independent experiments. M indicates marker, - indicates no Prx2, + indicates added Prx2.
5.4 Discussion
5.4.1 Effect of calcium on membrane binding of Prx2
I have demonstrated that Prx2 binds to erythrocyte membranes and this is dependent on intracellular calcium concentration, in accordance to previous reports [351, 426, 433]. An increase in intracellular calcium had no effect on the redox state of cytosolic Prx2. Increased binding of Prx2 in the presence of calcium was still present with ghosts and purified Prx2. One possibility is that calcium could stabilise Prx2 in a way that enforces binding. For example, it was observed that calcium induced conformation change and polymerisation of Prx2, which was reversed in the presence of sodium chloride [424]. Another possibility is activation of calcium-sensitive enzymes associated with the erythrocyte membrane. Syk kinase, a calcium-activated kinase, is known to phosphorylate Band3 and other membrane Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 110
proteins [545-547], which increases crosslinking and could change access to possible binding partners of Prx2. However, treatment of isolated ghosts with the thiol oxidant diamide, which is able to increase phosphorylation of membrane proteins [548], showed no effect on Prx2 membrane binding when Prx2 was added after the treatment [438]. Calcium could also regulate Calpain-1 activity, which is capable of cleaving Prx2 and Band3 [286, 434, 444, 549]. This could be hinted to by the observed small increase of Prx2 binding in the presence of EGTA, since the cellular calpain inhibitor calpastatin is also dependent on calcium [549, 550].
5.4.2 Redox state of peroxiredoxin 2 and effect of H2O2
I also demonstrated that H2O2, which oxidises Prx2, results in the loss of Prx2 from the erythrocyte membrane. This is in opposition with what Rocha et al. found [445], but in agreement with Matte et al. [406, 438]. The reasons behind the different findings are so far unclear. However, when the experiments were conducted in a purified system, the Prx2 redox state showed no influence on Prx2 membrane binding. This suggests that the observed changes in membrane binding in whole erythrocytes are unlikely to depend on Prx2 redox state alone. While it was observed that Prx2 was increased on erythrocyte membranes after heat stress [551], blood storage [464, 521], diamide [438] and in one case H2O2-treatment [445], it was expected to consist of reduced or hyperoxidised decamers. This expectation arose from the observation that hyperoxidised and reduced 2-Cys Prx2 decamers are more stabilised, while oxidation and dimerisation would destabilise the decamer [293, 355, 552]. Furthermore, Matte et al. also observed increased Prx2 binding with treatment of whole cells with diamide [438]. Diamide is a thiol probe that rapidly oxidises reactive thiols, including Prx2, which adds further weight to the observed redox state being independent of Prx2 membrane binding.
5.4.3 Effect of haemichromes on membrane binding of peroxiredoxin 2
I have revealed that haemichromes interfere with Prx2 binding to the membrane, while neither Hb nor C-terminal peptide showed a similar effect. Matte et al. reported a loss in Prx2 membrane binding with H2O2, and also with phenylhydrazine. Matte found that haemichromes, generated by H2O2 and phenylhydrazine, bind to Band3 with high affinity [69, 73] and displace Prx2 from Band3 [406, 438]. Attachment of haemichromes to Band3 leads to Band3 aggregation [70], which could also explain the observed decrease in membrane pellet size after incubation of erythrocyte ghosts with haemichromes. I have further shown that Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 111
small amounts of haemichromes are present in the lysate from erythrocytes treated with H2O2 concentrations where Prx2 binding was inhibited. I have also demonstrated that the increase of Prx2 binding caused by calcium is able to compensate for the loss of Prx2 binding caused by H2O2, which hints at two different, independent mechanisms, one that is causing binding, and one that is removing Prx2 from the membrane.
5.4.4 Specificity of peroxiredoxin 2 binding and possible binding partners
The preliminary results in this chapter do not clarify if Prx2 binds to membrane proteins. However, the interaction with the membrane seems to be specific to Prxs and haemichromes, since Prx3, a mitochondrial Prx, also bound to isolated erythrocyte ghosts, and neither Hb nor BSA (not shown) seemed to display a similar behaviour. While I was unable to show that Prx2 binds to Band3, it has been reported before [406]. Additionally, the cytosolic domain of Band3 is very unstable and is regularly found detached from the transmembrane domain in SDS-PAGE gels [437, 440, 553]. Band3 degradation has also been observed [437, 553], and can be induced by ROS [442], caspases [442], and calpain [444]. After storage the cytoplasmic domain appears free in the cytosol [124]. Prx2, when it was known as calpromotin, was also observed to bind to other membrane proteins, including stomatin, GAPDH, actin, and an unknown 40 kDa protein [439]. This unknown protein could represent the cleaved cytoplasmic domain of Band3, which has a size of 42 kDa [554]. Another possibility could be the membrane-bound and calcium-sensitive calpain activator protein dimer [555], which has a size of 44 kDa and increases the susceptibility of calpain for activation by calcium. Interaction of Prx2 with this protein would at least explain the aforementioned increase of Prx2 membrane binding with calcium, as well as the reported cleavage of Prx2 by calpain [286, 434]. In a preliminary search with glutardialdehyde, I identified several bands that could be associated proteins. It is important to further study the nature of the unknown interaction partners of Prx2 to elucidate the metabolical reason behind Prx2 membrane binding.
5.5 Summary
In this chapter, I have revealed that Prx2 binds to the erythrocyte membrane and is partially dependent on calcium but independent of its redox state. I have demonstrated that Prx2 is removed from the membrane during H2O2-induced oxidative stress, which is most likely caused by haemichrome formation. The removal of Prx2 from the membrane and the increase in Prx2 binding caused by calcium are two independent mechanisms. Further I have shown Chapter 5: Membrane binding of peroxiredoxin 2 and peroxiredoxin 2 redox state 112
that binding is likely due to interactions with membrane proteins or membrane-associated proteins, which might depend on the decameric structure of Prx2.
Chapter 6: General discussion 113
6 General discussion
6.1 Summary of major findings
Prx2 is a highly reactive antioxidant protein within erythrocytes, where is plays an important, nonredundant role within the antioxidant defence system. While it is known that no other enzyme is able to compensate for Prx2, the main role for its high presence in erythrocytes has been so far elusive. More specifically, its importance as an antioxidant within the erythrocyte, its redox state during oxidative stress in disease and blood storage, and how both are tied to its membrane binding, has not yet been clearly determined. The purpose of this thesis was to establish how disease and blood storage affects Prx2 redox state in erythrocytes, and how Prx2 redox state influences its membrane binding capability.
Prx2 oxidation increased with a decrease in pH levels commonly observed in disease, but was not affected by fever-like temperatures. Erythrocytes readily scavenged H2O2 generated by neutrophils, which led to Prx2 oxidation within the erythrocyte. Scavenging of MPO-derived ROS was less significant under the simulated ex vivo conditions. Oxidation of Prx2 was still observed when cells were stimulated with physiological activators of the neutrophil oxidative burst. Even during phagocytosis of bacteria, where ROS are generated inside the phagosome, Prx2 was still getting oxidised. Prx2 oxidation after neutrophil activation was completely inhibited by DPI, and affected by catalase. These findings led to the identification of H2O2 generated through superoxide production by the neutrophil NADPH oxidase as the main source of ROS. The observed Prx2 oxidation occurs at neutrophil:erythrocyte ratios seen in whole blood and could be more extensive in inflammatory sites, where neutrophils accumulate. In a mouse model of endotoxaemia, transient Prx2 oxidation was observed at an early time point. Prx2 oxidation was absent in mice whose neutrophils lacked the ability to generate an oxidative burst. This suggests Prx2 oxidation could be an excellent early marker for sepsis and neutrophil involvement in diseases.
To test whether Prx2 oxidation could also be a marker for the quality of blood storage, I stored erythrocytes under simulated blood banking conditions. Prx2 remained mainly reduced during the first three weeks of storage, after which Prx2 disulfide accumulated progressively, and Prx hyperoxidation increased after 28 days. NAC, ALA, or a mixture of NGQ was unable to prevent or delay the observed Prx2 oxidation when added to the storage buffer. After Chapter 6: General discussion 114
storage, rejuvenation of the cells by offering glucose, Rejuvesol™, ALA or NAC did not reverse Prx2 oxidation. Only DHLA seemed to have an effect on Prx2 oxidation, both when added during or after storage. Using an experimental buffer, EAS76v6, as well as the FDA- approved AS-7 in a preliminary experiment, it was possible to delay the observed Prx2 oxidation, which was significant in the case of EAS76v6. While stored erythrocytes are still able to cope with additional H2O2 induced stress, the oxidised Prx2 represents a new basal Prx2 oxidation level within the red cell. This new basal level of Prx2 oxidation appears not to be caused by an increased occurrence of Hb-autoxidation after storage, but could represent a population of old erythrocytes which are unable to recycle Prx2. However, I was unable to confirm this. It is also important to note that some Prx2 is expelled from the stored erythrocyte and packed into microvesicles (see Section 4.3.4).
To clarify some of the apparently conflicting hypotheses regarding to membrane binding of Prx2, I wanted to repeat some of the experiments that others had performed and test the effect of calcium and H2O2 on Prx2 membrane binding in erythrocytes. I confirmed that calcium increases Prx2 membrane binding, and that H2O2 decreases it, while both at the same time do not cancel each other out. This leads to the assumption that both are two independent mechanisms. After finding a working method to use with isolated Prx2 and erythrocyte membranes, my experiments showed that Prx2 membrane binding is independent of Prx2 oxidation state. Although oxyHb did not affect Prx2 membrane binding, addition of haemichromes significantly interfered with membrane binding of Prx2. I was also able to determine that calcium still increases Prx2 membrane binding in this limited system. This finding indicates that even though Prx2 membrane binding is independent of redox state, its activity and interaction with other proteins or processes might not be. Instead, membrane binding could theoretically be dependent on the decameric structure, which may be affected by calcium, as others have suggested [424, 556]. This can now be used to explore the physiological functions of the decameric structure and the possible differences in activity that might come with a change of the redox status of Prx2.
Chapter 6: General discussion 115
6.2 Implications and future work
6.2.1 Peroxiredoxin 2 as a biomarker for oxidative stress
I have shown that in the erythrocyte the level of Prx2 oxidation can be used as a biomarker for oxidative stress because the oxidised Prx2 is easy to distinguish by Western blotting, and because the protein is so sensitive to oxidation. Erythrocyte Prx2 redox status has advantages over the use of Prx2 membrane binding or even the redox state of membrane-bound Prx2 as a biomarker, as suggested by Rinalducci et al. [464, 521] and Marrocco et al. [557] for blood storage, and Rocha et al. for H2O2 induced oxidative stress [445]. In some of these studies, NEM was used to trap the redox state of Prx2, due to the high reactivity of the Prx2-thiols with H2O2 [233, 302]. However, alkylators also block other important thiol groups, which would be able to alter Prx2 membrane binding [558]. Plishker et al., for example, noticed that iodoacetic acid reduced Prx2 membrane binding [351], while it did not completely inhibit it [425], and reasoned that Prx2 could be bound to the membrane via disulfide linkage. Erythrocyte membrane stability is dependent on spectrin-thiols [559], which can be affected by alkylation and in turn could have effects on Prx2 membrane binding. Iodoacetic acid also alkylates thiols in GAPDH, Band4.2, Hb, and in Band3 [425]; all proteins that are involved in Prx2 interactions [302, 406, 560]. Alkylation also inhibits calpain proteolysis [561-563], which could theoretically interfere with Prx2 membrane binding as well [331, 434]. Because of the high reactivity of the Prx2 thiols towards H2O2, washing the cells in buffers without removing trace amounts of H2O2 with catalase first, as has been done in the aforementioned studies [464, 521, 557], already influences Prx2 redox state [233].
The membrane-bound proportion of Prx2 is also influenced by how the erythrocyte ghosts are purified. Prx2 binding to the membrane has been reported to decrease if the pH of the buffer used is higher than pH 7.8 [350, 424]. Prx2 is also removed from the membrane if EGTA or EDTA is present [351, 425, 426, 433], or if NaCl is used to wash membranes [424]. The use of both is documented in the studies of Rinalducci and Marrocco [464, 521, 557]. Furthermore, under conditions of oxidative stress which involves formation of haemichromes, Prx2 is removed from the membranes [85, 406, 438]. Thus, the use of membrane-bound Prx2 as a biomarker for oxidative stress is problematic and should be taken with care.
In contrast, the redox state of Prx2 is dependent on the amount of hydrogen peroxide present and the level of reduced thioredoxin (see Figure 6.1). The trapping of the Prx2 redox state by sufficient alkylation [233] before processing of the erythrocytes is crucial. The Prx2 redox Chapter 6: General discussion 116
assay that has been used in this thesis is straightforward and simple to comprehend. It is a real-time measure for oxidative load; even changes in ROS that occur in less than five minutes can be measured effectively (see Section 3.3.3). Additionally, the purification of erythrocyte ghosts is not applicable, which makes it faster and even easier to use as a biomarker to screen great numbers of subjects, for example in a clinical environment, as has been done with G6PD-deficient babies [105]. In this study, the assay was successfully used to measure impaired antioxidant activity. Not only did G6PD-deficient neonates show a significantly higher basal Prx2 oxidation level, but when erythrocytes were stressed with
H2O2, the G6PD-deficient cells were extremely impaired in their Prx2 recycling. This indicates that the compromised antioxidant activity of Prx2 could lead to haemolysis in G6PD deficiency. Furthermore, the Prx redox assay has been applied frequently to observe changes in redox status in other cell types [233, 564-566].
Figure 6.1 The balance of oxidation and reduction of peroxiredoxin 2 in erythrocytes. Schematic summary of physiologically relevant impacts on peroxiredoxin 2 redox state. For details see Section 1.5.
My results suggest that the erythrocyte Prx2 redox assay could be used as an early real time marker for human sepsis [510]. Other potential conditions that could use Prx2 redox state as a biomarker include ischaemia-reperfusion, diseases other than sepsis involving neutrophil activation and chronic inflammation like familial Mediterranean fever, inflammatory bowel disease, rheumatic arthritis, chronic obstructive pulmonary disease or granulomatosis with polyangiitis. Prx2 redox state could also be tested as a biomarker for measuring oxidative stress in metabolic syndrome, particularly obesity and diabetes, and aerobic or anaerobic training, or even in pre-eclampsia and eclampsia. Another field to explore would be everything that leads to haemolytic anaemias, for example glutathione peroxidase deficiency, glutathione reductase deficiency, or abnormal haemoglobins. Chapter 6: General discussion 117
6.2.2 Improvement of the quality of blood storage
Since the transfusion of blood is still highly important and transfusions based on the creation of erythrocytes from human embryonic stem cells is not yet available on a large scale [567], the enhancement of blood storage is important. Based on my results and the physiological influences on Prx2 redox state (Figure 6.1), the possibility of Prx2 being more than just a consequence of oxidative stress is high. It is likely that oxidised Prx2 reflects diminished antioxidant function in stored erythrocytes, since minimising the oxidative stress and offering nutrients for rejuvenation had only marginal effects and was unable to decrease Prx2 oxidation after storage. The use of the Prx2 redox assay as a measure of the quality of erythrocyte storage is a useful method to screen for possible ways to prolong blood storage, to measure the antioxidant quality of blood for transfusion, as well as to test the efficiency of rejuvenation methods. In particular, given this work has been conducted on simulated erythrocyte storage, it would be beneficial to see how Prx2 redox state varies with the different storage buffers that are available by testing erythrocytes stored by blood banks in AS1-AS7, CPD, CPDA-1, and other experimental additive solutions. It would also be valuable to investigate Prx2 redox state changes in aerobic vs. anaerobic blood storage. Gamma irradiation of blood packs is used for immunocompromised patients in need of transfusions [568, 569]. Gamma irradiation is expected to have a high impact on antioxidant systems within erythrocytes, including Prx2. However, Prx2 oxidation has not been measured yet in this case. With access to whole blood bags, other antioxidants, additives and biomarkers can be tested and measured, for example the use of DHLA or ascorbic acid during storage and its impact on erythrocyte deformability, haemolysis and vesiculation, on GSH and GSSG, haemichromes, as well as ATP levels. It would also be intriguing to see if Rejuvesol™ or erythrocyte storage in EAS76v6 have an effect on GSH synthesis.
My experiments have shown that in stored erythrocytes, reduction of Prx2 disulfides is impaired, but not absent. This might be due to different ages of the cell populations, with younger erythrocytes retaining their reductive capacity, and older cells being completely unable to cope with oxidative stress. With access to an improved method of separating cells by age, for example a flow cytometer with a cell sorter or by specific antibodies against glycated HbA1c or aggregated Band3, it would be possible to determine if the reason of the limited Prx2 reduction after storage lies in the complete antioxidant inactivity of a part of the erythrocyte population after storage, or inhibition of thioredoxin activity. Another beneficial experiment would be the transfusion of stored erythrocytes into an animal, and the Chapter 6: General discussion 118
determination of Prx2 oxidation over time in samples taken from the animal. This would establish how long oxidised Prx2 in erythrocytes survives in the blood stream due to erythrocyte clearance and dilution, which could verify if Prx2 redox state could represent a biomarker for blood doping by autologous transfusions.
6.2.3 Peroxiredoxin 2 membrane binding, decameric structure and calcium
My results from Section 5.3.2.2 have led to the conclusion that Prx2 membrane binding is independent on redox state. Taking into account that the Prx2 concentration within the erythrocyte is over 200 µM [399, 400, 418, 570], and that a concentration of 1-2 µM is enough to ensure decamer formation under oxidising conditions [282, 293, 354, 358], it is likely that Prx2 binds as the decameric structure, as proposed by Schroder et al. [293]. While it is questionable if oxidised Prx2 was present in its decameric form in the micromolar concentrations used in this thesis (0.5-4.5 µM, in contrast to approximately ¼ dimer at 10 µM [282]), the low binding of Prx2 at concentrations seven times higher than the calculated maximum suggest that it might not be the Prx2 dimer that binds, either. To elucidate whether only decamers bind, a recombinant Prx2 mutant unable to form decamers would be of interest. Another useful mutant would be a Prx2 that only forms decamers and is peroxidatively inactive.
Additionally, I have showed that calcium still increased Prx2 binding to the membrane in a limited system. The determination of how calcium increases Prx2 membrane binding would therefore be useful. In plant chloroplasts, ATP and magnesium are responsible for Prx2 structure transitions, which then form larger assemblies [556]. In that study, calcium was able to substitute for magnesium. Plant Prx2 also shows membrane binding behaviour [571, 572]. Valuable information could be gained by exploring if ATP and magnesium increase membrane-bound Prx2 in a purified system or if calcium induces a structural change within the Prx2 decamer by crystallography. Since C-terminal truncation of Prx2 was reported to affect membrane binding without interfering with peroxidase and chaperone activity [286], Prx2 mutants with a variety of deleted or exchanged amino acids in the C-terminal sequence could be assessed for their dependence on calcium for membrane binding.
6.2.4 Peroxiredoxin 2: holdase and thiol protector?
Since Prx2 did bind to the membrane independent of its redox state in this work, the role of Prx2 membrane binding within the erythrocytes, how it functions and what might govern its Chapter 6: General discussion 119
function needs to be rethought. Prx2 activity within the erythrocyte has been described as a holdase, a chaperone that prevents aggregation of proteins without actually folding the proteins [282, 354, 357, 360, 369]. Prx2 in erythrocytes has been observed to bind to haem or Hb [369, 426], and has been shown to protect Hb from exogenous oxidation [302] and degradation [369, 537]. While haemin inhibited peroxidase activity of Prx2 [287], it was not tested if Prx2 was in turn oxidised, or if it was a structural inhibition. A similar observation has been made with Prx6 [385], another Prx present in erythrocytes. Interestingly, in both cases thiol-groups are involved [385, 573]. Additionally, a Prx2-Prx6 complex of 92 kDa was detected [574]. In a Prx2 KO mouse model, Hb was more sensitive to aggregation, but was protected by addition of peroxidatic functional Prx2 decamers in the presence of >1 mM H2O2 [369], while addition of higher molecular weight structures of Prx2 and Prx2 disulfides did not. Some have reported that Prx2 holdase activity is independent of Prx2 redox state [357], whereas others found that the peroxidatic inactive Cys51 and Cys172 mutants were not protective as holdases [369, 575]. Other proteins that have been observed to form complexes with Prx2 are Band3 [406], alcohol dehydrogenase, flavin reductase (biliverdin reductase B), catalase and selenium-binding protein 1 [574]. All these proteins contain at least one free cysteine and the activity of some can be inhibited by alkylation or oxidation [576-579]. In the case of Band3, formation of disulfide crosslinked dimers result in the removal of the erythrocyte [441, 580]. When bound to Band3, Prx2 has been reported to undergo a structural change [406]. In a kinetic model approach, Benfeitas et al. found that the effective peroxidatic activity of Prx2 within the erythrocyte is lower than it should be according to the in vitro measurements [407]. The model revealed that Prx2 might be inhibited reversibly by a binding-dissociation balance.
All together these findings indicate that the idea that Prx2 is a thiol-protective holdase with a redox state dependent activity is attractive (Figure 6.2). Experiments to validate this hypothesis could include using a Prx2 KO mouse model, with erythrocytes being oxidatively stressed and the activities of flavin reductase and alcohol dehydrogenase being measured. It would also be of interest to test if the addition of cysteine to haemin-linked agarose beads interferes with Prx2 binding, and if that interference is dependent on Prx2 redox state. Another way to analyse interactions would be to pull down any of the mentioned thiol- proteins with beads crosslinked to Prx2 and to determine whether the interaction depends on Prx2 redox state. A completely different question that should be further explored is how the observed structural change of Prx2 when it binds to Band3 occurs, and if the C-terminal arm Chapter 6: General discussion 120
of Prx2 is involved. Crystallography of the cytosolic domain of Band3 with bound Prx2 could provide insight into these structural properties.
Figure 6.2 Hypothesis of oxidised peroxiredoxin 2 as holdase. Independent of its redox state, the Prx2 decamer in erythrocytes could function as a holdase, with its redox state determining its binding partners. See text for details.
6.2.5 Peroxiredoxin 2 and Gardos channel activation
Intriguingly, Prx2 needs to be present for Gardos channel activation [351, 399, 426]. The Gardos channel is a calcium-dependent potassium channel, which is responsible for erythrocyte dehydration in sickle cell disease [430], and eryptosis [581-583]. The Gardos channel needs a micromolar concentration of calcium to become activated [428, 430]. Additionally, channel activation is dependent on the presence of Prx2 decamers bound to the membrane [351, 399], and Prx2 membrane binding increases with calcium, as I have shown in whole cells and with purified erythrocyte ghosts. While it is clear that Gardos channel activation leads to erythrocyte dehydration [430], it is unclear if this is damaging [430, 584] or protective in nature by inhibiting haemolysis [581, 585-587]. A large variety of activators exist: the anti-inflammatory interleukin 10, the vasoconstrictor endothelin 1, the neutrophil activator RANTES, platelet activating factor [588, 589] and prostaglandin induced by erythrocyte shear stress [590, 591]. This suggests that dehydration of erythrocytes by the Gardos channel is a common physiological process which adapts erythrocyte size to physiological changes within the erythrocyte environment. Alkylation has been found to Chapter 6: General discussion 121
increase [427], as well as inhibit Gardos channel activity [425, 426] via different pathways. Since Prx2 is necessary for Gardos channel activation [351, 399, 425], it might simply depend on which protein is alkylated to cause inhibition or activation. Additionally, alkylated Prx2 was observed to still bind to the membrane, but it failed to activate the Gardos channel [425]. However, it has been observed that Gardos channel activation also occurs during conditions of increased oxidative stress, like depletion of GSH and production of metHb from accumulated sickle-Hb in sickle cells [592, 593], or with the oxidant generating system phenazine methosulfate (PMS) and ascorbate [594-596]. PMS and ascorbate failed to activate the Gardos channel when the medium did not contain calcium. With calcium, PMS induced cell dehydration, which was inhibited by flavonoids [594]. Given that oxidants can also cause cell swelling [594], Gardos channel activation is more complicated than just being activated by ROS. While I demonstrated that Prx2 membrane binding might be independent of redox state, I did not pursue the question if Gardos channel activation is dependent on Prx2 redox state. Since it was found that the Gardos channel contains a vital cysteine rich motif close to its pore region [434], which appears to be similar to a disulfide isomerase region [597], the necessity for the presence of Prx2 could lie in a thiol exchange function. This would link Prx2 activity to its redox state, rather than to its ability to bind to the membrane. Additionally, Hoffman et al. observed that neither purified Prx2 nor haemolysate activated the recombinant Gardos channel used in their experiments, and they explained that there is a possibility that more than Prx2 and calcium might be necessary to activate the channel [598]. To research whether Gardos channel activation is dependent on Prx2 redox state, an isolated system similar to what was used in this thesis could be applied, and channel opening could be analysed using the patch-clamp technique. Potassium efflux could be measured with closed inside-out-vesicles and either radioactive rubidium, or with a potassium electrode. The effect of oxidants on the dehydration of Prx2 KO mouse erythrocytes mentioned above could be tested with flow cytometry, as well as the Prx2 redox state and the phosphorylation state of the Gardos channel in wild type cells. Interaction of Prx2 and Gardos channel could be observed via surface plasmon resonance, or via a fluorophore and a dark quencher in combination, supressing the emission when bound together. Additionally, since Prx2 KO mouse erythrocytes still show blebbing [573], it would be of interest to see how these vesicles and erythrocytes differ from human micro- and nanovesicles, since only microvesicles contain Prx2 [534]. Chapter 6: General discussion 122
6.2.6 Peroxiredoxin 2: a major player in erythrocyte senescence?
The independence of Prx2 membrane binding from redox state and dependence on calcium, the reduction of Prx2 binding to erythrocyte membranes by haemichromes, the interaction of Prx2 and Band3, and Gardos channel activation, all indicate that Prx2 is involved in the modulation of erythrocyte metabolism, and could potentially be involved erythrocyte senescence and removal, in addition to being the main metaboliser of H2O2.
In the field of erythrocyte senescence, two major theories are discussed: the Band3 clustering model, and the eryptosis model (see [128] for a review on both models). In short, the Band3 hypothesis is based on the observation that when Band3 aggregates, naturally occurring anti- Band3-antibodies recognise these aggregates and the aged cell (or vesicle) is removed [2, 89, 169, 170]. While the complete mechanism is not yet fully comprehended, recent papers have tied Prx2 and haemichrome binding, as well as activation of protein kinases to Band3 aggregation [406, 573]. The eryptosis theory, however, is based on an increased influx of calcium [435, 599, 600], which leads to a variety of changes within the cell, like cell dehydration and vesiculation [153, 534, 601], PS exposure [582, 583, 602], and degradation by caspases and calpain [444, 562, 603, 604]. Prx2 membrane binding also increases with calcium, as I was able to show, and is involved in Gardos channel activation (see Sections 1.6.5 and 6.2.5). Therefore, I came to the hypothesis that Prx2 might represent the link between both models in erythrocyte senescence.
If Prx2 holdase function is taken into consideration, then most Prx2 could be bound to proteins, as proposed by Benfeitas et al. [407]. Prx2 could protect Band3 from aggregation [573] (Figure 6.3, 1), as well as Hb from haemichrome formation and Heinz body aggregation [302, 369] (also Dr. F. Low, unpublished results). Once the level of ROS, or the damage to Hb overwhelms the protective mechanisms and haemichromes are formed, Prx2 is replaced by haemichromes on the Band3 docking site [406] (Figure 6.3, 2). If increased calcium influx occurs at the same time, as a sign of failing cation homeostasis, the unbound Prx2 decamers could be accessible for proteolysis by calcium activated proteases [286, 406], or for interaction with the Gardos channel (see Section 6.2.5) (Figure 6.3, 3). Gardos channel activation is also dependent on calcium concentration, and this may or may not involve cleavage by a calcium-dependent protease to gain access to the necessary binding sites for Prx2 [603]. Additionally, phosphorylation of proteins is also involved in erythrocyte ageing. AMPK deficient mice developed anaemia due to increased calcium influx, PS exposure and increased erythrocyte clearance [599]. Prx2 KO mice showed increased vesiculation and Chapter 6: General discussion 123
Band3 aggregation [573], along with an increased activity of the tyrosine kinase syk, which increased phosphorylation of Band3 and the structural protein ankyrin.
Collectively this indicates that Prx2 might be an important player within erythrocyte senescence. More extensive research needs to be done to improve the knowledge on how all the various pathways interplay. Possible work should include Prx2 KO mice, as well as Band3 KO mice. Caspase and calpain activity in these KO erythrocytes could determine if Prx2 is necessary for activation. Gardos channel activity needs to be measured in Prx2 KO erythrocytes to understand how increased vesiculation is possible without Prx2 being present. In combination with the practicable experiments outlined in this chapter, the identification of how Prx2 redox state is involved in erythrocyte senescence should be achievable and will lead to a more thorough understanding of the erythrocyte metabolism.
Figure 6.3 Schematic diagram of the possible role of peroxiredoxin 2 in erythrocyte senescence. In clockwise order: (1) deoxyHb bound to Band3 autoxidises. ROS damage Band3 thiols, and Prx2 is recruited to Band3, or Prx2 protects Band3 from oxidation and aggregation. (2) Haemichromes displace Prx2 from Band3. ROS lead to aggregation and crosslinking of Band3, which is a removal signal. Band3 is unprotected from cleavage by calcium-activated calpain or caspases, ROS and phosphorylation by calcium-activated kinases. (3) Prx2 could now interact with the Gardos channel and activate it. This process might also involve cleavage, or redox-dependent thiol exchange. Chapter 6: General discussion 124
6.3 General conclusions
This study has shown that the Prx2 redox assay can be used as a biomarker for oxidative stress in sepsis and for the antioxidant status of stored erythrocytes. This study has also provided insight into the complicated mechanism of Prx2 membrane binding and the limited system which has been developed can now be used to examine other possible modulators of Prx2 membrane binding. In conclusion, this thesis has expanded our knowledge of how physiological effects impact on Prx2 redox status, and how Prx2 redox status might govern its function. However, the complex functions of Prx2 within erythrocytes remain unclear, especially in the field of erythrocyte senescence.
Chapter 7: References 125
7 References
1. Shemin, D. and D. Rittenberg, The life span of the human red blood cell. Journal of Biological Chemistry, 1946. 166(2): p. 627-636. 2. Pantaleo, A., G. Giribaldi, F. Mannu, P. Arese, and F. Turrini, Naturally occurring anti-band 3 antibodies and red blood cell removal under physiological and pathological conditions. Autoimmun Rev, 2008. 7(6): p. 457-62. 3. Halliwell, B. and J.M.C. Gutteridge, Erythrocytes, in Free Radicals in Biology and Medicine. 2007, Oxford University Press: Oxford. p. 346-350.
4. Winterbourn, C.C. and A. Stern, Human red cells scavenge extracellular hydrogen peroxide and inhibit formation of hypochlorous acid and hydroxyl radical. J Clin Invest, 1987. 80(5): p. 1486-91. 5. Richards, R.S., T.K. Roberts, N.R. McGregor, R.H. Dunstan, and H.L. Butt, The role of erythrocytes in the inactivation of free radicals. Med Hypotheses, 1998. 50(5): p. 363-7. 6. Stadtman, E.R. and R.L. Levine, Free radical-mediated oxidation of free amino acids and amino acid residues in proteins. Amino Acids, 2003. 25(3-4): p. 207-18. 7. Henle, E.S. and S. Linn, Formation, prevention, and repair of DNA damage by iron/hydrogen peroxide. J Biol Chem, 1997. 272(31): p. 19095-8. 8. Halliwell, B. and J.M.C. Gutteridge, The chemistry of free radicals and related 'reactive species', in Free Radicals in Biology and Medicine. 2007, Oxford University Press: Oxford. p. 30-78. 9. Sies, H., Oxidative stress: oxidants and antioxidants. Exp Physiol, 1997. 82(2): p. 291-5. 10. Jones, D.P., Redefining oxidative stress. Antioxidants & redox signaling, 2006. 8(9- 10): p. 1865-79. 11. Nicotera, P., J.M. Dypbukt, A.D. Rossi, L. Manzo, and S. Orrenius, Thiol modification and cell signalling in chemical toxicity. Toxicol Lett, 1992. 64-65 Spec No: p. 563-7. 12. Barouki, R. and Y. Morel, Repression of cytochrome P450 1A1 gene expression by oxidative stress: mechanisms and biological implications. Biochem Pharmacol, 2001. 61(5): p. 511-6.
13. Mikalsen, S.O. and T. Sanner, Increased gap junctional intercellular communication in Syrian hamster embryo cells treated with oxidative agents. Carcinogenesis, 1994. 15(2): p. 381-7. 14. Chen, Q.M., J.C. Bartholomew, J. Campisi, M. Acosta, J.D. Reagan, and B.N. Ames, Molecular analysis of H2O2-induced senescent-like growth arrest in normal human fibroblasts: p53 and Rb control G1 arrest but not cell replication. Biochem J, 1998. 332 ( Pt 1): p. 43-50. 15. Chandra, J., A. Samali, and S. Orrenius, Triggering and modulation of apoptosis by oxidative stress. Free Radical Biology and Medicine, 2000. 29(3–4): p. 323-333. Chapter 7: References 126
16. Buttke, T.M. and P.A. Sandstrom, Oxidative stress as a mediator of apoptosis. Immunology Today, 1994. 15(1): p. 7-10. 17. Higuchi, Y., Chromosomal DNA fragmentation in apoptosis and necrosis induced by oxidative stress. Biochem Pharmacol, 2003. 66(8): p. 1527-1535. 18. Galluzzi, L. and G. Kroemer, Necroptosis: A Specialized Pathway of Programmed Necrosis. Cell, 2008. 135(7): p. 1161-1163. 19. Martindale, J.L. and N.J. Holbrook, Cellular response to oxidative stress: Signaling for suicide and survival*. J Cell Physiol, 2002. 192(1): p. 1-15. 20. Poli, G. and M. Parola, Oxidative damage and fibrogenesis. Free Radic Biol Med, 1997. 22(1-2): p. 287-305. 21. Campian, J.L., M. Qian, X. Gao, and J.W. Eaton, Oxygen tolerance and coupling of mitochondrial electron transport. J Biol Chem, 2004. 279(45): p. 46580-7.
22. Gille, J.J. and H. Joenje, Chromosomal instability and progressive loss of chromosomes in HeLa cells during adaptation to hyperoxic growth conditions. Mutat Res, 1989. 219(4): p. 225-30. 23. Richards, R.S., T.K. Roberts, D. Mathers, R.H. Dunstan, N.R. McGregor, and H.L. Butt, Investigation of Erythrocyte Oxidative Damage in Rheumatoid Arthritis and Chronic Fatigue Syndrome. Journal of Chronic Fatigue Syndrome, 2000. 6(1): p. 37- 46. 24. Mohanty, J.G., E. Nagababu, and J.M. Rifkind, Red blood cell oxidative stress impairs oxygen delivery and induces red blood cell aging. Front Physiol, 2014. 5: p. 84. 25. Stocks, J. and T.L. Dormandy, The autoxidation of human red cell lipids induced by hydrogen peroxide. British journal of haematology, 1971. 20(1): p. 95-111.
26. Solans, R., C. Motta, R. Sola, A.E. La Ville, J. Lima, P. Simeon, N. Montella, L. Armadans-Gil, et al., Abnormalities of erythrocyte membrane fluidity, lipid composition, and lipid peroxidation in systemic sclerosis: evidence of free radical- mediated injury. Arthritis Rheum, 2000. 43(4): p. 894-900. 27. Memisogullari, R., S. Taysi, E. Bakan, and I. Capoglu, Antioxidant status and lipid peroxidation in type II diabetes mellitus. Cell Biochem Funct, 2003. 21(3): p. 291-6. 28. Hess, J.R., Red cell changes during storage. Transfus Apher Sci. 43(1): p. 51-9.
29. Weed, R.I., C.F. Reed, and G. Berg, Is hemoglobin an essential structural component of human erythrocyte membranes? J Clin Invest, 1963. 42: p. 581-8. 30. Hsia, C.C., Respiratory function of hemoglobin. N Engl J Med, 1998. 338(4): p. 239- 47.
31. Jia, L., C. Bonaventura, J. Bonaventura, and J.S. Stamler, S-nitrosohaemoglobin: a dynamic activity of blood involved in vascular control. Nature, 1996. 380(6571): p. 221-6. 32. Stryer L, B.J., Tymoczko JL, Hemoglobin: Portrait of a Protein in Action, in Biochemistry. 1995, W.H. Freeman: New York. p. 154-168. 33. Monod, J., J. Wyman, and J.P. Changeux, On the Nature of Allosteric Transitions: A Plausible Model. Journal of molecular biology, 1965. 12: p. 88-118. Chapter 7: References 127
34. Cao, Z., J.B. Bell, J.G. Mohanty, E. Nagababu, and J.M. Rifkind, Nitrite enhances RBC hypoxic ATP synthesis and the release of ATP into the vasculature: a new mechanism for nitrite-induced vasodilation. Am J Physiol Heart Circ Physiol, 2009. 297(4): p. H1494-503. 35. Christiansen, J., C.G. Douglas, and J.S. Haldane, The absorption and dissociation of carbon dioxide by human blood. J Physiol, 1914. 48(4): p. 244-71. 36. Bohr, C.H.K.K.A., Ueber einen in biologischer Beziehung wichtigen Einfluss, den die Kohlensäurespannung des Blutes auf dessen Sauerstoffbindung übt : aus dem physiologischen Laboratorium der Universität Kopenhagen. 1904, [Leipzig]. 37. Bunn, H.F. and J.H. Jandl, Control of hemoglobin function within the red cell. N Engl J Med, 1970. 282(25): p. 1414-21. 38. Bunn, H.F. and R.W. Briehl, The interaction of 2,3-diphosphoglycerate with various human hemoglobins. J Clin Invest, 1970. 49(6): p. 1088-95.
39. Lewis, I.A., M.E. Campanella, J.L. Markley, and P.S. Low, Role of band 3 in regulating metabolic flux of red blood cells. Proc Natl Acad Sci U S A, 2009. 106(44): p. 18515-20. 40. Wallace, W.J., J.C. Maxwell, and W.S. Caughey, The mechanisms of hemoglobin autoxidation. Evidence for proton-assisted nucleophilic displacement of superoxide by anions. Biochemical and biophysical research communications, 1974. 57(4): p. 1104- 10. 41. Winterbourn, C.C., Oxidative denaturation in congenital hemolytic anemias: the unstable hemoglobins. Semin Hematol, 1990. 27(1): p. 41-50. 42. Levy, A. and J.M. Rifkind, Low-temperature formation of a distal histidine complex in hemoglobin: a probe for heme pocket flexibility. Biochemistry, 1985. 24(22): p. 6050- 4.
43. Winterbourn, C.C., Free-radical production and oxidative reactions of hemoglobin. Environ Health Perspect, 1985. 64: p. 321-30. 44. Tsuruga, M., A. Matsuoka, A. Hachimori, Y. Sugawara, and K. Shikama, The molecular mechanism of autoxidation for human oxyhemoglobin. Tilting of the distal histidine causes nonequivalent oxidation in the beta chain. J Biol Chem, 1998. 273(15): p. 8607-15.
45. Rifkind, J.M., S. Ramasamy, P.T. Manoharan, E. Nagababu, and J.G. Mohanty, Redox reactions of hemoglobin. Antioxidants & redox signaling, 2004. 6(3): p. 657-66. 46. Shikama, K., The Molecular Mechanism of Autoxidation for Myoglobin and Hemoglobin: A Venerable Puzzle. Chem Rev, 1998. 98(4): p. 1357-1374.
47. Kiese, M., Methemoglobinemia: a Comprehensive Treatise: Causes, Consequences, and Correction of Increased Contents of Ferrihemoglobin in Blood. 1974: CRC Press Cleveland, Ohio. 48. Abugo, O.O. and J.M. Rifkind, Oxidation of hemoglobin and the enhancement produced by nitroblue tetrazolium. J Biol Chem, 1994. 269(40): p. 24845-53. 49. Halliwell, B. and J.M.C. Gutteridge, Superoxide radical, in Free Radicals in Biology and Medicine. 2007, Oxford University Press: Oxford. p. 46-50. Chapter 7: References 128
50. Giorgio, M., M. Trinei, E. Migliaccio, and P.G. Pelicci, Hydrogen peroxide: a metabolic by-product or a common mediator of ageing signals? Nat Rev Mol Cell Biol, 2007. 8(9): p. 722-8. 51. Bienert, G.P. and F. Chaumont, Aquaporin-facilitated transmembrane diffusion of hydrogen peroxide. Biochimica et biophysica acta, 2014. 1840(5): p. 1596-604. 52. Vieceli Dalla Sega, F., L. Zambonin, D. Fiorentini, B. Rizzo, C. Caliceti, L. Landi, S. Hrelia, and C. Prata, Specific aquaporins facilitate Nox-produced hydrogen peroxide transport through plasma membrane in leukaemia cells. Biochimica et biophysica acta, 2014. 1843(4): p. 806-14. 53. Rhee, S.G., T.S. Chang, Y.S. Bae, S.R. Lee, and S.W. Kang, Cellular regulation by hydrogen peroxide. Journal of the American Society of Nephrology : JASN, 2003. 14(8 Suppl 3): p. S211-5. 54. Halliwell, B. and J.M.C. Gutteridge, Hydrogen peroxide, in Free Radicals in Biology and Medicine. 2007, Oxford University Press: Oxford. p. 66-68. 55. Rhee, S.G., Cell signaling. H2O2, a necessary evil for cell signaling. Science, 2006. 312(5782): p. 1882-3. 56. Finkel, T., Oxidant signals and oxidative stress. Curr Opin Cell Biol, 2003. 15(2): p. 247-54. 57. Everse, J. and N. Hsia, The toxicities of native and modified hemoglobins. Free Radic Biol Med, 1997. 22(6): p. 1075-99. 58. Newman, E.S., C.A. Rice-Evans, and M.J. Davies, Identification of initiating agents in myoglobin-induced lipid peroxidation. Biochemical and biophysical research communications, 1991. 179(3): p. 1414-9. 59. Nagababu, E. and J.M. Rifkind, Reaction of hydrogen peroxide with ferrylhemoglobin: superoxide production and heme degradation. Biochemistry, 2000. 39(40): p. 12503-11. 60. Nagababu, E. and J.M. Rifkind, Formation of fluorescent heme degradation products during the oxidation of hemoglobin by hydrogen peroxide. Biochemical and biophysical research communications, 1998. 247(3): p. 592-6. 61. Wardman, P. and L.P. Candeias, Fenton chemistry: an introduction. Radiat Res, 1996. 145(5): p. 523-31.
62. Rifkind, J.M., O. Abugo, A. Levy, and J. Heim, Detection, formation, and relevance of hemichromes and hemochromes. Methods Enzymol, 1994. 231: p. 449-80. 63. Rachmilewitz, E.A., Formation of hemichromes from oxidized hemoglobin subunits. Annals of the New York Academy of Sciences, 1969. 165(1): p. 171-84.
64. Rachmilewitz, E.A., J. Peisach, and W.E. Blumberg, Studies on the stability of oxyhemoglobin A and its constituent chains and their derivatives. J Biol Chem, 1971. 246(10): p. 3356-66. 65. Rachmilewitz, E.A., Denaturation of the normal and abnormal hemoglobin molecule. Semin Hematol, 1974. 11(4): p. 441-62. 66. MacDonald, V.W. and S. Charache, Drug-induced oxidation and precipitation of hemoglobins A, S and C. Biochimica et biophysica acta, 1982. 701(1): p. 39-44. Chapter 7: References 129
67. Winterbourn, C.C. and R.W. Carrell, Studies of hemoglobin denaturation and Heinz body formation in the unstable hemoglobins. J Clin Invest, 1974. 54(3): p. 678-89. 68. Low, P.S., S.M. Waugh, K. Zinke, and D. Drenckhahn, The role of hemoglobin denaturation and band 3 clustering in red blood cell aging. Science, 1985. 227(4686): p. 531-3. 69. Waugh, S.M. and P.S. Low, Hemichrome binding to band 3: nucleation of Heinz bodies on the erythrocyte membrane. Biochemistry, 1985. 24(1): p. 34-9. 70. Waugh, S.M., B.M. Willardson, R. Kannan, R.J. Labotka, and P.S. Low, Heinz bodies induce clustering of band 3, glycophorin, and ankyrin in sickle cell erythrocytes. J Clin Invest, 1986. 78(5): p. 1155-60. 71. Balagopalakrishna, C., P.T. Manoharan, O.O. Abugo, and J.M. Rifkind, Production of superoxide from hemoglobin-bound oxygen under hypoxic conditions. Biochemistry, 1996. 35(20): p. 6393-8.
72. Chetrite, G. and R. Cassoly, Affinity of hemoglobin for the cytoplasmic fragment of human erythrocyte membrane band 3. Equilibrium measurements at physiological pH using matrix-bound proteins: the effects of ionic strength, deoxygenation and of 2,3- diphosphoglycerate. Journal of molecular biology, 1985. 185(3): p. 639-44. 73. Walder, J.A., R. Chatterjee, T.L. Steck, P.S. Low, G.F. Musso, E.T. Kaiser, P.H. Rogers, and A. Arnone, The interaction of hemoglobin with the cytoplasmic domain of band 3 of the human erythrocyte membrane. J Biol Chem, 1984. 259(16): p. 10238-46. 74. Piomelli, S., G. Lurinsky, and L.R. Wasserman, The mechanism of red cell aging. I. Relationship between cell age and specific gravity evaluated by ultracentrifugation in a discontinuous density gradient. J Lab Clin Med, 1967. 69(4): p. 659-74. 75. Mannu, F., P. Arese, M.D. Cappellini, G. Fiorelli, M. Cappadoro, G. Giribaldi, and F. Turrini, Role of hemichrome binding to erythrocyte membrane in the generation of band-3 alterations in beta-thalassemia intermedia erythrocytes. Blood, 1995. 86(5): p. 2014-20. 76. Linderkamp, O. and H.J. Meiselman, Geometric, osmotic, and membrane mechanical properties of density-separated human red cells. Blood, 1982. 59(6): p. 1121-7. 77. Kuross, S.A., B.H. Rank, and R.P. Hebbel, Excess heme in sickle erythrocyte inside- out membranes: possible role in thiol oxidation. Blood, 1988. 71(4): p. 876-82.
78. Bamm, V.V., V.A. Tsemakhovich, and N. Shaklai, Oxidation of low-density lipoprotein by hemoglobin-hemichrome. The international journal of biochemistry & cell biology, 2003. 35(3): p. 349-58. 79. Carrell, R.W., C.C. Winterbourn, and E.A. Rachmilewitz, Activated oxygen and haemolysis. British journal of haematology, 1975. 30(3): p. 259-64.
80. Chiu, D., F. Kuypers, and B. Lubin, Lipid peroxidation in human red cells. Semin Hematol, 1989. 26(4): p. 257-76. 81. Grinshtein, N., V.V. Bamm, V.A. Tsemakhovich, and N. Shaklai, Mechanism of low- density lipoprotein oxidation by hemoglobin-derived iron. Biochemistry, 2003. 42(23): p. 6977-85. 82. Turrini, F., G. Giribaldi, F. Carta, F. Mannu, and P. Arese, Mechanisms of band 3 oxidation and clustering in the phagocytosis of Plasmodium falciparum-infected erythrocytes. Redox Rep, 2003. 8(5): p. 300-3. Chapter 7: References 130
83. Kannan, R., R. Labotka, and P.S. Low, Isolation and characterization of the hemichrome-stabilized membrane protein aggregates from sickle erythrocytes. Major site of autologous antibody binding. J Biol Chem, 1988. 263(27): p. 13766-73. 84. Kannan, R., J. Yuan, and P.S. Low, Isolation and partial characterization of antibody- and globin-enriched complexes from membranes of dense human erythrocytes. Biochem J, 1991. 278 ( Pt 1): p. 57-62. 85. Antonelou, M.H., A.G. Kriebardis, K.E. Stamoulis, E. Economou-Petersen, L.H. Margaritis, and I.S. Papassideri, Red blood cell aging markers during storage in citrate-phosphate-dextrose-saline-adenine-glucose-mannitol. Transfusion, 2010. 50(2): p. 376-89. 86. Kuypers, F.A., Membrane lipid alterations in hemoglobinopathies. Hematology / the Education Program of the American Society of Hematology. American Society of Hematology. Education Program, 2007: p. 68-73.
87. Yuan, J., R. Kannan, E. Shinar, E.A. Rachmilewitz, and P.S. Low, Isolation, characterization, and immunoprecipitation studies of immune complexes from membranes of beta-thalassemic erythrocytes. Blood, 1992. 79(11): p. 3007-13. 88. Malowany, J.I. and J. Butany, Pathology of sickle cell disease. Seminars in Diagnostic Pathology, 2012. 29(1): p. 49-55. 89. Arese, P., F. Turrini, and E. Schwarzer, Band 3/complement-mediated recognition and removal of normally senescent and pathological human erythrocytes. Cell Physiol Biochem, 2005. 16(4-6): p. 133-46. 90. Fairbanks, V.F., Introduction to discussion of Heinz body anemias--unstable hemoglobinopathies. Exp Eye Res, 1971. 11(3): p. 365-75. 91. Jacob, H.S., S.H. Ingbar, and J.H. Jandl, Oxidative Hemolysis and Erythrocyte Metabolism in Hereditary Acatalasia. J Clin Invest, 1965. 44: p. 1187-99.
92. Fazi, A., U. Mancini, E. Piatti, A. Accorsi, and M. Magnani, Human red blood cells as bioreactors for the inactivation of harmful xenobiotics. Biotechnol Appl Biochem, 1991. 14(1): p. 60-8. 93. Dittrich, P., Ueber methämoglobinbildende Gifte. Archiv für experimentelle Pathologie und Pharmakologie, 1891. 29(3-4): p. 247-281. 94. Chevion, M., T. Navok, G. Glaser, and J. Mager, The chemistry of favism-inducing compounds. The properties of isouramil and divicine and their reaction with glutathione. Eur J Biochem, 1982. 127(2): p. 405-9. 95. De Flora, A., U. Benatti, L. Guida, G. Forteleoni, and T. Meloni, Favism: disordered erythrocyte calcium homeostasis. Blood, 1985. 66(2): p. 294-7.
96. Morelli, A., M. Grasso, T. Meloni, G. Forteleoni, E. Zocchi, and A. De Flora, Favism: impairment of proteolytic systems in red blood cells. Blood, 1987. 69(6): p. 1753-8. 97. Winterbourn, C.C., U. Benatti, and A. De Flora, Contributions of superoxide, hydrogen peroxide, and transition metal ions to auto-oxidation of the favism-inducing pyrimidine aglycone, divicine, and its reactions with haemoglobin. Biochem Pharmacol, 1986. 35(12): p. 2009-15. 98. Mansouri, A. and A.A. Lurie, Concise review: methemoglobinemia. American journal of hematology, 1993. 42(1): p. 7-12. Chapter 7: References 131
99. Ash-Bernal, R., R. Wise, and S.M. Wright, Acquired methemoglobinemia: a retrospective series of 138 cases at 2 teaching hospitals. Medicine (Baltimore), 2004. 83(5): p. 265-73. 100. McMillan, D.C., C.L. Powell, Z.S. Bowman, J.D. Morrow, and D.J. Jollow, Lipids versus proteins as major targets of pro-oxidant, direct-acting hemolytic agents. Toxicol Sci, 2005. 88(1): p. 274-83. 101. Geffner, M.E., D.R. Powars, and W.T. Choctaw, Acquired methemoglobinemia. West J Med, 1981. 134(1): p. 7-10. 102. Munday, R., Toxicity of thiols and disulphides: involvement of free-radical species. Free Radic Biol Med, 1989. 7(6): p. 659-73. 103. Jaffe, E.R., Methaemoglobinaemia. Clin Haematol, 1981. 10(1): p. 99-122. 104. Percy, M.J., N.V. McFerran, and T.R. Lappin, Disorders of oxidised haemoglobin. Blood Rev, 2005. 19(2): p. 61-8. 105. Cheah, F.C., A.V. Peskin, F.L. Wong, A. Ithnin, A. Othman, and C.C. Winterbourn, Increased basal oxidation of peroxiredoxin 2 and limited peroxiredoxin recycling in glucose-6-phosphate dehydrogenase-deficient erythrocytes from newborn infants. FASEB J, 2014. 28(7): p. 3205-10. 106. Meloni, T., G. Forteleoni, A. Dore, and S. Cutillo, Favism and hemolytic anemia in glucose-6-phosphate dehydrogenase-deficient subjects in North Sardinia. Acta Haematol, 1983. 70(2): p. 83-90. 107. Han, J.K., S.H. Kim, B.I. Choi, K.M. Yeon, and M.C. Han, Tuberculous colitis. Findings at double-contrast barium enema examination. Diseases of the colon and rectum, 1996. 39(11): p. 1204-9. 108. Hod, E.A. and S.L. Spitalnik, Stored red blood cell transfusions: Iron, inflammation, immunity, and infection. Transfusion clinique et biologique : journal de la Societe francaise de transfusion sanguine, 2012. 19(3): p. 84-9. 109. Holme, S., Current issues related to the quality of stored RBCs. Transfus Apher Sci, 2005. 33(1): p. 55-61. 110. Bennett-Guerrero, E., T.H. Veldman, A. Doctor, M.J. Telen, T.L. Ortel, T.S. Reid, M.A. Mulherin, H. Zhu, et al., Evolution of adverse changes in stored RBCs. Proc Natl Acad Sci U S A, 2007. 104(43): p. 17063-8.
111. Valeri, C.R. and N.M. Hirsch, Restoration in vivo of erythrocyte adenosine triphosphate, 2,3-diphosphoglycerate, potassium ion, and sodium ion concentrations following the transfusion of acid-citrate-dextrose-stored human red blood cells. J Lab Clin Med, 1969. 73(5): p. 722-33.
112. Raat, N.J., A.J. Verhoeven, E.G. Mik, C.W. Gouwerok, R. Verhaar, P.T. Goedhart, D. de Korte, and C. Ince, The effect of storage time of human red cells on intestinal microcirculatory oxygenation in a rat isovolemic exchange model. Crit Care Med, 2005. 33(1): p. 39-45; discussion 238-9. 113. Raat, N.J., P.M. Hilarius, T. Johannes, D. de Korte, C. Ince, and A.J. Verhoeven, Rejuvenation of stored human red blood cells reverses the renal microvascular oxygenation deficit in an isovolemic transfusion model in rats. Transfusion, 2009. 49(3): p. 427-4. Chapter 7: References 132
114. Tinmouth, A., D. Fergusson, I.C. Yee, and P.C. Hebert, Clinical consequences of red cell storage in the critically ill. Transfusion, 2006. 46(11): p. 2014-27. 115. D'Alessandro, A., G.M. D'Amici, S. Vaglio, and L. Zolla, Time-course investigation of SAGM-stored leukocyte-filtered red bood cell concentrates: from metabolism to proteomics. Haematologica, 2012. 97(1): p. 107-15. 116. Dern, R.J., G.J. Brewer, and J.J. Wiorkowski, Studies on the preservation of human blood. II. The relationship of erythrocyte adenosine triphosphate levels and other in vitro measures to red cell storageability. J Lab Clin Med, 1967. 69(6): p. 968-78. 117. D'Amici, G.M., S. Rinalducci, and L. Zolla, Proteomic analysis of RBC membrane protein degradation during blood storage. J Proteome Res, 2007. 6(8): p. 3242-55. 118. Knight, J.A., R.P. Voorhees, L. Martin, and H. Anstall, Lipid peroxidation in stored red cells. Transfusion, 1992. 32(4): p. 354-7.
119. Wagner, G.M., D.T. Chiu, J.H. Qju, R.H. Heath, and B.H. Lubin, Spectrin oxidation correlates with membrane vesiculation in stored RBCs. Blood, 1987. 69(6): p. 1777- 81. 120. Lockwood, W.B., R.W. Hudgens, I.O. Szymanski, R.A. Teno, and A.D. Gray, Effects of rejuvenation and frozen storage on 42-day-old AS-3 RBCs. Transfusion, 2003. 43(11): p. 1527-32. 121. Meyer, E.K., D.F. Dumont, S. Baker, and L.J. Dumont, Rejuvenation capacity of red blood cells in additive solutions over long-term storage. Transfusion, 2011. 51(7): p. 1574-9. 122. Yoshida, T., J.P. AuBuchon, L.J. Dumont, J.D. Gorham, S.C. Gifford, K.Y. Foster, and M.W. Bitensky, The effects of additive solution pH and metabolic rejuvenation on anaerobic storage of red cells. Transfusion, 2008. 48(10): p. 2096-105.
123. Chaudhary, R. and R. Katharia, Oxidative injury as contributory factor for red cells storage lesion during twenty eight days of storage. Blood Transfus, 2012. 10(1): p. 59- 62. 124. Kriebardis, A.G., M.H. Antonelou, K.E. Stamoulis, E. Economou-Petersen, L.H. Margaritis, and I.S. Papassideri, Progressive oxidation of cytoskeletal proteins and accumulation of denatured hemoglobin in stored red cells. J Cell Mol Med, 2007. 11(1): p. 148-55.
125. Dumaswala, U.J., M.J. Wilson, Y.L. Wu, J. Wykle, L. Zhuo, L.M. Douglass, and D.L. Daleke, Glutathione loading prevents free radical injury in red blood cells after storage. Free Radic Res, 2000. 33(5): p. 517-29. 126. Dumaswala, U.J., L. Zhuo, D.W. Jacobsen, S.K. Jain, and K.A. Sukalski, Protein and lipid oxidation of banked human erythrocytes: role of glutathione. Free Radic Biol Med, 1999. 27(9-10): p. 1041-9. 127. Pallotta, V., F. Gevi, A. D'Alessandro, and L. Zolla, Storing red blood cells with vitamin C and N-acetylcysteine prevents oxidative stress-related lesions: a metabolomics overview. Blood Transfus, 2014. 12(3): p. 376-87. 128. Lion, N., D. Crettaz, O. Rubin, and J.D. Tissot, Stored red blood cells: a changing universe waiting for its map(s). J Proteomics, 2010. 73(3): p. 374-85. Chapter 7: References 133
129. Bosman, G.J., J.M. Werre, F.L. Willekens, and V.M. Novotny, Erythrocyte ageing in vivo and in vitro: structural aspects and implications for transfusion. Transfus Med, 2008. 18(6): p. 335-47. 130. Kanias, T. and J.P. Acker, Biopreservation of red blood cells--the struggle with hemoglobin oxidation. FEBS J, 2010. 277(2): p. 343-56. 131. Roback, J.D., C.D. Josephson, E.K. Waller, J.L. Newman, S. Karatela, K. Uppal, D.P. Jones, J.C. Zimring, et al., Metabolomics of ADSOL (AS-1) Red Blood Cell Storage. Transfus Med Rev, 2014. 28(2): p. 41-55. 132. Yoshida, T., J.P. AuBuchon, L. Tryzelaar, K.Y. Foster, and M.W. Bitensky, Extended storage of red blood cells under anaerobic conditions. Vox Sang, 2007. 92(1): p. 22- 31. 133. Hogman, C.F., C.H. de Verdier, A. Ericson, K. Hedlund, and B. Sandhagen, Studies on the mechanism of human red cell loss of viability during storage at +4 degrees C in vitro. I. Cell shape and total adenylate concentration as determinant factors for posttransfusion survival. Vox Sang, 1985. 48(5): p. 257-68. 134. Whillier, S., J.E. Raftos, R.L. Sparrow, and P.W. Kuchel, The effects of long-term storage of human red blood cells on the glutathione synthesis rate and steady-state concentration. Transfusion, 2011. 51(7): p. 1450-9. 135. Hess, J.R., Storage of red blood cells under anaerobic conditions. Vox Sang, 2007. 93(2): p. 183; author reply 184. 136. Hess, J.R., N. Rugg, A.D. Knapp, J.F. Gormas, E.B. Silberstein, and T.J. Greenwalt, Successful storage of RBCs for 9 weeks in a new additive solution. Transfusion, 2000. 40(8): p. 1007-11. 137. Kim-Shapiro, D.B., J. Lee, and M.T. Gladwin, Storage lesion: role of red blood cell breakdown. Transfusion, 2011. 51(4): p. 844-51.
138. Messana, I., L. Ferroni, F. Misiti, G. Girelli, S. Pupella, M. Castagnola, B. Zappacosta, and B. Giardina, Blood bank conditions and RBCs: the progressive loss of metabolic modulation. Transfusion, 2000. 40(3): p. 353-60. 139. Gevi, F., A. D'Alessandro, S. Rinalducci, and L. Zolla, Alterations of red blood cell metabolome during cold liquid storage of erythrocyte concentrates in CPD-SAGM. J Proteomics, 2012.
140. Kurup, P.A., P. Arun, N.S. Gayathri, C.R. Dhanya, and A.R. Indu, Modified formulation of CPDA for storage of whole blood, and of SAGM for storage of red blood cells, to maintain the concentration of 2,3-diphosphoglycerate. Vox Sang, 2003. 85(4): p. 253-61. 141. Hess, J.R. and T.G. Greenwalt, Storage of red blood cells: new approaches. Transfus Med Rev, 2002. 16(4): p. 283-95. 142. Arun, P., K.G. Padmakumaran Nair, V. Manojkumar, K.V. Deepadevi, L.R. Lakshmi, and P.A. Kurup, Decreased hemolysis and lipid peroxidation in blood during storage in the presence of nicotinic acid. Vox Sang, 1999. 76(4): p. 220-5. 143. Rose, Z.B., Enzymes controlling 2,3-diphosphoglycerate in human erythrocytes. Fed Proc, 1970. 29(3): p. 1105-11. 144. Almac, E. and C. Ince, The impact of storage on red cell function in blood transfusion. Best Pract Res Clin Anaesthesiol, 2007. 21(2): p. 195-208. Chapter 7: References 134
145. Wolfe, L., The red cell membrane and the storage lesion. Clin Haematol, 1985. 14(1): p. 259-76. 146. Romero, P.J. and E.A. Romero, Effect of cell ageing on Ca2+ influx into human red cells. Cell Calcium, 1999. 26(3-4): p. 131-7. 147. Romero, P.J., V. Salas, and C. Hernandez, Calcium pump phosphoenzyme from young and old human red cells. Cell Biol Int, 2002. 26(11): p. 945-9. 148. CARDIAC arrest from potassium given in blood transfusions. Br Med J, 1960. 2(5198): p. 586-7. 149. Card, R.T., N. Mohandas, and P.L. Mollison, Relationship of post-transfusion viability to deformability of stored red cells. British journal of haematology, 1983. 53(2): p. 237-40. 150. Izzo, P., A. Manicone, A. Spagnuolo, V.M. Lauta, A. Di Pasquale, and D. Di Monte, Erythrocytes stored in CPD SAG-mannitol: evaluation of their deformability. Clin Hemorheol Microcirc, 1999. 21(3-4): p. 335-9. 151. Haradin, A.R., R.I. Weed, and C.F. Reed, Changes in physical properties of stored erythrocytes relationship to survival in vivo. Transfusion, 1969. 9(5): p. 229-37. 152. Berezina, T.L., S.B. Zaets, C. Morgan, C.R. Spillert, M. Kamiyama, Z. Spolarics, E.A. Deitch, and G.W. Machiedo, Influence of storage on red blood cell rheological properties. The Journal of surgical research, 2002. 102(1): p. 6-12. 153. Laczko, J., M. Szabolcs, and I. Jona, Vesicle release from erythrocytes during storage and failure of rejuvenation to restore cell morphology. Haematologia (Budap), 1985. 18(4): p. 233-48. 154. Willekens, F.L., J.M. Werre, Y.A. Groenen-Dopp, B. Roerdinkholder-Stoelwinder, B. de Pauw, and G.J. Bosman, Erythrocyte vesiculation: a self-protective mechanism? British journal of haematology, 2008. 141(4): p. 549-56. 155. Greenwalt, T.J., C.G. McGuinness, and U.J. Dumaswala, Studies in red blood cell preservation: 4. Plasma vesicle hemoglobin exceeds free hemoglobin. Vox Sang, 1991. 61(1): p. 14-7. 156. Bosman, G.J., E. Lasonder, M. Luten, B. Roerdinkholder-Stoelwinder, V.M. Novotny, H. Bos, and W.J. De Grip, The proteome of red cell membranes and vesicles during storage in blood bank conditions. Transfusion, 2008. 48(5): p. 827-35.
157. Greenwalt, T.J., D.J. Bryan, and U.J. Dumaswala, Erythrocyte membrane vesiculation and changes in membrane composition during storage in citrate-phosphate-dextrose- adenine-1. Vox Sang, 1984. 47(4): p. 261-70. 158. Cole, W.F., M.G. Rumsby, G.H. Longster, and L.A. Tovey, Changes in the inhibition of specific agglutination by plasma due to microvesicles released from human red cells during storage for transfusion. Vox Sang, 1979. 37(2): p. 73-7. 159. Rumsby, M.G., J. Trotter, D. Allan, and R.H. Michell, Recovery of membrane micro- vesicles from human erythrocytes stored for transfusion: a mechanism for the erythrocyte discocyte-to-spherocyte shape transformation. Biochem Soc Trans, 1977. 5(1): p. 126-8. 160. Rumsby, M.G., Y.W. Barnes, D.G. Hoare, and L.A. Tovey, The lecithin: cholesterol acyltransferase reaction, lysolecithin and red cell ageing in blood stored under normal transfusion service conditions. Vox Sang, 1976. 30(1): p. 23-35. Chapter 7: References 135
161. Dumaswala, U.J., M.J. Wilson, T. Jose, and D.L. Daleke, Effect of a glycerol- containing hypotonic medium on erythrocyte phospholipid asymmetry and aminophospholipid transport during storage. Biochimica et biophysica acta, 1997. 1330(2): p. 265-73. 162. Cardo, L.J., D. Wilder, and J. Salata, Neutrophil priming, caused by cell membranes and microvesicles in packed red blood cell units, is abrogated by leukocyte depletion at collection. Transfus Apher Sci, 2008. 38(2): p. 117-25. 163. Willekens, F.L., J.M. Werre, J.K. Kruijt, B. Roerdinkholder-Stoelwinder, Y.A. Groenen-Dopp, A.G. van den Bos, G.J. Bosman, and T.J. van Berkel, Liver Kupffer cells rapidly remove red blood cell-derived vesicles from the circulation by scavenger receptors. Blood, 2005. 105(5): p. 2141-5. 164. Dumaswala, U.J. and T.J. Greenwalt, Human erythrocytes shed exocytic vesicles in vivo. Transfusion, 1984. 24(6): p. 490-2.
165. Greenwalt, T.J. and U.J. Dumaswala, Effect of red cell age on vesiculation in vitro. British journal of haematology, 1988. 68(4): p. 465-7. 166. in Standards for blood bank transfusion services, T. Price, Editor. 2009, American Association of Blood Banks: Bethesda (MD). p. 25. 167. Karon, B.S., J.D. Hoyer, J.R. Stubbs, and D.D. Thomas, Changes in Band 3 oligomeric state precede cell membrane phospholipid loss during blood bank storage of red blood cells. Transfusion, 2009. 49(7): p. 1435-42. 168. Lutz, H.U., Naturally occurring anti-band 3 antibodies. Transfus Med Rev, 1992. 6(3): p. 201-11. 169. Lutz, H.U., F. Bussolino, R. Flepp, S. Fasler, P. Stammler, M.D. Kazatchkine, and P. Arese, Naturally occurring anti-band-3 antibodies and complement together mediate phagocytosis of oxidatively stressed human erythrocytes. Proc Natl Acad Sci U S A, 1987. 84(21): p. 7368-72. 170. Turrini, F., F. Mannu, M. Cappadoro, D. Ulliers, G. Giribaldi, and P. Arese, Binding of naturally occurring antibodies to oxidatively and nonoxidatively modified erythrocyte band 3. Biochimica et biophysica acta, 1994. 1190(2): p. 297-303. 171. D'Alessandro, A., A.G. Kriebardis, S. Rinalducci, M.H. Antonelou, K.C. Hansen, I.S. Papassideri, and L. Zolla, An update on red blood cell storage lesions, as gleaned through biochemistry and omics technologies. Transfusion, 2014.
172. Dumaswala, U.J., L. Zhuo, S. Mahajan, P.N. Nair, H.G. Shertzer, P. Dibello, and D.W. Jacobsen, Glutathione protects chemokine-scavenging and antioxidative defense functions in human RBCs. Am J Physiol Cell Physiol, 2001. 280(4): p. C867-73. 173. Brown, W.D. and L.B. Mebine, Autoxidation of Oxymyoglobins. Journal of Biological Chemistry, 1969. 244(24): p. 6696-6701. 174. Rael, L.T., R. Bar-Or, D.R. Ambruso, C.W. Mains, D.S. Slone, M.L. Craun, and D. Bar-Or, The effect of storage on the accumulation of oxidative biomarkers in donated packed red blood cells. The Journal of trauma, 2009. 66(1): p. 76-81. 175. Kriebardis, A.G., M.H. Antonelou, K.E. Stamoulis, E. Economou-Petersen, L.H. Margaritis, and I.S. Papassideri, RBC-derived vesicles during storage: ultrastructure, protein composition, oxidation, and signaling components. Transfusion, 2008. 48(9): p. 1943-53. Chapter 7: References 136
176. Nagababu, E., J.G. Mohanty, S. Bhamidipaty, G.R. Ostera, and J.M. Rifkind, Role of the membrane in the formation of heme degradation products in red blood cells. Life sciences, 2010. 86(3-4): p. 133-8. 177. Rifkind, J.M. and E. Nagababu, Hemoglobin redox reactions and red blood cell aging. Antioxidants & redox signaling, 2013. 18(17): p. 2274-83. 178. Hogman, C.F., C.H. de Verdier, A. Ericson, K. Hedlund, and B. Sandhagen, Effects of oxygen on red cells during liquid storage at +4 degrees C. Vox Sang, 1986. 51(1): p. 27-34. 179. Messana, I., F. Misiti, S. el-Sherbini, B. Giardina, and M. Castagnola, Quantitative determination of the main glucose metabolic fluxes in human erythrocytes by 13C- and 1H-MR spectroscopy. J Biochem Biophys Methods, 1999. 39(1-2): p. 63-84. 180. Untucht-Grau, R., R.H. Schirmer, I. Schirmer, and R.L. Krauth-Siegel, Glutathione reductase from human erythrocytes: amino-acid sequence of the structurally known FAD-binding domain. Eur J Biochem, 1981. 120(2): p. 407-19. 181. Jozwik, M., M. Jozwik, M. Jozwik, M. Szczypka, J. Gajewska, and T. Laskowska- Klita, Antioxidant defence of red blood cells and plasma in stored human blood. Clin Chim Acta, 1997. 267(2): p. 129-42. 182. van 't Erve, T.J., C.M. Doskey, B.A. Wagner, J.R. Hess, B.W. Darbro, K.K. Ryckman, J.C. Murray, T.J. Raife, et al., Heritability of glutathione and related metabolites in stored red blood cells. Free Radic Biol Med, 2014. 76: p. 107-13. 183. van 't Erve, T.J., B.A. Wagner, K.K. Ryckman, T.J. Raife, and G.R. Buettner, The concentration of glutathione in human erythrocytes is a heritable trait. Free Radic Biol Med, 2013. 65: p. 742-9. 184. Riedner, C., M.U. Heim, W. Mempel, and W. Wilmanns, Possibility to improve preservation of whole blood by leukocyte-depletion before storage. Vox Sang, 1990. 59(2): p. 78-82. 185. Heaton, W.A., S. Holme, K. Smith, M.E. Brecher, A. Pineda, J.P. AuBuchon, and E. Nelson, Effects of 3-5 log10 pre-storage leucocyte depletion on red cell storage and metabolism. British journal of haematology, 1994. 87(2): p. 363-8. 186. Kriebardis, A.G., M.H. Antonelou, K.E. Stamoulis, E. Economou-Petersen, L.H. Margaritis, and I.S. Papassideri, Membrane protein carbonylation in non- leukodepleted CPDA-preserved red blood cells. Blood Cells Mol Dis, 2006. 36(2): p. 279-82. 187. Hess, J.R., H.R. Hill, C.K. Oliver, L.E. Lippert, and T.J. Greenwalt, Alkaline CPD and the preservation of RBC 2,3-DPG. Transfusion, 2002. 42(6): p. 747-52. 188. Hess, J.R., H.R. Hill, C.K. Oliver, L.E. Lippert, N. Rugg, A.D. Joines, J.F. Gormas, P.G. Pratt, et al., Twelve-week RBC storage. Transfusion, 2003. 43(7): p. 867-872. 189. Hess, J.R., N. Rugg, J.K. Gormas, A.D. Knapp, H.R. Hill, C.K. Oliver, L.E. Lippert, E.B. Silberstein, et al., RBC storage for 11 weeks. Transfusion, 2001. 41(12): p. 1586- 90. 190. Hess, J.R., N. Rugg, A.D. Knapp, J.F. Gormas, E.B. Silberstein, and T.J. Greenwalt, Successful storage of RBCs for 10 weeks in a new additive solution. Transfusion, 2000. 40(8): p. 1012-6. Chapter 7: References 137
191. Cancelas, J.A., L.J. Dumont, L.A. Maes, N. Rugg, L. Herschel, P.H. Whitley, Z.M. Szczepiokowski, A.H. Siegel, et al., Additive solution-7 reduces the red blood cell cold storage lesion. Transfusion, 2014. 192. Richards, R.S., T.K. Roberts, R.H. Dunstan, N.R. McGregor, and H.L. Butt, Erythrocyte antioxidant systems protect cultured endothelial cells against oxidant damage. Biochemistry and molecular biology international, 1998. 46(5): p. 857-65. 193. Toth, K.M., D.P. Clifford, E.M. Berger, C.W. White, and J.E. Repine, Intact human erythrocytes prevent hydrogen peroxide-mediated damage to isolated perfused rat lungs and cultured bovine pulmonary artery endothelial cells. J Clin Invest, 1984. 74(1): p. 292-5. 194. van Asbeck, B.S., J. Hoidal, G.M. Vercellotti, B.A. Schwartz, C.F. Moldow, and H.S. Jacob, Protection against lethal hyperoxia by tracheal insufflation of erythrocytes: role of red cell glutathione. Science, 1985. 227(4688): p. 756-9.
195. Aoshiba, K., Y. Nakajima, S. Yasui, J. Tamaoki, and A. Nagai, Red blood cells inhibit apoptosis of human neutrophils. Blood, 1999. 93(11): p. 4006-10. 196. Zachee, P., M.P. Emonds, R. Lins, and M. Boogaerts, Red blood cells, the scavengers for toxic free radicals during haemodialysis. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association, 1988. 3(1): p. 104-105. 197. Test, S.T. and S.J. Weiss, Quantitative and temporal characterization of the extracellular H2O2 pool generated by human neutrophils. J Biol Chem, 1984. 259(1): p. 399-405. 198. Claster, S., D.T. Chiu, A. Quintanilha, and B. Lubin, Neutrophils mediate lipid peroxidation in human red cells. Blood, 1984. 64(5): p. 1079-84. 199. Weiss, S.J., The role of superoxide in the destruction of erythrocyte targets by human neutrophils. J Biol Chem, 1980. 255(20): p. 9912-7. 200. Weiss, S.J., Neutrophil-mediated methemoglobin formation in the erythrocyte. The role of superoxide and hydrogen peroxide. J Biol Chem, 1982. 257(6): p. 2947-53. 201. Yang, B.C., W.W. Nichols, and J.L. Mehta, Cardioprotective Effects of Red Blood Cells on Ischemia and Reperfusion Injury in Isolated Rat Heart: Release of Nitric Oxide as a Potential Mechanism. J Cardiovasc Pharmacol Ther, 1996. 1(4): p. 297- 306.
202. Brown, J.M., M.A. Grosso, A. Banerjee, G.J. Whitman, J.E. Repine, and A.H. Harken, Erythrocyte catalase decreases cardiac reperfusion injury. Current surgery, 1989. 46(2): p. 122-4. 203. Brown, J.M., M.A. Grosso, L.S. Terada, C.J. Beehler, K.M. Toth, G.J. Whitman, A.H. Harken, and J.E. Repine, Erythrocytes decrease myocardial hydrogen peroxide levels and reperfusion injury. The American journal of physiology, 1989. 256(2 Pt 2): p. H584-8. 204. Perrelli, M.G., P. Pagliaro, and C. Penna, Ischemia/reperfusion injury and cardioprotective mechanisms: Role of mitochondria and reactive oxygen species. World J Cardiol, 2011. 3(6): p. 186-200. 205. Chen, Q., A.K. Camara, D.F. Stowe, C.L. Hoppel, and E.J. Lesnefsky, Modulation of electron transport protects cardiac mitochondria and decreases myocardial injury Chapter 7: References 138
during ischemia and reperfusion. Am J Physiol Cell Physiol, 2007. 292(1): p. C137- 47. 206. Fonseca, A.M., C.F. Pereira, G. Porto, and F.A. Arosa, Red blood cells upregulate cytoprotective proteins and the labile iron pool in dividing human T cells despite a reduction in oxidative stress. Free Radic Biol Med, 2003. 35(11): p. 1404-16. 207. Fonseca, A.M., C.F. Pereira, G. Porto, and F.A. Arosa, Red blood cells promote survival and cell cycle progression of human peripheral blood T cells independently of CD58/LFA-3 and heme compounds. Cellular immunology, 2003. 224(1): p. 17-28. 208. Hampton, M.B., A.J. Kettle, and C.C. Winterbourn, Inside the neutrophil phagosome: oxidants, myeloperoxidase, and bacterial killing. Blood, 1998. 92(9): p. 3007-17. 209. Hickey, M.J. and P. Kubes, Intravascular immunity: the host-pathogen encounter in blood vessels. Nat Rev Immunol, 2009. 9(5): p. 364-75.
210. Levy, O., Antimicrobial proteins and peptides: anti-infective molecules of mammalian leukocytes. Journal of leukocyte biology, 2004. 76(5): p. 909-25. 211. Halliwell, B. and J.M.C. Gutteridge, Phagocytosis, in Free Radicals in Biology and Medicine. 2007, Oxford University Press: Oxford. p. 405-420. 212. Hirsch, J.G. and Z.A. Cohn, Degranulation of polymorphonuclear leucocytes following phagocytosis of microorganisms. The Journal of experimental medicine, 1960. 112: p. 1005-14. 213. Klebanoff, S.J., Oxygen metabolism and the toxic properties of phagocytes. Ann Intern Med, 1980. 93(3): p. 480-9. 214. Robinson, J.M. and J.A. Badwey, The NADPH oxidase complex of phagocytic leukocytes: a biochemical and cytochemical view. Histochem Cell Biol, 1995. 103(3): p. 163-80.
215. Makino, R., T. Tanaka, T. Iizuka, Y. Ishimura, and S. Kanegasaki, Stoichiometric conversion of oxygen to superoxide anion during the respiratory burst in neutrophils. Direct evidence by a new method for measurement of superoxide anion with diacetyldeuteroheme-substituted horseradish peroxidase. J Biol Chem, 1986. 261(25): p. 11444-7. 216. Roos, D., C.M. Eckmann, M. Yazdanbakhsh, M.N. Hamers, and M. de Boer, Excretion of superoxide by phagocytes measured with cytochrome c entrapped in resealed erythrocyte ghosts. J Biol Chem, 1984. 259(3): p. 1770-5. 217. Babior, B.M., J.D. Lambeth, and W. Nauseef, The neutrophil NADPH oxidase. Archives of biochemistry and biophysics, 2002. 397(2): p. 342-4. 218. Winterbourn, C.C. and A.J. Kettle, Redox reactions and microbial killing in the neutrophil phagosome. Antioxidants & redox signaling, 2013. 18(6): p. 642-60. 219. Nauseef, W.M., How human neutrophils kill and degrade microbes: an integrated view. Immunol Rev, 2007. 219: p. 88-102. 220. Remer, K.A., M. Brcic, and T.W. Jungi, Toll-like receptor-4 is involved in eliciting an LPS-induced oxidative burst in neutrophils. Immunol Lett, 2003. 85(1): p. 75-80. 221. Soler-Rodriguez, A.M., H. Zhang, H.S. Lichenstein, N. Qureshi, D.W. Niesel, S.E. Crowe, J.W. Peterson, and G.R. Klimpel, Neutrophil activation by bacterial Chapter 7: References 139
lipoprotein versus lipopolysaccharide: differential requirements for serum and CD14. Journal of immunology, 2000. 164(5): p. 2674-83. 222. Troelstra, A., P. Antal-Szalmas, L.A. de Graaf-Miltenburg, A.J. Weersink, J. Verhoef, K.P. Van Kessel, and J.A. Van Strijp, Saturable CD14-dependent binding of fluorescein-labeled lipopolysaccharide to human monocytes. Infect Immun, 1997. 65(6): p. 2272-7. 223. Troelstra, A., L.A. de Graaf-Miltenburg, T. van Bommel, J. Verhoef, K.P. Van Kessel, and J.A. Van Strijp, Lipopolysaccharide-coated erythrocytes activate human neutrophils via CD14 while subsequent binding is through CD11b/CD18. Journal of immunology, 1999. 162(7): p. 4220-5. 224. Troelstra, A., B.N. Giepmans, K.P. Van Kessel, H.S. Lichenstein, J. Verhoef, and J.A. Van Strijp, Dual effects of soluble CD14 on LPS priming of neutrophils. Journal of leukocyte biology, 1997. 61(2): p. 173-8.
225. Kitchens, R.L., R.J. Ulevitch, and R.S. Munford, Lipopolysaccharide (LPS) partial structures inhibit responses to LPS in a human macrophage cell line without inhibiting LPS uptake by a CD14-mediated pathway. The Journal of experimental medicine, 1992. 176(2): p. 485-94. 226. Fuchs, T.A., U. Abed, C. Goosmann, R. Hurwitz, I. Schulze, V. Wahn, Y. Weinrauch, V. Brinkmann, et al., Novel cell death program leads to neutrophil extracellular traps. The Journal of cell biology, 2007. 176(2): p. 231-41. 227. Wartha, F., K. Beiter, S. Normark, and B. Henriques-Normark, Neutrophil extracellular traps: casting the NET over pathogenesis. Curr Opin Microbiol, 2007. 10(1): p. 52-6. 228. Akiba, Y., T. Kimura, T. Kitaoka, E. Toyoshima, S. Fujiuchi, S. Osanai, H. Nakano, Y. Ohsaki, et al., [Respiratory disorders in type-1 hereditary motor and sensory neuropathy]. Nihon Kyobu Shikkan Gakkai zasshi, 1996. 34(8): p. 850-5.
229. Parker, H., A.M. Albrett, A.J. Kettle, and C.C. Winterbourn, Myeloperoxidase associated with neutrophil extracellular traps is active and mediates bacterial killing in the presence of hydrogen peroxide. Journal of leukocyte biology, 2012. 91(3): p. 369-76. 230. Hampton, M.B., A.J. Kettle, and C.C. Winterbourn, Involvement of superoxide and myeloperoxidase in oxygen-dependent killing of Staphylococcus aureus by neutrophils. Infect Immun, 1996. 64(9): p. 3512-7.
231. Marquez, L.A. and H.B. Dunford, Kinetics of oxidation of tyrosine and dityrosine by myeloperoxidase compounds I and II. Implications for lipoprotein peroxidation studies. J Biol Chem, 1995. 270(51): p. 30434-40. 232. Winterbourn, C.C., Comparative reactivities of various biological compounds with myeloperoxidase-hydrogen peroxide-chloride, and similarity of the oxidant to hypochlorite. Biochimica et biophysica acta, 1985. 840(2): p. 204-10. 233. Low, F.M., M.B. Hampton, A.V. Peskin, and C.C. Winterbourn, Peroxiredoxin 2 functions as a noncatalytic scavenger of low-level hydrogen peroxide in the erythrocyte. Blood, 2007. 109(6): p. 2611-7. 234. Stacey, M.M., A.V. Peskin, M.C. Vissers, and C.C. Winterbourn, Chloramines and hypochlorous acid oxidize erythrocyte peroxiredoxin 2. Free Radic Biol Med, 2009. 47(10): p. 1468-76. Chapter 7: References 140
235. Peskin, A.V., F.M. Low, L.N. Paton, G.J. Maghzal, M.B. Hampton, and C.C. Winterbourn, The high reactivity of peroxiredoxin 2 with H(2)O(2) is not reflected in its reaction with other oxidants and thiol reagents. J Biol Chem, 2007. 282(16): p. 11885-92. 236. Kettle, A.J., A.M. Albrett, A.L. Chapman, N. Dickerhof, L.V. Forbes, I. Khalilova, and R. Turner, Measuring chlorine bleach in biology and medicine. Biochimica et biophysica acta, 2014. 1840(2): p. 781-93. 237. Ney, P.A., M.M. Christopher, and R.P. Hebbel, Synergistic effects of oxidation and deformation on erythrocyte monovalent cation leak. Blood, 1990. 75(5): p. 1192-8. 238. Johnston, C.S., C.G. Meyer, and J.C. Srilakshmi, Vitamin C elevates red blood cell glutathione in healthy adults. Am J Clin Nutr, 1993. 58(1): p. 103-5. 239. Rose, R.C. and A.M. Bode, Biology of free radical scavengers: an evaluation of ascorbate. FASEB J, 1993. 7(12): p. 1135-42.
240. Smirnoff, N., Ascorbic acid: metabolism and functions of a multi-facetted molecule. Curr Opin Plant Biol, 2000. 3(3): p. 229-35. 241. Traber, M.G. and H. Sies, Vitamin E in humans: demand and delivery. Annu Rev Nutr, 1996. 16: p. 321-47. 242. Li, H., H. Tu, Y. Wang, and M. Levine, Vitamin C in mouse and human red blood cells: an HPLC assay. Analytical biochemistry, 2012. 426(2): p. 109-17. 243. Bjornson, L.K., H.J. Kayden, E. Miller, and A.N. Moshell, The transport of alpha- tocopherol and beta-carotene in human blood. Journal of lipid research, 1976. 17(4): p. 343-52. 244. Fotouhi, N., M. Meydani, M.S. Santos, S.N. Meydani, C.H. Hennekens, and J.M. Gaziano, Carotenoid and tocopherol concentrations in plasma, peripheral blood mononuclear cells, and red blood cells after long-term beta-carotene supplementation in men. Am J Clin Nutr, 1996. 63(4): p. 553-8. 245. Rubinstein, H.M., A.A. Dietz, and R. Srinavasan, Relation of vitamin E and serum lipids. Clin Chim Acta, 1969. 23(1): p. 1-6. 246. Halliwell, B. and J.M.C. Gutteridge, Antioxidant defence enzymes: superoxide dismutases (SOD), in Free Radicals in Biology and Medicine. 2007, Oxford University Press: Oxford. p. 81-95.
247. Bertini, I., S. Manganl, and M.S. Viezzoli, Structure and Properties of Copper-Zinc Superoxide Dismutases, in Advances in Inorganic Chemistry, A.G. Sykes, Editor. 1998, Academic Press. p. 127-250. 248. Cho, C.S., G.J. Kato, S.H. Yang, S.W. Bae, J.S. Lee, M.T. Gladwin, and S.G. Rhee, Hydroxyurea-induced expression of glutathione peroxidase 1 in red blood cells of individuals with sickle cell anemia. Antioxidants & redox signaling, 2010. 13(1): p. 1- 11. 249. Kurata, M., M. Suzuki, and N.S. Agar, Antioxidant systems and erythrocyte life-span in mammals. Comp Biochem Physiol B, 1993. 106(3): p. 477-87. 250. Dunford, H.B., Catalases, in Heme peroxidases. 1999, John Wiley: New York. p. 435- 451. Chapter 7: References 141
251. Halliwell, B. and J.M.C. Gutteridge, Antioxidant defence enzymes: catalases, in Free Radicals in Biology and Medicine. 2007, Oxford University Press: Oxford. p. 105- 110. 252. Chance, B., H. Sies, and A. Boveris, Hydroperoxide metabolism in mammalian organs. Physiol Rev, 1979. 59(3): p. 527-605. 253. Gaetani, G.F., A.M. Ferraris, M. Rolfo, R. Mangerini, S. Arena, and H.N. Kirkman, Predominant role of catalase in the disposal of hydrogen peroxide within human erythrocytes. Blood, 1996. 87(4): p. 1595-9. 254. Mueller, S., H.D. Riedel, and W. Stremmel, Direct evidence for catalase as the predominant H2O2 -removing enzyme in human erythrocytes. Blood, 1997. 90(12): p. 4973-8. 255. Johnson, R.M., G. Goyette, Jr., Y. Ravindranath, and Y.S. Ho, Hemoglobin autoxidation and regulation of endogenous H2O2 levels in erythrocytes. Free Radic Biol Med, 2005. 39(11): p. 1407-17. 256. Mills, G.C., Hemoglobin catabolism. I. Glutathione peroxidase, an erythrocyte enzyme which protects hemoglobin from oxidative breakdown. J Biol Chem, 1957. 229(1): p. 189-97. 257. Gaetani, G.F., S. Galiano, L. Canepa, A.M. Ferraris, and H.N. Kirkman, Catalase and glutathione peroxidase are equally active in detoxification of hydrogen peroxide in human erythrocytes. Blood, 1989. 73(1): p. 334-9. 258. Cohen, G. and P. Hochstein, Glutathione Peroxidase: The Primary Agent for the Elimination of Hydrogen Peroxide in Erythrocytes. Biochemistry, 1963. 2: p. 1420-8. 259. Epp, O., R. Ladenstein, and A. Wendel, The refined structure of the selenoenzyme glutathione peroxidase at 0.2-nm resolution. Eur J Biochem, 1983. 133(1): p. 51-69.
260. Cheng, W., Y.X. Fu, J.M. Porres, D.A. Ross, and X.G. Lei, Selenium-dependent cellular glutathione peroxidase protects mice against a pro-oxidant-induced oxidation of NADPH, NADH, lipids, and protein. FASEB J, 1999. 13(11): p. 1467-75. 261. Ho, Y.S., J.L. Magnenat, R.T. Bronson, J. Cao, M. Gargano, M. Sugawara, and C.D. Funk, Mice deficient in cellular glutathione peroxidase develop normally and show no increased sensitivity to hyperoxia. J Biol Chem, 1997. 272(26): p. 16644-51. 262. Fu, Y., W.H. Cheng, J.M. Porres, D.A. Ross, and X.G. Lei, Knockout of cellular glutathione peroxidase gene renders mice susceptible to diquat-induced oxidative stress. Free Radic Biol Med, 1999. 27(5-6): p. 605-11. 263. Cheng, W.H., Y.S. Ho, B.A. Valentine, D.A. Ross, G.F. Combs, Jr., and X.G. Lei, Cellular glutathione peroxidase is the mediator of body selenium to protect against paraquat lethality in transgenic mice. J Nutr, 1998. 128(7): p. 1070-6.
264. Toppo, S., L. Flohe, F. Ursini, S. Vanin, and M. Maiorino, Catalytic mechanisms and specificities of glutathione peroxidases: variations of a basic scheme. Biochimica et biophysica acta, 2009. 1790(11): p. 1486-500. 265. Fahey, R.C. and A. Sundquist, Evolution of glutathione metabolism. Adv. Enzymol. Relat. Areas Mol. Biol, 1991. 64(1): p. 53. 266. Asensi, M., J. Sastre, F.V. Pallardo, A. Lloret, M. Lehner, J. Garcia-de-la Asuncion, and J. Vina, Ratio of reduced to oxidized glutathione as indicator of oxidative stress status and DNA damage. Methods Enzymol, 1999. 299: p. 267-76. Chapter 7: References 142
267. Meister, A., Glutathione metabolism. Methods Enzymol, 1995. 251: p. 3-7.
268. Kosower, N.S., The Glutathione Status of Cells1. International review of cytology, 1978. 54: p. 109. 269. May, J.M., Z.C. Qu, and C.E. Cobb, Human erythrocyte recycling of ascorbic acid: relative contributions from the ascorbate free radical and dehydroascorbic acid. J Biol Chem, 2004. 279(15): p. 14975-82. 270. Wells, W.W. and D.P. Xu, Dehydroascorbate reduction. J Bioenerg Biomembr, 1994. 26(4): p. 369-77. 271. Meister, A., Glutathione biosynthesis and its inhibition. Methods Enzymol, 1995. 252: p. 26-30. 272. Putnam, C.D., A.S. Arvai, Y. Bourne, and J.A. Tainer, Active and inhibited human catalase structures: ligand and NADPH binding and catalytic mechanism. Journal of molecular biology, 2000. 296(1): p. 295-309. 273. Gilboa, H., B.E. Chapman, and P.W. Kuchel, 19F NMR magnetization transfer between 5-FBAPTA and its complexes. An alternative means for measuring free Ca2+ concentration, and detection of complexes with protein in erythrocytes. NMR Biomed, 1994. 7(7): p. 330-8. 274. Scott, M.D., B.H. Lubin, L. Zuo, and F.A. Kuypers, Erythrocyte defense against hydrogen peroxide: preeminent importance of catalase. J Lab Clin Med, 1991. 118(1): p. 7-16. 275. Ho, Y.S., Y. Xiong, W. Ma, A. Spector, and D.S. Ho, Mice lacking catalase develop normally but show differential sensitivity to oxidant tissue injury. J Biol Chem, 2004. 279(31): p. 32804-12. 276. Halliwell, B. and J.M.C. Gutteridge, Antioxidant defence enzymes: the glutathione peroxidase family, in Free Radicals in Biology and Medicine. 2007, Oxford University Press: Oxford. p. 110-114. 277. Wood, Z.A., E. Schroder, J. Robin Harris, and L.B. Poole, Structure, mechanism and regulation of peroxiredoxins. Trends Biochem Sci, 2003. 28(1): p. 32-40. 278. Hofmann, B., H.J. Hecht, and L. Flohe, Peroxiredoxins. Biol Chem, 2002. 383(3-4): p. 347-64.
279. Poole, L.B., Bacterial defenses against oxidants: mechanistic features of cysteine- based peroxidases and their flavoprotein reductases. Archives of biochemistry and biophysics, 2005. 433(1): p. 240-54. 280. Rhee, S.G., H.Z. Chae, and K. Kim, Peroxiredoxins: a historical overview and speculative preview of novel mechanisms and emerging concepts in cell signaling. Free Radic Biol Med, 2005. 38(12): p. 1543-52. 281. Trujillo, M., G. Ferrer-Sueta, L. Thomson, L. Flohe, and R. Radi, Kinetics of peroxiredoxins and their role in the decomposition of peroxynitrite. Subcell Biochem, 2007. 44: p. 83-113. 282. Manta, B., M. Hugo, C. Ortiz, G. Ferrer-Sueta, M. Trujillo, and A. Denicola, The peroxidase and peroxynitrite reductase activity of human erythrocyte peroxiredoxin 2. Archives of biochemistry and biophysics, 2009. 484(2): p. 146-54. Chapter 7: References 143
283. Limauro, D., E. Pedone, I. Galdi, and S. Bartolucci, Peroxiredoxins as cellular guardians in Sulfolobus solfataricus: characterization of Bcp1, Bcp3 and Bcp4. FEBS J, 2008. 275(9): p. 2067-77. 284. Johnson, N.A., Y. Liu, and H.M. Fletcher, Alkyl hydroperoxide peroxidase subunit C (ahpC) protects against organic peroxides but does not affect the virulence of Porphyromonas gingivalis W83. Oral Microbiol Immunol, 2004. 19(4): p. 233-9. 285. Kawakami, R., H. Sakuraba, S. Kamohara, S. Goda, Y. Kawarabayasi, and T. Ohshima, Oxidative stress response in an anaerobic hyperthermophilic archaeon: presence of a functional peroxiredoxin in Pyrococcus horikoshii. Journal of biochemistry, 2004. 136(4): p. 541-7. 286. Cha, M.K., C.H. Yun, and I.H. Kim, Interaction of human thiol-specific antioxidant protein 1 with erythrocyte plasma membrane. Biochemistry, 2000. 39(23): p. 6944-50. 287. Chae, H.Z., H.J. Kim, S.W. Kang, and S.G. Rhee, Characterization of three isoforms of mammalian peroxiredoxin that reduce peroxides in the presence of thioredoxin. Diabetes Res Clin Pract, 1999. 45(2-3): p. 101-12. 288. Seo, M.S., S.W. Kang, K. Kim, I.C. Baines, T.H. Lee, and S.G. Rhee, Identification of a new type of mammalian peroxiredoxin that forms an intramolecular disulfide as a reaction intermediate. J Biol Chem, 2000. 275(27): p. 20346-54. 289. Immenschuh, S. and E. Baumgart-Vogt, Peroxiredoxins, oxidative stress, and cell proliferation. Antioxidants & redox signaling, 2005. 7(5-6): p. 768-77. 290. Neumann, C.A., D.S. Krause, C.V. Carman, S. Das, D.P. Dubey, J.L. Abraham, R.T. Bronson, Y. Fujiwara, et al., Essential role for the peroxiredoxin Prdx1 in erythrocyte antioxidant defence and tumour suppression. Nature, 2003. 424(6948): p. 561-5. 291. Marzocchi, B., L. Ciccoli, C. Tani, S. Leoncini, V. Rossi, L. Bini, S. Perrone, and G. Buonocore, Hypoxia-induced post-translational changes in red blood cell protein map of newborns. Pediatric research, 2005. 58(4): p. 660-5. 292. Chae, H.Z., K. Robison, L.B. Poole, G. Church, G. Storz, and S.G. Rhee, Cloning and sequencing of thiol-specific antioxidant from mammalian brain: alkyl hydroperoxide reductase and thiol-specific antioxidant define a large family of antioxidant enzymes. Proc Natl Acad Sci U S A, 1994. 91(15): p. 7017-21. 293. Schroder, E., J.A. Littlechild, A.A. Lebedev, N. Errington, A.A. Vagin, and M.N. Isupov, Crystal structure of decameric 2-Cys peroxiredoxin from human erythrocytes at 1.7 A resolution. Structure, 2000. 8(6): p. 605-15. 294. Weinreich, W., J. Acker, and I. Graber, Determination of total fluoride in HF/HNO3/H2SiF6 etch solutions by new potentiometric titration methods. Talanta, 2007. 71(5): p. 1901-5.
295. Ellis, H.R. and L.B. Poole, Novel application of 7-chloro-4-nitrobenzo-2-oxa-1,3- diazole to identify cysteine sulfenic acid in the AhpC component of alkyl hydroperoxide reductase. Biochemistry, 1997. 36(48): p. 15013-8. 296. Ellis, H.R. and L.B. Poole, Roles for the two cysteine residues of AhpC in catalysis of peroxide reduction by alkyl hydroperoxide reductase from Salmonella typhimurium. Biochemistry, 1997. 36(43): p. 13349-56. 297. Lim, Y.S., M.K. Cha, H.K. Kim, T.B. Uhm, J.W. Park, K. Kim, and I.H. Kim, Removals of hydrogen peroxide and hydroxyl radical by thiol-specific antioxidant Chapter 7: References 144
protein as a possible role in vivo. Biochemical and biophysical research communications, 1993. 192(1): p. 273-80. 298. Poole, L.B., C.M. Reynolds, Z.A. Wood, P.A. Karplus, H.R. Ellis, and M. Li Calzi, AhpF and other NADH:peroxiredoxin oxidoreductases, homologues of low Mr thioredoxin reductase. Eur J Biochem, 2000. 267(20): p. 6126-33. 299. Wood, Z.A., L.B. Poole, R.R. Hantgan, and P.A. Karplus, Dimers to doughnuts: redox-sensitive oligomerization of 2-cysteine peroxiredoxins. Biochemistry, 2002. 41(17): p. 5493-504. 300. Chae, H.Z., S.J. Chung, and S.G. Rhee, Thioredoxin-dependent peroxide reductase from yeast. J Biol Chem, 1994. 269(44): p. 27670-8. 301. Low, F.M., M.B. Hampton, and C.C. Winterbourn, Peroxiredoxin 2 and peroxide metabolism in the erythrocyte. Antioxidants & redox signaling, 2008. 10(9): p. 1621- 30.
302. Lim, Y.S., M.K. Cha, C.H. Yun, H.K. Kim, K. Kim, and I.H. Kim, Purification and characterization of thiol-specific antioxidant protein from human red blood cell: a new type of antioxidant protein. Biochemical and biophysical research communications, 1994. 199(1): p. 199-206. 303. Sauri, H., L. Butterfield, A. Kim, and H. Shau, Antioxidant function of recombinant human natural killer enhancing factor. Biochemical and biophysical research communications, 1995. 208(3): p. 964-9. 304. Evrard, C., A. Capron, C. Marchand, A. Clippe, R. Wattiez, P. Soumillion, B. Knoops, and J.P. Declercq, Crystal structure of a dimeric oxidized form of human peroxiredoxin 5. Journal of molecular biology, 2004. 337(5): p. 1079-90. 305. Zhou, S., Y.C. Lien, T. Shuvaeva, K. DeBolt, S.I. Feinstein, and A.B. Fisher, Functional interaction of glutathione S-transferase pi and peroxiredoxin 6 in intact cells. The international journal of biochemistry & cell biology, 2013. 45(2): p. 401-7. 306. Ralat, L.A., Y. Manevich, A.B. Fisher, and R.F. Colman, Direct evidence for the formation of a complex between 1-cysteine peroxiredoxin and glutathione S- transferase pi with activity changes in both enzymes. Biochemistry, 2006. 45(2): p. 360-72. 307. Manevich, Y., S.I. Feinstein, and A.B. Fisher, Activation of the antioxidant enzyme 1- CYS peroxiredoxin requires glutathionylation mediated by heterodimerization with pi GST. Proc Natl Acad Sci U S A, 2004. 101(11): p. 3780-5. 308. Baker, L.M., A. Raudonikiene, P.S. Hoffman, and L.B. Poole, Essential thioredoxin- dependent peroxiredoxin system from Helicobacter pylori: genetic and kinetic characterization. Journal of bacteriology, 2001. 183(6): p. 1961-73.
309. Munhoz, D.C. and L.E. Netto, Cytosolic thioredoxin peroxidase I and II are important defenses of yeast against organic hydroperoxide insult: catalases and peroxiredoxins cooperate in the decomposition of H2O2 by yeast. J Biol Chem, 2004. 279(34): p. 35219-27. 310. Dietz, K.J., Plant peroxiredoxins. Annu Rev Plant Biol, 2003. 54: p. 93-107. 311. Flohe, L., H. Budde, and B. Hofmann, Peroxiredoxins in antioxidant defense and redox regulation. Biofactors, 2003. 19(1-2): p. 3-10. Chapter 7: References 145
312. Jang, H.H., Y.H. Chi, S.K. Park, S.S. Lee, J.R. Lee, J.H. Park, J.C. Moon, Y.M. Lee, et al., Structural and functional regulation of eukaryotic 2-Cys peroxiredoxins including the plant ones in cellular defense-signaling mechanisms against oxidative stress. Physiologia Plantarum, 2006. 126(4): p. 549-559. 313. Chae, H.Z., S.W. Kang, and S.G. Rhee, Isoforms of mammalian peroxiredoxin that reduce peroxides in presence of thioredoxin. Methods Enzymol, 1999. 300: p. 219-26. 314. Parsonage, D., D.S. Youngblood, G.N. Sarma, Z.A. Wood, P.A. Karplus, and L.B. Poole, Analysis of the link between enzymatic activity and oligomeric state in AhpC, a bacterial peroxiredoxin. Biochemistry, 2005. 44(31): p. 10583-92. 315. Ogusucu, R., D. Rettori, D.C. Munhoz, L.E. Netto, and O. Augusto, Reactions of yeast thioredoxin peroxidases I and II with hydrogen peroxide and peroxynitrite: rate constants by competitive kinetics. Free Radic Biol Med, 2007. 42(3): p. 326-34. 316. Lillig, C.H. and A. Holmgren, Thioredoxin and related molecules--from biology to health and disease. Antioxidants & redox signaling, 2007. 9(1): p. 25-47. 317. Gilbert, H.F., Molecular and cellular aspects of thiol-disulfide exchange. Adv Enzymol Relat Areas Mol Biol, 1990. 63: p. 69-172. 318. Sun, Q.A., Y. Wu, F. Zappacosta, K.T. Jeang, B.J. Lee, D.L. Hatfield, and V.N. Gladyshev, Redox regulation of cell signaling by selenocysteine in mammalian thioredoxin reductases. J Biol Chem, 1999. 274(35): p. 24522-30. 319. Holmgren, A. and J. Lu, Thioredoxin and thioredoxin reductase: current research with special reference to human disease. Biochemical and biophysical research communications, 2010. 396(1): p. 120-4. 320. Holmgren, A., Thioredoxin structure and mechanism: conformational changes on oxidation of the active-site sulfhydryls to a disulfide. Structure, 1995. 3(3): p. 239-43.
321. Holmgren, A., Antioxidant function of thioredoxin and glutaredoxin systems. Antioxidants & redox signaling, 2000. 2(4): p. 811-20. 322. Du, Y., H. Zhang, X. Zhang, J. Lu, and A. Holmgren, Thioredoxin 1 is inactivated due to oxidation induced by peroxiredoxin under oxidative stress and reactivated by glutaredoxin system. J Biol Chem, 2013. 323. Watson, W.H., J. Pohl, W.R. Montfort, O. Stuchlik, M.S. Reed, G. Powis, and D.P. Jones, Redox potential of human thioredoxin 1 and identification of a second dithiol/disulfide motif. J Biol Chem, 2003. 278(35): p. 33408-15. 324. Watson, W.H., X. Yang, Y.E. Choi, D.P. Jones, and J.P. Kehrer, Thioredoxin and its role in toxicology. Toxicol Sci, 2004. 78(1): p. 3-14. 325. Gromer, S., S. Urig, and K. Becker, The thioredoxin system--from science to clinic. Med Res Rev, 2004. 24(1): p. 40-89. 326. Wood, Z.A., L.B. Poole, and P.A. Karplus, Peroxiredoxin evolution and the regulation of hydrogen peroxide signaling. Science, 2003. 300(5619): p. 650-3. 327. Rabilloud, T., M. Heller, F. Gasnier, S. Luche, C. Rey, R. Aebersold, M. Benahmed, P. Louisot, et al., Proteomics analysis of cellular response to oxidative stress. Evidence for in vivo overoxidation of peroxiredoxins at their active site. J Biol Chem, 2002. 277(22): p. 19396-401. Chapter 7: References 146
328. Wagner, E., S. Luche, L. Penna, M. Chevallet, A. Van Dorsselaer, E. Leize-Wagner, and T. Rabilloud, A method for detection of overoxidation of cysteines: peroxiredoxins are oxidized in vivo at the active-site cysteine during oxidative stress. The Biochemical journal, 2002. 366(Pt 3): p. 777-85. 329. Yang, K.S., S.W. Kang, H.A. Woo, S.C. Hwang, H.Z. Chae, K. Kim, and S.G. Rhee, Inactivation of human peroxiredoxin I during catalysis as the result of the oxidation of the catalytic site cysteine to cysteine-sulfinic acid. J Biol Chem, 2002. 277(41): p. 38029-36. 330. Baty, J.W., M.B. Hampton, and C.C. Winterbourn, Proteomic detection of hydrogen peroxide-sensitive thiol proteins in Jurkat cells. Biochem J, 2005. 389(Pt 3): p. 785- 95. 331. Koo, K.H., S. Lee, S.Y. Jeong, E.T. Kim, H.J. Kim, K. Kim, K. Song, and H.Z. Chae, Regulation of thioredoxin peroxidase activity by C-terminal truncation. Archives of biochemistry and biophysics, 2002. 397(2): p. 312-8.
332. Ackerstaff, R.G., Cerebral circulation monitoring in carotid endarterectomy and carotid artery stenting. Front Neurol Neurosci, 2006. 21: p. 229-38. 333. Peskin, A.V., N. Dickerhof, R.A. Poynton, L.N. Paton, P.E. Pace, M.B. Hampton, and C.C. Winterbourn, Hyperoxidation of peroxiredoxins 2 and 3: rate constants for the reactions of the sulfenic acid of the peroxidatic cysteine. J Biol Chem, 2013. 288(20): p. 14170-7. 334. Hall, A., P.A. Karplus, and L.B. Poole, Typical 2-Cys peroxiredoxins--structures, mechanisms and functions. FEBS J, 2009. 276(9): p. 2469-77. 335. Biteau, B., J. Labarre, and M.B. Toledano, ATP-dependent reduction of cysteine- sulphinic acid by S. cerevisiae sulphiredoxin. Nature, 2003. 425(6961): p. 980-4. 336. Woo, H.A., H.Z. Chae, S.C. Hwang, K.S. Yang, S.W. Kang, K. Kim, and S.G. Rhee, Reversing the inactivation of peroxiredoxins caused by cysteine sulfinic acid formation. Science, 2003. 300(5619): p. 653-6. 337. Chang, T.S., W. Jeong, H.A. Woo, S.M. Lee, S. Park, and S.G. Rhee, Characterization of mammalian sulfiredoxin and its reactivation of hyperoxidized peroxiredoxin through reduction of cysteine sulfinic acid in the active site to cysteine. J Biol Chem, 2004. 279(49): p. 50994-1001. 338. Liu, X.P., X.Y. Liu, J. Zhang, Z.L. Xia, X. Liu, H.J. Qin, and D.W. Wang, Molecular and functional characterization of sulfiredoxin homologs from higher plants. Cell Res, 2006. 16(3): p. 287-96. 339. Cho, C.S., S. Lee, G.T. Lee, H.A. Woo, E.J. Choi, and S.G. Rhee, Irreversible inactivation of glutathione peroxidase 1 and reversible inactivation of peroxiredoxin II by H2O2 in red blood cells. Antioxidants & redox signaling, 2010. 12(11): p. 1235- 46. 340. Woo, H.A., W. Jeong, T.S. Chang, K.J. Park, S.J. Park, J.S. Yang, and S.G. Rhee, Reduction of cysteine sulfinic acid by sulfiredoxin is specific to 2-cys peroxiredoxins. J Biol Chem, 2005. 280(5): p. 3125-8. 341. Jeong, W., S.J. Park, T.S. Chang, D.Y. Lee, and S.G. Rhee, Molecular mechanism of the reduction of cysteine sulfinic acid of peroxiredoxin to cysteine by mammalian sulfiredoxin. J Biol Chem, 2006. 281(20): p. 14400-7. Chapter 7: References 147
342. Jonsson, T.J., M.S. Murray, L.C. Johnson, L.B. Poole, and W.T. Lowther, Structural basis for the retroreduction of inactivated peroxiredoxins by human sulfiredoxin. Biochemistry, 2005. 44(24): p. 8634-42. 343. Lee, D.Y., S.J. Park, W. Jeong, H.J. Sung, T. Oho, X. Wu, S.G. Rhee, and J.M. Gruschus, Mutagenesis and modeling of the peroxiredoxin (Prx) complex with the NMR structure of ATP-bound human sulfiredoxin implicate aspartate 187 of Prx I as the catalytic residue in ATP hydrolysis. Biochemistry, 2006. 45(51): p. 15301-9. 344. Barranco-Medina, S., J.J. Lazaro, and K.J. Dietz, The oligomeric conformation of peroxiredoxins links redox state to function. FEBS letters, 2009. 583(12): p. 1809-16. 345. Harris, J.R., E. Schroder, M.N. Isupov, D. Scheffler, P. Kristensen, J.A. Littlechild, A.A. Vagin, and U. Meissner, Comparison of the decameric structure of peroxiredoxin-II by transmission electron microscopy and X-ray crystallography. Biochimica et biophysica acta, 2001. 1547(2): p. 221-34.
346. Lee, W., K.S. Choi, J. Riddell, C. Ip, D. Ghosh, J.H. Park, and Y.M. Park, Human peroxiredoxin 1 and 2 are not duplicate proteins: the unique presence of CYS83 in Prx1 underscores the structural and functional differences between Prx1 and Prx2. J Biol Chem, 2007. 282(30): p. 22011-22. 347. Cao, Z., A.W. Roszak, L.J. Gourlay, J.G. Lindsay, and N.W. Isaacs, Bovine mitochondrial peroxiredoxin III forms a two-ring catenane. Structure, 2005. 13(11): p. 1661-4. 348. Kitano, K., Y. Niimura, Y. Nishiyama, and K. Miki, Stimulation of peroxidase activity by decamerization related to ionic strength: AhpC protein from Amphibacillus xylanus. Journal of biochemistry, 1999. 126(2): p. 313-9. 349. Chauhan, R. and S.C. Mande, Characterization of the Mycobacterium tuberculosis H37Rv alkyl hydroperoxidase AhpC points to the importance of ionic interactions in oligomerization and activity. Biochem J, 2001. 354(Pt 1): p. 209-15.
350. Kristensen, P., D.E. Rasmussen, and B.I. Kristensen, Properties of thiol-specific anti- oxidant protein or calpromotin in solution. Biochemical and biophysical research communications, 1999. 262(1): p. 127-31. 351. Plishker, G.A., D. Chevalier, L. Seinsoth, and R.B. Moore, Calcium-activated potassium transport and high molecular weight forms of calpromotin. J Biol Chem, 1992. 267(30): p. 21839-43.
352. Allen, D.W. and S. Cadman, Calcium-induced erythrocyte membrane changes. The role of adsorption of cytosol proteins and proteases. Biochimica et biophysica acta, 1979. 551(1): p. 1-9. 353. Jang, H.H., S.Y. Kim, S.K. Park, H.S. Jeon, Y.M. Lee, J.H. Jung, S.Y. Lee, H.B. Chae, et al., Phosphorylation and concomitant structural changes in human 2-Cys peroxiredoxin isotype I differentially regulate its peroxidase and molecular chaperone functions. FEBS letters, 2006. 580(1): p. 351-5. 354. Barranco-Medina, S., S. Kakorin, J.J. Lazaro, and K.J. Dietz, Thermodynamics of the dimer-decamer transition of reduced human and plant 2-cys peroxiredoxin. Biochemistry, 2008. 47(27): p. 7196-204. 355. Schroder, E., A.C. Willis, and C.P. Ponting, Porcine natural-killer-enhancing factor- B: oligomerisation and identification as a calpain substrate in vitro. Biochimica et biophysica acta, 1998. 1383(2): p. 279-91. Chapter 7: References 148
356. Lim, J.C., H.I. Choi, Y.S. Park, H.W. Nam, H.A. Woo, K.S. Kwon, Y.S. Kim, S.G. Rhee, et al., Irreversible oxidation of the active-site cysteine of peroxiredoxin to cysteine sulfonic acid for enhanced molecular chaperone activity. J Biol Chem, 2008. 283(43): p. 28873-80. 357. Ogasawara, Y., T. Ohminato, Y. Nakamura, and K. Ishii, Structural and functional analysis of native peroxiredoxin 2 in human red blood cells. The international journal of biochemistry & cell biology, 2012. 44(7): p. 1072-7. 358. Harris, J.R., Release of a macromolecular protein component from human erythrocyte ghosts. Biochimica et biophysica acta, 1968. 150(3): p. 534-7. 359. Jang, H.H., K.O. Lee, Y.H. Chi, B.G. Jung, S.K. Park, J.H. Park, J.R. Lee, S.S. Lee, et al., Two enzymes in one; two yeast peroxiredoxins display oxidative stress-dependent switching from a peroxidase to a molecular chaperone function. Cell, 2004. 117(5): p. 625-35.
360. Moon, J.C., Y.S. Hah, W.Y. Kim, B.G. Jung, H.H. Jang, J.R. Lee, S.Y. Kim, Y.M. Lee, et al., Oxidative stress-dependent structural and functional switching of a human 2-Cys peroxiredoxin isotype II that enhances HeLa cell resistance to H2O2-induced cell death. J Biol Chem, 2005. 280(31): p. 28775-84. 361. Chuang, M.H., M.S. Wu, W.L. Lo, J.T. Lin, C.H. Wong, and S.H. Chiou, The antioxidant protein alkylhydroperoxide reductase of Helicobacter pylori switches from a peroxide reductase to a molecular chaperone function. Proc Natl Acad Sci U S A, 2006. 103(8): p. 2552-7. 362. Phalen, T.J., K. Weirather, P.B. Deming, V. Anathy, A.K. Howe, A. van der Vliet, T.J. Jonsson, L.B. Poole, et al., Oxidation state governs structural transitions in peroxiredoxin II that correlate with cell cycle arrest and recovery. The Journal of cell biology, 2006. 175(5): p. 779-89. 363. Kumsta, C. and U. Jakob, Redox-regulated chaperones. Biochemistry, 2009. 48(22): p. 4666-76. 364. Phillips, A.J., J. Littlejohn, N.A. Yewdall, T. Zhu, C. Valery, F.G. Pearce, A.K. Mitra, M. Radjainia, et al., Peroxiredoxin is a versatile self-assembling tecton for protein nanotechnology. Biomacromolecules, 2014. 15(5): p. 1871-81. 365. Harris, J.R., Some negative contrast staining features of a protein from erythrocyte ghosts. Journal of molecular biology, 1969. 46(2): p. 329-35.
366. Harris, J.R., The isolation and purification of a macromolecular protein component from the human erythrocyte ghost. Biochimica et biophysica acta, 1969. 188(1): p. 31- 42. 367. Harris, J.R. and I. Naeem, Further studies on the characterization of cylindrin and torin, two extrinsic proteins of the erythrocyte membrane. Biochimica et biophysica acta, 1981. 670(2): p. 285-90. 368. Meissner, U., E. Schroder, D. Scheffler, A.G. Martin, and J.R. Harris, Formation, TEM study and 3D reconstruction of the human erythrocyte peroxiredoxin-2 dodecahedral higher-order assembly. Micron, 2007. 38(1): p. 29-39. 369. Han, Y.H., S.U. Kim, T.H. Kwon, D.S. Lee, H.L. Ha, D.S. Park, E.J. Woo, S.H. Lee, et al., Peroxiredoxin II is essential for preventing hemolytic anemia from oxidative stress through maintaining hemoglobin stability. Biochemical and biophysical research communications, 2012. 426(3): p. 427-32. Chapter 7: References 149
370. Lee, T.H., S.U. Kim, S.L. Yu, S.H. Kim, D.S. Park, H.B. Moon, S.H. Dho, K.S. Kwon, et al., Peroxiredoxin II is essential for sustaining life span of erythrocytes in mice. Blood, 2003. 101(12): p. 5033-8. 371. Egler, R.A., E. Fernandes, K. Rothermund, S. Sereika, N. de Souza-Pinto, P. Jaruga, M. Dizdaroglu, and E.V. Prochownik, Regulation of reactive oxygen species, DNA damage, and c-Myc function by peroxiredoxin 1. Oncogene, 2005. 24(54): p. 8038-50. 372. Moghaddam, D.A., A. Heber, D. Capin, T. Kreutz, D. Opitz, E. Lenzen, W. Bloch, K. Brixius, et al., Training increases peroxiredoxin 2 contents in the erythrocytes of overweight/obese men suffering from type 2 diabetes. Wien Med Wochenschr, 2011. 161(21-22): p. 511-8. 373. Eismann, T., N. Huber, T. Shin, S. Kuboki, E. Galloway, M. Wyder, M.J. Edwards, K.D. Greis, et al., Peroxiredoxin-6 protects against mitochondrial dysfunction and liver injury during ischemia-reperfusion in mice. Am J Physiol Gastrointest Liver Physiol, 2009. 296(2): p. G266-74.
374. Nagy, N., G. Malik, A.B. Fisher, and D.K. Das, Targeted disruption of peroxiredoxin 6 gene renders the heart vulnerable to ischemia-reperfusion injury. Am J Physiol Heart Circ Physiol, 2006. 291(6): p. H2636-40. 375. Halliwell, B. and J.M.C. Gutteridge, Reactive species and disease: fact, fiction or filibuster?, in Free Radicals in Biology and Medicine. 2007, Oxford University Press: Oxford. p. 488-613. 376. Bae, S.H., S.H. Sung, E.J. Cho, S.K. Lee, H.E. Lee, H.A. Woo, D.Y. Yu, I.S. Kil, et al., Concerted action of sulfiredoxin and peroxiredoxin I protects against alcohol- induced oxidative injury in mouse liver. Hepatology, 2011. 53(3): p. 945-53. 377. Poole, L.B. and K.J. Nelson, Discovering mechanisms of signaling-mediated cysteine oxidation. Curr Opin Chem Biol, 2008. 12(1): p. 18-24.
378. Veal, E.A., A.M. Day, and B.A. Morgan, Hydrogen peroxide sensing and signaling. Mol Cell, 2007. 26(1): p. 1-14. 379. Rudden, M., B. Milrod, M. Target, S. Ackerman, and E. Graf, Reflective functioning in panic disorder patients: a pilot study. J Am Psychoanal Assoc, 2006. 54(4): p. 1339-43. 380. Rabilloud, T., R. Berthier, M. Vincon, D. Ferbus, G. Goubin, and J.J. Lawrence, Early events in erythroid differentiation: accumulation of the acidic peroxidoxin (PRP/TSA/NKEF-B). Biochem J, 1995. 312 ( Pt 3): p. 699-705. 381. Rhee, S.G. and H.A. Woo, Multiple functions of peroxiredoxins: peroxidases, sensors and regulators of the intracellular messenger H(2)O(2), and protein chaperones. Antioxidants & redox signaling, 2011. 15(3): p. 781-94.
382. Winterbourn, C.C., The biological chemistry of hydrogen peroxide. Methods Enzymol, 2013. 528: p. 3-25. 383. Winterbourn, C.C., Reconciling the chemistry and biology of reactive oxygen species. Nat Chem Biol, 2008. 4(5): p. 278-86. 384. Winterbourn, C.C. and M.B. Hampton, Thiol chemistry and specificity in redox signaling. Free Radic Biol Med, 2008. 45(5): p. 549-61. Chapter 7: References 150
385. Stuhlmeier, K.M., J.J. Kao, P. Wallbrandt, M. Lindberg, B. Hammarstrom, H. Broell, and B. Paigen, Antioxidant protein 2 prevents methemoglobin formation in erythrocyte hemolysates. Eur J Biochem, 2003. 270(2): p. 334-41. 386. Fourquet, S., M.E. Huang, B. D'Autreaux, and M.B. Toledano, The dual functions of thiol-based peroxidases in H2O2 scavenging and signaling. Antioxidants & redox signaling, 2008. 10(9): p. 1565-76. 387. Toledano, M.B., A. Delaunay, L. Monceau, and F. Tacnet, Microbial H2O2 sensors as archetypical redox signaling modules. Trends Biochem Sci, 2004. 29(7): p. 351-7. 388. Jarvis, R.M., S.M. Hughes, and E.C. Ledgerwood, Peroxiredoxin 1 functions as a signal peroxidase to receive, transduce, and transmit peroxide signals in mammalian cells. Free Radic Biol Med, 2012. 53(7): p. 1522-30. 389. Tester, D.J. and M.J. Ackerman, Postmortem long QT syndrome genetic testing for sudden unexplained death in the young. Journal of the American College of Cardiology, 2007. 49(2): p. 240-6. 390. Okazaki, S., T. Tachibana, A. Naganuma, N. Mano, and S. Kuge, Multistep disulfide bond formation in Yap1 is required for sensing and transduction of H2O2 stress signal. Mol Cell, 2007. 27(4): p. 675-88. 391. O'Neill, J.S. and A.B. Reddy, Circadian clocks in human red blood cells. Nature. 469(7331): p. 498-503. 392. Cho, C.S., H.J. Yoon, J.Y. Kim, H.A. Woo, and S.G. Rhee, Circadian rhythm of hyperoxidized peroxiredoxin II is determined by hemoglobin autoxidation and the 20S proteasome in red blood cells. Proc Natl Acad Sci U S A, 2014. 111(33): p. 12043-8. 393. Kil, I.S., S.K. Lee, K.W. Ryu, H.A. Woo, M.C. Hu, S.H. Bae, and S.G. Rhee, Feedback control of adrenal steroidogenesis via H2O2-dependent, reversible inactivation of peroxiredoxin III in mitochondria. Mol Cell, 2012. 46(5): p. 584-94.
394. Chang, T.S., C.S. Cho, S. Park, S. Yu, S.W. Kang, and S.G. Rhee, Peroxiredoxin III, a mitochondrion-specific peroxidase, regulates apoptotic signaling by mitochondria. J Biol Chem, 2004. 279(40): p. 41975-84. 395. Yang, H.Y., D.K. Jeong, S.H. Kim, K.J. Chung, E.J. Cho, U. Yang, S.R. Lee, and T.H. Lee, The role of peroxiredoxin III on late stage of proerythrocyte differentiation. Biochemical and biophysical research communications, 2007. 359(4): p. 1030-6.
396. Leavey, P.J., C. Gonzalez-Aller, G. Thurman, M. Kleinberg, L. Rinckel, D.W. Ambruso, S. Freeman, F.A. Kuypers, et al., A 29-kDa protein associated with p67phox expresses both peroxiredoxin and phospholipase A2 activity and enhances superoxide anion production by a cell-free system of NADPH oxidase activity. J Biol Chem, 2002. 277(47): p. 45181-7.
397. Manevich, Y. and A.B. Fisher, Peroxiredoxin 6, a 1-Cys peroxiredoxin, functions in antioxidant defense and lung phospholipid metabolism. Free Radic Biol Med, 2005. 38(11): p. 1422-32. 398. Ambruso, D.R., M.A. Ellison, G.W. Thurman, and T.L. Leto, Peroxiredoxin 6 translocates to the plasma membrane during neutrophil activation and is required for optimal NADPH oxidase activity. Biochimica et biophysica acta, 2012. 1823(2): p. 306-15. Chapter 7: References 151
399. Moore, R.B., M.V. Mankad, S.K. Shriver, V.N. Mankad, and G.A. Plishker, Reconstitution of Ca(2+)-dependent K+ transport in erythrocyte membrane vesicles requires a cytoplasmic protein. J Biol Chem, 1991. 266(28): p. 18964-8. 400. Shau, H. and A. Kim, Identification of natural killer enhancing factor as a major antioxidant in human red blood cells. Biochemical and biophysical research communications, 1994. 199(1): p. 83-8. 401. Arrigo, A.P., K. Tanaka, A.L. Goldberg, and W.J. Welch, Identity of the 19S 'prosome' particle with the large multifunctional protease complex of mammalian cells (the proteasome). Nature, 1988. 331(6152): p. 192-4. 402. Lupas, A. and W. Baumeister, The 20 S Proteasome, in Ubiquitin and the Biology of the Cell. 1998, Springer. p. 127-146. 403. Harris, J.R., Torin and cylindrin, two extrinsic proteins of the erythrocyte membrane: a review. Nouv Rev Fr Hematol, 1980. 22(4): p. 411-48.
404. Mitsumoto, A., Y. Takanezawa, K. Okawa, A. Iwamatsu, and Y. Nakagawa, Variants of peroxiredoxins expression in response to hydroperoxide stress. Free Radic Biol Med, 2001. 30(6): p. 625-35. 405. Seo, J.H., J.C. Lim, D.Y. Lee, K.S. Kim, G. Piszczek, H.W. Nam, Y.S. Kim, T. Ahn, et al., Novel protective mechanism against irreversible hyperoxidation of peroxiredoxin: Nalpha-terminal acetylation of human peroxiredoxin II. J Biol Chem, 2009. 284(20): p. 13455-65. 406. Matte, A., M. Bertoldi, N. Mohandas, X. An, A. Bugatti, A.M. Brunati, M. Rusnati, E. Tibaldi, et al., Membrane association of peroxiredoxin-2 in red cells is mediated by the N-terminal cytoplasmic domain of band 3. Free Radic Biol Med, 2013. 55: p. 27- 35. 407. Benfeitas, R., G. Selvaggio, F. Antunes, P.M. Coelho, and A. Salvador, Hydrogen peroxide metabolism and sensing in human erythrocytes: a validated kinetic model and reappraisal of the role of peroxiredoxin II. Free Radic Biol Med, 2014. 74: p. 35- 49. 408. Goth, L. and M. Vitai, Hungarian hereditary acatalasemia and hypocatalasemia are not associated with chronic hemolysis. Clin Chim Acta, 1995. 233(1-2): p. 75-9. 409. Aebi, H., J.P. Heiniger, R. Buetler, and A. Haessig, Two cases of acatalasia in Switzerland. Experientia, 1961. 17: p. 466.
410. Takahara, S., Actalasemia and hypocatalasemia in the Orient. Semin Hematol, 1971. 8(4): p. 397-416. 411. Johnson, R.M., G. Goyette, Jr., Y. Ravindranath, and Y.S. Ho, Red cells from glutathione peroxidase-1-deficient mice have nearly normal defenses against exogenous peroxides. Blood, 2000. 96(5): p. 1985-8. 412. Paglia, D.E., Erythroenzymopathies: 1960s in Retrospect. Current Contents/ life sciences/ clinical medicine, 1989. 32(8): p. 1. 413. Beutler, E., Glutathione Peroxidase Deficiency, in Hemolytic anemia in disorders of red cell metabolism. 1978, Springer. p. 222-224. 414. Beutler, E., GLUCOSE-6-PHOSPHATE DEHYDROGENASE DEFICIENCY in Hemolytic anemia in disorders of red cell metabolism. 1978, Springer. p. 23-108. Chapter 7: References 152
415. Wang, X., S.A. Phelan, K. Forsman-Semb, E.F. Taylor, C. Petros, A. Brown, C.P. Lerner, and B. Paigen, Mice with targeted mutation of peroxiredoxin 6 develop normally but are susceptible to oxidative stress. J Biol Chem, 2003. 278(27): p. 25179-90. 416. Li, L., W. Shoji, H. Takano, N. Nishimura, Y. Aoki, R. Takahashi, S. Goto, T. Kaifu, et al., Increased susceptibility of MER5 (peroxiredoxin III) knockout mice to LPS- induced oxidative stress. Biochemical and biophysical research communications, 2007. 355(3): p. 715-21. 417. Woo, H.A., S.H. Yim, D.H. Shin, D. Kang, D.Y. Yu, and S.G. Rhee, Inactivation of peroxiredoxin I by phosphorylation allows localized H(2)O(2) accumulation for cell signaling. Cell, 2010. 140(4): p. 517-28. 418. Shau, H., L.H. Butterfield, R. Chiu, and A. Kim, Cloning and sequence analysis of candidate human natural killer-enhancing factor genes. Immunogenetics, 1994. 40(2): p. 129-34.
419. Kim, K., I.H. Kim, K.Y. Lee, S.G. Rhee, and E.R. Stadtman, The isolation and purification of a specific "protector" protein which inhibits enzyme inactivation by a thiol/Fe(III)/O2 mixed-function oxidation system. J Biol Chem, 1988. 263(10): p. 4704-11. 420. Zhang, P., B. Liu, S.W. Kang, M.S. Seo, S.G. Rhee, and L.M. Obeid, Thioredoxin peroxidase is a novel inhibitor of apoptosis with a mechanism distinct from that of Bcl-2. J Biol Chem, 1997. 272(49): p. 30615-8. 421. Palek, J., P.A. Liu, and S.C. Liu, Polymerisation of red cell membrane protein contributes to spheroechinocyte shape irreversibility. Nature, 1978. 274(5670): p. 505-7. 422. Palek, J., S.C. Liu, and L.M. Snyder, Metabolic dependence of protein arrangement in human erythrocyte membranes. I. Analysis of spectrin-rich complexes in ATP-depleted red cells. Blood, 1978. 51(3): p. 385-95. 423. Riggs, M.G. and V.M. Ingram, Differences in erythrocyte membrane proteins and glycoproteins in sickle cell disease. Biochemical and biophysical research communications, 1977. 74(1): p. 191-8. 424. Mizukami, H., D.E. Bartnicki, and L. Chaplin, Band-8 protein of human erythrocyte membrane: another Ca++ binding protein? Prog Clin Biol Res, 1984. 159: p. 47-55.
425. Plishker, G.A., Iodoacetic acid inhibition of calcium-dependent potassium efflux in red blood cells. The American journal of physiology, 1985. 248(5 Pt 1): p. C419-24. 426. Plishker, G.A., P.H. White, and E.D. Cadman, Involvement of a cytoplasmic protein in calcium-dependent potassium efflux in red blood cells. The American journal of physiology, 1986. 251(4 Pt 1): p. C535-40.
427. Blum, R.M. and J.F. Hoffman, The membrane locus of Ca-stimulated K transport in energy depleted human red blood cells. J Membr Biol, 1971. 6(4): p. 315-28. 428. Gardos, G., The function of calcium in the potassium permeability of human erythrocytes. Biochimica et biophysica acta, 1958. 30(3): p. 653-4. 429. Lew, V.L., Comparative biochemistry and physiology of transport, in Proceedings of the 1973 meeting of the International Conference on Biological Membranes. 1974, North-Holland Pub. Co. p. 310-316. Chapter 7: References 153
430. Maher, A.D. and P.W. Kuchel, The Gardos channel: a review of the Ca2+-activated K+ channel in human erythrocytes. The international journal of biochemistry & cell biology, 2003. 35(8): p. 1182-97. 431. Fabry, M.E., L. Benjamin, C. Lawrence, and R.L. Nagel, An objective sign in painful crisis in sickle cell anemia: the concomitant reduction of high density red cells. Blood, 1984. 64(2): p. 559-63. 432. Bookchin, R.M., O.E. Ortiz, and V.L. Lew, Activation of calcium-dependent potassium channels in deoxygenated sickled red cells. Prog Clin Biol Res, 1987. 240: p. 193-200. 433. Moore, R.B., S.K. Shriver, L.D. Jenkins, V.N. Mankad, A.K. Shah, and G.A. Plishker, Calpromotin, a cytoplasmic protein, is associated with the formation of dense cells in sickle cell anemia. American journal of hematology, 1997. 56(2): p. 100-6. 434. De Franceschi, L., R.S. Franco, M. Bertoldi, C. Brugnara, A. Matte, A. Siciliano, A.J. Wieschhaus, A.H. Chishti, et al., Pharmacological inhibition of calpain-1 prevents red cell dehydration and reduces Gardos channel activity in a mouse model of sickle cell disease. FASEB J, 2013. 27(2): p. 750-9. 435. Lang, P.A., S. Kaiser, S. Myssina, T. Wieder, F. Lang, and S.M. Huber, Role of Ca2+- activated K+ channels in human erythrocyte apoptosis. Am J Physiol Cell Physiol, 2003. 285(6): p. C1553-60. 436. Salzer, U., P. Hinterdorfer, U. Hunger, C. Borken, and R. Prohaska, Ca(++)- dependent vesicle release from erythrocytes involves stomatin-specific lipid rafts, synexin (annexin VII), and sorcin. Blood, 2002. 99(7): p. 2569-77. 437. Low, T.Y., T.K. Seow, and M.C. Chung, Separation of human erythrocyte membrane associated proteins with one-dimensional and two-dimensional gel electrophoresis followed by identification with matrix-assisted laser desorption/ionization-time of flight mass spectrometry. Proteomics, 2002. 2(9): p. 1229-39.
438. Matte, A., P.S. Low, F. Turrini, M. Bertoldi, M.E. Campanella, D. Spano, A. Pantaleo, A. Siciliano, et al., Peroxiredoxin-2 expression is increased in beta-thalassemic mouse red cells but is displaced from the membrane as a marker of oxidative stress. Free Radic Biol Med, 2010. 49(3): p. 457-66. 439. Moore, R.B. and S.K. Shriver, Protein 7.2b of human erythrocyte membranes binds to calpromotin. Biochemical and biophysical research communications, 1997. 232(2): p. 294-7.
440. Bosman, G.J., J.M. Klaarenbeek, M. Luten, and H.J. Bos, Storage-related changes in erythrocyte band 3: not a case for the Diego blood group antigens. Cell Mol Biol (Noisy-le-grand), 2005. 51(2): p. 195-200. 441. Low, P.S., Structure and function of the cytoplasmic domain of band 3: center of erythrocyte membrane-peripheral protein interactions. Biochimica et biophysica acta, 1986. 864(2): p. 145-67. 442. Rinalducci, S., E. Ferru, B. Blasi, F. Turrini, and L. Zolla, Oxidative stress and caspase-mediated fragmentation of cytoplasmic domain of erythrocyte band 3 during blood storage. Blood Transfus, 2012. 10 Suppl 2: p. s55-62. 443. Goll, D.E., V.F. Thompson, H. Li, W. Wei, and J. Cong, The calpain system. Physiol Rev, 2003. 83(3): p. 731-801. Chapter 7: References 154
444. Schwarz-Ben Meir, N., T. Glaser, and N.S. Kosower, Band 3 protein degradation by calpain is enhanced in erythrocytes of old people. Biochem J, 1991. 275 ( Pt 1): p. 47- 52. 445. Rocha, S., E. Costa, S. Coimbra, H. Nascimento, C. Catarino, P. Rocha-Pereira, A. Quintanilha, L. Belo, et al., Linkage of cytosolic peroxiredoxin 2 to erythrocyte membrane imposed by hydrogen peroxide-induced oxidative stress. Blood Cells Mol Dis, 2009. 43(1): p. 68-73. 446. Rocha, S., R.M. Vitorino, F.M. Lemos-Amado, E.B. Castro, P. Rocha-Pereira, J. Barbot, E. Cleto, F. Ferreira, et al., Presence of cytosolic peroxiredoxin 2 in the erythrocyte membrane of patients with hereditary spherocytosis. Blood Cells Mol Dis, 2008. 41(1): p. 5-9. 447. Longo, V., A. D'Alessandro, and L. Zolla, Deoxygenation of leucofiltered erythrocyte concentrates preserves proteome stability during storage in the blood bank. Blood Transfus, 2014. 12(4): p. 599-604.
448. Zaccarin, M., M. Falda, A. Roveri, V. Bosello-Travain, L. Bordin, M. Maiorino, F. Ursini, and S. Toppo, Quantitative label-free redox proteomics of reversible cysteine oxidation in red blood cell membranes. Free Radic Biol Med, 2014. 71: p. 90-8. 449. Schroder, E., M.N. Isupov, A. Naran, and J.A. Littlechild, Crystallization and preliminary X-ray analysis of human thioredoxin peroxidase-B from red blood cells. Acta Crystallogr D Biol Crystallogr, 1999. 55(Pt 2): p. 536-8. 450. Kettle, A.J. and C.C. Winterbourn, Superoxide-dependent hydroxylation by myeloperoxidase. J Biol Chem, 1994. 269(25): p. 17146-51. 451. Green, J.N., C.C. Winterbourn, and M.B. Hampton, Analysis of Neutrophil Bactericidal Activity. 2007. p. 319-332. 452. Wilkie, R.P., M.C. Vissers, M. Dragunow, and M.B. Hampton, A functional NADPH oxidase prevents caspase involvement in the clearance of phagocytic neutrophils. Infect Immun, 2007. 75(7): p. 3256-63. 453. Jackson, S.H., J.I. Gallin, and S.M. Holland, The p47phox mouse knock-out model of chronic granulomatous disease. The Journal of experimental medicine, 1995. 182(3): p. 751-8. 454. Pollock, J.D., D.A. Williams, M.A. Gifford, L.L. Li, X. Du, J. Fisherman, S.H. Orkin, C.M. Doerschuk, et al., Mouse model of X-linked chronic granulomatous disease, an inherited defect in phagocyte superoxide production. Nat Genet, 1995. 9(2): p. 202-9. 455. Wang, Y., H. Liu, G. McKenzie, P.K. Witting, J.P. Stasch, M. Hahn, D. Changsirivathanathamrong, B.J. Wu, et al., Kynurenine is an endothelium-derived relaxing factor produced during inflammation. Nat Med, 2010. 16(3): p. 279-85.
456. Suru, I.P.L. Blood Bag System CE 0120. 2010 [cited 2010 11.05.2010]; Features of the Blood Bag System including recipes of the available pre-filles storage buffers]. Available from: http://www.suru.com/frmblo.htm. 457. Beers, R.F., Jr. and I.W. Sizer, A spectrophotometric method for measuring the breakdown of hydrogen peroxide by catalase. J Biol Chem, 1952. 195(1): p. 133-40. 458. Danon, D. and V. Marikovsky, Determination of Density Distribution of Red Cell Population. J Lab Clin Med, 1964. 64: p. 668-74. Chapter 7: References 155
459. Waugh, R.E., M. Narla, C.W. Jackson, T.J. Mueller, T. Suzuki, and G.L. Dale, Rheologic properties of senescent erythrocytes: loss of surface area and volume with red blood cell age. Blood, 1992. 79(5): p. 1351-8. 460. Salvo, G., P. Caprari, P. Samoggia, G. Mariani, and A.M. Salvati, Human erythrocyte separation according to age on a discontinuous "Percoll" density gradient. Clin Chim Acta, 1982. 122(2): p. 293-300. 461. Bosch, F.H., J.M. Werre, B. Roerdinkholder-Stoelwinder, T.H. Huls, F.L. Willekens, and M.R. Halie, Characteristics of red blood cell populations fractionated with a combination of counterflow centrifugation and Percoll separation. Blood, 1992. 79(1): p. 254-60. 462. Rennie, C.M., S. Thompson, A.C. Parker, and A. Maddy, Human erythrocyte fraction in "Percoll" density gradients. Clin Chim Acta, 1979. 98(1-2): p. 119-25. 463. Alderman, E.M., H.H. Fudenberg, and R.E. Lovins, Binding of immunoglobulin classes to subpopulations of human red blood cells separated by density-gradient centrifugation. Blood, 1980. 55(5): p. 817-22. 464. Rinalducci, S., G.M. D'Amici, B. Blasi, and L. Zolla, Oxidative stress-dependent oligomeric status of erythrocyte peroxiredoxin II (PrxII) during storage under standard blood banking conditions. Biochimie, 2011. 93(5): p. 845-53. 465. Winterbourn, C.C., Oxidative reactions of hemoglobin. Methods Enzymol, 1990. 186: p. 265-72. 466. Ladner, C.L., J. Yang, R.J. Turner, and R.A. Edwards, Visible fluorescent detection of proteins in polyacrylamide gels without staining. Analytical biochemistry, 2004. 326(1): p. 13-20. 467. Beutler, E., Glutathione, in Red Cell Metabolism. 1984, Grune & Stratton, Inc: Orlando, FL 32887. p. 132-134.
468. Harwood, D.T., A.J. Kettle, S. Brennan, and C.C. Winterbourn, Simultaneous determination of reduced glutathione, glutathione disulphide and glutathione sulphonamide in cells and physiological fluids by isotope dilution liquid chromatography-tandem mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci, 2009. 877(28): p. 3393-9. 469. Dodge, J.T., C. Mitchell, and D.J. Hanahan, The preparation and chemical characteristics of hemoglobin-free ghosts of human erythrocytes. Archives of biochemistry and biophysics, 1963. 100: p. 119-30. 470. Jacobson, B.S. and D. Branton, Plasma membrane: rapid isolation and exposure of the cytoplasmic surface by use of positively charged beads. Science, 1977. 195(4275): p. 302-4.
471. Steck, T.L. and J.A. Kant, Preparation of impermeable ghosts and inside-out vesicles from human erythrocyte membranes. Methods Enzymol, 1974. 31: p. 172-80. 472. Lowry, O.H., N.J. Rosebrough, A.L. Farr, and R.J. Randall, Protein measurement with the Folin phenol reagent. J Biol Chem, 1951. 193(1): p. 265-75. 473. Peterson, G.L., Review of the Folin phenol protein quantitation method of Lowry, Rosebrough, Farr and Randall. Analytical biochemistry, 1979. 100(2): p. 201-20. Chapter 7: References 156
474. Hall, J.E. and A.C. Guyton, Transport of Oxygen and Carbon Dioxide in Blood and Tissue Fluids, in Guyton and Hall textbook of medical physiology. 2011, Saunders/Elsevier: Philadelphia, Pa. p. 502-512. 475. Pasquali, M.L., N. , Newborn screening and inborn errors of metabolism, in Tietz textbook of clinical chemistry and molecular diagnostics, N.W. Tietz, et al., Editors. 2012, Elsevier Saunders: St. Louis, Mo. p. 2045-2082. 476. Tasker, R.C., D. Lutman, and M.J. Peters, Hyperventilation in severe diabetic ketoacidosis. Pediatr Crit Care Med, 2005. 6(4): p. 405-11. 477. Kraut, J.A. and N.E. Madias, Metabolic acidosis: pathophysiology, diagnosis and management. Nat Rev Nephrol, 2010. 6(5): p. 274-285. 478. Gabow, P.A., W.D. Kaehny, P.V. Fennessey, S.I. Goodman, P.A. Gross, and R.W. Schrier, Diagnostic Importance of an Increased Serum Anion Gap. New England Journal of Medicine, 1980. 303(15): p. 854-858.
479. Decoursey, T.E. and E. Ligeti, Regulation and termination of NADPH oxidase activity. Cell Mol Life Sci, 2005. 62(19-20): p. 2173-93. 480. O'Donnell, B.V., D.G. Tew, O.T. Jones, and P.J. England, Studies on the inhibitory mechanism of iodonium compounds with special reference to neutrophil NADPH oxidase. The Biochemical journal, 1993. 290 ( Pt 1): p. 41-9. 481. Kettle, A.J., C.A. Gedye, and C.C. Winterbourn, Mechanism of inactivation of myeloperoxidase by 4-aminobenzoic acid hydrazide. Biochem J, 1997. 321 ( Pt 2): p. 503-8. 482. Stendahl, O., B.I. Coble, C. Dahlgren, J. Hed, and L. Molin, Myeloperoxidase modulates the phagocytic activity of polymorphonuclear neutrophil leukocytes. Studies with cells from a myeloperoxidase-deficient patient. J Clin Invest, 1984. 73(2): p. 366-73.
483. Nauseef, W.M., J.A. Metcalf, and R.K. Root, Role of myeloperoxidase in the respiratory burst of human neutrophils. Blood, 1983. 61(3): p. 483-92. 484. Forehand, N.a.J., The Metabolic Basis of Inherited Disease. Ann Intern Med, 1989. 114(6): p. 2779-2801. 485. Kutter, D., P. Devaquet, G. Vanderstocken, J.M. Paulus, V. Marchal, and A. Gothot, Consequences of total and subtotal myeloperoxidase deficiency: risk or benefit ? Acta Haematol, 2000. 104(1): p. 10-5. 486. DeLeo, F.R., J. Renee, S. McCormick, M. Nakamura, M. Apicella, J.P. Weiss, and W.M. Nauseef, Neutrophils exposed to bacterial lipopolysaccharide upregulate NADPH oxidase assembly. J Clin Invest, 1998. 101(2): p. 455-63.
487. Hailman, E., H.S. Lichenstein, M.M. Wurfel, D.S. Miller, D.A. Johnson, M. Kelley, L.A. Busse, M.M. Zukowski, et al., Lipopolysaccharide (LPS)-binding protein accelerates the binding of LPS to CD14. The Journal of experimental medicine, 1994. 179(1): p. 269-77. 488. Kouoh, F., B. Gressier, M. Luyckx, C. Brunet, T. Dine, M. Cazin, and J.C. Cazin, Antioxidant properties of albumin: effect on oxidative metabolism of human neutrophil granulocytes. Farmaco, 1999. 54(10): p. 695-9. Chapter 7: References 157
489. Vissers, M.C., W.A. Day, and C.C. Winterbourn, Neutrophils adherent to a nonphagocytosable surface (glomerular basement membrane) produce oxidants only at the site of attachment. Blood, 1985. 66(1): p. 161-6. 490. Bedard, K. and K.H. Krause, The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology. Physiol Rev, 2007. 87(1): p. 245-313. 491. Lundqvist, H., A. Karlsson, P. Follin, C. Sjolin, and C. Dahlgren, Phagocytosis following translocation of the neutrophil b-cytochrome from the specific granule to the plasma membrane is associated with an increased leakage of reactive oxygen species. Scand J Immunol, 1992. 36(6): p. 885-91. 492. Achterberg, A., M. Ackermann, J. Adams, J. Ahrens, K. Andeen, D.W. Atlee, J.N. Bahcall, X. Bai, et al., Limits on the high-energy gamma and neutrino fluxes from the SGR 1806-20 giant flare of 27 December 2004 with the AMANDA-II detector. Physical review letters, 2006. 97(22): p. 221101.
493. Agar, N.S., S.M. Sadrzadeh, P.E. Hallaway, and J.W. Eaton, Erythrocyte catalase. A somatic oxidant defense? J Clin Invest, 1986. 77(1): p. 319-21. 494. Segal, A.W., How neutrophils kill microbes. Annu Rev Immunol, 2005. 23: p. 197- 223. 495. Nathan, C., Neutrophils and immunity: challenges and opportunities. Nat Rev Immunol, 2006. 6(3): p. 173-82. 496. Kono, H. and K.L. Rock, How dying cells alert the immune system to danger. Nat Rev Immunol, 2008. 8(4): p. 279-89. 497. McDonald, B., K. Pittman, G.B. Menezes, S.A. Hirota, I. Slaba, C.C. Waterhouse, P.L. Beck, D.A. Muruve, et al., Intravascular danger signals guide neutrophils to sites of sterile inflammation. Science. 330(6002): p. 362-6.
498. Bone, R.C., C.J. Grodzin, and R.A. Balk, Sepsis: a new hypothesis for pathogenesis of the disease process. Chest, 1997. 112(1): p. 235-43. 499. George, A., S. Pushkaran, D.G. Konstantinidis, S. Koochaki, P. Malik, N. Mohandas, Y. Zheng, C.H. Joiner, et al., Erythrocyte NADPH oxidase activity modulated by Rac GTPases, PKC, and plasma cytokines contributes to oxidative stress in sickle cell disease. Blood, 2013. 121(11): p. 2099-107. 500. Victor, V.M. and M. De la Fuente, Immune cells redox state from mice with endotoxin-induced oxidative stress. Involvement of NF-kappaB. Free Radic Res, 2003. 37(1): p. 19-27. 501. Victor, V.M., M. Rocha, and M. De la Fuente, Immune cells: free radicals and antioxidants in sepsis. Int Immunopharmacol, 2004. 4(3): p. 327-47.
502. Yang, C.S., D.S. Lee, C.H. Song, S.J. An, S. Li, J.M. Kim, C.S. Kim, D.G. Yoo, et al., Roles of peroxiredoxin II in the regulation of proinflammatory responses to LPS and protection against endotoxin-induced lethal shock. The Journal of experimental medicine, 2007. 204(3): p. 583-94. 503. Koch, C.G., L. Li, D.I. Sessler, P. Figueroa, G.A. Hoeltge, T. Mihaljevic, and E.H. Blackstone, Duration of red-cell storage and complications after cardiac surgery. N Engl J Med, 2008. 358(12): p. 1229-39. 504. Wang, D., J. Sun, S.B. Solomon, H.G. Klein, and C. Natanson, Transfusion of older stored blood and risk of death: a meta-analysis. Transfusion, 2012. 52(6): p. 1184-95. Chapter 7: References 158
505. Raat, N.J., F. Berends, A.J. Verhoeven, D. de Korte, and C. Ince, The age of stored red blood cell concentrates at the time of transfusion. Transfus Med, 2005. 15(5): p. 419-23. 506. AABB, in Standards for blood bank transfusion services, A.A.o.B. Banks, Editor. 2012, American Association of Blood Banks: Washington, DC. p. 25-26. 507. Walker, R., Blood Components: Preparation, Storage and Shipment, in Technical Manual, A.A.o.B. Banks, Editor. 1993, American Association of Blood Banks: Bethesda. p. 51-52. 508. Kor, D.J., C.M. Van Buskirk, and O. Gajic, Red blood cell storage lesion. Bosn J Basic Med Sci, 2009. 9 Suppl 1: p. 21-7. 509. Ellman, G.L., Tissue sulfhydryl groups. Archives of biochemistry and biophysics, 1959. 82(1): p. 70-7.
510. Bayer, S.B., G. Maghzal, R. Stocker, M.B. Hampton, and C.C. Winterbourn, Neutrophil-mediated oxidation of erythrocyte peroxiredoxin 2 as a potential marker of oxidative stress in inflammation. FASEB J, 2013. 27(8): p. 3315-22. 511. Dumont, L.J. and J.P. AuBuchon, Evaluation of proposed FDA criteria for the evaluation of radiolabeled red cell recovery trials. Transfusion, 2008. 48(6): p. 1053- 60. 512. Klein, H.G. and D.J. Anstee, Mollison’s Blood Transfusion in Clinical Medicine. 2005. 513. Constantinescu, A., U. Pick, G.J. Handelman, N. Haramaki, D. Han, M. Podda, H.J. Tritschler, and L. Packer, Reduction and transport of lipoic acid by human erythrocytes. Biochem Pharmacol, 1995. 50(2): p. 253-61. 514. Arner, E.S., J. Nordberg, and A. Holmgren, Efficient reduction of lipoamide and lipoic acid by mammalian thioredoxin reductase. Biochemical and biophysical research communications, 1996. 225(1): p. 268-74. 515. Packer, L., E.H. Witt, and H.J. Tritschler, alpha-Lipoic acid as a biological antioxidant. Free Radic Biol Med, 1995. 19(2): p. 227-50. 516. Rushworth, G.F. and I.L. Megson, Existing and potential therapeutic uses for N- acetylcysteine: The need for conversion to intracellular glutathione for antioxidant benefits. Pharmacol Ther, 2013.
517. Reas, H.W., The Use of N-Acetylcysteine in the Treatment of Cystic Fibrosis. J Pediatr, 1964. 65: p. 542-57. 518. Hess, J.R., N. Rugg, A.D. Joines, J.F. Gormas, P.G. Pratt, E.B. Silberstein, and T.J. Greenwalt, Buffering and dilution in red blood cell storage. Transfusion, 2006. 46(1): p. 50-4. 519. Lutz, H.U., P. Stammler, S. Fasler, M. Ingold, and J. Fehr, Density separation of human red blood cells on self forming Percoll gradients: correlation with cell age. Biochimica et biophysica acta, 1992. 1116(1): p. 1-10. 520. Harper, V.M., J.Y. Oh, R. Stapley, M.B. Marques, L. Wilson, S. Barnes, C.W. Sun, T. Townes, et al., Peroxiredoxin-2 recycling is inhibited during erythrocyte storage. Antioxidants & redox signaling, 2015. 22(4): p. 294-307. Chapter 7: References 159
521. Rinalducci, S., G.M. D'Amici, B. Blasi, S. Vaglio, G. Grazzini, and L. Zolla, Peroxiredoxin-2 as a candidate biomarker to test oxidative stress levels of stored red blood cells under blood bank conditions. Transfusion. 51(7): p. 1439-49. 522. Meryman, H.T., M.L. Hornblower, and R.L. Syring, Prolonged storage of red cells at 4 degrees C. Transfusion, 1986. 26(6): p. 500-5. 523. Luten, M., B. Roerdinkholder-Stoelwinder, H.J. Bost, and G.J. Bosman, Survival of the fittest?--survival of stored red blood cells after transfusion. Cell Mol Biol (Noisy- le-grand), 2004. 50(2): p. 197-203. 524. Hogman, C.F. and H.T. Meryman, Storage parameters affecting red blood cell survival and function after transfusion. Transfus Med Rev, 1999. 13(4): p. 275-96. 525. D'Alessandro, A., B. Blasi, G.M. D'Amici, C. Marrocco, and L. Zolla, Red blood cell subpopulations in freshly drawn blood: application of proteomics and metabolomics to a decades-long biological issue. Blood Transfus, 2013. 11(1): p. 75-87.
526. Tiffert, T., N. Daw, Z. Etzion, R.M. Bookchin, and V.L. Lew, Age decline in the activity of the Ca2+-sensitive K+ channel of human red blood cells. J Gen Physiol, 2007. 129(5): p. 429-36. 527. Tuo, W.W., D. Wang, W.J. Liang, and Y.X. Huang, How cell number and cellular properties of blood-banked red blood cells of different cell ages decline during storage. PLoS One, 2014. 9(8): p. e105692. 528. Jones, W., X. Li, Z.C. Qu, L. Perriott, R.R. Whitesell, and J.M. May, Uptake, recycling, and antioxidant actions of alpha-lipoic acid in endothelial cells. Free Radic Biol Med, 2002. 33(1): p. 83-93. 529. Holmgren, A., Thioredoxin catalyzes the reduction of insulin disulfides by dithiothreitol and dihydrolipoamide. J Biol Chem, 1979. 254(19): p. 9627-32.
530. Gleason, F.K. and A. Holmgren, Thioredoxin and related proteins in procaryotes. FEMS Microbiol Rev, 1988. 4(4): p. 271-97. 531. Werre, J.M., F.L. Willekens, F.H. Bosch, L.D. de Haans, S.G. van der Vegt, A.G. van den Bos, and G.J. Bosman, The red cell revisited--matters of life and death. Cell Mol Biol (Noisy-le-grand), 2004. 50(2): p. 139-45. 532. Greenwalt, T.J., The how and why of exocytic vesicles. Transfusion, 2006. 46(1): p. 143-52.
533. Rubin, O., D. Crettaz, G. Canellini, J.D. Tissot, and N. Lion, Microparticles in stored red blood cells: an approach using flow cytometry and proteomic tools. Vox Sang, 2008. 95(4): p. 288-97. 534. Allan, D., P. Thomas, and A.R. Limbrick, The isolation and characterization of 60 nm vesicles ('nanovesicles') produced during ionophore A23187-induced budding of human erythrocytes. Biochem J, 1980. 188(3): p. 881-7. 535. Antonelou, M.H., I.S. Papassideri, F.J. Karababa, D.J. Stravopodis, A. Loutradi, and L.H. Margaritis, Defective organization of the erythroid cell membrane in a novel case of congenital anemia. Blood Cells Mol Dis, 2003. 30(1): p. 43-54. 536. Margetis, P., M. Antonelou, F. Karababa, A. Loutradi, L. Margaritis, and I. Papassideri, Physiologically important secondary modifications of red cell membrane in hereditary spherocytosis-evidence for in vivo oxidation and lipid rafts protein variations. Blood Cells Mol Dis, 2007. 38(3): p. 210-20. Chapter 7: References 160
537. Nagababu, E., J.G. Mohanty, J.S. Friedman, and J.M. Rifkind, Role of peroxiredoxin-2 in protecting RBCs from hydrogen peroxide-induced oxidative stress. Free Radic Res, 2013. 47(3): p. 164-71. 538. Haematology Erythrocyte Ghost., in Dorland's Medical Dictionary for Health Consumers. 2015, Saunders, an imprint of Elsevier. 539. Haematology Erythrocyte Ghost, in Farlex Partner Medical Dictionary. 2015. 540. Stokke, B.T., A. Mikkelsen, and A. Elgsaeter, Spectrin, human erythrocyte shapes, and mechanochemical properties. Biophysical journal, 1986. 49(1): p. 319-27. 541. Bennett, V. and D. Branton, Selective association of spectrin with the cytoplasmic surface of human erythrocyte plasma membranes. Quantitative determination with purified (32P)spectrin. J Biol Chem, 1977. 252(8): p. 2753-63. 542. Marchesi, V.T. and E. Steers, Jr., Selective solubilization of a protein component of the red cell membrane. Science, 1968. 159(3811): p. 203-4. 543. Shinar, E. and E.A. Rachmilewitz, Oxidative denaturation of red blood cells in thalassemia. Semin Hematol, 1990. 27(1): p. 70-82. 544. Rice-Evans, C. and E. Baysal, Iron-mediated oxidative stress in erythrocytes. Biochem J, 1987. 244(1): p. 191-6. 545. Siciliano, A., F. Turrini, M. Bertoldi, A. Matte, A. Pantaleo, O. Olivieri, and L. De Franceschi, Deoxygenation affects tyrosine phosphoproteome of red cell membrane from patients with sickle cell disease. Blood Cells Mol Dis, 2010. 44(4): p. 233-42. 546. Pantaleo, A., E. Ferru, F. Carta, F. Mannu, L.F. Simula, A. Khadjavi, P. Pippia, and F. Turrini, Irreversible AE1 tyrosine phosphorylation leads to membrane vesiculation in G6PD deficient red cells. PLoS One, 2011. 6(1): p. e15847.
547. Pantaleo, A., E. Ferru, G. Giribaldi, F. Mannu, F. Carta, A. Matte, L. de Franceschi, and F. Turrini, Oxidized and poorly glycosylated band 3 is selectively phosphorylated by Syk kinase to form large membrane clusters in normal and G6PD-deficient red blood cells. Biochem J, 2009. 418(2): p. 359-67. 548. Bordin, L., A.M. Brunati, A. Donella-Deana, B. Baggio, A. Toninello, and G. Clari, Band 3 is an anchor protein and a target for SHP-2 tyrosine phosphatase in human erythrocytes. Blood, 2002. 100(1): p. 276-82.
549. Maddock, K.R., E. Huff-Lonergan, L.J. Rowe, and S.M. Lonergan, Effect of pH and ionic strength on mu- and m-calpain inhibition by calpastatin. J Anim Sci, 2005. 83(6): p. 1370-6. 550. Wendt, A., V.F. Thompson, and D.E. Goll, Interaction of calpastatin with calpain: a review. Biol Chem, 2004. 385(6): p. 465-72.
551. Sharma, S., S.M. Zingde, and S.M. Gokhale, Identification of human erythrocyte cytosolic proteins associated with plasma membrane during thermal stress. J Membr Biol, 2013. 246(8): p. 591-607. 552. Acker, J.I., F. Annoni, I. Carreno, G.V. Hahn, and C.R. Medeiros, [Midwives and care in childbirth]. Rev Bras Enferm, 2006. 59(5): p. 647-51. 553. Bosman, G.J., F.L. Willekens, and J.M. Werre, Erythrocyte aging: a more than superficial resemblance to apoptosis? Cell Physiol Biochem, 2005. 16(1-3): p. 1-8. Chapter 7: References 161
554. Low, P.S., M.A. Westfall, D.P. Allen, and K.C. Appell, Characterization of the reversible conformational equilibrium of the cytoplasmic domain of erythrocyte membrane band 3. J Biol Chem, 1984. 259(21): p. 13070-6. 555. Salamino, F., R. De Tullio, P. Mengotti, P.L. Viotti, E. Melloni, and S. Pontremoli, Site-directed activation of calpain is promoted by a membrane-associated natural activator protein. Biochem J, 1993. 290 ( Pt 1): p. 191-7. 556. Aran, M., D. Ferrero, A. Wolosiuk, S. Mora-Garcia, and R.A. Wolosiuk, ATP and Mg2+ promote the reversible oligomerization and aggregation of chloroplast 2-Cys peroxiredoxin. J Biol Chem. 286(26): p. 23441-51. 557. Marrocco, C., V. Pallotta, A. D'Alessandro, G. Alves, and L. Zolla, Red blood cell populations and membrane levels of peroxiredoxin 2 as candidate biomarkers to reveal blood doping. Blood Transfus, 2012. 10 Suppl 2: p. s71-7. 558. Lauf, P.K., J. Bauer, N.C. Adragna, H. Fujise, A.M. Zade-Oppen, K.H. Ryu, and E. Delpire, Erythrocyte K-Cl cotransport: properties and regulation. The American journal of physiology, 1992. 263(5 Pt 1): p. C917-32. 559. Chasis, J.A. and N. Mohandas, Erythrocyte membrane deformability and stability: two distinct membrane properties that are independently regulated by skeletal protein associations. The Journal of cell biology, 1986. 103(2): p. 343-50. 560. Iwahara, S., H. Satoh, D.X. Song, J. Webb, A.L. Burlingame, Y. Nagae, and U. Muller-Eberhard, Purification, characterization, and cloning of a heme-binding protein (23 kDa) in rat liver cytosol. Biochemistry, 1995. 34(41): p. 13398-406. 561. Runge-Morris, M. and R.F. Novak, Effects of phenelzine and hydralazine on hydrogen peroxide production and proteolysis in human red blood cells. The Journal of pharmacology and experimental therapeutics, 1993. 267(3): p. 1401-6. 562. Mellgren, R.L. and T. Murachi, Intracellular calcium-dependent proteolysis. 1990: CRC Press. 563. Sigma-Aldrich, Protease inhibition and detection, in Lofe Science Bio Files, S.-. Aldrich, Editor. 2009, Sigma-Alrdich. 564. Cox, A.G., C.C. Winterbourn, and M.B. Hampton, Measuring the redox state of cellular peroxiredoxins by immunoblotting. Methods Enzymol, 2010. 474: p. 51-66. 565. Poynton, R.A. and M.B. Hampton, Peroxiredoxins as biomarkers of oxidative stress. Biochimica et biophysica acta, 2014. 1840(2): p. 906-12. 566. Kumar, V., N. Kitaeff, M.B. Hampton, M.B. Cannell, and C.C. Winterbourn, Reversible oxidation of mitochondrial peroxiredoxin 3 in mouse heart subjected to ischemia and reperfusion. FEBS letters, 2009. 583(6): p. 997-1000.
567. Mountford, J.C. and M. Turner, In vitro production of red blood cells. Transfus Apher Sci, 2011. 45(1): p. 85-9. 568. Button, L.N., W.C. DeWolf, P.E. Newburger, M.S. Jacobson, and S.V. Kevy, The effects of irradiation on blood components. Transfusion, 1981. 21(4): p. 419-26. 569. Leitman, S.F. and P.V. Holland, Irradiation of blood products. Indications and guidelines. Transfusion, 1985. 25(4): p. 293-303. Chapter 7: References 162
570. Zito, E., E.P. Melo, Y. Yang, A. Wahlander, T.A. Neubert, and D. Ron, Oxidative protein folding by an endoplasmic reticulum-localized peroxiredoxin. Mol Cell, 2010. 40(5): p. 787-97. 571. Dietz, K.J., F. Horling, J. Konig, and M. Baier, The function of the chloroplast 2- cysteine peroxiredoxin in peroxide detoxification and its regulation. J Exp Bot, 2002. 53(372): p. 1321-9. 572. Konig, J., M. Baier, F. Horling, U. Kahmann, G. Harris, P. Schurmann, and K.J. Dietz, The plant-specific function of 2-Cys peroxiredoxin-mediated detoxification of peroxides in the redox-hierarchy of photosynthetic electron flux. Proc Natl Acad Sci U S A, 2002. 99(8): p. 5738-43. 573. Matte, A., A. Pantaleo, E. Ferru, F. Turrini, M. Bertoldi, F. Lupo, A. Siciliano, C. Ho Zoon, et al., The novel role of peroxiredoxin-2 in red cell membrane protein homeostasis and senescence. Free Radic Biol Med, 2014. 76C: p. 80-88.
574. Pallotta, V., A. D'Alessandro, S. Rinalducci, and L. Zolla, Native protein complexes in the cytoplasm of red blood cells. J Proteome Res, 2013. 12(7): p. 3529-46. 575. Moon, E.Y., Y.W. Noh, Y.H. Han, S.U. Kim, J.M. Kim, D.Y. Yu, and J.S. Lim, T lymphocytes and dendritic cells are activated by the deletion of peroxiredoxin II (Prx II) gene. Immunol Lett, 2006. 102(2): p. 184-90. 576. Takeda, A., T. Miyahara, A. Hachimori, and T. Samejima, The interactions of thiol compounds with porcine erythrocyte catalase. Journal of biochemistry, 1980. 87(2): p. 429-39. 577. Cheng, L.Y. and L.H. Lek, Inhibition of alcohol dehydrogenases by thiol compounds. FEBS letters, 1992. 300(3): p. 251-3. 578. Kutty, R.K. and M.D. Maines, Purification and characterization of biliverdin reductase from rat liver. Journal of Biological Chemistry, 1981. 256(8): p. 3956-3962.
579. Raucci, R., G. Colonna, E. Guerriero, F. Capone, M. Accardo, G. Castello, and S. Costantini, Structural and functional studies of the human selenium binding protein-1 and its involvement in hepatocellular carcinoma. Biochimica et biophysica acta, 2011. 1814(4): p. 513-22. 580. Zhang, D., A. Kiyatkin, J.T. Bolin, and P.S. Low, Crystallographic structure and functional interpretation of the cytoplasmic domain of erythrocyte membrane band 3. Blood, 2000. 96(9): p. 2925-33.
581. Allan, D. and P. Thomas, Ca2+-induced biochemical changes in human erythrocytes and their relation to microvesiculation. Biochem J, 1981. 198(3): p. 433-40. 582. Nicolay, J.P., P.J. Bentzen, M. Ghashghaeinia, T. Wieder, and F. Lang, Stimulation of erythrocyte cell membrane scrambling by amiodarone. Cell Physiol Biochem, 2007. 20(6): p. 1043-50. 583. Sopjani, M., M. Foller, and F. Lang, Gold stimulates Ca2+ entry into and subsequent suicidal death of erythrocytes. Toxicology, 2008. 244(2-3): p. 271-9. 584. Lew, V.L., Z. Etzion, and R.M. Bookchin, Dehydration response of sickle cells to sickling-induced Ca(++) permeabilization. Blood, 2002. 99(7): p. 2578-85. 585. Udden, M.M. and C.S. Patton, Butoxyacetic acid-induced hemolysis of rat red blood cells: effect of external osmolarity and cations. Toxicol Lett, 2005. 156(1): p. 81-93. Chapter 7: References 163
586. Halperin, J.A., C. Brugnara, and A. Nicholson-Weller, Ca2+-activated K+ efflux limits complement-mediated lysis of human erythrocytes. J Clin Invest, 1989. 83(5): p. 1466-71. 587. De Franceschi, L., A. Rivera, M.D. Fleming, M. Honczarenko, L.L. Peters, P. Gascard, N. Mohandas, and C. Brugnara, Evidence for a protective role of the Gardos channel against hemolysis in murine spherocytosis. Blood, 2005. 106(4): p. 1454-9. 588. Rivera, A., P. Jarolim, and C. Brugnara, Modulation of Gardos channel activity by cytokines in sickle erythrocytes. Blood, 2002. 99(1): p. 357-603. 589. Rivera, A., M.A. Rotter, and C. Brugnara, Endothelins activate Ca(2+)-gated K(+) channels via endothelin B receptors in CD-1 mouse erythrocytes. The American journal of physiology, 1999. 277(4 Pt 1): p. C746-54. 590. Dyrda, A., U. Cytlak, A. Ciuraszkiewicz, A. Lipinska, A. Cueff, G. Bouyer, S. Egee, P. Bennekou, et al., Local membrane deformations activate Ca2+-dependent K+ and anionic currents in intact human red blood cells. PLoS One. 5(2): p. e9447. 591. Oonishi, T., K. Sakashita, N. Ishioka, N. Suematsu, H. Shio, and N. Uyesaka, Production of prostaglandins E1 and E2 by adult human red blood cells. Prostaglandins Other Lipid Mediat, 1998. 56(2-3): p. 89-101. 592. Shartava, A., J. McIntyre, A.K. Shah, and S.R. Goodman, The Gardos channel is responsible for CDNB-induced dense sickle cell formation. American journal of hematology, 2000. 64(3): p. 184-9. 593. Muzyamba, M.C. and J.S. Gibson, Effect of 1-chloro-2,4-dinitrobenzene on K+ transport in normal and sickle human red blood cells. J Physiol, 2003. 547(Pt 3): p. 903-11. 594. Lisovskaya, I.L., I.M. Shcherbachenko, R.I. Volkova, and V.P. Tikhonov, Modulation of RBC volume distributions by oxidants (phenazine methosulfate and tert-butyl hydroperoxide): role of Gardos channel activation. Bioelectrochemistry, 2008. 73(1): p. 49-54. 595. Shcherbachenko, I.M., I.L. Lisovskaya, and V.P. Tikhonov, Oxidation-induced calcium-dependent dehydration of normal human red blood cells. Free Radic Res, 2007. 41(5): p. 536-45. 596. Gibson, J.S. and M.C. Muzyamba, Modulation of Gardos channel activity by oxidants and oxygen tension: effects of 1-chloro-2,4-dinitrobenzene and phenazine methosulphate. Bioelectrochemistry, 2004. 62(2): p. 147-52. 597. Woycechowsky, K.J. and R.T. Raines, The CXC motif: a functional mimic of protein disulfide isomerase. Biochemistry, 2003. 42(18): p. 5387-94. 598. Hoffman, J.F., W. Joiner, K. Nehrke, O. Potapova, K. Foye, and A. Wickrema, The hSK4 (KCNN4) isoform is the Ca2+-activated K+ channel (Gardos channel) in human red blood cells. Proc Natl Acad Sci U S A, 2003. 100(12): p. 7366-71. 599. Foller, M., M. Sopjani, S. Koka, S. Gu, H. Mahmud, K. Wang, E. Floride, E. Schleicher, et al., Regulation of erythrocyte survival by AMP-activated protein kinase. FASEB J, 2009. 23(4): p. 1072-80. 600. Lang, F. and S.M. Qadri, Mechanisms and significance of eryptosis, the suicidal death of erythrocytes. Blood Purif, 2012. 33(1-3): p. 125-30. Chapter 7: References 164
601. Lew, V.L., Membrane Transport in Red Cells Edited by J. C. Ellory and V. L. Lew. Pp. 470. Academic Press Inc. (London) Ltd. 1977. . Biochemical Education, 1978: p. 93-100. 602. Woon, L.A., J.W. Holland, E.P. Kable, and B.D. Roufogalis, Ca2+ sensitivity of phospholipid scrambling in human red cell ghosts. Cell Calcium, 1999. 25(4): p. 313- 20. 603. Foller, M., S.M. Huber, and F. Lang, Erythrocyte programmed cell death. IUBMB Life, 2008. 60(10): p. 661-8. 604. Berg, C.P., I.H. Engels, A. Rothbart, K. Lauber, A. Renz, S.F. Schlosser, K. Schulze- Osthoff, and S. Wesselborg, Human mature red blood cells express caspase-3 and caspase-8, but are devoid of mitochondrial regulators of apoptosis. Cell Death Differ, 2001. 8(12): p. 1197-206.
Chapter 8: Appendix 165
8 Appendix
8.1 Calculations used in Chapter 4
Quotation marks denote names of samples.
Figure 4.3
Band intensities of hyperoxidised Prx were measured.
Relative intensities for “7 days” = Intensity for “7 days” / intensity “0 days”
8.2 Calculations used in Chapter 5
Quotation marks denote names of samples.
Figure 5.1, Figure 5.2 and Figure 5.5
For simplification, the word Band3 is omitted from this explanation, even though all measurements include Band3 intensities. Quotation marks denote names of samples. Band intensities of Prx2 bands on Western blot and Spectrin on Coomassie stained gel were measured.
(1) Calculation of relative intensities of Spectrin in all samples: Relative intensity of “Spectrin in 5 µM Ca” = Intensity of “Spectrin in 5 µM Ca” / Intensity of “Spectrin in EGTA” (2) Application of results from 1 to the measured band intensities of Prx2: Normalised intensity of “Prx2 in 5 µM Ca” = Intensity of “Prx2 in 5 µM Ca” / Relative intensity of “Spectrin in 5 µM Ca” (3) Calculation of relative intensity of Prx2 in 5 µM Ca: Relative intensity of “Prx2 in 5 µM Ca” = Normalised intensity of “Prx2 in 5 µM Ca” / Intensity of “Prx2 in EGTA” (4) Calculation of relative units of membrane bound Prx2: Relative units of membrane bound “Prx2 in 5 µM Ca” = Sum of (Relative intensity of “Prx2 in 5 µM Ca”) / Number of individual experiments
Chapter 8: Appendix 166
Figure 5.9
Band intensities of Prx2 bands on Western blot and 5 ng running standard of Prx2 on Western blot were measured.
(1) Calculation of Prx2 bound per 100 µg membrane protein Prx2 bound = ( (Intensity of “Prx2 in 1 µg Prx2 added sample”) / (Intensity of “5 ng Prx2 standard”)) * 100
Figure 5.12 and Figure 5.13
Band intensities of Prx2 bands on Western blot were measured.
(1) Calculation of average binding for 10 µg Prx2 control Average “10 µg Prx2 control” = Sum of (Intensity for “10 µg Prx2 control” from all individual samples) / Number of individual samples (2) Calculation of relative intensities of membrane-bound Prx2 Relative intensities for “1 µg Prx2 + Ca” = Intensity for “1 µg Prx2 + Ca” / Average “10 µg Prx2 control” (3) Calculation of relative units of membrane-bound Prx2 Relative units of membrane bound “1 µg Prx2 + Ca” = Sum of (Relative intensity of “1 µg Prx2 + Ca”) / Number of individual experiments