UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The regulation of apoptosis by toxic shock syndrome toxin-1 and associated mutants Hung, Ryan Wei Yan 2004

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2004-931331.pdf [ 28.78MB ]
JSON: 831-1.0091873.json
JSON-LD: 831-1.0091873-ld.json
RDF/XML (Pretty): 831-1.0091873-rdf.xml
RDF/JSON: 831-1.0091873-rdf.json
Turtle: 831-1.0091873-turtle.txt
N-Triples: 831-1.0091873-rdf-ntriples.txt
Original Record: 831-1.0091873-source.json
Full Text

Full Text

THE R E G U L A T I O N OF APOPTOSIS B Y TOXIC SHOCK S Y N D R O M E TOXIN-1 A N D ASSOCIATED MUTANTS by R Y A N WEI Y A N H U N G B.Sc , The University of British Columbia, 1995 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Department of Experimental Medicine) We accept this thesis as conforming to the required standard i THE UNIVERSITY OF BRITISH C O L U M B I A M A Y 2004 © Ryan Wei Yan Hung, 2004 Abstract Toxic shock syndrome toxin-1 (TSST-1), the primary virulence factor in toxic shock syndrome (TSS), is an exotoxin produced by Staphylococcus ^aureus and a member of the superantigen family of T cell mitogens. Activation-induced cell death (AICD) of superantigen-reactive T cells is thought to play a role in modulating the hyperactive immune response and promoting host survival. While this form of apoptotic cell death occurs with the related staphylococcal enterotoxins (SE) A and B, conflicting reports exist for TSST-1. Therefore, the aim of this project was to investigate the regulation of apoptosis by TSST-1 and site-directed mutants. TSST-1 was found to exert either pro- or anti-apoptotic effects depending on the dose, thus resolving some of the contradictory reports in the literature. At commonly used concentrations (<1 /xM), TSST-1 was less potent at inducing apoptosis than SEB. Moreover, there was evidence of early induction of anti-apoptotic pathways. The glycine 31-to-arginine site-directed mutant of TSST-1 (G31R) was serendipitously found to be potently pro-apoptotic at doses as low as 10-nM. This apoptosis was very rapid and broadly directed, affecting T cells, monocytes and T lymphoma-derived cell lines. Surprisingly, high doses of TSST-1 (>1 juM) also exhibited similarly potent, non-specific apoptosis much like G31R. The mechanism behind this novel form of apoptosis was examined by dissecting the involvement of components of the extrinsic and intrinsic apoptotic pathways. When ligated, classical death receptors like Fas receptor utilize the extrinsic pathway. In contrast, conditions unfavourable to cell survival such as growth factor withdrawal, ionizing radiation, and other cell damage can converge at the level of the mitochondria to trigger death through the intrinsic pathway. G31R-induced apoptosis was dependent on the intrinsic, but not the extrinsic, pathway. ii Because TSST-1 is not known to possess direct enzymatic effects that could account for its cytotoxicity, the most likely site of action for the induction of apoptosis by G31R and high-dose TSST-1 was considered to be a non-classical death receptor coupled to the intrinsic pathway, a phenomenon previously described. However, the upstream pathways remain unclear, so that other possibilities may include direct cell membrane disruption and/or induction of reactive oxygen species. Either way, this previously unknown apoptotic activity of TSST-1 may be relevant in pathogenesis, acting as a local immunosuppressant at infection sites by deleting lymphocytes and monocytes that transit into the region. With identification of the apical mechanism for apoptosis in this system, possibilities exist for the therapeutic use of G31R or its derivatives in the treatment of tumours and autoimmune diseases. iii Table of Contents Abstract ii Table of Contents iv Table of Contents (Detailed) v List of Tables viii List of Figures viii List of Abbreviations x Preface xii Acknowledgements xiii Chapter 1 1 Introduction 1 Chapter 2 42 Materials and Methods 42 Chapter 3 60 Apoptosis Induced by TSST-1 Compared with SEB 60 Chapter 4 79 Characterization of Apoptosis Induced by the G31R Mutant of TSST-1 79 Chapter 5 112 The Extrinsic Apoptotic Pathway Is Not A Predominant Route for G31R-Mediated Apoptosis 112 Chapter 6 127 Intrinsic Pathway is the Predominant Pathway for G31R-Mediated Apoptosis 127 Chapter 7 150 G31R-Induced Apoptosis: A Working Model, Significance and Conclusions 150 Appendix A 163 Apoptotic Activity of TSST-1 Mutants: Structure and Function 163 Appendix B 179 Publications 179 Bibliography 181 iv Table of Contents (Detailed) Abstract ii Table of Contents iv Table of Contents (Detailed) v List of Tables viii List of Figures viii List of Abbreviations x Preface xii Acknowledgements xiii Chapter 1 1 Introduction 1 1.1. Apoptosis 1 1.1.1. General features of apoptosis 1 1.1.2. Apoptosis in health and disease 4 1.1.3. Molecular pathways of apoptosis 6 Caspases 8 Extrinsic pathway for apoptosis 11 Intrinsic pathway for apoptosis 12 Alternative pathways for apoptosis 15 1.1.4. Regulation of apoptosis 18 1.1.5. Approaches to assaying apoptosis 20 1.2. Superantigens 23 1.2.1. Superantigens in disease 24 1.2.2. Structure-function relationships in superantigen activity 28 1.2.3. T cell responses to superantigens 32 1.2.4. Secondary effects of superantigen on other cell types 36 1.3. Connection between apoptosis and superantigens 37 1.4. Objectives and specific aims 41 Chapter 2 42 Materials and Methods 42 2.1. Generation of TS ST-1 and mutant toxins 42 2.2. Purification of human peripheral blood mononuclear cells 44 2.3. Growth of MOLT-16 and Jurkat cell lines 44 2.4. Isolation of murine splenocytes 44 2.5. Flow cytometry 45 2.6. Characterization of apoptosis 46 2.6.1. Hypodiploid population by propidium iodide 46 2.6.2. Membrane integrity by 7-aminoactinomycin D 46 2.6.3. Phosphatidylserine exposure by Annexin V 47 2.6.4. DNA fragmentation by TdT incorporation 47 2.6.5. Caspase 3 activation 48 2.6.6. Detection of apoptotic changes by microscopy 48 2.7. Testing of extrinsic mechanisms for apoptosis 49 2.7.1. Receptor blocking antibodies 49 2.7.2. Caspase inhibition 50 2.8. Testing of intrinsic mechanisms for apoptosis 50 2.8.1. CD47 neutralization 51 v 2.8.2. Mitochondrial transmembrane potential determination 51 2.8.3. Immunoblot analysis of mitochondrial and apoptosis proteins 51 2.8.4. Fluorescence microscopy for Bax, cytochrome c and AIF translocation 53 2.8.5. Flow cytometry for B ax activation 54 2.9. Anti-apoptotic pathways 54 2.9.1. Immunoblot analysis for Akt activation and caspase 3 expression 55 2.9.2. Flow cytometry for Akt activation 55 2.10. Statistical Analyses 55 2.11. Three Dimensional Protein Structures 56 Chapter 3 60 Apoptosis Induced by TSST-1 Compared with SEB 60 3.1. Introduction 60 3.2. Results 60 3.2.1. TSST-1-induced apoptosis: dose changes everything 60 3.2.2. Apoptosis: time response to TSST-1 and SEB 65 3.2.3. Fas expression does not correlate with TSST-1-induced apoptosis 65 3.2.4. TSST-1 activates the Akt pro-survival pathway 68 3.2.5. TSST-1 downregulates caspase 3 69 3.2.6. High dose TSST-1 induced apoptosis is specific ' 72 3.3. Discussion and conclusions 74 Chapter 4 79 Characterization of Apoptosis Induced by the G31R Mutant of TSST-1 79 4.1. Introduction 79 4.2. Results 81 4.2.1. G3 IR causes apoptosis in a dose-dependent manner, corresponding to its decreased activity in mitogenesis 81 4.2.2. G3 lR-induced apoptosis is rapid 81 4.2.3. Characteristics of G3lR-induced apoptosis in human PBMC 84 G31R induces apoptotic cellular morphology 84 G31R induces chromatin condensation 88 G31R induces DNA fragmentation 90 G3IR induces membrane permeabilization 93 G31R induces phosphatidylserine inversion 93 4.2.4. G31R-induced apoptosis is specific 94 4.2.5. G3lR-induced apoptosis requires minimal exposure 97 4.2.6. G3 lR-induced apoptosis affects diverse cell subpopulations 99 4.2.7. G3 IR causes apoptosis in cell types other than human PBMC 102 4.2.8. TSST-1 does not suppress G31R-induced apoptosis 105 4.3. Discussion and conclusion 107 Chapter 5 112 The Extrinsic Apoptotic Pathway Is Not A Predominant Route for G31R-Mediated Apoptosis ....112 5.1. Introduction 112 5.2. Results 114 5.2.1. G31R-induced apoptosis is not mediated by soluble factors 114 5.2.2. G31R-induced apoptosis does not depend upon Fas ligand 117 5.2.3. Caspase 8 inhibition fails to prevent G31R-induced apoptosis 119 5.2.4. G31R causes caspase 3 activation but caspase 3 inhibition only minimally reduces G31R-mediated apoptosis 121 vi 5.3. Discussion and conclusions 125 Chapter 6 127 Intrinsic Pathway is the Predominant Pathway for G31R-Mediated Apoptosis 127 6.1. Introduction 127 6.2. Results 130 6.2.1. G31R does not primarily signal apoptosis through CD47 130 6.2.2. G3 IR mediates Bax activation and translocation to the mitochondria 132 6.2.3. G31R causes loss of mitochondrial transmembrane potential 137 6.2.4. G31R causes cytochrome c and AIF translocation 139 6.2.5. G31R-induced apoptosis is mostly independent of caspase 9 145 6.3. Discussion of data and conclusions 147 Chapter 7 150 G31R-Induced Apoptosis: A Working Model, Significance and Conclusions 150 7.1. Introduction: the possibilities 150 7.2. Evidence for a putative non-classical death receptor 153 7.3. Precedence for important low-affinity interactions 154 7.4. Properties of a putative death receptor for G3 IR 155 7.5. Concentration dependent effects of TSST-1 in the pathogenesis of staphylococcal toxic shock syndrome 155 7.6. Possible applications of G31R in cytotoxic therapies 159 7.7. Potential application of G3 IR for research in cell biology 160 7.8. Future directions 160 7.9. Conclusion 162 Appendix A 163 Apoptotic Activity of TSST-1 Mutants: Structure and Function 163 8.1. Introduction 163 8.1.1. Mutations around MHC class II binding sites 165 8.1.2. Mutations around TCR interaction site 170 8.1.3. Mutation not related to MHC class II or TCR interaction sites 170 8.2. Experimental Approach 172 8.3. Results 172 8.3.1. All TSST-1 mutants induce apoptosis to varying extents, and there is no correlation between apoptotic and superantigenic activities 172 8.4. Conclusions and discussion of data 177 Appendix B 179 Publications 179 9.1. Papers 179 9.2. Abstracts 179 Bibliography 181 vii List of Tables Table 2.1: List of Reagents 57 Table 8.1: Summary of TSST-1 mutants and their apoptotic vs. superantigenic activities 176 List of Figures Figure 1.1: Apoptosis vs. Necrosis 3 Figure 1.2: Apoptosis in Health and Disease 5 Figure 1.3: Schematic of Apoptotic Pathways 7 Figure 1.4: Two Pathways for Apoptosis 10 Figure 1.5: Regulation of Apoptosis 17 Figure 1.6: 3D Ribbon Structure of TSST-1 27 Figure 1.7: TSST-1 :HLA-DR Contact 31 Figure 1.8: T Cell Responses to Superantigen 35 Figure 1.9: Two Routes to Activated T Cell Death 40 Figure 2.1: Preparation of Site-Directed Mutant Toxins 43 Figure 3.1: Apoptosis Dose Response for TSST-1 62 Figure 3.2: Apoptosis Dose Response for High TSST-1 Concentrations 63 Figure 3.3: Apoptosis Time Response for TSST-1 64 Figure 3.4: Fas Expression with TSST-1 and SEB 67 Figure 3.5: TSST-1-induced Akt Activation 70 Figure 3.6: Caspase 3 Expression 71 Figure 3.7: Specificity of TSST-1-induced Apoptosis 73 Figure 3.8: Role of Anti-Apoptotic TSST-1 Activity in Pathogenesis 78 Figure 4.1: T Cell Proliferation vs. Apoptosis 80 Figure 4.2: Apoptosis Dose Response for G31R 82 Figure 4.3: Apoptosis Time Response for G31R 83 Figure 4.4: Kinetics of G31R Response 86 Figure 4.5: Cell Morphology with G31R 87 Figure 4.6: Annexin V-FITC and Hoechst 33258 Staining 89 Figure 4.7: Propidium Iodide Staining 91 Figure 4.8: TUNEL Staining 92 Figure 4.9: Specificity of G31R-induced Apoptosis 96 Figure 4.10: Exposure Requirement for G31R-induced Apoptosis 98 Figure 4.11: Apoptosis in Human PBMC Subpopulations 101 Figure 4.12: Apoptosis in Cell Lines 103 Figure 4.13: Apoptosis in Murine Splenocytes 104 Figure 4.14: TSST-1 Effect on G31R-induced Apoptosis 106 Figure 4.15: Mechanisms for G31R and/or TSST-1 Induced Apoptosis I l l Figure 5.1: Involvement of the Extrinsic Pathway 113 Figure 5.2: Dependency on Soluble Factor for Apoptosis 116 Figure 5.3: Fas Ligand Neutralization 118 Figure 5.4: Caspase 8 Inhibition 120 Figure 5.5: Caspase 3 Expression .' 123 Figure 5.6: Caspase 3 Inhibition 124 viii Figure 6.1: Involvement of the Intrinsic Pathway 129 Figure 6.2: CD47 Involvement in G31R-induced Apoptosis 131 Figure 6.3: Bax Involvement in G31R-induced Apoptosis 135 Figure 6.4: Bax Localization by Fluorescence Microscopy 136 Figure 6.5: Loss of Mitochondrial Transmembrane Potential 138 Figure 6.6: Cytochrome c and AIF Immunblot Analysis 141 Figure 6.7: Cytochrome c Localization by Fluorescence Microscopy 142 Figure 6.8: AIF Localization by Fluorescence Microscopy After 30 Minutes 143 Figure 6.9: AIF Localization by Fluorescence Microscopy After 3 Hours 144 Figure 6.10: Caspase 9 Inhibition 146 Figure 7.1: Working Model 152 Figure 7.2: Apoptosis in Superantigen-Mediated Pathogenesis 158 Figure 8.1: Mutation Sites on TSST-1 Molecule 164 Figure 8.2: Random Coil Region From F47-E60 167 Figure 8.3: TSST-1 :HLA-DR Contact 168 Figure 8.4: Y51 Residue Buried in E173-P180 Cleft 169 Figure 8.5: TCR Interaction Region 171 Figure 8.6: Apoptosis with TSST-1 Mutants 174 Figure 8.7: Summary of TSST-1 Mutant Apoptotic Activity 175 ix List of Abbreviations Abbreviation Definition 7-AAD 7-aminoactinomycin D ACAD Activated T cell autonomous death AICD Activation-induced cell death AIF Apoptosis inducing factor ALPS Autoimmune lymphoproliferative syndrome ANOVA Analysis of variance Apaf-1 Apoptotic protease activating factor-1 APC Antigen presenting cell BCR B cell receptor BH3 Bcl-2 homology domain 3 BIR Baculovirus IAP repeats BTK Bruton agammaglobulinaemia tyrosine kinase CAD Caspase-activated DNase CARD Caspase recruitment domain CD95 Fas receptor CED Caenorhabditis elegans death gene COX4 Cytochrome oxidase subunit IV CTL Cytotoxic T lymphocyte DD Death domain DED Death effector domain DISC Death inducing signaling complex DN Double negative DP Double positive DR Death receptor EBV Epstein Barr virus EGL-1 Egg-laying defective gene 1 endoG Endonuclease G ER Endoplasmic reticulum EtBr Ethidium bromide FBS Fetal bovine serum FITC Fluorescein isothiocyanate FLIP FLICE-inhibitory protein GAB2 Grb2-associated binding protein 2 G31R Glycine 31 to arginine mutant of TSST-1 gtBid Granzyme B-truncated product of Bid HHV-8 Human herpes virus 8 IAP Inhibitor of apoptosis protein ICE Interleukin-l/3-converting enzyme I F N 7 Interferon gamma Ig Immunoglobulin IKK IKB kinase complex IL- Interleukin-ITK IL-2-inducible T-cell kinase LPS Lipopolysaccharide X Abbreviation Definition M A S Mycoplasma arthritidis-derived superantigen M H C Major histocompatibility complex M O M P Mitochondrial outer membrane permeabilization N U M A Nuclear mitotic apparatus protein 1 PARP Poly(ADP-ribose) polymerase P B L Peripheral blood lymphocyte P B M C Peripheral blood mononuclear cell PBS Phosphate buffered saline PE Phycoerythrin PH Pleckstrin-homology PI Propidium iodide PI3K Phosphoinositide-3' -kinase PtdInsP2 Phosphatidylinositol-(4,5)-bisphosphate PtdInsP3 Phosphatidylinositol-(3,4,5)-trisphosphate sAg Superantigen SDS-PAGE SDS-polyacrylamide gel electrophoresis SEA Staphylococcal enterotoxin A SEB Staphylococcal enterotoxin B SEC 1,2,3 Staphylococcal enterotoxin C 1,2,3 SED Staphylococcal enterotoxin D SEE Staphylococcal enterotoxin E SP Single positive SPEA Streptococcus pyogenes exotoxin A SPEC Streptococcus pyogenes exotoxin C tBid Truncated Bid TBS Tris-buffered saline TCR T cell receptor TdT Terminal deoxynucleotidyl transferase T N F R 1 Tumour necrosis factor receptor 1 TNFo! Tumour necrosis factor a TNF/3 Tumour necrosis factor 6 TRIM T cell receptor-interacting molecule TSS Toxic shock syndrome TSST-1 Toxic shock syndrome toxin-1 T U N E L TdT fluorescein-dUTP nick end labeing VB Variable region of the T cell receptor B chain A\j/m Change in mitochondrial transmembrane potential xi Preface At the human level, we find our reasons in life and disease unto death; at the cellular level, we again find our reasons in the intricate intertwining of life and death. Thus, to know life, we examine death, and what dying can teach us. xii Acknowledgements I wish to acknowledge first and foremost my family for their stalwart love and support throughout these many years. I would also like to express my gratitude to my supervisor, Dr. Anthony W. Chow, for putting up with me for all this time. He has mentored me through the long learning and maturation process, beginning as a wet-behind-the-ears new graduate and culminating now as clinician scientist in training. In the first several years Dr. Donna Hogge, and more recently Dr. Barrett Benny, have both kindly gifted our lab with regular leukopheresis packs from healthy donors at the Vancouver General Hospital's Cell Separator Unit. Without this source of human cells, most of this work would not have been possible. Next, I thank all my committee members, Dr. Vincent Duronio, Dr. Hung-Sia Teh, and Dr. Mike Gold for their extensive guidance and advice throughout this doctorate degree program. It would be remiss to neglect all the present and former members of both our laboratory and neighbouring laboratories who have selflessly contributed their indispensable help, advice, time and effort to aid me throughout this process. In no particular order, I would like to acknowledge: Dr. Winnie Kum, who as lab supervisor had a hand in every aspect of my progress from start to finish, be it at the planning, experimental or reporting stages; Dr. Robert McMaster, who generously donated the use of crucial equipment; Dr. Jennifer Kong, who provided invaluable expertise on a variety of techniques; and Scott Cameron, who aided during many cogent discussions to distill many of the concepts presented here. Finally and foremost, I attribute my ultimate source of inspiration to God, whose attributes of verity and omniscience have driven my twin desire to seek after truth and knowledge throughout these years. xin Chapter 1 Introduction "A journey of a thousand mUes begins with a single step. " -Lao Tzu 1.1. Apoptosis The 2002 Nobel Prize in Physiology or Medicine was awarded to Brenner, Horvitz and Sulston "for their discoveries concerning genetic regulation of organ development and programmed cell death", commonly synonymous with apoptosis (Lendahl et al., 2002; Marx, 2002; Check, 2002). Beginning with their seminal studies on the nematode Caenorhabditis elegans, the natural extension has been made to regulated cell death in mammalian systems. The conserved morphological and biochemical features of apoptosis distinguish it from its dysregulated counterpart, necrosis, wherein cell lysis and dispersal of intracellular components lead to an inflammatory response (Wyllie et al., 1980). However, it is increasingly being recognized that forms of programmed cell death exist outside the framework of apoptosis, often as a 'backup' process to remove unwanted or redundant cells (Jaattela et al., 2003). 1.1.1. General features of apoptosis Apoptosis, a term derived from the Greek word ontoTTUOLC; meaning 'a falling off in the context leaves falling off a tree, was first used in modern science by Kerr, Wyllie and Currie in 1972 (Kerr et al., 1972) to describe the cell death process characterized histologically as ending in cytoplasmic bodies that contain speckled patterns of condensed chromatin (Kerr, 2002). Apoptosis is a well-ordered process, requiring active metabolic involvement of the dying cell, including energy expenditure through the use of ATP, activation of cellular enzymes, and cytoskeletal rearrangement (Figure 1.1). Once committed to programmed cell death, a flurry of biochemical processes generate the archetypal features of apoptosis including chromatin 1 condensation toward the periphery of the nucleus, cell shrinkage around organelles and membrane blebbing or zeimosis (Hengartner, 2000). In the late stages, the continued blebbing results in the budding of so-called apoptotic bodies, which are plasma membrane-bound and contain one or more organelles. These very specific changes contrast with necrosis, where with severe insult or damage, cells undergo osmotic swelling and lysis (Figure 1.1), rather than retaining membrane integrity. The chromatin tends not to condense toward the periphery, but may instead clump within the nucleus or become dispersed rapidly with dissolution of the nuclear membrane. Thus, cellular contents tend to be dispersed into the interstitium during necrotic cell death in tissues. At the molecular level, activation of endonucleases during apoptosis causes characteristic cleavage of genomic DNA still wrapped on histone proteins, resulting in double-stranded fragments that are multiples of 180-200 base pairs (Enari et al., 1998). The normally asymmetric distribution of phosphatidylserine, with a preponderance on the inner plasma membrane, is lost (Fadok et al., 1992b). Increasing external expression of phosphatidylserine and vitronectin alerts neighbouring phagocytes of the apoptotic cell's impending demise and recruits them to migrate near (Fadok et al., 1992a). In vivo, this results in rapid phagocytosis of apoptotic bodies, leaving very little cellular debris behind (Savill et al., 2000). Again, this contrasts with necrosis where the dispersal of cellular contents promotes an inflammatory response. 2 Active process In vivo "clean" removal of fragments by phagocytosis Figure 1.1: Apoptosis vs. Necrosis. Apoptosis is an active process whereby upon execution of the apoptotic death program, a cell undergoes specific biochemical and morphological changes that result in a 'clean' removal of cells without instituting an inflammatory response. Necrosis describes any of a variety of pathological forms of cell death where rapid cell swelling and lysis result in dispersal of cellular contents, promoting inflammation. 3 1.1.2. Apoptosis in health and disease In a healthy state, apoptosis is important toward the maintenance of homeostasis for such processes as regulation of the immune response, tissue development and remodelling, and normal cell turnover (Figure 1.2). During embryonic and fetal development, virtually all organs acquire their final structure through a careful orchestration of growth and apoptotic death. Alterations in apoptosis may result in congenital abnormalities ranging from the cosmetic, such as interdigital webs, to the critical, such as endocardial cushion or cardiac outflow tract defects (Lindsten et al., 2000; Rothenberg et al., 2003). During fetal and infant life, massive neuronal apoptosis is essential for maturation of the neocortex (Chan et al., 2002). In disease states, either an overabundance of cell death or insufficient removal of undesirable or redundant cells may be culpable (Figure 1.2). Examples of the former include: neurodegenerative disorders, where neuronal apoptosis leads to permanent neurologic deficits (Jellinger, 2001); ischemic injury, where the initial anoxia may not be as damaging as inappropriate apoptosis in hypoxic cells following reperfusion (Rami, 2003; Zheng et al., 2003; Borutaite et al., 2003); and viral infections, such as with HIV deletion of CD4+ T cells leading to profound immunosuppression (Gougeon, 2003). Disease states with inadequate cell death include: tumour development, where either tumour suppressor genes become mutated, that would otherwise commit a damaged cell to apoptosis, or anti-apoptotic genes are overexpressed (Cappello et al., 2002); viral infections, such as with Epstein Barr virus (EBV) and human herpes virus 8 (HHV-8), which encode anti-apoptotic Bcl-2 homologues that may contribute to survival of virally infected cells (Henderson et al., 1993; Cheng et al., 1997); chronic inflammation as in asthma where persistence of eosinophils drives the reactive airway disease (Walsh et al., 2003); and autoimmunity, where deletion of autoreactive T or B cells fails (Siegel et al., 2003). 4 M o r a Embryogenesis and fetal growth Haematopoiesis Epithelial cell replacement Lymphocyte activation Pcancerand neoplasia Chronic inflammation: e.g. asthma Autoimmunity: e.g. rheumatoid arthritis, lupus Viral infections e.g. EBV, HHV-8 J £ £ j 1 Terminal neurons More Apoptosis Embryonic and fetal tissue re-modeling Immune downregulation Cell turnover Negative selection Ischemia and R ischemic re-perfusion odegenerative diseases: e .g . Alzheimer's, 1^Parkinson's, Huntington's Too Much Apoptosis Figure 1.2: Apoptosis in Health and Disease. Apoptosis is required during normal tissue development and homeostasis, but is inhibited during cell growth and proliferation. Excessive apoptosis is present during HIV infection, following ischemia, and with neurodegenerative diseases. Inhibition of normal apoptosis can produce abnormally persistent inflammation, autoimmunity and neoplasia. 5 1.1.3. Molecular pathways of apoptosis While the biochemical pathways for apoptosis are variable between different cell lineages and types, incorporating much redundancy, general themes extend through most cellular model systems and include both normal and transformed cells. Harkening back to early mutational studies done in the nematode C. elegans, a family of genes now found to be conserved from the nematode to mammals appears critical for apoptosis. In the nematode, these constitute two of the C. elegans death (CED) genes, CED-3 and CED-4, and a BH3-domain containing protein EGL-1 (egg-laying defective gene 1). CED-3 is homologous to the mammalian interleukin-1/3-converting enzyme (ICE) (Yuan et al., 1993). ICE (now called caspase 1) is primarily involved in LPS-mediated inflammation by processing pro-IL-1/3 into its active secreted form, but not in apoptosis (Creagh et al., 2003). However, a family of related aspartate-specific cysteine-dependent proteases, later known as caspases, were found to be essential for apoptosis in multicellular organisms ranging from hydras to mammals (Budihardjo et al., 1999; Cikala et al., 1999). This large repertoire of caspases operates in a proteolytic cascade similar to the coagulation and complement activation pathways to carry out the demolition/destruction program, by dismantling cellular structures, halting metabolic pathways, inactivating inhibitors of apoptosis, and activating other enzymes crucial for the death program (Thornberry et al., 1998; Slee et al., 1999a) (Figure 1.3). 6 Cytokine withdrawal DNA damage p53 BH3-only proteins Bcl-2 family anti-apoptotic proteins (Bcl-2. Bcl-xL) Cytotoxic insults / Bax/Bak Mitochondrial permeabilization Cytochrome c release Caspase 9 activation (initiator) IAP Smac • and Omi Caspase 3/7 activation (executioner) — I I Cleavage of cellular substrates (Enzymes, lamins DNA, cytoskeletal proteins) • ^^ ^^ ^^ ^^ ^^ ^^ ^^  ^^^^^H Apoptosis Death ligand (FasL, TNF, > t Death receptor (Fas, TNFR1, DR3, 4, 5, 6) > f Death adapter (FADD, TRADD) Caspase 8 activation (initiator) FLIP Figure 1.3: Schematic of Apoptotic Pathways. The two major branches of apoptotic pathways are the intrinsic mitochondrial pathway (left) and the extrinsic death receptor pathway (right). 7 Caspases Most of the caspases exist as inactive zymogens that need to be cleaved to be rendered active. One group of caspases (2, 8, 9, and 10) have conserved N-terminal pro-domains that allow them to associate with the death effector domains (DED) of adapter molecules such as FADD or caspase recruitment domains (CARDs) of molecules like the apoptotic protease activating factor-1 (Apaf-1) (Earnshaw et al., 1999). These require proteolytic processing, usually through proximity-mediated autocatalysis, to produce the polypeptide fragments that, once dimerized, constitute the active caspase. Another group (caspase 3, 6 and 7) lacks the extended pro-domain, and is dependent on the proteolysis by other caspases or proteases such as the aspartate-specific granzyme B to become activated (Thornberry et al., 1998). This stratification places the former group into an apical class called 'initiators' that usually kick-start the caspase cascade, and the latter into the effector or 'executioner' class of caspases that cleave most of the vital downstream substrates to carry out the death program (Figure 1.3). The remaining human caspases (1, 4, 5, and 14) are thought to be involved in inflammation, with caspase l's proteolytic activity known to be required for IL-1/3 and IL-18 activation from precursors. It is also important for IL-la production, for unknown reasons. Caspases are all cysteine-dependent proteases that recognize a four-amino acid motif, starting with aspartate (Asp) in the PI (peptide-1) position (Thornberry et al., 1997). An active caspase will cleave target polypeptide chains after the tetrapeptide motif. The variation among caspases for recognition of the remaining three residues determines their target specificity, but also has permitted the development of peptide and chemical inhibitors for specific caspases. The list of cellular substrate proteins that possess caspase recognition sites now encompasses more than 100 genes. While not all are necessarily critical for the death program, a few key substrates include ICAD, PARP and the lamins. ICAD normally sequesters the DNase CAD/DFF40 in an inactive state until caspase 3 cleaves ICAD, allowing CAD to carry out 8 internucleosomal degradation of the genomic DNA (Thornberry et al., 1998). However, CAD does not appear to be necessary for the first stage of chromatin condensation during apoptosis, involving cleavage of high molecular weight DNA (Samejima et al., 2001); instead, other factors such as apoptosis-inducing factor (AIF) may be important for the initial chromatin changes (Susin et al., 2000). Poly(ADP-ribose) polymerase (PARP), which is involved in DNA repair, was among the first identified substrates of caspase 3 (Tewari et al., 1995). Nuclear structural proteins like the lamins that constitute the nuclear lamina and nuclear mitotic apparatus protein 1 (TSJUMA), which helps to organize the mitotic spindle, are also targets for caspase 3 and 6 activity (Hirata et al., 1998; Ruchaud et al., 2002; Taimen et al., 2003). Together, these appear to coordinate and execute the nuclear morphology changes that take place during apoptosis. 9 Intrinsic Pathway Extrinsic Pathway Figure 1.4: Two Pathways for Apoptosis. Typical activation through the extrinsic pathway (right) involves ligation of Fas receptor by Fas ligand, which recruits FADD to form a death-inducing signaling complex (DISC) that processes caspase 8 into its active form. Caspase 8 then cleaves caspase 3 and other executioner caspases to engage the death program. In contrast, a variety of cellular insults promote activation or expression of BH3-only proteins that inhibit anti-apoptotic Bcl-2 and Bcl-xL, allowing the freed pro-apoptotic Bax and Bak to mediate pore formation in the mitochondrial outer membrane. This allows leakage of constitutes like cytochrome c, which promotes assembly of the heptameric apoptosome consisting of its adapter APAF-1 and caspase 9. Caspase 9 in turn cleaves downstream executioner caspases like caspase 3. Other released mitochondrial constituents involved in apoptosis include AIF. Smac/Diablo, Omi and endonuclease G (discussed in text). 10 Extrinsic pathway for apoptosis Typically, a cell undergoing apoptosis receives either an external (extrinsic) or internal (intrinsic) signal sufficient to cause it to commit to the death program (Figure 1.3) (Strasser et al., 2000; Schultz et al., 2003). Extrinsic signals include those derived through ligation of death receptors of the TNF receptor superfamily, consisting at last count of Fas (CD95), TNFR1, DR3, DR4, DR5 and DR6 (Ashkenazi et al., 1998). Upon Fas ligand binding, Fas receptor forms mini-aggregates that recruit the adapter molecule FADD to associate its death domain (DD) with the DD on the intracytoplasmic tail of the Fas receptor (Boldin et al., 1995; Chinnaiyan et al., 1995) (Figure 1.4). This assembled death inducing signaling complex (DISC) (Kischkel et al., 1995) binds to pro-caspase 8 through the respective DEDs on FADD and pro-caspase 8 to promote proximity-mediated autocatalytic processing (Muzio et al., 1996; Salvesen et al., 1999). Similarly, TNFR1 ligation following TNF binding forms a DISC via its adapter molecule TRADD to activate caspase 8 and/or caspase 10 in humans (Boldin et al., 1996). Activated caspase 8 or 10 can then proceed either to cleave and activate pro-caspase 3, initiating the downstream death program, or to cleave Bid, promoting apoptosis through the intrinsic pathway (Li et al., 1998b). Cells susceptible to the former direct pathway to downstream caspase activation are termed type I cells, and include lymphocytes undergoing AICD in vitro via Fas ligation (Scaffidi et al., 1998). Type I cells are not protected from death-receptor mediated apoptosis by overexpression of the anti-apoptotic molecules Bcl-2 and Bcl-xL (Strasser et al., 1995). In contrast, cells such as hepatocytes that require amplification of Fas signaling through the intrinsic pathway are termed type II cells. However, in human peripheral blood T lymphocytes specifically, Fas receptor ligation can also produce mitochondrial dysfunction through an as yet unidentified mechanism independent of caspase 3 or 8 (Gollapudi et al., 2003); this suggests that the simplistic delineation between extrinsic and intrinsic pathways may instead be blurred in vivo. In humans, caspase 10 is important toward death signaling through TNFR1, 11 but can fulfill a similar role to caspase 8 (Tibbetts et al., 2003; Kischkel et al., 2001). Defects in Fas, FasL, caspase 10 and caspase 8 in humans can give rise to autoimmune lymphoproliferative syndrome (ALPS) (Rieux-Laucat et al., 2003). Humans lacking caspase 8 also show immunodeficiency due to defects in lymphocyte activation, underscoring the dual role of caspase 8 in apoptosis and antigen receptor-mediated signaling (Chun et al., 2002). Intrinsic pathway for apoptosis The intrinsic pathway is somewhat of a catch-all for death signals that culminate at the level of the mitochondria to promote mitochondrial outer membrane permeabilization (MOMP) (Marsden et al., 2003). Not only do many cytotoxic drugs, oxidative stress, ionizing radiation, DNA damage and growth factor deprival integrate into this pathway, but a sublethal stimulus through extrinsic death receptors can also exert an effect by tilting the balance toward MOMP (Figure 1.3). MOMP is controlled by antagonistic anti-apoptotic and pro-apoptotic members of the Bcl-2 family (Vaux et al, 1988). These proteins share significant homology and interact both with themselves (forming homodimers) and each other (forming heterodimers). As with caspases, there appear to be more apical members (e.g. Bid, Bad, Bim, PUMA, NOXA) of the Bcl-2 family containing only the Bcl-2 homology region 3, termed BH3-only proteins. Other Bcl-2 family proteins contain multiple domains and possess either anti-apoptotic (Bcl-2, Bcl-xL, Bfl-1, Bcl-w, Boo, Mcl-1, etc.) or pro-apoptotic (Bax, Bak, Bok, etc.) activity. The expression levels and activation state of anti-apoptotic vs. pro-apoptotic members of the Bcl-2 family constitute the critical fulcrum point upon which is balanced a cell's fate. The BH3-only proteins play an important role by 'sensing' conditions that should initiate cellular commitment to suicide (Figure 1.4). Increased functional availability of BH3-only proteins by activation, increased expression or subcellular redistribution tilt the balance in favour 12 of the pro-apoptotic 'team' (Puthalakath et al., 2002). The BH3-only protein Bim, for instance, is essential for negative selection of double positive (DP) thymocytes that react too strongly to peptide-MHC (Bouillet et al., 1999; Bouillet et al., 2002). Signals transduced through TCR ligation result in the release of Bim from its inactive association with cytoskeletal proteins, whereupon it can initiate MOMP and subsequent apoptosis (Puthalakath et al, 1999). Similarly, Bim appears to be essential for apoptosis of T lymphocytes in response to cytokine deprivation (Bouillet et al., 1999) by becoming dephosphorylated and active (Seward et al, 2003). The alternative pathway for Fas-mediated cell death alluded to in the previous section operates through Bid (Figure 1.4). In cells where insufficient caspase 8 activation is promoted, or which are specifically resistant to apoptosis induction through caspase 8 activation, the little active caspase 8 amplifies the signal by cleaving Bid into a truncated, active form called tBid (Li et al., 1998b; Luo et al., 1998). tBid displaces Bax from inactive heterodimers with Bcl-2 or Bcl-xL, permitting it to carry out MOMP and apoptosis (Desagher et al., 1999; Eskes et al., 2000; Roucou et al, 2002a; Roucou et al., 2002b). A special case of integration between the extrinsic and intrinsic pathways exists with granzyme B (Figure 1.4). Cytotoxic T cells and NK cells can kill target cells through either Fas ligand/Fas receptor-mediated death or through directional deposition of the contents of cytotoxic granules by a process called degranulation (Froelich et al., 1998). Among the granular contents are perforin, which mediates the formation of pores in the target cell membrane, facilitating the entry of the serine protease granzyme B, which counts among its targets caspase 3, 8 and Bid (Pinkoski et al., 2001). Upon cell entry, granzyme B initiates but cannot complete processing of pro-caspase 3 (Sutton et al., 2003); because partially processed pro-caspase 3 is efficiently inhibited by the binding of IAPs, release of mitochondrial factors such as Smac/Diablo and Omi is required to relieve inhibition (Goping et al., 2003). To this end, cleavage of Bid into a 13 truncated form called gtBid (granzyme B-truncated product of Bid) serves to activate the mitochondrial-dependent intrinsic pathway (Sutton et al., 2000). The pro-apoptotic protein Bad at first glance seemed to provide an important bridge between the survival pathways promoted by cytokines and costimulation, and the apoptotic pathways by having phosphorylation sites that modify its activity. Akt/PKB activation downstream of the PI3 -kinase pathway can phosphorylate Bad to sequester it on 14-3-3 protein and render it inactive (Zha et al., 1996; Blume-Jensen et al., 1998; Datta et al., 1997; del Peso et al., 1997). Recently, however, the actual importance of this phosphorylation site has been questioned in vivo, since bad1' mice show no visible phenotype (Putcha et al., 2002). Moreover, Bad phosphorylation does not necessarily correlate with cell survival (Hinton et al., 1999). Instead, Bad may indirectly contribute to Bim signaling by binding and inactivating anti-apoptotic molecules such as Bcl-2 (Letai et al., 2002). This potentiates the action of Bim on Bax oligomerization as described below, which is a critical step for cytokine-withdrawal mediated cell death (Bouillet et al., 1999). DNA damage causes activation of the tumour suppressor gene p53. As a transcription factor, p53 modifies the expression of a variety of proteins including the BH3-only proteins PUMA and NOXA. Upregulation of PUMA and NOXA triggers cell death to avoid perpetuation of genetic errors (Oda et al, 2000; Nakano et al., 2001). An increase in the abundance, activation state or availability of the multi-domain pro-apoptotic proteins Bax (Bcl-2-associated X protein) and Bak lead to their translocation from the cytosol and oligomerization on the outer mitochondrial membrane, forming pores that allow some constituents in the mitochondrial intermembrane space to translocate to the cytosol (Antonsson, 2001). Bax and Bak are normally sequestered by the formation of inactive heterodimers in the cytoplasm with the anti-apoptotic proteins Bcl-2 and Bcl-xL. Thus, most of the BH3-only protein activated pathways for apoptosis are susceptible to inhibition by Bcl-2 14 overexpression. Recent studies with Bax^ Bak"'" haematopoetic cell reconstitution have suggested that Bax and Bak are required for both thymic negative selection and peripheral deletion of activated T cells (Rathmell et al., 2002). Whatever the death stimulus, once MOMP is achieved, leakage of constituents such as cytochrome c, Smac/Diablo, apoptosis-inducing factor (AIF), endonuclease G (endoG), and Omi facilitate activation of the death program. Cytochrome c binds to monomelic Apaf-1 in an ATP-dependent manner, which promotes assembly of the so-called apoptosome, a heptameric complex (Liu et al., 1996; Acehan et al., 2002). With the relief of the inhibitory effect of WD40 repeats on Apaf-1 to recruitment of pro-caspase 9 (Zou et al., 1999; Adrain et al., 1999), autocatalytic processing to the active form of caspase 9 ensues, which can then activate downstream caspase 3 and 7 to carry out apoptosis (Slee et al., 1999b). Smac/Diablo, once released from the mitochondria, relieves the inhibition of the caspase pathway by stoichiometrically displacing the inhibitor of apoptosis proteins (IAPs) from their binding to caspases (Du et al., 2000). Apoptosis-inducing factor translocates directly to the nucleus where it can initiate chromatin condensation (Susin et al., 1999). Once released from the mitochondria, endonuclease G participates directly in nucleosomal fragmentation independent of caspase activity, unlike CAD (Section (Li et al., 2001a). Finally, Omi, a serine protease, catalytically targets and irreversibly inactivates IAPs by proteolytic cleavage, thus efficiently removing inhibition of caspase activity by the IAPs (Yang et al., 2003). Alternative pathways for apoptosis Several other pathways have been identified that apparently defy straightforward categorization into either the caspase-dependent extrinsic or intrinsic pathways. One school of thought holds that the MOMP pathway may be activated only secondary to the caspase cascade, since stress-mediated caspase 2 activation can lead to MOMP (Lassus et al., 15 2002). Unlike caspase 8, caspase 2 has no known downstream caspase targets, although it can cleave Bid (Guo et al., 2002). In this view, MOMP would serve merely to amplify these caspase-initiated death signals. Recent identification of apoptosis triggered by stressors to the endoplasmic reticulum (ER) 2_|_ , that disrupt Ca homeostasis have implicated caspase 12 in coupling ER dysfunction to mitochondrial-mediated apoptosis (Rao et al., 2002). However, caspase 12 can also directly cleave and activate pro-caspase 9, bypassing the requirement for cytochrome c release (Morishima et al., 2002). Released calcium itself may activate calpains, that can in turn cleave and activate Bax directly (Wang, 2000; Toyota et al., 2003). Reported instances of caspase-independent cell death may arise apically from the generation of reactive oxygen species, and downstream with the release of mitochondrial constituents such as AIF that can drive chromatin condensation irrespective of caspase activation (Susin et al., 1999). ROS play a critical role in activated T cell autonomous death (ACAD) to drive Bim-dependent mitochondrial destabilization and death (Hildeman et al., 1999). Downstream, the kinetics of mitochondrial constituent leakage may determine whether caspase-dependent or caspase-independent pathways are recruited (Leist et al., 2001). In the presence of cytochrome c translocation, the caspase cascade in primary T cells may dominate by virtue of the rapid rate of activation, while in its absence or with inhibition of the caspase cascade, non-caspase dependent AIF activity may serve to guarantee continued execution of cell death. 16 riuMii.mcii Anti-Apoptotic Figure 1.5: Regulation of Apoptosis. Apoptosis is regulated by both anti-apoptotic proteins (right) and pro-survival signaling (left). Anti-apoptotic molecules include FLIP, IAP and various members of the Bcl-2 family. Pro-survival signals can derive through PI3K activation, which results in Akt activation. Akt promotes downstream effectors of cell survival ( N F - K B ) and inhibits pro-apoptotic transcription factors (FOX03). 17 1.1.4. Regulation of apoptosis The regulation of apoptosis is accomplished in many ways, often times with redundancy, as would be expected for a response with such finality as cell suicide. At the level of protein expression, both gene transcription and protein degradation are important in modifying the susceptibility of cells to apoptosis. The prototypical Bcl-2 anti-apoptotic protein was first identified as overexpressed in a B cell lymphoma (hence the name), and was later found to confer resistance to apoptosis in response to many physiological and pharmacologic inducers of cell death (Marsden et al., 2003). Bcl-2 and its related anti-apoptotic analogues contain Bcl-2 homology (BH) domains 1-4, permitting them to heterodimerize non-specifically with the pro-apoptotic multidomain proteins such as Bax and Bak to render them inactive (Figure 1.5). Apart from the Bcl-2 family, caspases and their endogenous inhibitors or competitors are themselves differentially expressed in response to a broad array of transcription factors. FLICE-inhibitory proteins (FLIPs) resemble pro-caspase 8 in containing DEDs, but lack the catalytic domains (Thome et al., 2001) (Figure 1.5). Both viral and endogenous FLIPs can therefore inhibit death receptor signaling by usurping pro-caspase 8's place on the DISC (Thome et al., 1997). The two mammalian splice variants of cellular F L I P consist of the long C - F L I P L , which allows partial but incomplete pro-caspase 8 processing, and small c-FLIPs, which completely prevents pro-caspase 8 cleavage (Krueger et al., 2001). C - F L I P L is initially expressed at high levels in activated T cells (Irmler et al., 1997), preventing AICD, before continued IL-2 signaling suppresses the production of c-FLIPL and confers sensitivity to Fas-mediated death (Refaeli et al., 1998). The inhibitor of apoptosis proteins (IAPs) contain baculovirus IAP repeats (BIR) that bind to active caspases and render them inactive (Verhagen et al., 2001) (Figure 1.5). These are targeted for ubiquitination and degradation (Huang et al., 2000). As mentioned previously, IAPs potently suppress caspase activation, so that inhibitors of IAPs such as Smac/Diablo and Omi are needed 18 to permit engagement of the apoptotic program. The expression of IAPs such as survivin are frequently elevated in tumours, possibly downstream of the Ras family of oncogenes (Sommer et al, 2003). Survival signals transmitted by antigen receptor ligation, growth factors and costimulatory molecules on lymphocytes simultaneously provide important anti-apoptotic signals, both immediately through the action of kinases, as well as downstream through the activation or inactivation of transcription factors involved in regulating pro- and anti-apoptotic genes. For example, early on, activated T cells depend upon IL-2 and other related cytokines for their survival and proliferation (Vella et al., 1998). Signaling through the Y c chain of the IL-2 receptor increases Bcl-2 and Bcl-xL expression, protecting against Bim-mediated cell death (Li et al., 2001b; Kreisel et al., 2002). However, continued exposure to IL-2 specifically also inhibits c-F L I P L expression and increases FasL expression, conferring susceptibility to Fas-mediated AICD (Refaeli et al., 1998). In T lymphocytes, IL-7 can similarly produce an elevation in c-IAP-2 expression through the PI3K/Akt pathway, as well as increased expression of B C I - X L through an independent pathway (Sade et al., 2003). Details of the signal transduction events following TCR ligation and CD28 costimulation remain elusive, but it has been suggested that they achieve their rapid activation of PI3K through the action of conserved YXXM tetrapeptide motifs either on adapter proteins such as the T-cell-receptor-interacting molecule (TRIM), or directly on the intracytoplasmic tail of receptors such as CD28 (Okkenhaug et al., 2003) (Figure 1.5). IL-2 receptor is similarly thought to mediate PI3K activation through Grb-2 and the YXXM-containing Grb2-associated binding protein 2 (GAB2) (Ward et al., 2001). In both cases, membrane-localized YXXM-containing proteins, when tyrosine phosphorylated, are able to bind to the Src-homology 2 domains of the p85 regulatory subunit of PI3K. This recruits the pi 10 catalytic subunit of PI3K to the membrane, allowing it to convert phosphatidylinositol-(4,5)-bisphosphate (PtdInsP2) to phosphatidylinositol-19 (3,4,5)-trisphosphate (PtdInsP3). PtdrnsP3 in turn binds to pleckstrin-homology (PH) domain-containing proteins including the most important of these, the serine/threonine kinase Akt/PKB, as well as the lymphocyte-specific Tec-family tyrosine kinases BTK (Bruton agammaglobulinemia tyrosine kinase in B cells) and ITK (IL-2-inducible T-cell kinase in T cells) (Schaeffer et al., 2000). Akt plays a conserved role in cell survival, growth, proliferation, and development across many cell types (Vanhaesebroeck et al., 2000). The transcription factor N F - K B , normally inactive and associated with I K B in the cytoplasm, translocates to the nucleus after I K B phosphorylation by the Akt-activated I/cB kinase complex (IKK), causing it to dissociate from N F - K B and be targeted for degradation (Ozes et al., 1999). NF-/cB promotes the expression of anti-apoptotic genes including the TNFR-associated factors 1 and 2, cIAP-1 and 2, XIAP, c-FLIP and Bcl-xL (Wang et al., 1998; Xiao et al, 2002). As alluded to earlier, Akt can phosphorylate and inactivate the pro-apoptotic molecule Bad (Datta et al., 1997) and inhibit conformational changes needed for Bax to translocate to the mitochondria and produce MOMP (Yamaguchi et al., 2001). Finally, Akt inactivates Forkhead transcription factors of the FOXO subfamily by direct phosphorylation (Burgering et al., 2003). The FOXO subfamily can elicit cell cycle arrest by promoting p27kipl expression and inhibiting cyclin D transcription (Schmidt et al, 2002). FOX03 in particular is likely involved in inducing transcription of the pro-apoptotic protein Bim during cytokine deprivation (Stahl et al., 2002; Dijkers et al., 2000), while FOX03a and FOX04 suppress the expression of B C I - X L through induction of Bcl-6 (Schmidt et al, 2002; Tang et al., 2002). 17.5. Approaches to assaying apoptosis The characteristic features of apoptosis have led to the development of many approaches to identifying apoptotic cells and distinguishing them from those undergoing necrosis. These may be divided into several broad categories, although some approaches may simultaneously utilize 20 methods from multiple categories: cell morphology, cell viability, nuclear changes and apoptosis-associated proteins. The simplest of morphologic approaches is direct microscopic visualization of cells, with or without staining. Apoptotic cells exhibit shrinkage and membrane blebbing compared to their healthy counterparts, while necrotic cells swell and rapidly undergo lysis. The cell shrinkage is the result of intracellular fluid loss, and can be detected by flow cytometry by the corresponding decrease in forward scatter signal (an indicator of cell size) and slight increase in side scatter signal (an indicator of cell granularity). Cell viability is commonly assessed using membrane impermeant dyes such as ethidium bromide (EtBr), propidium iodide (PI) and 7-aminoactinomycin D (7-AAD). All of these fluoresce in association with nucleic acids and are generally excluded from live cells with intact plasma membranes. Cells stained with these dyes can therefore be analyzed using either fluorescence microscopy or flow cytometry. However these do not, in and of themselves, distinguish apoptosis from necrosis. 7-AAD's far red fluorescence makes it attractive for multi-colour flow cytometry, since it offers less spectral interference with other fluorochromes such as phycoerythrin (PE). Combined with flow cytometric information about cell size and granularity, as described above, 7-AAD offers a rapid and specific method for detecting apoptosis (Schmid et al, 1992). EtBr and PI bind both RNA and DNA, thus requiring the use of an RNase for accurate DNA quantitation and cell cycle analysis (Nicoletti et al., 1991). While 7-AAD selectively binds to GC-rich regions of DNA, but not to RNA, it is less precise for DNA quantitation than the two former dyes, and thus is not used for cell cycle analysis. Cycling cells progress from diploid (2N; Go/Gi phase) to DNA synthesis (>2N; S phase), then complete replication and undergo mitosis (4N; G2/M phase), before returning to diploid (2N). Apoptotic cells undergo DNA degradation and have a lower DNA content, termed hypodiploid (<2N). As 21 a result, fixed and RNase-treated cell populations stained with PI can be assayed for their relative proportions of apoptotic, resting, and proliferating cells using flow cytometry. DNA cleavage during apoptosis occurs specifically in the internucleosomal linker regions. As a result, the 180-200 bp segments wrapped around histone proteins are intact, and give rise to fragments consisting of multiples of these 180-200 bp segments. Therefore, gel electrophoresis of extracted DNA provides a specific method for assaying apoptotic cells, which exhibit a laddering pattern of these low molecular weight fragments. An in situ method for detecting DNA strand breaks using the enzyme terminal deoxynucleotidyl transferase (TdT) has also been developed. Called the TdT Fluorescein-dUTP Nick End Labeling (TUNEL) method, apoptotic cells containing DNA strand breaks are preferentially labeled and fluoresce more intensely than healthy cells (Gorczyca et al., 1993b; Gorczyca et al., 1993a). However, this method does not distinguish apoptotic from necrotic cells, since both undergo DNA degradation. While healthy cells have a preponderance of phosphatidylserine on their inner plasma membrane leaflet, apoptotic cells rapidly lose the asymmetry and equalize distribution of phosphatidylserine across the inner and outer faces of the plasma membrane. As a result, the phosphatidylserine-specific Annexin V protein has been employed to detect exposure of phosphatidylserine that occurs on apoptotic cells (Vermes et al., 1995). Fluorochrome conjugates of Annexin V such as Annexin V-FITC have allowed both visualization by fluorescence microscopy and population determinations by flow cytometry. With the detailed understanding of the biochemistry of apoptosis that has emerged, including knowledge about the proteins involved in the extrinsic, intrinsic and common pathways, more direct methods for studying the activation of these pathways have been developed. The caspase processing required for caspase activity results in the generation of lower molecular weight (active) fragments. Therefore immunoblot methods have been used to detect the activation of specific caspases by the relative abundance of pro-caspase (zymogen) and active caspases 22 (processed fragments). The availability of antibodies specific for the active form of caspases such as caspase 3 has also permitted in situ measurement of caspase 3 activation by flow cytometry and fluorescence microscopy. Coupled with the use of cell-permeable peptide inhibitors which are either specific to certain caspases or general to all, these reagents have presented a useful toolkit for examination of the downstream caspase cascades. Specifically for the intrinsic pathway, detection of mitochondrial membrane permeabilization has been aided by the use of dyes which preferentially aggregate in cationic regions. These dyes, including JC-1, Mitotracker™ and Mitosensor™, aggregate in healthy mitochondria, which have a proton gradient that establishes a difference in membrane potential between the inside of the mitochondria and the cytoplasm (mitochondrial transmembrane potential, Av|/m). These dyes produce a red fluorescence under these conditions. In contrast, pore formation on the outer mitochondrial membrane during activation of the intrinsic pathway results in a loss of A\|/m. Without a cationic region, these dyes remain as monomers instead of aggregating, and produce a green fluorescence. As the mitochondrial membrane integrity is lost, mitochondrial constitutents translocate to the cytoplasm. This translocation can be assessed using differential centrifugation of cell lysates to separate intact mitochondria from cytosolic fractions, allowing immunoblot analysis for molecules such as cytochrome c and AIF. Immunohistochemical staining of cytochrome c, AIF and other mitochondrial constituents with visualization by regular or fluorescence microscopy allows in situ localization of these molecules within cells. Similarly, translocation of Bcl-2 family members between the cytoplasm and mitochondria can be detected by both of these methods. 1.2. Superantigens Superantigens historically derived their name from their propensity to act like antigens in eliciting T cell responses, but with far greater potency and less specificity than for conventional 23 antigen responses. Whereas only 1 in 104-106 T cells might recognize an antigen and become activated, superantigens can activate between 5-20% of the entire host repertoire of T cells. While the first superantigens identified were of viral origin (mouse mammary tumor virus-derived minor lymphocyte stimulating (Mis) antigens), an assortment of exotoxins produced by Staphylococcus aureus and Streptococcus pyogenes have gained notoriety as the most prominent and best-studied of superantigens (Marrack et al., 1990). 1.2.1. Superantigens in disease The staphylococcal superantigens include the staphylococcal enterotoxins (SEA, SEB, SEC1-3, D, E, G-Q) and toxic shock syndrome toxin-1 (TSST-1), some of which have been implicated as virulence factors in staphylococcal toxic shock syndrome (TSS). TSST-1 is the dominant toxin involved in TSS, especially for menstrual cases (Bergdoll et al., 1981; Schlievert et al., 1981) where its ability to cross the vaginal mucosa contributes to its pathogenic activity (Schlievert et al., 2000; Hamad et al., 1997; Shupp et al., 2002); in non-menstrual cases, SEB and the SECs are also important (Bohach et al., 1990; Schlievert, 1986; Kain et al, 1993). The massive polyclonal T cell response triggered by superantigens is chiefly Thl-polarized, leading in TSS to an extreme pro-inflammatory cytokine response that precipitates a severe hypotensive crisis, hyperthermia, and the catastrophic dysregulation of multiple physiological systems and organs, potentially resulting in death. Concomitant dysregulation by superantigens of the humoral and cell-mediated immune responses, accompanied by T cell repertoire deletion or anergy lead to a state of immunosuppression favouring the bacterial infection. Patients with TSS can present with fever above 39°C, nausea and/or vomiting, watery diarrhea with abdominal pain, syncope, myalgias, pharyngitis, and confusion. Clinical signs include a diffuse erythematous rash, generalized edema and profound, though transient lymphocytopenia (Chow et al., 1984; Chesney et al., 1981). If progressive, multi-organ involvement may include 24 disseminated intravascular coagulation, ventricular arrythmias, acute renal and/or hepatic failure, acute respiratory distress syndrome, necrotizing fasciitis, altered state of consciousness, and mucosal inflammation. Initially associated with super-absorbant tampon use in menstruating women, the incidence of menstrual TSS has diminished considerably after the withdrawal of these products from the market (McCormick et al., 2001). Its incidence is now estimated at around 5-15 cases per 100,000 (Hajjeh et al., 1999). However, the case fatality rate for staphylococcal toxic shock syndrome remains high, at around 3-5%. Moreover, in survivors there is a propensity for recurrence and clinical sequelae. Thus, investigation into pathogenesis and mechanisms surrounding the activity of superantigens continues to be relevant (Andrews et al., 2001). Furthermore, non-menstrual TSS involving both TSST-1 and the staphylococcal enterotoxins and streptococcal TSS involving the superantigens Streptococcus pyogenes exotoxin A (SPEA) and C (SPEC) are increasingly prevalent in post-operative infections around wound sites in both sexes and all ages. Curiously, there is little inflammation at the site of these infections (Strausbaugh, 1993). Streptococcal TSS is especially aggressive (30-70%) mortality) and often presents in the context of invasive S. pyogenes infections involving spreading cellulitis, necrotizing fasciitis and/or bacteremia (Demers et al., 1993). Aside from the involvement in life-threatening TSS, superantigens have been implicated in diverse human diseases. The staphylococcal enterotoxins are the causative agents of staphylococcal food poisoning, characterized by a self-limiting emetic illness following ingestion of preformed toxin. Staphylococcal enterotoxins stimulate nerve centers in the gut through activation of inflammatory mediators to produce nausea, vomiting, diarrhea and abdominal discomfort. The staphylococcal exfoliative toxins A and B were previously thought to be superantigens (Zollner et al., 1996; Monday et al., 1999), but recent evidence using purified recombinant proteins have demonstrated no T cell mitogenic activity, suggesting contamination 25 in previous preparations (Piano et al, 2000). Staphylococcal superantigens are suspected to be responsible for both sudden unexpected nocturnal death (SUND) in adults (Al Madani et al., 1999) and sudden infant death syndrome (SUDS) (Zorgani et al, 1999). SPEA and SPEC are causative agents in scarlet fever, a complication of streptococcal pharyngitis (Schlievert, 1981). In scarlet fever, a diffuse erythematous rash along with characteristic 'strawberry tongue' appears transiently, disappearing after 5-7 days, and is followed by desquamation. Finally, both staphylococcal and streptococcal superantigens are now thought to be involved in atopic dermatitis (Taskapan et al., 2000), Kawasaki disease (Leung et al., 1995), and guttate psoriasis (Yarwood et al., 2000), although the causal relationships remain to be proven (Leung et al., 2002). 26 Figure 1.6: A. 3D ribbon diagram of TSST-1 in the conventional orientation showing the N-terminal domain (dotted yellow line) and C-terminal domain (dotted green line), with secondary structural elements labelled. B. An approximately 180° rotation of A. about the vertical axis. 27 1.2.2. Structure-function relationships in superantigen activity A typical superantigen acts primarily on T cells and APCs by cross-linking T cell receptors (TCR) to major histocompatibility complex (MHC) class II molecules on APCs without a requirement for prior antigen processing. While the interaction is polyclonal, affecting between 5-20% of the T cells in the host, each superantigen has characteristic specificity for relatively non-polymorphic regions of TCR /3-chain's variable region (VB) subsets (Kappler et al., 1989). For example, TSST-1 produces selective activation of the VB2+ T cell population (Choi et al., 1990). Both the MHCII binding capacity and V/3-specificity of superantigens stem from distinct regions of the proteins that have been mapped by crystallography and mutational studies. TSST-1 is the smallest of the bacterial superantigens, consisting of only 194 amino acids for a molecular mass of 22 kilodaltons. Based on sequence homology, three main subgroups of staphylococcal superantigens have been identified: the first comprises SEA, SEE, SED and SEH, sharing 53-81% homology; the second SEB, the SECs, and SEG, sharing 50-66% homology; and the third TSST-1, without significant sequence similarity to the staphylococcal enterotoxins (Marrack et al, 1990). In spite of the limited sequence homology, all of the staphylococcal superantigens share remarkable similarities in structure consisting of two distinct domains separated by centrally located a helices (Alouf et al., 2003). Circular dichroism indicates that all possess predominantly /3-sheet structures, with low ohelical content (Singh et al., 1988a; Singh et al., 1988b). TSST-1 lacks cysteines and is therefore missing the 'disulphide loop' present in most, but not all, the staphylococcal enterotoxins, that is thought to be responsible for their emetic properties. X-ray crystallographic studies of TSST-1 compared with the staphylococcal enterotoxins indicate that it contains far fewer a-helices than the other superantigens (Prasad et al, 1993; Prasad et al., 1997; Acharya et al., 1994; Papageorgiou et al, 1996). The secondary 28 structural content of TSST-1 comprises 2 alpha helices and 12/3 strands, divided into residues according to the designation of Papageorgiou et al. as follows: al (7-15), /31 (18-29), /32 (32-37), |83 (41-47), BA (60-75), B5 (79-89), /36 (101-106), 81 (109-111), /38 (119-124), a2 (124-140), 69 (152-158), /310 (161-166), BU (181-182), and /312 (186-193) (Figure 1.6). Discrepancies in the secondary structural elements within the crystal structure described by different authors account for shifts in the assignment of TSST-1 residues by less than one amino acid. The two major regions of the TSST-1 molecule are the N-terminal domain (1-89) and the C-terminal domain (101-194) (Figure 1.6). The N-terminal domain begins with the short al helix and transitions into a characteristic, concaved /3-barrel formed by the subsequent B strands (/31-5). This /3-barrel, common to other superantigens, resembles the oligosaccharide/oligonucleotide binding (OB) fold found in proteins of unrelated sequence such as the staphylococcal nuclease and E. coli verotoxin (Murzin, 1993). The C-terminal domain of TSST-1 consists of the long o2 helix surrounded by a highly twisted 5-strand /3-sheet formed by the strands B6, 1,9, 10 and 12. The central o2 helix borders the N-terminal domain al helix and forms important intramolecular hydrogen bond interactions (Papageorgiou et al., 1996) both with the al helix and the linker region between Bl and BS (Figure 1.6). SEA, SEE and SED are all dependent on the formation of a coordination complex with Zn 2 + ions for high-affinity binding to MHC class II, although they possess an alternate low affinity binding site. In contrast, SEB, SEC1.3 and TSST-1 all bind independently of metal ions, although Zn 2 + binding by TSST-1 in a crystal structure has been demonstrated and seems to potentiate mitogenicity at low toxin doses (Prasad et al., 1997). While binding affinity to MHC class II does not directly correlate with the potency of the superantigens, TSST-1 possesses the lowest affinity with a IQ of 440 nM (Krakauer, 1999). X-ray crystallography of TSST-1/HLA-DR1 complexes has revealed that for TSST-1, the N-terminal beta-pleated sheet domain constitutes the majority of its MHC class II interaction region 29 (Kim et al., 1994). This study demonstrated that TSST-1 appears to interact not only with the a chain of human HLA-DR1, but also spans the peptide-binding groove to interact with the peptide contained therein and the B chain of HLA-DR1 (Figure 1.7). In contrast, SEB interacts only with the a chain of HLA-DR1 (Jardetzky et al., 1994). That the nature of the peptide presented could affect TSST-1 binding may explain why neither TSST-1 nor SEB are able to completely inhibit each other's binding. Both superantigens, however, cover residues on the DR1 cd involved in TCR recognition, suggesting that superantigens may bridge MHC class II molecules and TCR without the latter two directly interacting, unlike conventional antigen recognition (McCormick et al., 2001). Functional confirmation of the N-terminal domain's involvement in MHC class II interaction has been obtained through the use of site-directed mutants (Kum et al., 1996). Unlike SEA and SEB, TSST-1 binds not only to both the a and B chains of the HLA-DR complex, but also to the peptide presented in the peptide-binding groove. Thus, it has been postulated that the type of peptide presented may influence TSST-1 's binding and activity. Mutational studies have localized the putative TCR interaction region for TSST-1 to the shallow groove between the alpha helices on the opposite side of the molecule from the MHC class II interaction domain (Deresiewicz et al., 1994b; Blanco et al., 1990; Bonventre et al., 1995; Cullen et al., 1995; Earhart et al, 1998; Kum et al., 1996) (Figure 1.6). Studies using synthetic peptide have localized residues 125-158 on TSST-1 as being critical for its VB specificity (Hu et al., 1998). Recently, surface plasmon resonance has been used to demonstrate TSST-1 binding to V/32 TCR with a Kd of 2.3 [iM (McCormick et al, 2003). 30 Peptide Figure 1.7: TSST-1 :HLA-DR Contact. A. Ribbon diagram of TSST-1 (light green) complexed with human HLA-DR1 (spacefilling). TSST-1 residues involved in binding to HLA-DR1 are shown with ball-and-stick models. TSST-1 contacts primarily the HLA-DR1 a chain (yellow), with a small overlap on to the HLA-DR1 B chain (orange) and the peptide presented in the antigen-presentation groove (magenta). B. Detail of dotted region from A. 31 1.2.3. T cell responses to superantigens Superantigen-activated T cells proceed through a frenzy of proliferation accompanied by the release of cytokines such as IL-2, IX-12, TNFa, TNF/3 and IFN7, which in turn elicit inflammatory responses from APCs that generate more TNFa, IL-1/3 and IL-6 (Figure 1.8). Continued loop amplification of this Thl-polarized milieu produces a systemic acute phase response that leads to characteristic clinical features: hypotension from capillary leakage, high fever from endogenous pyrogens, and a disseminated, desquamating erythematous rash. The T cell response centers on the two-stage model for activation: first, stimulation through the TCR provided by the superantigen crosslinking of TCR to MHC class II molecules, and secondly, through costimulation through a variety of receptors. Thus, CD4+ T cells, which physiologically recognize MHC class II, are preferentially involved and activated, although CD8+ T cells bearing the correct VB specificity can also respond to superantigen stimulation (Makida et al., 1996). For TSST-1, the drive to clonal proliferation and production of inflammatory cytokines such as TNFa is much weaker in responsive CD8+ T cells (Akatsuka et al., 1994); instead, sometimes it results in anergy (Zhao et al., 1997) or induction of suppressor cells (Saha et al, 1996b). Following APC-assisted presentation of TSST-1 to VB2+ T cells, the T cells rapidly internalize their TCRs, limiting further antigen receptor signaling (Makida et al., 1996). Tyrosine phosphorylation events mediated through both the TCR and costimulatory molecules lead to activation of PKC0, phosphatidylinositol 3'-kinase (PI3K), Akt (Bauer et al, 2003), p38 mitogen-activated protein kinase (MAPK) (Ramirez et al., 1999) and NFAT (Murphy et al, 1999), which are essential for TNFa production by T cells (Chen et al., 2003). The T cells quickly but transiently upregulate CD 154, providing costimulatory signals through CD40 on concomitantly activated APCs (discussed in the next section), leading to the production, within hours, of an initial burst of proinflammatory cytokines including IL-2, IFN7, TNFa and IL-8 (Krakauer, 1998; Kum et al., 2001). Over the next several days, this is followed by the secretion 32 of a second wave of both pro- (IL-1/3, IL-12, TNF/3) and anti-inflammatory (IL-6, IL-10) cytokines (Kum et al., 2001). A range of co-stimulatory signals appear to be involved in T cell activation by TSST-1. For example, while TSST-1 stimulation of V/32+ T cells alone is insufficient to induce clonal expansion, the provision of costimulation with anti-CD28 antibody is sufficient to cause T cell proliferation (Hu et al., 1996). The importance of CD28 costimulation has been highlighted by in vitro antibody neutralization studies of its ligand, CD80/86, for both murine (Muraille et al., 1995) and human cells (Kum et al., 2002), and in vitro with CD28"7" transgenic mice that were shown to be protected against TSST-1-induced lethal shock (Saha et al., 1996a). For B cell presentation of TSST-1 to T cells, ICAM-1 on the B cells appears to be important for providing costimulatory signals through LFA-1 on T cells (Dennig et al, 1994). However, the relative contribution of these various costimulatory receptors upon TSST-l-mediated T cell activation in vivo remains unclear. While superantigen-mediated T cell stimulation and induction of cytokine production appears dominantly polarized toward the inflammatory Thl phenotype, arguments have been made concerning roles for humoral Th2 polarization in superantigen-exacerbated allergy and atopy, and regulatory Tri polarization in counteracting the toxic effects of superantigens (Cameron et al., 2001). Both the nature of the APC presenting the superantigen and the dose can sway the T cell response. Low doses of TSST-1 favour Th2 polarization, producing B cell polyclonal activation (Hofer et al., 1996). B cells added to purified naive CD4+ T cells also favour Th2 polarization, while monocytes used as APCs polarize toward a Thl response (Brandt et al., 2002). Whereas naive ThO cells tend to be strongly polarized toward a Thl response, a host undertaking an active Th2 response, for example during allergic flare-ups or atopic episodes, may merely have its Th2 responses exacerbated by the presence of superantigen. Along with the 3 3 polyclonal B cell activity of TSST-1 at lower doses, this has been proposed as the mechanism for involvement of TSST-1 in worsening atopic dermatitis (Lester et al., 1995). 34 Figure 1.8: T cell responses to superantigen. Superantigens promote cognate interaction between VP-specific T cells and APC, resulting in either T cell activation and proliferation (A.), or induction of anergy (B.) and/or apoptosis (C). Activated T cells produce proinflammatory cytokines such as IFNy and TNFa (D.), but eventually succumb to activation-induced cell death (E.). The combined proinflammatory cytokines from T cells (D.) and APCs (F.) can result in hypotension, shock and ultimately death of the host (G.). 35 1.2.4. Secondary effects of superantigen on other cell types Often overlooked is the bystander effect of superantigens on diverse cell types due to the plethora of circulating cytokines during the response. While TSST-1 binds to MHC class II on APCs, the signaling and induction of cytokine production is limited without either induced dimerization of MHC class II (e.g., using anti-TSST-1 antibodies) (Scholl et al., 1992), or the engagement of additional costimulatory molecules such as CD40 and/or LFA-1 (Mehindate et al, 1996). Disruption of CD40 interaction on APCs with its ligand CD 154 on T cells markedly impairs the production of IL-2, TNFa and EFNYin human PBMC treated with TSST-1 (Kum et al., 2002), highlighting the importance of T cell/APC cooperation for maximal cytokine induction. In human monocytes, TSST-1 causes rapid tyrosine phosphorylation and activation of kinases, leading to downstream activation of PLCy and PKC (Trede et al., 1994). These in turn cause translocation of transcription factors such as N F - K B (Trede et al., 1993a) to the nucleus to promote production of TNFa, IL-1/3 and IL-8 (Krakauer, 1998), possibly by transcriptional activation of AP-1 (Trede et al., 1993b). Besides the typical professional APCs, other MHC class II-bearing cells, including endothelial cells (Murphy et al, 1999) and bronchial epithelial cells, have been shown to be responsive to TSST-1 stimulation, producing TNFa; and IL-8 (Aubert et al., 2000). Even basophils from atopic dermatitis patients have been demonstrated to secrete histamine and leukotriene C4 in response to TSST-1 stimulation (Wehner et al., 2001). Internalization of TSST-1 on epithelial cells by receptor-mediated endocytosis has been suggested as the mechanism for both transcytosis of the toxin across mucosal barriers (Kushnaryov et al., 1984) and cytotoxicity toward endothelium, leading to capillary leakage (Lee et al., 1991). However, the possible cytotoxic effect of activated T cells themselves on endothelium cannot be overlooked (Krakauer, 1996). Taken together, the traditional understanding of TSST-1 primarily as a T cell-directed mitogen likely underestimates its true diversity of action in vivo. 36 Besides direct receptor-mediated interactions, TSST-1 also exerts indirect effects by promoting the release of cytokines and chemokines. Thus, it has also been described as anti-apoptotic toward monocytes because of its induction of GM-CSF (Bratton et al., 1999). Furthermore, TSST-1 has alternating pro-activation and pro-apoptotic effects on B cells, depending on the dose regime. At low doses, TSST-1 promotes B cell polyclonal activation and expansion, immunoglobulin production, and even isotype switching; however, at high doses, TSST-1 causes B cell apoptosis, probably secondary to production of IFN7 (Hofer et al., 1996). In contrast, the production of IFN7 and GM-CSF by T cells and APCs appears to promote neutrophil survival (Moulding et al., 1999). These indirect cytokine-mediated actions of TSST-1 probably complement the direct actions via receptor interactions to exacerbate the host's already exaggerated response to the toxin. 1.3. Connection between apoptosis and superantigens As a counterpoint to the proliferative response, a portion of superantigen-reactive T cells in vitro undergo rapid deletion following stimulation with SEA, SED or SEE, regardless of the presence of APCs (Kabelitz et al, 1992). Similarly, administration of SEB in vivo is accompanied by a contraction of the V/3-specific repertoire attributable to clonal deletion (Gonzalo et al., 1992; Wahl et al., 1993). The term 'activation-induced cell death' (AICD) was originally coined by Green et al. to describe apoptosis of immature or mature T cells in response to TCR ligation (Shi et al, 1989), following observations that TCR ligation in hybridomas could lead to cell autonomous death by Fas-Fas ligand interaction (Brunner et al., 1995). Once an immune response is complete, it has been suggested that the majority of the activated T cells undergo AICD by a Fas-dependent mechanism (Mogil et al., 1995), and the few remaining cells differentiate into memory or regulatory T cells (Cameron et al., 2001). With superantigens such as SEB, this process takes place after several proliferation cycles, when the majority of the 37 activated T cells die by AICD (Renno et al., 1999) in response to Fas-Fas ligand ligation among neighbouring cells (Ettinger et al., 1995; Gorak-Stolinska et al., 2002). For SEB, Fas receptor upregulation in human PBMC is noted by day 2, and significant apoptosis by day 3 (Weber et al., 2000). Elevation in Fas ligand has also been demonstrated on both lymphoid and non-lymphoid cells in response to superantigens, the latter in response to TNFa production (Pinkoski et al., 2002). This Fas-dependent mechanism constitutes a critical element in survival of superantigen-mediated shock, since lethality is higher in mice lacking Fas (Ipr) or FasL (gld) (McKallip et al., 2002; Mountz et al., 1995). However, downstream activation of the pro-apoptotic molecule Bim appears to be required for in vivo superantigen-mediated AICD, and corresponding overexpression of the anti-apoptotic Bcl-2 can also protect against apoptosis (Hildeman et al., 2002a). Despite these extensive investigations into activated T cell death with superantigens such as SEB, comparatively little information is available for the more clinically relevant TSST-1 with regards to apoptosis. However, a body of evidence exists to suggest differences between TSST-1 and SEB for their activity toward apoptosis, including TSST-1 's lack of tolerigenic activity compared to other superantigens (Ochi et al., 1993) and its failure to induce T cell-specific apoptosis (Hofer et al., 1996). The traditional understanding of AICD has recently been modified to include two distinct pathways by which activated T cells can die (Hildeman et al., 2002b). One branch, traditional AICD, depends upon re-stimulation of the activated T cell through its antigen receptor, and is dependent upon Fas ligation by neighbouring cells bearing Fas ligand. The other type is termed 'activated T cell autonomous death' (ACAD), and requires neither bystander cells nor repeated TCR ligation to take place. ACAD is the primary mechanism for negative selection of strongly self-reactive thymocytes where, upon strong TCR signaling, the thymocytes die according to a Bim-dependent pathway without the need for additional Fas receptor signaling. Despite the prior evidence pointing to involvement of Fas-mediated traditional AICD in superantigen-mediated T 38 cell apoptosis, recent studies with limiting dilutions have demonstrated that in the absence of cell contact, SEB-stimulated T cells of both CTL (Tel and Tc2) and T helper (Thl and Th2) subsets can undergo autonomous apoptosis in an ACAD fashion (Gorak-Stolinska et al., 2002). Both mechanisms may be at play under different circumstances: in conditions where superantigens continue to be produced or are reintroduced to activated T cells, the primary mechanism of T cell death may be AICD; likewise, when high cell densities of superantigen-activated CD8+ CTLs are present, these may kill other neighbouring superantigen-activated CD4+ T cells by AICD; however, in the absence of restimulation with superantigens or with insufficient pro-survival cytokines, the T cell death that ensues after the first superantigen stimulus may be primarily ACAD (Figure 1.9). In vivo, Fas-mediated apoptosis appears only partially responsible for removal of activated T cells, as transgenic mice with over-expression of C - F L I P L in their T cell compartment can still delete activated T cells (Lens et al., 2002). A detailed understanding of AICD and ACAD must take into the account the regulatory apparatus that determines cell susceptibility to death by either Fas or Bim (Figure 1.9). IL-2 signaling drives not only mitogenesis, but also up-regulation of both Fas ligand and Fas receptor, as well as downregulation of FLIPs (Algeciras-Schimnich et al., 1999), so that the same cytokine that initially promotes T cell survival and proliferation also sets the stage for these cells' eventual demise. Activated T cells are initially resistant to Fas-mediated death despite increasing expression of the Fas receptor. During and after the initial TCR ligation, co-stimulation through CD28 and growth cytokine signaling activate the PI3-kinase pathway, which phosphorylates and activates downstream Akt/PKB (Okkenhaug et al., 2003). As discussed in Section 1.1.4, Akt in turn modifies the activity of transcription factors to increase the production of anti-apoptotic proteins such as FLIPs, Bcl-2 and IAPs and suppresses the production of pro-apoptotic proteins. 39 Figure 1.9: Two Routes to Activated T Cell Death. A superantigen-activated T cell ultimately succumbs to either activated cell autonomous death (ACAD; left) or activation-induced cell death (AICD; right). A C A D ensues via a Bim-dependent intrinsic mechanism, either directly following activation (A.) or after cell growth outstrips the available growth cytokines (B.). In contrast, AICD proceeds via Fas receptor ligation after proliferation (C) . 40 1.4. Objectives and specific aims Given the paucity of information available on TSST-1 and apoptosis, the overall objective of these studies are to better understand the effect of TSST-1 on apoptosis, and the role this might play in TSS pathogenesis. The specific aims for this project were: 1. To determine whether TSST-1 affects apoptosis in a human in vitro model system 2. To characterize and test hypotheses concerning potential mechanisms for apoptosis induced by the G31R mutant of TSST-1 3. To draw structure-function correlations for the regulation of apoptosis activity using a broad set of TSST-1 mutants 41 Chapter 2 Materials and Methods "Though this be madness, yet there is method in 't. " -William Shakespeare; Hamlet, II, ii, 211 2.1. Generation of TSST-1 and mutant toxins The wild-type recombinant TSST-1 (wtTSST-1) and the mutant toxins S14N, G31R, S49N, P50S, S53R, S53K, A55T, T57S, P95F, and H135A were obtained by random and site-direct mutagenesis as described previously (Kum et al., 1996; Kum et al., 2001). After transformation of wild-type and mutant TSST-1 genes into Staphylococcus aureus RN4220, toxins expressed in pyrogen-free brain heart infusion (BHI) supernatants were purified by a combination of preparative isoelectric focusing and chromatofocusing, described previously (Kum et al., 1993) (Figure 2.1). Commercially purchased partially purified SEB (Toxin Technologies, Madison, WI) was further purified by chromato focusing (Kum et al., 1993). The purity of the toxins obtained were assessed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and silver staining; absence of lipopolysaccharide (LPS; <10 pg/ml) was verified by the Limulus amoebocyte lysate assay (Kum et al., 1993). 42 Ampr pBS (tef) 4561 bp ColE1 ori Afl\\\ ^ tst^ Step 1 : Denature plasmid DNA and anneal mutagenic (TST-MUT) & selection (TAB) primers. tef signal pBS (ts*) *A 4561 bp ¥ ] > ColE1 ori tstiL Afn„ ^—~tst signal Step 2: Primers extension and ligation. Step 3: | Primary selection with AflVW and first transformation into S Epicurian coli (E. coli). pBS (tef) 4561 bp ColE1 ori Step 5: DNA sequencing of mutant plasmids to confirm presence of desired mutations. tef signal Restrict DNA with Afllll digestion and second transformation into mut S E. coli. ColE1 ori pSK265 [pBS (tef: MUTANT]] 7561 bp Kpnl pSK265 3000 bp REP Construct shuttle plasmid by ligating Kpnl digested pBS (tst: MUTANT) with Kpnl digested pSK265 [plasmid of Staphylococcus aureus (5. aureus) origin]. AT REP • tef signal Kpnl Step 7: CAT Purify mutant plasmid and electroporate into 5. aureus RN4220 (tsr). Figure 2.1: Preparation of Site-Directed Mutant Toxins. The flow diagram describes the procedure for generating single amino acid substitutions by site-directed mutagenesis (1-4), verifying the mutations (5), and then constructing shuttle plasmids for mutant toxin purification (6-7). Amp r is the ampicillin resistance gene for antibiotic selection; pBS(tst) ColEl ori is the plasmid origin of replication; CAT of pSK265 represents the chloramphenicol acetyltransferase structural gene; and REP of pSK265 is the origin of replication of pUC19 from which pSK265 was derived. Adapted with permission from Kum et al., 1996. 43 2.2. Purification of human peripheral blood mononuclear cells Peripheral blood mononuclear cells (PBMC) were obtained by density gradient centrifugation of day-old leukopheresis packs from random, healthy donors over Ficoll-Paque™ (Amersham Biosciences Corp, Piscataway, NJ), as described previously (See et al., 1992). Cells from the interface layer were washed four times with RPMI 1640 medium (Stem Cell Technologies Inc., Vancouver, British Columbia, Canada). Purified cells (2xl05) were resuspended in 200 /xl of RPMI 1640 culture medium (Stem Cell) supplemented with 10% fetal bovine serum (FBS; heat inactivated at 56°C for 30 min; HyClone Laboratories, Inc., Logan, Utah), 2 mM L-glutamine (Stem Cell), 25 mM HEPES buffer (Stem Cell) and 2 /xg/ml polymyxin B sulphate (to neutralize any endotoxin contamination; Sigma Chemical Co., St. Louis, MO) in 96-well U-bottom tissue culture plates (Falcon Labware; Becton Dickinson Canada Inc., Mississauga, Ontario, Canada), and incubated at 37°C in 5% CO2 with the wtTSST-1 or mutants. 2.3. Growth of MOLT-16 and Jurkat cell lines The human T cell leukemia derived cell lines MOLT-16 (Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSMZ), Braunschweig, Germany) and Jurkat E6-1 (TIB-152; American Type Culture Collection, Manassas, VA) were cultured at 105-106 cells/ml in T75 tissue culture flasks (Becton Dickinson) with 30 ml of medium. RPMI 1640 medium supplemented with 10% FBS, 2 mM L-glutamine, 10 mM HEPES buffer, 1 mM sodium pyruvate (Stem Cell), 100 U/ml penicillin and 0.1 mg/ml streptomycin (Stem Cell) was used to culture the cells at 37°C in 5% C0 2 . 2.4. Isolation of murine splenocytes Spleens were removed from healthy 6-8 week old female Balb/c mice (Charles River Laboratories, Inc., Wilmington, MA) sacrificed by cervical dislocation. After homogenization 44 through a homogenization cell strainer (Becton Dickinson) in a 5 ml volume of RPMI 1640, the single cell suspension was centrifuged at 800 xg for 7 min to pellet the cells, then resuspended in 1 ml of RPMI 1640. To lyse red blood cells, 4 ml of Gey's lysis buffer were added to the cells on ice for 1 minute, followed by dilution with 5 ml of RPMI 1640 and centrifugation at 4°C at 800 xg. The cells were washed twice more with 10 ml of RPMI 1640, then resuspended at 106 cells/ml and cultured on 96-well U-bottom tissue culture plates in RPMI growth medium containing 10% FBS, 2 mM L-glutamine, 10 mM HEPES buffer, 1 mM sodium pyruvate, 100 U/ml penicillin and 0.1 mg/ml streptomycin at 37°C in 5% CO2. 2.5. Flow cytometry Antibodies and stains used for flow cytometry were: anti-CD4-phycoerythrin (PE) for CD4+ T cells, anti-CD8-PE for CD8+ T cells, anti-CD 14-PE for monocytes, anti-CD 19-PE for B cells, anti-CD95-PE for Fas receptor, and Annexin V-fluorescein isothiocyanate (FITC) (all from BD Biosciences Canada, Mississauga, Ontario, Canada); anti-V/32-PE for Vj32+ T cells (Beckman Coulter Inc., Miami, FL); propidium iodide (PI) and 7-aminoactinomycin D (7-AAD) (both from Sigma). Cells were cultured with wtTSST-1 or a mutant toxin in 96-well U-bottom tissue culture plates. After centrifuging at 1250 xg to pellet the cells, the supernatant was discarded and the cells were washed with phosphate-buffered saline (PBS; Stem Cell). Fluorescent conjugates of antibodies against cell surface markers (1:10 for anti-CD4, anti-CD8, anti-CD14, anti-CD19, anti-CD95; 1:20 for anti-V/52) were added in PBS containing 2% FBS, incubated for 20 minutes at 4°C, then either resuspended in PBS to a final volume of 400 /xl for analysis or washed for further staining. Flow cytometry was performed using either a single-laser (488 nm) EPICS™ XL-MCL™ (Beckman Coulter) or a dual-laser (488 nm, 633 nm) FACSCalibur™ (Becton Dickinson). In all cases, forward and side scattered light intensity were monitored for all 10,000 events recorded, 45 along with the appropriate fluorescent light intensity channels. Sample data were analyzed using either WinMDI 2.8 (The Scripps Research Institute, La Jolla, CA), Expo 32 (Beckman Coulter) or Cytomation Summit (DakoCytomation Inc., Mississauga, Ontario, Canada) software. 2.6. Characterization of apoptosis To ascertain that the cell death observed was apoptosis and not necrosis, multiple methods were employed, as described below. 2.6.1. Hypodiploid population by propidium iodide PI staining for the hypodiploid cell population indicative of apoptosis was performed using a hypotonic lysis buffer method (Nicoletti et al., 1991). Briefly, the toxin-treated cells were pelleted, washed, and fixed in 70% ethanol in microcentrifuge tubes at 4°C for 30 minutes. The cells were then pelleted and washed, resuspended in 1 ml DNA staining reagent (PBS, pH 7.4, with 0.1% Triton X-100 (Sigma), 0.1 mM EDTA (pH 7.4; Sigma), 0.05 mg/ml RNase A (50 units/mg; Sigma), 50 tig/ml PI), allowed to incubate at room temperature for 1 hour, and analyzed on the flow cytometer. 2.6.2. Membrane integrity by 7-aminoactinomycin D Apoptotic cell determination by combination of decrease in cell size and increase in membrane permeability was performed using 7-AAD staining (Schmid et al., 1992). The cells (2xl05) were treated with toxin, centrifuged for 2 minutes at 1250 *g, then washed with PBS and stained using 1 fig 7-AAD in 100 /xl PBS. In cases where determination of cell subsets was desired, the antibody staining for cell surface markers was performed as above, prior to 7-AAD staining. The cells were diluted to 400 \i\ volume with PBS and analyzed on the flow cytometer. 46 2.6.3. Phosphatidylserine exposure by Annexin V For Annexin V staining, toxin-treated cells were harvested from the 96-well plates at the end of the incubation period by centrifugation for 2 minutes at 1250 *g, washed once with 200 [il of PBS, and resuspended in 100 id Annexin V binding buffer (10 mM HEPES, 140 mM NaCI (Sigma), 2.5 mM CaCl2 (Sigma), pH 7.4) containing 5 /xl Annexin V-FITC and 1 jig 7-AAD for 15 minutes at room temperature (Vermes et al., 1995). The use of 7-AAD together with Annexin V-FITC allowed early and late apoptosis, as well as necrotic cells, to be distinguished (Lecoeur et al., 1997): Annexin V-FITC77-AAD" cells were considered live; Annexin V-FITC" /7-AAD + cells were considered necrotic; Annexin V-FITC /7-AAD" cells constituted the early apoptotic population; and Annexin V-FITC+/7-AAD+ cells were either late apoptotic/secondary necrotic or late necrotic, although this could be distinguished by observing the kinetics of the death response. The cells were suspended in a final volume of 400 til of Annexin V binding buffer. 2.6.4. DNA fragmentation by TdTincorporation To confirm the presence of DNA fragmentation, a characteristic of apoptosis, cells were assayed for DNA strand breaks using the Terminal deoxynucleotidyl Transferase (TdT) Fluorescein-dUTP Nick End Labeling (TUNEL) method (Gorczyca et al., 1993b; Gorczyca et al., 1993a). Cells were stained with the In Situ Cell Death Detection Kit, Fluorescein (Boehringer Mannheim Biochemicals, Indianapolis, IN) using the provided procedure. Briefly, cells (5><106) in 96-well U-bottom plates were washed and fixed with paraformaldehyde (4% paraformaldehyde (Sigma) and 1% BSA (Sigma) in PBS) at room temperature for 30 minutes. The cells were washed again and permeabilized with a PBS solution containing 0.1% Triton X-100 and 0.1 % sodium citrate (Sigma) for 2 minutes at 4°C. After washing twice, the cells were 47 incubated with the label buffer solution (45 ul) and TdT enzyme solution (5 ul) at 37°C for 1 hour. The cells were then resuspended in 1 ml PBS and analyzed on the flow cytometer. 2.6.5. Caspase 3 activation The presence of the active, cleaved form of caspase 3 in human PBMC, indicating activation of the caspase cascade, was determined by intracellular staining with a PE-labelled anti-active caspase 3-specific antibody (BD Biosciences). Toxin treated cells (2xl05) in 96-well U-bottom plates were washed with PBS, then fixed and permeabilized with 200 /xl of Cytofix/Cytoperm™ (BD Biosciences) for 20 minutes at ambient temperature. The cells were washed with PBS containing 1% FBS, then resuspended in Perm/Wash™ Buffer (BD Biosciences) for 5 minutes. Anti-caspase 3 antibody was added (1:10 dilution) in Perm/Wash™ Buffer to a volume of 100 /xl. The cells were washed again with PBS-1%FBS, then resuspended into 400 /xl of Perm/Wash™ Buffer for flow cytometry. 2.6.6. Detection of apoptotic changes by microscopy Human PBMC (2xl05) cultured with toxin on 96-well flat-bottom tissue culture plates (Becton Dickinson) were observed and photographed under an inverted light microscope (model TMS-F, Nikon, Missasauga, Ontario, Canada) at ambient temperature between 5 minutes to 24 hours after stimulation. For fluorescence microscopy, cells were stained as in Section 2.6.3 for Annexin V-FITC, but with the addition of 2 /xg/ml Hoechst 33258 dye (Sigma) along with the Annexin V-FITC and 7-AAD in Annexin V binding buffer. The Hoechst 33258 is cell-permeant and binds to the minor-groove of DNA, emitting a blue fluorescence and allowing visualization of nuclear morphology during apoptosis. After 15 minutes, the cells were pelleted at 1250 xg for 2 minutes, then resuspended in Cytofix™ buffer (BD Biosciences) for 20 minutes at 4°C to fix the cells. The cells were washed again in 200 /xl of Annexin V binding buffer, then resuspended in 100 /xl of Annexin V binding buffer and transferred to a 24-well tissue culture plate (Becton 48 Dickinson) containing a round 15 mm cover glass (Fisher Scientific Limited, Nepean, Ontario, Canada) with 300 ill of Annexin V binding buffer. The 24-well plate was spun at 1000 xg for 10 minutes to cause the cells to adhere to the cover glass. Then the cover glasses were inverted onto microscope slides with 10-15 / x l of Prolong Anti-Fade mounting medium (Molecular Probes, Eugene, OR). After overnight drying, the slides were observed and photographed using an Axioplan 2 upright fluorescence microscope (Zeiss, Thornwood, NY). 2 .7. Testing of extrinsic mechanisms for apoptosis Two key methods were employed to test for involvement of the extrinsic pathway in G31R-mediated apoptosis. First, a neutralizing antibody against Fas ligand was employed to determine whether Fas receptor might be involved. Second, general and specific caspase inhibitors were used to pre-treat cells before toxin exposure. 2.7.1. Receptor blocking antibodies Human PBMC were pre-incubated with either the anti-Fas ligand clone NOK-1 (10 /xg/ml; BD Biosciences) or with a mouse IgGi isotype control (R&D Systems Inc., Minneapolis, MN) for 15 minutes in 96-well U-bottom plates, before addition of toxin. The cells were assayed using Annexin V-FITC staining and flow cytometry as described previously (Section 2.6.3). To verify the activity of the NOK-1 clone, it was used to demonstrate inhibition of vesicular Fas ligand (Upstate Group Inc., Waltham, MA) induced apoptosis of Jurkat cells. Here, Jurkat cells (2xl05) were pre-incubated with the anti-Fas ligand clone NOK-1 at doses ranging from 0-10 jUg/ml for 15 minutes, before addition of the vesicular Fas ligand at 1:500 dilution for 3 hours. The Jurkat cells were then assessed for apoptosis by Annexin V-FITC staining and flow cytometry (Section 2.6.3) 49 2.7.2. Caspase inhibition The peptide caspase inhibitors used were cell permeable inhibitors solubilized at 20 mM in DMSO (Sigma): the general inhibitors Z-VAD-FMK (BD Biosciences) and Boc-D-FMK (Calbiochem®, EMD Biosciences Inc., San Diego, CA); the specific inhibitors for caspase 8 (Z-IETD-FMK) and caspase 9 (Z-LEHD-FMK) (both from Calbiochem). All were cell permeable, and thus were pre-incubated with the intact cells for 15 minutes prior to addition of an apoptotic stimulus. Where possible, these inhibitors were tested for their ability to inhibit Fas-mediated apoptosis in Jurkat cells. At 100 uM, the caspase 3 and 8 specific inhibitors and the Boc-D-FMK general inhibitor all reduced apoptosis to basal levels. In contrast, the caspase 9 specific inhibitor was unable to inhibit Fas-mediated apoptosis at all. Equivalent concentrations of caspase inhibitors had no effect in the absence of vesicular Fas ligand. 2.8. Testing of intrinsic mechanisms for apoptosis To investigate the intrinsic mechanisms regulating MOMP, both upstream and downstream elements of the signaling cascade were assessed. The CD47 receptor, which can transmit a death signal, was neutralized in an attempt to prevent apoptosis. Because the final endpoint for MOMP involvement is the loss of mitochondrial permeability, changes in mitochondrial transmembrane potential were monitored as an indicator of permeability. Furthermore, as MOMP results in the translocation of pro-apoptotic proteins from the intermembrane space, cytochrome c and AIF were assessed by immunoblot analysis of mitochondrial and cytosolic fractions. These were also confirmed by fluorescence microscopy. Finally, to identify possible upstream mediators for MOMP, activation of the candidate pro-apoptotic protein Bax, a multidomain member of the Bcl-2 thought to play a direct role in pore-formation, was assessed by flow cytometry. 50 2.8.1. CD47 neutralization Human PBMC were pre-incubated with either the monoclonal anti-CD47 neutralizing antibody B6H12 (BD Biosciences) or an IgGi isotype control (R&D Systems) for 15 minutes in 96-well U-bottom plates before addition of toxin. The cells were stained using Annexin V-FITC as described in Section 2.6.3, and assayed for apoptosis by flow cytometry. 2.8.2. Mitochondrial transmembrane potential determination For assessment of changes in the mitochondrial transmembrane potential (Av|/m), the ApoAlert™ Mitochondrial Membrane Sensor Kit (BD Biosciences) was used according to manufacturer instructions. The included Mitosensor dye is similar to others like JC-1, in that they preferentially accumulate and aggregate in the cationic environment of the mitochondria. Mitosensor fluoresces red when aggregated in healthy mitochondria, and green when dispersed in monomeric form in cells undergoing MOMP. Human PBMC (2xl05 cells) were treated with toxin in 96-well U-bottom plates (Becton Dickinson), then washed with 200 /xl of the included Incubation Buffer before resuspension with 1:1000 Mitosensor dye in Incubation Buffer at 37°C for 15 minutes. The cells were washed again with 200 /xl of Incubation Buffer before resuspension in 400 /xl of Incubation Buffer for flow cytometry. Percentage of green positive cells was taken to indicate cells undergoing MOMP. 2.8.3. Immunoblot analysis of mitochondrial and apoptosis proteins Mitochondrial and cytosolic fractions were prepared using the ApoAlert™ Cell Fractionation Kit (BD Biosciences) according to manufacturer instructions. Human PBMC (3-5xl07 cells) were treated with the toxin, stopped at the appropriate time point with the addition of an equal volume of ice-cold RPMI 1640 medium, then spun at 600 xg for 5 minutes at 4°C to pellet the cells. After resuspension in 1 ml of ice-cold Cell Wash buffer (provided), the cells were transferred to a 1.5 ml microcentrifuge tube. The cells were spun and resuspended in 0.8 ml of 51 ice-cold Fractionation buffer (provided) and homogenized on ice with 100 passes of a Dounce tissue grinder (Kontes; VWR International, Edmonton, Alberta, Canada). The homogenate was transferred to a fresh microcentrifuge tube and spun at 700 xg for 10 minutes at 4°C to pellet unlysed cells, nuclei and debris. The supernatant was transferred to a fresh tube and spun again at 10,000 xg for 25 minutes at 4°C to pellet the mitochondria. This supernatant was taken as the cytosolic fraction, while the pellet was resuspended with 100 ill Fractionation buffer and considered to be the mitochondrial-enriched fraction. To assess the purity of the preparations, the included anti-cytochrome oxidase subunit IV (COX4) antibody was used to probe the fractions during immunoblotting. COX4 is present only in the mitochondria, and does not leak out during MOMP. Thus, the presence of COX4 in the cytosolic fraction would suggest contamination by lysed mitochondria. The protein content of cell lysates was determined using the BCA Protein Assay Kit (Pierce Biotechnology Inc., Rockford, IL) prior to addition of SDS sample buffer (2x sample buffer consisted of 125 mM Tris-HCl, 4% SDS w/v, 20% glycerol v/v, 100 mM DTT, and 0.2% Bromophenol blue in deionized water (all reagents from Sigma)). Whole cell lysates were prepared by treating human PBMC (l-2xl07 cells) with toxin, stopping the reaction with ice-cold RPMI 1640 medium, then centrifuging the cells at 2000 xg for 10 minutes at 4°C to pellet the cells. The cells were washed in 1 ml of ice-cold Tris-buffered saline (TBS) in a microcentrifuge tube, then resuspended in 100 /xl of ice-cold TBS. An equal volume of 2X SDS sample buffer was added to lyse the cells. The cell lysates were sonicated twice for 15 seconds on ice at ~6-8 W power to decrease viscosity and disrupt DNA. All samples were heated to 95°C for 5 minutes prior to loading for electrophoresis on 12% SDS-PAGE with a 5% polyacrylamide stacking layer on a Mini-PROTEAN III apparatus (Bio-Rad Laboratories Ltd., Mississauga, Ontario, Canada). For mitochondrial and cytosolic fractions, 5 /xg of protein was loaded, while for whole cell lysates, 10 /xl of lysate was loaded, since the protein concentration could not be directly assayed. The protein was transferred to a 52 PVDF membrane (Immobilon-P; Sigma), blocked with TBS containing 2% bovine serum albumin (Sigma) or 2% skim milk powder (BD Biosciences) and blotted with primary antibody in the manufacturer-recommended buffer overnight. The primary antibodies used were: mouse monoclonal anti-cytochrome c and rabbit polyclonal anti-AD? (both BD Biosciences); rabbit polyclonal anti-Bax (Cell Signaling Technologies, Beverly, MA); anti-actin (Sigma) as the housekeeping gene to compare between samples. The secondary antibodies were goat anti-mouse IgG horseradish peroxidase conjugate or goat anti-rabbit IgG horse-radish peroxidase conjugate (BD Biosciences or Cell Signaling Technology, matching the source of the primary antibody), and streptavidin horseradish peroxidase (Sigma) for biotinylated protein markers (Cell Signaling). The West Pico Chemiluminescent Substrate (Pierce) was used on the blots at this point, and the areas of chemiluminescence identified using Hyperfilm™ ECL™ (Amersham). When necessary, blots were stripped of antibody using Western Blot Stripping Buffer (Pierce) for 15 minutes at 37°C, then rinsed in TBS before re-probing. Densitometric analysis was performed using Scion Image (Scion Corporation, Frederick, MD). 2.8.4. Fluorescence microscopy for Bax, cytochrome c and AIF translocation To confirm the immunoblot observations for the translocation of cytochrome c and AEF from the mitochondria, in situ staining and fluorescence microscopy were performed. The antibodies and stains used were: mouse monoclonal IgGi anti-Bax monomer specific (6A7 clone), mouse monoclonal IgGi anti-cytochrome c, rabbit polyclonal anti-AIF (all from BD Biosciences); mouse monoclonal IgG2a anti-cytochrome oxidase IV subunit I (serving as the positive control for the mitochondria), Alexa Fluor® 488 goat anti-mouse IgG2a, Alexa Fluor® 594 goat anti-mouse IgGi, Alexa Fluor® 594 goat anti-rabbit IgG (all from Molecular Probes); Hoechst 33258 (Sigma). Human PBMC (2xl05 cells) were cultured with toxin in 96-well U-bottom plates (BD Biosciences), centrifuged at 2000 xg to pellet the cells, washed in 200 /xl PBS, and fixed and 53 permeabilized with 100 /xl Cytofix™/Cytoperm™ (BD Biosciences) for 20 minutes at ambient temperature. After washing once with 200 /xl Perm/Wash™ Buffer (BD Biosciences), the cells were resuspended in Perm/Wash™ Buffer for 5 minutes to continue permeabilization. The cells were incubated with primary antibodies in 50 /xl of Perm/Wash™ Buffer for 20 minutes at ambient temperature, before washing again in Perm/Wash™ Buffer. Cells were resuspended in 100 /xl of Perm/Wash™ Buffer containing the secondary antibodies and 2 /xg/ml Hoechst 33258 for 20 minutes. Finally, after washes with Perm/Wash™ Buffer and PBS, the cells were transferred in 100 /xl of PBS into 300 /xl of PBS over a 15mm round glass coverslip in a 24-well flat bottom tissue culture plate (as in Section 2.6.6). The cells were spun onto the coverslip and glued to glass slides using Prolong Anti-Fade mounting medium (Molecular Probes) before visualization on an Axioplan 2 fluorescence microscope (Zeiss). 2.8.5. Flow cytometry for Bax activation The 6A7 antibody clone against the N-terminal region of Bax (BD Biosciences) detects only the monomelic form of Bax (Hsu et al., 1998; Hsu et al., 1997), whereas the rabbit polyclonal anti-Bax (Cell Signaling) detects Bax whether or not it is complexed. Human PBMC were treated with the toxin and stained as in Section 2.8.4 with anti-Bax 6A7 and rabbit polyclonal anti-Bax as the primary antibodies, and Alexa Fluor® 488 goat anti-rabbit IgG and allophycocyanin goat anti-mouse IgG (Molecular Probes) as the secondary antibodies. The cells were resuspended in 400 /xl of Perm/Wash™ buffer after the last wash with Perm/Wash™ buffer. This allowed for two-colour visualization of cells by flow cytometry for both total Bax protein and the conformational epitope. 2.9. Anti-apoptotic pathways The PI-3 kinase/Akt pathway was selected as a likely candidate for the anti-apoptotic action of TSST-1. This was investigated by immunoblot analysis for the phosphorylated form of Akt at 54 phosphoserine 473, and confirmed by flow cytometry. Furthermore, since IAPs can target caspases for ubiquitination and degradation, the relevance of this second potential anti-apoptotic pathway was assessed by immunoblot analysis for pro-caspase 3 expression. 2.9.7. Immunoblot analysis for Akt activation and caspase 3 expression Whole cell lysates were prepared after toxin treatment and immunoblotted as described in Section 2.8.3. The PhophoPlus Akt (Ser473) Antibody Kit (Cell Signaling) was used to provide primary and secondary antibodies for Akt, as well as positive and negative control cell lysates. The primary antibodies included a rabbit polyclonal anti-phosphoserine 473 Akt-specific antibody as well as a rabbit polyclonal anti-Akt antibody that was not dependent on phosphorylation. The horseradish peroxidase-conjugated goat anti-rabbit secondary antibody was provided in the kit. As in Section 2.8.3, anti-actin was used as the control for variations in protein content of the cell lysates. A rabbit polyclonal antibody against caspase 3 (BD Biosciences) was paired with a horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (BD Biosciences) to determine the total caspase 3 content of cell lysates. 2.9.2. Flow cytometry for Akt activation Human PBMC (2xl05 cells) were treated with toxin, then stained as described in Section 2.5 with anti-V/32-PE to identify the TSST-1-reactive T cell population. These cells were fixed and permeabilized as in Section 2.8.4, using the anti-phosphoserine 473 Akt-specific antibody as the primary and allophycocyanin goat anti-rabbit Ig (Molecular Probes) as the secondary, before resuspension in 400 til of Perm/Wash™ buffer for flow cytometry. 2.10. Statistical Analyses Statistical analyses were performed using Microsoft® Excel and Prism® 3.02 (GraphPad Software, San Diego, CA). Microsoft® Excel was used to collate and assemble data for further analysis in GraphPad Prism®, as well as to display some graphical plots. GraphPad Prism® was 55 used to perform all statistical tests. For experiments requiring correlation, such as protein concentration determinations, linear regression analysis was used, with calculated regression coefficient values (r2) always above 0.9. For replicate donors (at least 3 independent healthy human blood donors) where a treatment and a control group were compared for differences, paired two-tailed Student's t-test was performed with the cut-off for statistical significance requiring a p value less than 0.05. In situations where multiple treatment groups were compared for any differences, one-way analysis of variance (ANOVA) was performed using a p value of 0.05 as the cut-off for significance, before further post-testing of significant pairs by Tukey's test. 2.11. Three Dimensional Protein Structures The three dimensional depictions of protein structures were rendered using WebLab ViewerPro 4.0 software (Accelrys Inc., San Diego, CA). Molecular coordinate data was obtained from the Research Collaboratory for Structural Bioinformatics (RCSB; http://www.rcsb.org). 56 5 a* a o o o I' o O cu I Co cd C O 03 c3 rtc3c3cQcdMcdo3 *"0 , - 0 r O ""O ""O r O '"O , - 0 r O *"0 r O o o o Pi Pi p£ Pi o o o o 3 o CD C J CD CD oo oo 3 3 CD <D 3 ~ O cd 5 ca co CO £ £ £ • o o o o o o o o o o o !zf £ £ 2f £ 55 Z Z O O O O O O O O cd ca ca ca ca ca ca ca 3 O H-1 0 0 0 0 0 0 * ? 0 0 0 0 0 0 0 0 0 0 0 0 3 3 3 , l < 3 3 3 3 3 3 ca ca ca CO CO 3 ca co CO <u .„ .„ .. .„ .„ .„ .„ .„ .„ .„ .„ c c o c o c D c o c o c o c o c o c o c o c o C D C O C O > C O C O C « C O C O C O C O C O 0 0 ca CO CO ca 0 0 3 ca CO CO 0 0 3 O cd T3 cd C cd O o 005 j ca ca -a T3 cd cd c c cd cd O O t« « a _ O CU 3 cd CO CO ^ '53 CD g CO CO 'co CO o o "cd o cu w co 3 3 cd 3 o O g O J£ JD 5 5 ^ 5 o Q . a o 2 S CQ pq cu ft o co O O r n o o o 1—1 o U S3 w , 2?l 03 CD o o o ccj c3 ccj 03 S3 S3 u u z ; £ o o ^ ^ S oo oo <^  o o CO CO CO 3 o H-1 3 o CO CD CO CO -5 .2 £ H-1 03 I 00 O . 00 00 00 00 00 O . . . . H-1 PH 3 O H-1 - H > 00 ccj c« u rt o 03 KJ^  CO i>~> .22 T: CO CD CO > 2 PQ 03 03 u CD > 3 O o S3 03 > 00 g o « < ... CCJ § g a . *=3 O HJ O £ PH 00 H3 2 > CQ 03 3 CCJ O 1-1 cd co c * CO I 3 T  03 ccj 03 03 "2 TO 1 « 03 J H U TO 3 3 o H-1 00 > S CD > CD o o o S3 HH CO s CD 00 00 o u o a — w CO 43 CD CD Q ^ 3 3 B J O o o PH H5 HS 00 oo CQ oo u S X ) 3 to 5 o ,2 2 £ 43 S S " » o § > § ^ O C o3 CQ ffl U 3 s o 00 CD X X «H «H o o U U U HH l a S o co CD 43 "S •S .2 co p CD £> ^ s. > OS O PH - H ' 33 rt • o m CD CD +^ +^ ta 03 3 3 'HT'E? o o o o CD CD co co cd 03 * K ' * o o S H ~ CD CD OH OH 45 43 CO CO •3 '-3 03 03 I H CD CD co co H - i -O O 43 43 o o 00 00 HH HH CD CD CO CO =3 3 O O B B 1 i f t ' t < y 03 03 m O O - H o co CD — O CD 1 0 O CD V~t S PH HJ Q U > S o m H O O H w O Q Q P H f e O a O O f f i f f i f f i CtH 13 42 00 W PH  C N CO « H (D PH PH I S3 O s H^J c + H o CD +^ CD Q 43 CD Q U 3 +^  i2 CN IT ) ffl HH H CD S3 CD S3 1 CD S3 O VO HH S3 SH HH HH HH 03 3 H-5 ooO HJ S 00 IT ) o -4—> cd o o h-1 "3, I 1 o O CD C SPI Co CD P< o o Pi o CD" c CD 0 0 W oo 3 o cd cd u C Q CD > O o bo > cd -a cd u H ° „ 0 0 0 0 =3 O h-1 <<<< ^ ^ ^ ^ o co cd cd 0 0 3 GC CO 'co CO CO 03 CO ' 3 ' 3 ' 3 o o o h-1 h-1 h-1 0 0 0 0 C/3 3 O h-1 cd T3 cd u o CQ >-r CD > =3 O CD > oo O O ^ h J h J < H H CO " 3 o h- l =3 o h-1 =3 O h-1 i A. 0 0 0 0 -s o o pd, < < U U d d 0 0 0 0 CD CD cd T3 cd C cd U o oo CO CO o cd -a cd co S CD td 0 0 o M o CD E-i 0 0 oo 5 5 h 3 h 3 •3 £ X o H CO CD o c CD 'o co O CQ Q OQ o o o o O U u U cd cd cd cd o ' »-H o ' r H o B CD s CD s CD 3 CD o U u u cd cd cd cd oo 0 0 0 0 OO h H O 0 OO s 2 '-3 CD 0 0 .2 S3 =3 CD O 3 cd CO UH CD •3 o OQ -43 £ , .2 C cd o 2 2 % PH C H PH PH oo . 3 o CD T 3 CD * • C CD ^ ^ .-s cd CO CD < CD ' *o O T 3 PH" OO £ W <^  m ^ in in 3 '•3 o 2 •3 o 2 -3 o CD Id 0 0 =3 'E? o o CD co cd •a '* 2 CD a, .a CO •3 cd i -CD CD co cd o 2 C CD Id 1--*—» CO X ) =3 OO c CD O co CD C o o oo fl CD u S 3 CO cd UH B 2 •3 o cd 4-* _ & 0 0 CD co b m * C 1 b X ^ iS O O O O O O O O O O O O O O O O O O O O O O O H H a o -t-* •a H 0 0 oo H c2 =3 OQ 0 0 c .& CQ ft, S Q » H t o L d CD " 0 0 0 0 h ^ UH i4 Q h4 > N N N Os •n Chapter 3 Apoptosis Induced by TSST-1 Compared with SEB "Try to learn something about everything and everything about something. " -Thomas H. Huxley 3.1. Introduction Apoptosis induced by superantigens has been studied in a number of cell types and linked to both beneficial and detrimental host responses (McKallip et al., 2002; Mountz et al., 1995). The bulk of work on AICD in superantigen-reactive T cells has been undertaken with SEB. Some limited investigation of B cell apoptosis induced by TSST-1 has been reported previously. Whereas high doses of TSST-1 cause B cell apoptosis, low doses of TSST-1 appear to trigger polyclonal B cell activation (Hofer et al., 1996). Therefore, the first aim of this study was to clarify whether, and under what circumstances, TSST-1 can cause apoptosis of human peripheral blood mononuclear cells (PBMC). Then, noting the dependency of SEB-mediated T cell apoptosis on Fas receptor, the next aim was to determine whether there was a correlation between Fas receptor expression on TSST-1 treated cells and apoptosis (Weber et al., 2000). Finally, whatever the outcome, elucidation of the possible mechanisms for the regulation of apoptosis induced by TSST-1 constitutes the last specific aim of this Chapter. 3 .2. Results 3.2.1. TSST-1-induced apoptosis: dose changes everything The ability of TSST-1 to induce apoptosis in human PBMC was assessed in two stages: initially with various doses ranging from 100 pM to 0.5 uM, and subsequently at high concentrations at or above 1 uM. Annexin V-FITC and 7-AAD staining were performed as detailed in Chapter 2.6.2 and 2.6.3, with comparisons made to treatment with equivalent doses of SEB. As described in Chapter 1.1.5, the goal of these methods for characterizing cell death was 60 to distinguish apoptosis from necrosis. To review, 7-AAD as a membrane-impermeable viability indicator does not in itself distinguish apoptotic from necrotic cells until combined with information about cell size obtained by flow cytometry. In each sample population, the number of cells with both a decrease in cell size and an increase in 7-AAD uptake were divided by the total to obtain a percentage apoptosis. For later studies, Annexin V-FITC was used as a more specific early indicator of apoptosis, by measuring phosphatidylserine exposure on the cell surface. The number of cells with an increase in phosphatidylserine exposure was divided by the total cell count to obtain a percentage apoptosis. Unless otherwise specified, the results were plotted as bar graphs with standard error of the mean, and statistical significance was established at the p<0.05 level by paired Student's t-test between toxin treatments and RPMI medium control. The dose response for TSST-1 mediated apoptosis was not significantly different from RPMI medium control for all concentrations below 1 fiM when examined up to 3 days after stimulation (Figure 3.1). This contrasted with SEB at 3 days, which showed a dose-dependent elevation in apoptosis (Figure 3.1 A). However, for TSST-1 doses above 1 /xM, a strikingly rapid induction of apoptosis was seen within 3 hours of stimulation (Figure 3.2). This novel effect of TSST-1 was also extremely potent, with an EC50 between 2-5 /xM and affecting upwards of 70% of the mixed population of cells comprising human PBMC. 61 A . Figure 3.1: A. Percentage apoptosis by PI staining in human PBMC after 3 days of treatment with TSST-1 or SEB at the indicated doses. B. Percentage apoptosis by Annexin V-FITC staining in human PBMC after 3 h treatment with TSST-1 at the indicated doses. The 0 nM dose points correspond to RPMI medium control. 62 Figure 3.2: Percentage apoptosis by Annexin V-FITC staining in human PBMC after 3 h TSST-1 treatment. The 0 nM dose point corresponds to RPMI medium control. 63 5 0 -I , , , , , , 1 0 1 2 3 4 5 6 7 Time (days) Figure 3.3: The apoptotic response in human PBMC after treatment with 100 nM TSST-1 was not different from RPMI medium control within the first 5h after stimulation (A) and was significantly higher than RPMI medium control only after 3 days of stimulation (B). 64 3.2.2. Apoptosis: time response to TSST-1 and SEB At doses of TSST-1 below 1 iiM, TSST-1 did not induce apoptosis of human PBMC until after 3 days (Figure 3.3B). Furthermore, this apoptotic effect was significantly lower than that induced by SEB at equivalent concentrations (Figure 3.1 A). At doses of TSST-1 above 1 uM, TSST-1 induced apoptosis rapidly (within 3 h) and in a dose-dependent manner, demonstrating a maximal apoptotic effect (70%) at a concentration of 10 uM (Figure 3.2). It should be noted that at equivalent time points, the percentage apoptosis demonstrated by Annexin V-FITC staining (Figure 3.3A) was higher for all samples compared to that detected by permeability to 7-AAD alone (data not shown), as would be expected due to early phosphatidylserine exposure to Annexin V. A direct comparison of the Annexin V-FITC vs. 7-AAD stained cells shown (Figure 3.3A & B) is misleading, since they represent vastly disparate time scales. 3.2.3. Fas expression does not correlate with TSST-1-induced apoptosis Because most prior studies have arbitrarily restricted the concentrations of superantigens below 1 fiM, this was the range in which the mechanisms surrounding regulation of apoptosis were studied. Several studies have shown that SEB produced a dose and time dependent increase in Fas-mediated apoptosis, both in vivo (Yang et al, 1998) and in vitro (Weber et al., 2000). In both systems, Fas expression among superantigen-activated T cells was highest around day 2 after stimulation, while susceptibility to AICD did not increase until day 3. Therefore, the expression of Fas receptor (CD95) on superantigen-treated human PBMC was assessed at specific time points by antibody staining with an anti-CD95-PE antibody as described in Chapter 2.5. The results from flow cytometry for each sample were expressed as the mean fluorescence intensity of the cell population. Co-staining with Annexin V-FITC, as above, permitted correlation with the percentage apoptosis. 65 TSST-1 produced a clear increase in Fas expression with time as indicated by the mean fluorescent intensity of anti-CD95-PE staining, with maximal expression at 6 days (Figure 3 . 4 A ) . Similarly, both SEB and TSST-1 caused a dose-dependent increase in Fas receptor expression after 3 days stimulation (Figure 3 . 4 B ) . However, a comparison of the apoptotic responses to SEB and TSST-1 demonstrated that only SEB-treated cells were more susceptible to apoptosis at 3 days (Figure 3 . 4 B ) . This indicates that an increase in Fas receptor expression alone does not necessarily confer increased susceptibility to apoptosis. Instead, other factors such as Fas ligand availability, expression of endogenous caspase inhibitors (e.g. the IAPs) and the presence of intracellular modulators of caspase 8 activity (i.e. the FLIPs) may play an important role in determining the fate of superantigen-stimulated cells. 66 A . Kinetics of Fas Expression at High & Low Doses of TSST-1 Day 1 Day 2 Day 3 Day 6 B. Q. 1 + 100 pM 1 nM 10 nM 100 nM 1 uM 100 pM 1 nM 1 0 n M 1 0 0 n M 1 uM Figure 3.4: A. The expression of Fas (CD95) is shown as a mean fluorescence intensity (MFI) index, showing the response to high (1 uM) and low (1 pM) doses of TSST-1. Results are expressed as a the ratio of MFI in treated vs. untreated PBMC and are representative of three separate donors. B. The line graphs correspond to Fas expression on the right hand Y-axis and are plotted along with index of apoptosis (ratio of percentage apoptosis in treated vs. untreated PBMC) in bars on the left hand Y-axis, both after stimulation for 3 days with TSST-1 or SEB. Apoptosis with TSST-1 at or below 1 nM is not significantly different than RPMI medium control. 67 3.2.4. TSST-1 activates the Akt pro-survival pathway In order to account for the delay and reduction in apoptosis seen with conventional doses of TSST-1 (<1 uM) compared to SEB, two key candidates for anti-apoptotic activity were selected: the PI3K/Akt pathway, known to increase production of anti-apoptotic proteins and to decrease production of pro-apoptotic proteins; and caspase 3 regulation, a central target for IAPs. Whole cell lysates from TSST-1-treated human PBMC were blotted with specific antibodies against either total cellular Akt or Akt phosphorylated at serine 473 as described in Chapter 2.9.1. The amount of phosphorylated Akt in each sample, determined by densitometry, was divided by the total Akt content to yield a percentage of activated Akt. Jurkat cell lysates with or without LY294002 treatment were provided by the antibody manufacturer (Cell Signaling) as negative and positive controls, respectively. Because immunoblot analysis was relatively insensitive to the small amount of activated Akt from the limited (-5%) population of V/3-specific TSST-1-reactive T cells within the human PBMC, the results were confirmed by flow cytometry as detailed in Chapter 2.9.2. The results were expressed as a normalized index of phosphoserine 473-Akt expression in either the VB2+ or V82~ T cell population. This was calculated by taking the ratio of percentage of cells positive for phosphoserine 473-Akt with and without toxin treatment. The normalization was performed to account for variation in the baseline levels of Akt phosphorylation among donors. The cell lysates from human PBMC showed a rapid increase in phosphorylated Akt within 1 minute of TSST-1 treatment that was sustained for at least the first hour after stimulation (Figure 3.5A). Methodological limitations prevented accurate sampling before 1 minute. As a control, the total cellular Akt content was probed with another antibody, showing little variation across samples. However, Akt phosphorylation was not consistently seen in immunoblot experiments across different donors. Flow cytometry confirmed that TSST-1 caused a marked elevation of Akt phosphorylation in V/32+ T cells, but not in V/32" T cells, within 10 minutes after stimulation 68 (Figure 3.5B). This effect was consistent across multiple donors, unlike the immunoblotting of whole cell lysates that could not specifically probe for Akt phosphorylation in the small (-5%) V/32+ T cell population. Therefore the TSST-1-reactive T cells alone respond to TSST-1 stimulation by activating the PI3K/Akt pathway. 3.2.5. TSST-1 downregulates caspase 3 The second anti-apoptotic pathway investigated was that of caspase 3 regulation, a key component of the apoptotic machinery. Pro-caspase 3 expression was assessed by immunoblot using a rabbit polyclonal antibody as described in Chapter 2.9.1. The results were expressed as cellular content of pro-caspase 3, determined from densitometry. Caspase 3 activation was measured on a cell-by-cell basis by flow cytometry, using intracellular staining with an antibody specific for the cleaved, active form of caspase 3. The number of cells containing activated caspase 3 was divided by the total to obtain the percentage of cells with activated caspase 3. Immunoblotting for total cellular caspase 3 content in human PBMC demonstrated a profound loss of cellular pro-caspase 3 (-80% within 3 h) (Figure 3.6A). However, in the same time frame, no change was observed in caspase 3 activation compared to RPMI medium control (Figure 3.6B). This is consistent with our earlier data demonstrating that TSST-1 at 100 nM did not induce apoptosis in human PBMC. It also raises the possibility that the lack of apoptotic activity may be related to the depletion of procaspase 3 by unknown mechanisms. 69 A. Total Akt (60kD) P(473)-Akt(60kD) Actin (42kD) > 7? T l > n=1 B. 8 c o w 7 o I 6 0 . 2 5 n 0 . 2 -: A k t 0 . 1 5 -<C 0 . 1 -O _£= Q . tn 0 . 0 6 -O _s= O - 0 -10J 103 LU CL C<J 102 RPMI (min) TSST-1 (min) 1 5 10 30 60 1 5 10 30 60 TSST-1 100 nM \ 10° 101 102 103 104 Phospho-Akt (APC) VP2- VP2+ Figure 3.5: A. Immunoblot analysis of total Akt, phosphoserine-473 Akt and actin in whole cell lysates of human PBMC with 100 nM TSST-1 or RPMI medium control for specified times in min (1-60). The percentage of total Akt that is phosphorylated is shown by densitometry. B. Phosphoserine-473 Akt in the V(32" and Vp2" populations of human PBMC after treatment with 100 nM TSST-1 for 10 min, as determined by flow cytometry (see insert), was expressed as a ratio compared to RPMI medium control. 70 2 1.8 1.6 > 1.4 ** to Den 1.2 -O 03 0.8 © 0.6 0.4 0.2 0 B. 32kD pro-caspase-3 RPMI TSST-1 RPMI TSST-1 Figure 3.6: Caspase 3 Expression. A. Cellular pro-caspase 3 content in human PBMC as determined by immunoblot and densitometric analysis after treatment with 100 nM TSST-1 or RPMI medium control. B. Activated caspase 3 (percentage positive) in human PBMC as determined by intracellular staining and flow cytometry after treatment with 100 nM TSST-1 or RPMI medium control for 3h. 71 3.2.6. High dose TSST-1 induced apoptosis is specific When using high doses (>1 /xM) of TSST-1, a contaminant present in the toxin preparation could be responsible for apoptotic activity. Therefore, the last aim was to determine whether TSST-1 's pro-apoptotic effect at high doses (>1 juM) was specific to the toxin. A rabbit polyclonal anti-TSST-1 antibody previously generated in our lab (Rosten et al., 1987) was pre-incubated at 1:10 with 10 [iM TSST-1 for 15 minutes to neutralize its apoptotic activity. A negative control using pooled normal rabbit serum for pre-incubation with TSST-1 was performed alongside. Both mixtures were assessed for their ability to induce apoptosis in human PBMC by Annexin V-FITC staining and flow cytometry, as described in Chapter 2.6.3. The results were expressed as a normalized index of apoptosis, calculated by dividing the percentage apoptosis in treated samples with the corresponding percentage apoptosis in the RPMI medium control sample. TSST-1-induced apoptosis at 10 /xM was significantly inhibited by pre-incubation with a rabbit polyclonal anti-TSST-1 antibody (Figure 3.7A). No change was observed with the non-apoptotic lower dose of TSST-1 at 100 nM (Figure 3.7A). In contrast, normal rabbit serum had minimal effects on TSST-1-induced apoptosis (Figure 3.7B). 72 A . TSST-1 (10uM) TSST-1 (100nM) TSST-1 (10uM) + Polyclonal anti-TSST-1 TSST-1 (100nM) + Polyclonal anti-TSST-1 B. TSST-1 (10uM) TSST-1 (100nM) TSST-1 (10uM) + Normal Rabbit Serum TSST-1 (100nM) + Normal Rabbit Serum Figure 3.7: A. The apoptosis index in human PBMC with or without rabbit polyclonal anti-TSST-1 serum ( 1 : 1 0 ) was determined by Annexin V-FITC staining. The results were expressed as the ratio of percentage apoptosis after toxin treatment to percentage after treatment with RPMI medium control. B. The apoptosis index in human PBMC with or without pre-incubation with normal rabbit serum ( 1 : 1 0 ) . 73 3.3. Discussion and conclusions Within the lower concentration ranges that are commonly studied £iM), TSST-1 was noted to have little effect on apoptosis of human PBMC in vitro both in the short and long term. The small decrease in apoptosis at two days is probably relative, reflecting the rapid expansion of the TSST-1-reactive T cell subset. The increase in apoptosis after 3 days may represent delayed AICD, compared with SEB which begins to exhibit AICD within 3 days. However, this could also be the result of non-specific cell death from nutrient depletion or waste accumulation, since the highly metabolically active and proliferating T cells would utilize more nutrients, invariably causing a pH shift in the medium attributable to buildup of waste products. Visual observation of the pH indicator contained in the RPMI 1640 medium corroborates this explanation (data not shown). Other individuals in our lab who have performed long-term PBMC cultures with TSST-1 with successive changes of media report a sustained but low-level apoptosis throughout the culture period (unpublished data), although this may be confounded by the addition of IL-2 as a growth factor for the T cells. The apparent decrease in apoptosis during the first 5 hours after stimulation is attributable to an experimental artifact due to the order in which samples were analyzed on the flow cytometer (Figure 3.3A). All time points were coordinated to end at the same time for convenience of simultaneous staining, but acquisition proceeded from the longest time points to the shortest, meaning that the stained, but still live cells continued to die during the flow cytometric analysis. The sustained, low-level apoptosis exhibited by TSST-1-treated cells in the face of rising Fas expression appears incongruous at first. In previous studies, the early rise in Fas expression among the superantigen-reactive T cells without concomitant rise in apoptosis was attributed to the early activation of anti-apoptotic factors (Yang et al., 1998), such as Akt-induced downregulation of Bim and upregulation of FLIPs. IL-2 receptor signaling in particular causes Akt activation that inactivates the Forkhead transcription factor Fox03, resulting in 74 downregulation of Bim production and a decrease in sensitivity to Bim-dependent ACAD (Stahl et al., 2002). Akt-mediated expression of FLIPs directly antagonize AICD by interfering with signaling through Fas receptor (Panka et al., 2001). Thus, the confirmation that rapid Akt activation appears to occur in TSST-1-reactive V/32+ T cells affirms the involvement of the PI3K/Akt pathway, paving the way for further studies that would aim to establish the role of downstream targets of Akt in prevention of apoptosis. This would include the use of chemical inhibitors of the PBK/Akt pathway such as wortmannin and LY294002. The reduction in cellular content of pro-caspase 3 after TSST-1 treatment is surprising, considering the short time frame involved (3 h) and lack of caspase 3 activation. The absence of a shift from pro-caspase 3 to the active form of caspase 3 suggests that pro-caspase 3 is lost, possibly by degradation. Although the cell lysates represented the protein content of the mixture of cells in human PBMC, the magnitude of the reduction (-80%) is substantially greater than could be explained by a loss of pro-caspase 3 from only the V/32+ population (-5% of total PBMC). The effect, therefore, would seem to affect a much wider population than the superantigen-reactive T cells alone. One possible explanation is that cIAP2 can promote monoubiquitination of caspase 3 and 7 (Huang et al., 2000). This suggests that TSST-1 might initiate caspase 3 processing in many cell types, as with other forms of TCR ligation (Alam et al., 1999), but immediately targets caspase 3 for degradation. The corresponding lack of caspase 3 activation coincides with the lack of apoptosis in human PBMC seen at these lower doses of TSST-1. Most investigators have traditionally set arbitrary limits on the dose ranges examined for superantigens. In this study, a one order of magnitude increase above the usual ceiling of 1 fjM revealed a previously unknown, early and potent apoptotic effect of TSST-1. The induction of apoptosis could be neutralized using a polyclonal anti-TSST-1 antibody, suggesting, but not proving, that the activity was specific to the toxin rather than a contaminant. The lack of batch-75 to-batch variability in the apoptotic dose response (data not shown) and the description of potently pro-apoptotic mutants of TSST-1 prepared by the same method (Chapters 4 & Appendix A) more definitively establish the specificity. The mechanism for this apoptotic activity is unclear, but the effect appears unrestricted to the V/32+ TSST-1-reactive T cells, considering the magnitude of apoptosis. Thus, it should be clarified that descriptions of low-dose TSST-1 as 'anti-apoptotic' reflect only its activity on the limited V/32+ T cell subset, in contrast to the high-dose 'pro-apoptotic' effects on this wider cell population. To place all these findings in perspective, a pathogen such as S. aureus seeking to evade immune detection might benefit from superantigen-derived T cell responses by the diversion of specific immune responses away from the organism (Figure 3.8). In conjunction with either TSST-1-induced polyclonal B cell activation and/or B cell apoptosis, this would render ineffectual immune responses against S. aureus by respectively re-directing immunoglobulin production against non-specific targets or by deleting B cells that might produce S. aureus-specific antibodies (Figure 3.8C-F). From the host perspective, TSST-1 further subverts the host response by recruiting most of the superantigen-reactive T cells into a proliferating and inflammatory Thl state without the recourse of dying gracefully (Hofer et al., 1996), as would otherwise be the case (Figure 3.8G-H). The lethality of SEB is enhanced in animal model systems where AICD is negated (McKallip et al., 2002; Mountz et al., 1995), so it would be reasonable to suppose that lack of or delay in AICD contributes in this manner to TSST-1 's virulence. This is supported by observations made with the non-lethal MHC class II binding-deficient glycine 31 to arginine mutant (G31R) of TSST-1, more fully described in the next Chapter. Initial characterization of this mutant demonstrated a loss of activity in both T cell mitogenesis and cytokine secretion (Kum et al., 1996). This reduction in T cell proliferation and cytokine production was subsequently found to be associated with apoptosis induced by G31R, in distinct 76 contrast to the inhibitory activity against apoptosis demonstrated by TSST-1. Indeed, the early apoptotic activity observed with high doses of TSST-1 above 1 uM mirrors the apoptotic activity induced by G31R at much lower doses below 1 uM, as described in the next Chapter. 77 Figure 3.8: Roles of TSST-1 in Pathogenesis. Besides the induction of T cell activation and proliferation (A.) that leads to proinflammatory cytokine production and a detrimental host response (B.), S. aureus may also use TSST-1 to engage in diversionary tactics. Thus, low doses of TSST-1 elicit polyclonal B cell activation (C), while higher doses cause IFNy-dependent B cell apoptosis (D-F). Finally, TSST-1 also exacerbates the proinflammatory response by inhibiting T cell AICD (G.), possibly via Akt activation (H.). 78 C h a p t e r 4 Characterization of Apoptosis Induced by the G31R Mutant of TSST-1 "T'is strange, but true; for truth is always strange,—stranger than fiction. " -Lord Byron 4.1. In t roduct ion Before the pro-apoptotic effect of high dose TSST-1 (>1 uM) was recognized (Chapter 3), the first indication that studies of apoptosis may be relevant came from our earlier characterization of the glycine 31 to arginine (G31R) site-directed mutant of TSST-1. We previously observed that G31R was ~4 orders of magnitude less potent than TSST-1 in inducing mitogenesis and TNFa production (Kum et al., 1996). This was attributed to its defect in binding to MHC class II, rendering the toxin largely incapable of promoting the cognate T cell to APC interactions necessary to recruit costimulation for full activation. Furthermore, at doses between 10-1000 nM, the T cell proliferative activity dropped precipitously, unlike TSST-1 which showed sustained proliferative activity in this dose range (Figure 4.1). This could be the result of anergy or apoptosis in the G31R-reactive T cell population, in light of prior studies demonstrating AICD and anergy with other superantigens. The aim in this chapter was therefore to determine whether the loss of proliferative activity in human PBMC with G31R at concentrations ranging from 10-1000 nM was due to apoptosis. 79 A. ~c o TO i_ o a i_ o o _ ^ £ c o, '•B E • I CO 25000 20000 : 15000 10000 5000 0 *"» ^ K» Q ^ G31R mutant * * # # # # wild-type TSST-1 B. o o CL U o o _ ^ S <D CL C O, T 3 E CO IN* ^ G31R mutant wild-type TSST-1 Figure 4.1: A. Proliferation of human PBMC demonstrated by 3H-thymidine incorporation following treatment with G3 IR vs. TSST-1 for 48h (0.1 fM - 1 uM by 10-fold increments); (reproduced from Kum et al., 1996, with permission). B. Overlay of proliferation (bars, left Y-axis) and percentage apoptosis by PI staining (lines, right Y-axis) in human PBMC after toxin treatment for 3 days. 80 4.2. Results 4.2.1. G31R causes apoptosis in a dose-dependent manner, corresponding to its decreased activity in mitogenesis Unlike TSST-1, which sustained proliferative activity up to 1 juM, G31R did not induce proliferation in human PBMC at doses above 10 nM (Figure 4.1 A). An overlay of the percentage apoptosis induced by G31R after 3 days shows a direct concordance of apoptotic activity with the loss of proliferation at doses above 100 pM (Figure 4. IB). The same pattern was observed at earlier time points, such as 3 hours, where the EC50 for apoptotic activity measured by Annexin V-FITC was -50 nM (Figure 4.2B). Furthermore, the magnitude of the response affecting upwards of 60% of the total PBMC population, suggested that more than one subpopulation was involved. By contrast, wild type TSST-1 had no significant effect on apoptosis compared with RPMI medium control over the dose range from 10-1000 nM. 4.2.2. G31 R-induced apoptosis is rapid Treatment of human PBMC with 100 nM G31R or TSST-1 for up to 5 hours revealed a time-dependent increase in apoptosis (Figure 4.3B). A significant increase in Annexin V staining with G31R, compared with the RPMI medium control, was seen after only 30 minutes (Figure 4.3B). At 100 nM, the time to half maximal apoptosis (ty2) for G31R was ~1 hour. In contrast, 100 nM TSST-1 produced no significant change in apoptosis compared with RPMI medium control during the 5 hour time course (Figure 4.3B). 81 B. -•-G31R -•-TSST-1 OnM 10 nM 20 nM 50 nM 100 nM 200 nM 500 nM 1000 nM Concentration ^ G 3 1 R -•-TSST-1 OnM 10 nM 20 nM 50 nM 100 nM 200 nM 500 nM 1000 nM Concentration Figure 4.2: A. Percentage apoptosis indicated by 7-AAD staining in human PBMC after 3 h toxin treatment. B. Percentage apoptosis indicated by Annexin V-FITC staining in human PBMC after 3 h toxin treatment. In both cases 0 nM corresponds to the RPMI medium control. 82 A . 100 nM G31R o -l 1 1 1 1 — | 0 1 2 3 4 5 Time (h) Figure 4.3: A. Time-dependent increase in apoptosis in human PBMC after treatment with 100 nM G3 IR as shown by flow cytometry after double-staining with 7-AAD (X-axis) and Annexin V-FITC (Y-axis), hiitially healthy cells (0.0 h; region I) begin to stain with Annexin V-FITC (2.5 h; region II) before finally losing plasma membrane integrity and incorporating 7-AAD (5 h; region III). B. Percentage apoptosis indicated by Annexin V-FITC staining for human PBMCs after treatment with 100 nM G3 IR, 100 nM TSST-1 or RPMI medium control. 83 4.2.3. Characteristics of G31 R-induced apoptosis in human PBMC To ascertain whether G31 R-induced cell death in human PBMC was apoptosis, the treated cells were assessed using some of the methods described in Chapter 1.1.5. These relied on examination of cells for various hallmarks of apoptosis: cell morphology, nuclear changes, cell membrane integrity, and phosphatidylserine exposure. G31R induces apoptotic cellular morphology Changes in cell morphology were observed by treating human PBMC with 100 nM G31R and monitoring the cells under a light microscope as described in Chapter 2.6.6. Healthy cells were identified by their smooth, round appearance, whereas apoptotic cells were distinguished by cell shrinkage and membrane blebbing or zeiosis (Figures 4.4A and 4.6A). Time lapse photography was performed to document these changes during the first 24 hours after toxin stimulation. In order to ensure the same microscopic field was imaged in each successive frame, the tissue culture plate was kept on the microscope stage at ambient temperature, rather than at 37°C as described elsewhere. The cell populations were also assessed by flow cytometry, as described in Chapter 2.5, for morphologically distinguished populations of live and dead cells (Figures 4.4B and 4.5A). Healthy cells are shown at time 0 in both Figure 4.5A and Figure 4.5B. With induction of apoptosis, the population of dying cells shifts toward decreased forward scattered light intensity, indicating cell shrinkage, and increased side scattered light intensity, indicating increased granularity (i.e. shifting from Live to Dead dotted regions at 5.0 h in Figure 4.5A). For an accurate time response, the cells were incubated with toxin in tubes at ambient temperature without washing or staining prior to analysis by flow cytometry. The results were expressed as a percentage apoptosis by dividing the number of cells in the Dead region by the total number of cells in both the Live and Dead regions of the forward scatter-side scatter dotplot (Figure 4.5B). 84 Time-lapse photography demonstrated increasing cell shrinkage and zeiosis (membrane blebbing) in human PBMC congruent with an apoptotic process following treatment with 100 nM G31R (Figure 4.4A). The cell photographed and observed up to 5 hours is representative of the population examined, in that it demonstrates distinct roughening of the plasma membrane, unlike the smooth surface of healthy cells seen at 0.5 h. Since time is required to add toxin and centrifuge the cells on to the bottom of tissue culture plate wells, and to prepare for photography, the earliest observable time point was 15 minutes after toxin stimulation. While photographs were taken every 5 minutes, changes in cell morphology observable by microscopy occurred over hours, so only selected time points at half-hour intervals are shown (Figure 4.4A). With only one microscope stage available, the plate could not be moved, so that a simultaneous matched RPMI medium control was unavailable; the early time points were therefore used to represent RPMI medium control. Also, due to technical issues, the focus gradually deteriorated toward the later time points. Similar changes were observed in the overall cell populations by flow cytometry. Cells treated with 100 nM G31R, even at ambient temperatures, rapidly underwent shrinkage and a slight increase in granularity (Figure 4.5A). The rise in the proportion of cells undergoing these morphological changes was evident within 30 minutes after stimulation (Figure 4.5B). In contrast, treatment with 100 nM TSST-1 or RPMI medium control had no effect on the percentage apoptosis (Figure 4.5B). 85 A>.5 > — *» * - — ** 3.0 h 2.0 h 0 1023 Forward Scatter - 2.5 h 0 1023 Forward Scatter 0 1023 Forward Scatter 4.0 h 0 1023 Forward Scatter 0 1023 Forward Scatter 1.5 h 2.0 h 3.0 h 1 -MnU O 3.5 h d m •s 10 0 10' 10- 10 ' 10' 7-AAD 5 CO < i 10 u 10' 10 : 10 : 10" I 10" 10' 10 : 10- 10 ; 7-AAD • 7-AAD 4.0 h 4.5 h 5.0 h l a I — s o { 10' 10- 10 s 7-AAD Figure 4.4: Time dependent cellular morphology of human PBMC (after treatment with 100 nM G3 IR). A. Progressive membrane blebbing (zeiosis) observed by light microscopy. B. Flow cytometry demonstrates decreasing cell size (forward scatter) and increasing granularity (side scatter). C. Simultaneously, the apoptotic cells first express phosphatidylserine (Annexin V-FTTC staining), then become permeable (7-AAD). 86 A . 0 1023 ^ ^ ^ H 0 1023 Forward Scatter ^ ^ ^ H Forward Scatter B. 0 20 40 60 80 100 Time (minutes) Figure 4.5: A. Time-dependent cellular morphology of human PBMC by flow cytometric analysis after treatment with 100 nM G31R at 37°C showing decreasing cell size by forward scatter (X-axis) and increasing granularity by side scatter (Y-axis). B. Percentage apoptosis indicated by changes in cell morphology after treatment with 100 nM G3 IR at ambient temperature. 87 G31R induces chromatin condensation The nuclear morphology of human PBMC treated with 100 nM G31R, 100 nM TSST-1 or RPMI medium control is shown in Figure 4.6B. Examination for chromatin condensation was conducted using the DNA minor groove-binding dye Hoechst 33258 and fluorescence microscopy, as described in Chapter 2.6.6. Healthy cells have euchromatin, which stains uniformly to give the appearance of speckled nuclei. By contrast, apoptotic cells have condensation of chromatin around the periphery of their nuclei, leaving characteristically decreased staining in the centre. As seen in Figure 4.6B, RPMI and TSST-1-treated cells had speckled, mostly uniform nuclei, indicative of healthy euchromatin. In contrast, cells treated with 100 nM G31R had significant chromatin condensation around the periphery of the nucleus. 88 RPMI G31R TSST-1 (3%) (61%) (5%) Figure 4.6: Micrographs of human PBMC after treatment with 100 nM G3 IR, 100 nM TSST-1 or RPMI medium control for 3 h (n=3). A. Brightfield image. B. Blue fluorescence from Hoescht 33258 nuclear stain showing chromatin condensation with G3 IR. C. Green fluorescence from Annexin V-FITC stain showing phosphatidylserine exposure after treatment with G31R but not with TSST-1 or RPMI medium control. D. Overlay of green and blue fluorescence. E. Overlay of green and blue fluorescence on brightfield image. 89 G31R induces DNA fragmentation Internucleosomal DNA degradation, another hallmark of apoptosis, was assessed using two surrogate markers, DNA content and TUNEL. Whereas healthy resting cells have diploid (2N) DNA content, apoptotic cells lose DNA and thus have hypodiploid (<2N) content. Propidium iodide was used to stain human PBMC treated with either G31R or TSST-1 as described in Chapter 2.6.1. Percentage apoptosis was defined as the proportion of the total cell population with hypodiploid DNA content (Figure 4.7A). TUNEL staining for DNA strand-breaks secondary to internucleosomal DNA cleavage was performed on toxin-treated human PBMC according to Chapter 2.6.4. The results were expressed as the percentage of cells in the total population containing elevated DNA fragmentation (Figure 4.8A). Asssessing the DNA content of cells treated with G31R by PI staining demonstrated a dose-dependent increase in the hypodiploid population after 3 days in culture (Figure 4.7B). In contrast, TSST-1 exhibited no change in the proportion of cells with decreased DNA content (Figure 4.7B). TUNEL staining for DNA strand breaks demonstrated a similar increase in cells with DNA fragmentation for 500 nM G31R-treated human PBMC (Figure 4.8B). Here, the topoisomerase I inhibitor camptothecin was used as a positive control for apoptosis. Again, TSST-1 treated cells were no different than the RPMI medium control (Figure 4.8B). 90 A. s 0 nM 0.05 nM 0.5 nM 5 nM 50 nM 500 nM Figure 4.7: A. A sample histogram showing analysis of Pl-stained cells for the hypodiploid population (<2N, apoptotic). B. Percentage apoptosis indicated by PI staining for the hypodiploid population in human PBMC treated with 500 nM G3 IR or TSST-1 for 3 days. 91 10° 10' 102 103 FITC-dUTP Fluorescence 104 B. RPMI G31R500nM TSST-1 500 nM Camptothecin 2.5 MM Figure 4.8: A. Sample histogram showing TUNEL-stained apoptotic human PBMC after treatment with the topoisomerase I inhibitor camptothecin. B. Percentage apoptosis of human PBMC after treatment with various toxins or RPMI medium control for 3 days, as indicated by DNA fragmentation and flow cytometric analysis after TUNEL staining. 92 G31R induces membrane permeabilization Loss of plasma membrane integrity, a late marker for apoptosis, was assessed by 7-AAD staining in conjugation with cell morphology. Human PBMC were treated with G31R or TSST-1 in the dose range from 10-1000 nM for 3 hours before staining as described in Chapter 2.6.2. The results from flow cytometry were expressed as a percentage apoptosis, calculated from the number of cells that uptake 7-AAD but have decreased forward scatter (cell size), divided by the total number of cells. This combined approach (membrane integrity and cell size) helped to minimize the inclusion of necrotic cells that would also incorporate 7-AAD. Human PBMC began to uptake 7-AAD following G31R exposure, in a dose-dependent manner (Figure 4.2A). This indicated an increase in plasma membrane permeability and therefore inability to exclude this dye. RPMI medium control and TSST-1 had no effect on membrane permeability to 7-AAD. In dual staining experiments where both 7-AAD and Annexin V-FITC were used, it can be seen that the increase in permeability to 7-AAD occurred after phosphatidylserine exposure (Figure 4.3A and 4.4C). G31R induces phosphatidylserine inversion The final method for apoptosis characterization involved an assay for phosphatidylserine exposure on the plasma membrane using Annexin V-FITC staining, as discussed in greater detail in Chapter 1.1.5. Annexin V-FITC, which binds specifically to phosphatidylserine exposed on the cell surface, provided an early and specific indicator of apoptosis. Human PBMC treated with G31R or TSST-1 were stained as described in Chapter 2.6.3. The results were expressed as the percentage apoptosis, calculated as the number of cells with increased staining for Annexin V-FITC divided by the total. While the combined staining with 7-AAD can be used to distinguish early apoptotic cells (population II in Figure 4.3A) from late apoptotic or necrotic cells (population III in Figure 4.3A), for the purposes of this analysis, only the total percentage apoptosis (early and late) was used for comparisons. These human PBMC were also stained with 93 Annexin V-FITC and Hoechst 33258 for fluorescence microscopy, as described above and in Chapter 2.6.6. This permitted localization of these stains within the cells, as well as correlations with their morphological appearance. Annexin V-FITC staining became the assay of choice due to its sensitivity for apoptosis during the early stages. G31R induced both dose- (Figure 4.2B) and time- (Figure 4.3B) dependent increases in phosphatidylserine exposure, as described above, whereas RPMI and TSST-1 had no effect within the time (0-5 h) and dose (0-1000 nM) range studied. Fluorescence photomicrographs of these same cells after 3 hours showed definite differences in surface Annexin V-FITC staining (Figure 4.6C). G31R-treated human PBMC acquired a green fluorescent corona, corresponding to extensive surface exposure of phosphatidylserine. TSST-1 and RPMI treated cells do not exhibit green fluorescence. An overlay of the Hoechst 33258 nuclear stain with the Annexin V-FITC image showed the centrally-located nucleus surrounded by the bright green-staining plasma membrane around the periphery (Figure 4.6D, centre). Finally, an overlay of both fluorescent images on the brightfield photomicrograph showed a direct concordance between the membrane blebbing and green fluorescence after treatment with G31R (Figure 4.6E, centre). This contrasts with cells after treatment with RPMI or TSST-1, both of which showing round, smooth plasma membranes and euchromatic nuclei (Figure 4.6E, left and right). 4.2.4. G31R-induced apoptosis is specific In order to determine whether the G31R-mediated apoptosis was specific to the toxin rather than the result of a contaminant, a rabbit polyclonal anti-TSST-1 antiserum was employed (Rosten et al., 1987), as described in Chapter 3.2.6. Either the polyclonal antibody or a matched normal rabbit serum control at a 1:10 dilution were pre-incubated with 100 nM G31R or TSST-1 before addition to human PBMC. If the apoptosis is specific to the G31R, then it would be 94 expected that the polyclonal antibody would be able to neutralize its apoptotic activity. As above, the cells were stained with Annexin V-FITC and analyzed by flow cytometry. Results were expressed as an index of apoptosis, calculated by dividing the percentage apoptosis in treated samples by the percentage apoptosis in matched samples treated with RPMI medium control. With the addition of neutralizing rabbit polyclonal anti-TSST-1 antibody, G31 R-induced apoptosis was reduced to basal levels (Figure 4.9A). By contrast, using normal rabbit serum to pre-incubate toxin as a control demonstrated no effect on G31R-induced apoptosis (Figure 4.9B). These results, coupled with the finding that treatment with wild-type TSST-1, which was expressed and purified in identical fashion as G31R but did not induce apoptosis at the same concentrations, indicated that the apoptotic effect of G31R is likely specific and not attributable to the presence of a contaminant. 95 TSST-1 (100nM) G31R(100nM) TSST-1 (100nM) + G31R (100nM) + Polyclonal anti- Polyclonal anti-TSST-1 TSST-1 TSST-1 (100nM) G31R (100nM) TSST-1 (100nM) G31R(100nM) + + Normal Rabbit Normal Rabbit Serum Serum Figure 4.9: A. The apoptosis index (ratio of percentage apoptosis after toxin treatment to percentage apoptosis after treatment with RPMI medium control) in human PBMC after treatment with G3 IR or TSST-1 for 3 h and analyzed by Annexin V-FITC staining and flow cytometry, with or without pre-incubation with rabbit polyclonal anti-TSST-1 serum (1:10). B. The apoptosis index in human PBMC after treatment with G3 IR or TSST-1 for 3 h and analyzed by Annexin V-FITC staining and flow cytometry with or without pre-incubation with normal rabbit serum (1:10). 96 4.2.5. G31 R-induced apoptosis requires minimal exposure A similar approach was taken to ascertain whether human PBMC require prolonged exposure to the toxin for it to achieve cytotoxicity. That is, we wished to address whether the signal for cells to die was delivered completely at the outset at G31R exposure, or was instead accumulated with different cells reaching the death signaling threshold at different times. Human PBMC were therefore treated with 100 nM G31R or TSST-1 and immediately pelleted and washed to remove the majority of the toxin. Any residual unbound toxin was neutralized using the rabbit polyclonal anti-TSST-1 antibody, while control cells received normal rabbit serum, both at a 1:10 dilution. The cells were then stained with Annexin V-FITC as above to assess apoptosis. If G31R uniformly activated a death signal in the target cells immediately upon binding, then washing the toxin off should have had no effect on the final percentage apoptosis. Once human PBMC were treated with G31R, washing off the toxin immediately and neutralizing any residual free toxin with rabbit polyclonal anti-TSST-1 antibody had no effect on reducing the extent of apoptosis (Figure 4.1 OA). Curiously, normal rabbit serum seemed to minimally reduce apoptosis in washed, G31R-treated cells, but this level of apoptosis was no different from G31R-treated cells that were just washed without the addition of any serum (data not shown). Cells treated with TSST-1 and RPMI medium control were unaffected by removal of toxin, with or without addition of either serum (Figures 4.1 OA & B). 97 • Untreated (left) • Anti-TSST-1 rabbit serum (right) RPMI G31R100nM TSST-1 100 nM H Untreated (left) • Normal rabbit serum (right) RPMI G31R100nM TSST-1 100 nM Figure 4.10: A. Percentage apoptosis in human PBMC after treatment for 3 h with G3 IR, TSST-1 or RPMI medium control and analyzed by Annexin V-FITC staining and flow cytometry. Cells were either untreated (left bar of each pair) or were washed and treated with a rabbit polyclonal anti-TSST-1 antibody (right bar of each pair). B. Percentage apoptosis in human PBMC by Annexin V-FITC staining, with the same treatments as A, except that a normal rabbit serum control is used instead of anti-TSST-1 antibody. 98 4.2.6. G31 R-induced apoptosis affects diverse cell subpopulations The next specific aim was to identify the cell types susceptible to G31R-mediated apoptosis. Human PBMC comprise a mixed population containing mostly T and B lymphocytes and monocytes. The ability of G31R to induce apoptosis in CD4+ and CD8+ T cells, B cells and monocytes were assessed by combining specific antibody staining, described in Chapter 2.5, with Annexin V-FITC staining, described in Chapter 2.6.3. The CD4+ T cells, CD8+ T cells, B cells and monocytes were identified using anti-CD4-PE, anti-CD8-PE, anti-CD 19-PE and anti-CD 14-PE, respectively. In each case, logical gating was performed during the analysis to include only the positively stained cell subpopulation. The percentage apoptosis was calculated within this subpopulation as above, dividing the number of cells with phosphatidylserine exposure by the total number of cells in the subpopulation. The TSST-1-reactive V/32+ T cell population was identified using anti-V/32-PE as above. However, to determine whether G31 R-induced apoptosis was specific to V/32+ T cells, the percentage apoptosis in both the V/32+ and V/32" cell populations were calculated and compared. Because VB2+ T cells undergo TCR downregulation by internalization following TSST-1 stimulation (Makida et al., 1996), rendering all cells V/32" shortly after TSST-1-treatment (Kum et al., 2002), the TSST-1 treatment group could not be analyzed in this manner. Therefore, the comparison in apoptosis among V/32+ and V/32" cells was only available between G31R and RPMI treatment groups. Similar levels of apoptosis were observed among different cell populations within human PBMC following the treatment with 100 nM G31R (Figure 4.11). CD4+ and CD8+ T cells, as well as monocytes, all demonstrated 80-90% apoptosis as measured by Annexin V-FITC staining by 3 hours (Figure 4.11A-C). While attempts were made to determine the susceptibility of the B cell population to G31R-mediated apoptosis, this was unsuccessful due to their rapid die-off under normal culture conditions. Both the V/32+ and V/32" populations within human PBMC had similarly high levels of apoptosis (60-70%) following treatment with 100 nM G31R, indicating 99 that the apoptotic effect is definitely not specific to the conventional TSST-1-reactive V/32+ subset of T cells (Figure 4.1 ID). As mentioned above, a comparison with 100 nM TSST-1 treatment could not be made for the V/32+ T cell population, since following TSST-1 treatment the surface expression of Vp2 among TSST-1-reactive T cells was immediately down-regulated, rendering them indistinguishable from the V/32" population. 100 A . 100 CD4 + T cells B. 100 CD8 + T cells RPMI G31R TSST-1 RPMI G31R TSST-1 c . 100 90 80 70 60 in o Q. O < o 50 « V 40 o i _ o OL 30 20 i 10 ^ CD14+ monocytes D. n=3 VP2+ and VP2T cells I RPMI G31R TSST-1 RPMIVB2+ RPMIVp2- G31RVB2+ G31RVp2-Figure 4.11: Percentage apoptosis in human P B M C after treatment for 3 h with 100 nM G3 IR, 100 nM TSST-1 or RPMI medium control, and analyzed by staining with Annexin V -FITC and PE-conjugated antibodies against: A. CD4~ T cells; B. C D r T cells; C. CD14+ monocytes; and D. VP2 T and Vp2" T cells. 101 4.2.7. G31R causes apoptosis in cell types other than human PBMC To understand the extent of G31R's effects, we sought to determine whether G31R could induce apoptosis in other cell types apart from human PBMC. Two non-V/32 bearing human T leukemia-derived cell lines that express TCR, MOLT-16 (Dao et al., 1993) and Jurkat E6-1, were chosen to eliminate bystander activity from either APCs or V/32+ T cells that might be reactive to TSST-1. These cells were cultured as described in Chapter 2.3. Jurkat cells in particular are commonly used with SEB stimulation, since they bear SEB-reactive V/38-specific TCRs. Murine splenocytes were selected to determine species specificity of the apoptotic activity, and were purified according to Chapter 2.4. As with human PBMC, the cells were treated with G31R or TSST-1, and stained with Annexin V-FITC to identify apoptotic cells. The results in all cases were expressed as percentage apoptosis, calculated by dividing the number of cells with phosphatidylserine exposure by the total number of cells examined. G31R was found to induce dose-dependent apoptosis in both the T leukemia cell lines MOLT-16 and Jurkat E6-1 (Figure 4.12). MOLT-16 T cells were considerably less susceptible to the apoptotic effect of G31R compared to human PBMC, as nearly an order of magnitude greater concentrations were required to demonstrate any apoptotic effect (Figure 4.12A). TSST-1 and RPMI medium control had no effect on apoptosis up to 1000 nM. In contrast, Jurkat cells were similar to human PBMC in terms of their response to G31R, requiring only that the concentration be above 10 nM (Figure 4.12B). The apparent decrease in apoptosis at low doses of G31R compared with TSST-1 was the result of an experimental artifact caused by the order of sample acquisition on the flow cytometer, reflecting the gradual rise in background apoptosis over time. Murine splenocytes were also less susceptible to G31R-induced apoptosis than human PBMC by an order of magnitude, requiring above 200 nM to induce significant apoptosis above background (Figure 4.13). 102 0 10 20 50 100 200 500 1000 Dose (nM) 0 J< 1 1 1 1 1 1 0 0.1 1 10 100 1000 Dose (nM) Figure 4.12: A. Dose-dependent percentage apoptosis in MOLT-16 cells treated with either G3 IR or TSST-1 for 3 h and analyzed by Annexin V-FITC staining and flow cytometry. B. Dose-dependent percentage apoptosis in Jurkat cells treated with either G3 IR or TSST-1 for 3 h and analyzed by Annexin V-FITC staining and flow cytometry. In both cases, 0 nM corresponds to RPMI medium control. 103 Concentration (nM) Figure 4.13: Dose-dependent percentage apoptosis in murine splenocytes treated with either G3 IR or TSST-1 for 3 h and analyzed by Annexin V-FITC staining and flow cytometry. A dose of 0 nM corresponds to RPMI medium control. 104 4.2.8. TSST-1 does not suppress G31 R-induced apoptosis Finally, in light of the findings from Chapter 3 that low doses of TSST-1 (<1 fiM) were protective against apoptosis in human PBMC, it was of interest to determine whether TSST-1 could suppress G31R-mediated apoptosis. The cells were therefore treated simultaneously with G31R, in doses ranging from 0-1000 nM, along with TSST-1, from 0-1000 nM. After staining with Annexin V-FITC, the percentage apopotosis was determined, and the results were expressed by plotting G31R dose response curves showing the effect of adding either no TSST-1 (RPMI medium control) or 10 nM, 100 nM and 1000 nM TSST-1 (Figure 4.13A). For each G31R dose response curve, the EC5o was determined using a variable-slope sigmoidal curve fit. These were graphed and compared by one-way ANOVA, then Tukey's post test (Figure 4.13B). TSST-1 causes the activation of at least two anti-apoptotic pathways, described in Chapter 3: the PI3K/Akt cascade in VB2+ TSST-1-reactive T cells, and loss of pro-caspase 3 in a non-V/32 restricted manner possibly by a degradative process. If G31R represents a loss of function mutation for anti-apoptotic activity, co-treatment with TSST-1 should restore anti-apoptotic signaling and protect against G31 R-induced death. However, contrary to expectations, co-administration of TSST-1 failed to rescue G31R-treated human PBMC from apoptosis (Figure 4.14A). Instead, at the highest dose of TSST-1 tested (1000 nM) that did not induce apoptosis alone (Chapter 3.2.1), the combination of this dose of TSST-1 actually potentiated G31R-induced apoptosis, shifting the G31R dose response curve to the left (Figure 4.14A). A comparison of the EC50 for each G31R dose response curve according to the dose of co-administered TSST-1 demonstrated a significant reduction in the EC50 of G31R in the presence of 1000 nM TSST-1 (Figure 4.14B). The intermediate doses of TSST-1, 10 nM and 100 nM, also demonstrated a downward trend in terms of decreasing the EC50 of G31R and thus potentiating the effect of G31 R-induced apoptosis, but this effect was not statistically significant (Figure 4.14B). 105 A. 100 Cfl a o a o a < a> c a o i_ o Q. B. RPMI TSST-1 10nM TSST-1 100nM TSST-1 1000nM 0 1 2 5 10 20 50 100 200 500 1000 G31R Concentration (nM) 10 100 1000 TSST-1 Dose(nM) Figure 4.14: A. Dose-dependent percentage apoptosis in human PBMC after treatment for 3 h with a combination of G31R and different concentrations of TSST-1 or RPMI medium control and analyzed by Annexin V-FITC staining and flow cytometry. B. A significant reduction in the E C 5 0 of G3 IR was demonstrated in the presence of 1000 nM TSST-1 (*p<0.01, 1-way ANOVA with Tukey post-test), but not at other concentrations of TSST-1. 106 4.3. Discussion and conclusion The main finding from the characterization of apoptosis is that the G31R mutant of TSST-1 induces rapid and potent apoptosis in human PBMC within hours after stimulation. This was not limited to the superantigen-reactive T cell subset (V/32+), but rather affected multiple cell subpopulations. Some of the hallmark apoptotic changes associated with G31R stimulation included cell morphological changes, chromatin condensation, DNA fragmentation, early phosphatidylserine exposure, and eventual plasma membrane permeabilization. TSST-1 and RPMI medium control, in all cases, exhibited a lack of apoptotic changes by all these methods. One notable irregularity in the data was that PI staining for DNA content after 3 days of toxin stimulation yielded apparently lower apoptosis than either 7-AAD or Annexin V-FITC staining after only 3 hours of toxin stimulation. This may reflect either insensitivity of PI staining or the high cell loss after 3 days of G31R treatment due to cell disintegration in late stages of apoptosis combined with the flow cytometric gating parameters that would interpret apoptotic bodies as debris. Cell permeability to 7-AAD also underestimated apoptosis, compared with Annexin V-FITC; dual staining indicated that phosphatidylserine exposure occurred well before membrane permeability to 7-AAD (Figure 4.3A & 4.4C). That is, only the populations in region III of Figure 4.3A would have been identified as apoptotic by 7-AAD, while both regions II and III were identified as such by Annexin V-FITC staining. While TUNEL staining is frequently used as an indicator of apoptosis, caution should be applied in its use, since many necrotic systems also show increased TUNEL staining indicative of DNA fragmentation (Grasl-Kraupp et al., 1995; Yasuda et al., 1995). However, by all the other measures, including morphological changes visualized microscopically, cell death in this system was predominantly apoptotic, not necrotic. Taken together, these data strongly imply that the form of cell death induced, while very rapid, conforms to the conventional model of apoptosis. 107 From these characterization studies, the dose of 100 nM was chosen for subsequent experiments to determine the molecular mechanisms for G31R induced apoptosis. This was the concentration at which G31R lost proliferative activity (Figure 4.1 A), while not saturating the apoptotic response according to the earliest apoptosis assays used, PI nuclear staining and 7-AAD permeability (Figures 4.IB, 4.7B & 4.8B). However, phosphatidylserine exposure was ultimately selected as the apoptosis detection method of choice for subsequent studies because it was the most sensitive method for detecting early apoptotic changes within the first several hours after stimulation (Figure 4.4C). Human PBMC were originally chosen for this model system due to several considerations. First, this in vitro model is well accepted and studied in application to superantigenic effects. Human PBMC contain a mixture of mononuclear cell types including T cells and monocytes that act as APCs; thus the two key cellular requirements for superantigenic responses are available in the approximate physiologic ratios present in human blood. Furthermore, much of the proliferation and cytokine production data obtained in our lab that served as the impetus for studying apoptosis was also performed in the human PBMC model. Availability of leukopheresis packs through the local hospital's Cell Separator Unit was another contributor to this decision. Finally, as far as relevance to human disease models, human PBMC serve to more realistically emulate pathogenesis in humans than murine models or transformed cell lines. However, extension of the apoptosis measurements to both T leukemia cell lines and murine splenocytes have shown that G31R can induce apoptosis in all these diverse cell types, to varying extents. This would suggest that the mechanism for apoptosis is likely ubiquitous and conserved even across different animal species. Since washing off the toxin could not prevent apoptosis, G31R would seem to produce an immediate death signal in its target cells. This suggests that this mechanism, once triggered, causes a cell to become irreversibly committed to apoptosis. Furthermore, the inability of co-108 administered TSST-1 to counteract G31R's apoptotic activity suggests that although intermediate doses of TSST-1 (<1 fiM) may exert anti-apoptotic effects, the G31R apoptotic activity behaves in a dominant fashion and overrides any residual anti-apoptotic effects still present in the mutant toxin. Putting these findings together, several hypotheses concerning the mechanism may be put forth (Figure 4.15). These may be divided into the classical superantigenic V/32-specific mechanisms (Figure 4.15, left) and non-V/32-specific mechanisms (Figure 4.15, right). As described in Chapter 1.3, V/32-specific mechanisms would include ACAD and AICD (Figure 4.15A). These would depend upon the presence of VB2+ T cells and APCs that can present the superantigen. However, given that apoptosis is seen not only in populations containing VB2+ T cells, but also in cells completely free of VB2+ cells such as the V/38+ Jurkat cell line, the mechanism for G31R- and/or high-dose TSST-1-induced apoptosis is not dependent upon superantigenic mechanisms acting through V/32+ T cells. Instead, G31R may be triggering apoptosis by signaling a ubiquitous death receptor besides the TCR, thereby inducing cell death. As far as candidates for death receptors are concerned, the classical death receptor Fas is most likely to be involved, considering its role in the downregulation not only of T cell responses (Krueger et al., 2003), but also in monocyte apoptosis (Perlman et al., 2001). G31R could either bind directly to receptors such as Fas, TNFRI, and/or other members of the TNFR superfamily (Figure 4.15B), or act indirectly by promoting Fas ligand expression or susceptibility to Fas-mediated apoptosis (Figure 4.15C). The latter mechanism might behave analogously to the TSST-1 duality in regard to its ability to induce either B cell activation or apoptosis in a dose-dependent fashion based on the production of IFN7 (Chapter 1.2.4; Figure 4.15D). Here, a broader effect might be achieved by production of death-inducing ligands other than IFN7 (Figure 4.15C). Either way, the extrinsic pathway for apoptosis would be primarily involved in 109 classical death receptor ligation. Determining whether the extrinsic pathway is involved is the primary objective of the next chapter, Chapter 5. A second possibility is that G31R acts non-specifically either on cell plasma membranes or intracellular targets (Figure 4.15E). Neither is considered likely, but the latter must be considered in light of the ability of TSST-1 to transcytose across epithelial and endothelial cell barriers by receptor-based endocytosis, thus killing these cells (Hamad et al., 1997; Lee et al., 1991). Such mechanisms would likely involve the intrinsic apoptotic pathway, and will be considered in Chapter 6. All these contrast with the known pro-survival effects of TSST-1, at equivalent concentrations, on V/3-specific T cells, monocytes and B cells (Figure 4.15F, G & H, respectively; see Chapter 3.2.4 & 1.2.4). 110 Figure 4.15: Possible mechanisms for G3IR and/or TSST-1 Induced Apoptosis. VP2-specific mechanisms for apoptosis include ACAD and AICD (A). However, G3 IR also causes non-VP-specific apoptosis, suggesting other mechanisms such as direct death receptor binding (B), induction of death ligands (C) or death-inducing cytokines (D), or non-specific cytotoxicity (E). This contrasts with the pro-survival activities of TSST-1 on Vp2' T cells (F), monocytes (G) and B cells (H). I l l Chapter 5 The Extrinsic Apoptotic Pathway Is Not A Predominant Route for G31 R-Mediated Apoptosis "It is a good morning exercise for a research scientist to discard a pet hypothesis every day before breakfast. It keeps him young. " -Konrad Lorenz 5.1. Introduction The glycine 31 to arginine site-directed mutant of TSST-1 (G31R) has been shown to induce early and potent apoptosis in a range of different cell types, independently of its superantigenic effects. Given the susceptibility of different haematopoietic cell lineages to G31R-mediated apoptosis, the objective of the studies in this Chapter was to determine whether a ubiquitous, classical death receptor such as the Fas receptor in the extrinsic apoptotic pathway is involved in transmitting the death signal. 112 Figure 5.1: Involvement of the Extrinsic Pathway. The approach to determining whether the extrinsic pathway is involved in G31R-mediated apoptosis was: A. checking whether soluble factor was involved; B. neutralizing FasL to see if Fas-mediated apoptosis was involved; C. inhibiting caspase 8 activation, which is essential for transmitting apoptotic signalling for the extrinsic pathway; and D. measuring caspase 3 processing and activation. 113 5.2. Results 5.2.1. G31R-induced apoptosis is not mediated by soluble factors The first specific aim was to test whether G31 R-induced apoptosis was dependent on the induction of soluble factors such as TNFa or sFasL that function as death ligands (Figure 5.1 A). Human PBMC were treated with 100 nM G31R for 3 h and pelleted by centrifugation to yield a supernatant that could be examined for the presence of any soluble apoptogenic factors. Unbound, residual G31R in the supernatant was neutralized with rabbit polyclonal anti-TSST-1 or normal rabbit serum (1:10 dilution; Chapter 4.2.4). This supernatant was then incubated with fresh human PBMC at 37°C for 3 h. Unless otherwise specified, all cells were assessed for apoptosis using Annexin V-FITC staining and flow cytometry, as described in Chapter 2.6.3. If the mechanism of apoptosis induced by G31R is mediated by soluble factors, then the fresh human PBMC would be expected to undergo apoptosis. If, however, the mechanism requires direct killing of target cells by G31R, then no induction of apoptosis should be seen in the fresh human PBMC treated with neutralized supernatant. Similarly, if G31R is recruiting a population of cytotoxic cells to kill target cells, again apoptosis would not be induced in these fresh PBMC treated with neutralized supernatant. Both the response of the original cells and of fresh cells treated with non-neutralized supernatant (i.e., addition of normal rabbit serum rather than the polyclonal anti-TSST-1) was evaluated as controls. The results were expressed as an index of apoptosis, obtained by dividing the percentage of apoptosis in samples treated with toxin by the percentage apoptosis in RPMI medium control. While human PBMC treated with 100 nM G31R underwent apoptosis, supernatants from these cells that had been neutralized with rabbit polyclonal anti-TSST-1 to remove any residual G31R were unable to induce apoptosis in fresh human PBMC (Figure 5.2A). TSST-1 at the same dose had little effect on apoptosis compared with RPMI medium control, whether on the 114 original cells or with supernatant transferred to the fresh cells (Figures 5.2A and 5.2B). Sham neutralization itself, using normal rabbit serum, did not alter apoptosis with any treatment condition including G31R, indicating that insufficient residual G31R was present in these supernatants to induce apoptosis on fresh cells (Figure 5.2B). 115 A. 4 3.5 X 3 o -o _c 2.5 m 2 o Q. O 1.5 a < 1 0.5 0 PBMC treated with toxin • Fresh PBMC treated with supernatant plus anti-TSST-1 rabbit serum TSST-1 (100nM) G31R(100nM) B. 3.5 3 2.5 2 x Q • D o Q- 1.5 o < 1 0.5 0 n=3 1 PBMC treated with toxin • Fresh PBMC treated with supernatant plus normal rabbit serum TSST-1 (100nM) G31R(100nM) Figure 5.2: A. Apoptosis index in human PBMC assessed by Annexin V-FITC staining and flow cytometry after treatment for 3 h with G3 IR or TSST-1 (left bar of each pair), or with supernatant neutralized with polyclonal anti-TSST-1 rabbit serum (right bar of each pair). The apoptosis index was obtained by dividing the percentage apoptosis in samples treated with toxin or supernatant by the percentage apoptosis in RPMI medium control. B. Apoptosis index in human PBMC after treatment with G3 IR or TSST-1 (left bar of each pair) or the supernatant pre-treated with normal rabbit serum (right bar of each pair). 116 5.2.2. G31 R-induced apoptosis does not depend upon Fas ligand The second specific aim was to determine whether Fas ligand (FasL) was involved in mediating G31R-induced apoptosis (Figure 5.IB). Human PBMC were pre-treated with a monoclonal anti-FasL neutralizing antibody (NOK-1, 10 /xg/ml) to neutralize any soluble or membrane-bound FasL, as described in Chapter 2.7.1 (Kayagaki et al., 1995). Pre-incubation with an IgGi isotype control was used as a control. 100 nM G31R or TSST-1 were then administered for 3 hours before staining for apoptosis. The results were expressed as index of apoptosis as above. To ensure that the NOK-1 antibody was active at neutralizing Fas-mediated apoptosis, it was applied to a positive control for Fas-mediated apoptosis involving the use of vesicular-bound Fas ligand on Jurkat cells (Chapter 2.7.1). The anti-Fas ligand neutralizing antibody NOK-1 completely impaired Fas ligand-bearing vesicles from inducing apoptosis of Jurkat cells at concentrations as low as 100 ng/ml (Figure 5.3A). A concentration two orders of magnitude higher (10 itg/ml) was chosen to assure neutralization of Fas ligand in the human PBMC system. However, pre-incubation with NOK-1 did not prevent G31R-induced apoptosis in human PBMC (Figure 5.3B). The IgGi isotype control antibody similarly had no effect on apoptosis (Figure 5.3C). 117 0 i 1 1 , , 1 0 ng/ml 1 ng/ml 10 ng/ml 100 ng/ml 1 ug/ml 10 ug/ml TSST-1 G31R(100nM) TSST-1 G31R(100nM) (100nM) (100nM) + lgG1 + lgG1 Control Control Figure 5.3: A. Percentage apoptosis assessed by Annexin V-FITC staining and flow cytometry in Jurkat cells pre-incubated with varying doses of anti-Fas ligand neutralizing antibody (NOK-1) before treatment with vesicular Fas ligand (FasL) or RPMI medium for 3 h. B. Apoptosis index in human PBMC assessed by Annexin V-FITC staining and flow cytometry after treatment for 3 h with G3 IR or TSST-1, with or without pre-incubation with NOK-1. Results were expressed as a ratio of percentage apoptosis in treatment to percentage in RPMI medium control. C. Apoptosis index in human PBMC after treatment for 3 h with G3 IR or TSST-1, with or without pre-incubation with an IgG! isotype control antibody. 1 1 8 5.2.3. Caspase 8 inhibition fails to prevent G31 R-induced apoptosis The next step in determining the possible involvement of the extrinsic pathway focused on the intracellular first-messengers of the death signal, the initiator caspase 8 (Figure 5.1C). To determine whether caspase 8 was involved in transmitting the death signal for G31R-mediated apoptosis, the relatively caspase 8-specific inhibitor Z-IETD-FMK was used to selectively bind to and antagonize its activity (Chapter 2.7.2). Human PBMC were pre-incubated for 15 minutes with either the caspase inhibitor at 100 [iM or DMSO, then treated with 100 nM G31R or TSST-1. Any confounding effects of the DMSO solvent used to dissolve the caspase inhibitor were controlled for by treating the non-inhibited samples with DMSO alone. The results were expressed as a percentage apoptosis, calculated by dividing the number of cells staining with Annexin V-FITC by the total number of cells. Fas receptor-mediated apoptosis in Jurkat cells was again used as a control to ensure that the caspase 8 inhibitor was active. Since Fas-mediated apoptosis is dependent upon caspase 8 activity for downstream activation of executioner caspases, prior caspase 8 inhibition should abrogate this form of apoptosis. Pre-incubation of Jurkat cells with the relatively caspase 8-specific inhibitor Z-IETD-FMK prevented vesicular Fas ligand-mediated apoptosis at a dose of 100 juM (Figure 5.4A). At the same concentration, the caspase inhibitor did not prevent apoptosis in human PBMC in response to 100 nM G31R treatment (Figure 5.4B). Again, no effect on apoptosis was observed with 100 nM TSST-1 treatment, compared with RPMI medium control (Figure 5.4B). 119 60 50 in 3 40 a. o Q. < o 30 Ui & e o o m a. 20 10 n=3 -+-1:500 Fas Ligand -o-RPMI OuM 10 nM 100 nM 1 uM 10 uM 100 uM Caspase 8 Inhibitor (Z-IETD-FMK) y DMSO carrier control (left) • Caspase 8 Inhibitor (right) RPMI G31R TSST-1 Figure 5.4: A. Percentage apoptosis by Annexin V-FITC staining and flow cytometry in Jurkat cells pre-incubated with varying doses of the caspase 8 inhibitor Z-IETD-FMK for 15 min before administration of vesicular Fas ligand (1:500) to induce apoptosis. B. Percentage apoptosis in human PBMC assessed by Annexin V-FITC staining and flow cytometry, pre-incubated either with the DMSO carrier control (left in each pair) or the caspase 8 inhibitor Z-IETD-FMK (100 uM; right in each pair) before treatment with 100 nM toxin for 3 h. 120 5.2.4. G31R causes caspase 3 activation but caspase 3 inhibition only minimally reduces G31R-mediated apoptosis Finally, to determine whether downstream caspase activation was itself required for G31R-mediated apoptosis, activation of the executioner caspase 3 was studied (Figure 5.ID). To measure the loss of pro-caspase 3 that would indicate caspase processing, human PBMC treated with 100 nM toxin for 3 hours were lysed for immunoblot analysis as described in Chapter 2.9.1. These results were analyzed by densitometry to compare the average cell content of pro-caspase 3. To ascertain the cellular content of activated caspase 3 directly, the toxin-treated cells were stained intracellularly using an active caspase-3 specific antibody and analyzed by flow cytometry (Chapter 2.6.5). These results were expressed as a percentage of cells containing active caspase 3. Lastly, the general caspase inhibitor Z-VAD-FMK (100 /xM) was used to determine whether complete caspase inhibition could prevent apoptosis, as detected by phosphatidylserine exposure. To this end, human PBMC were pre-incubated with either Z-VAD-FMK or DMSO for 15 minutes, as for the caspase 8 inhibitor, before addition of 100 nM G31R or TSST-1 for 3 hours. These cells were assessed either for active caspase 3 content, as above, or for percentage apoptosis by Annexin V-FITC staining. Immunoblot analysis for pro-caspase 3 content in the human PBMC lysates after G31R treatment demonstrated a significant decrease in pro-caspase 3 content, suggesting that pro-caspase 3 was being proteolytically processed to its active form (Figure 5.5A). This correlated with the increased content of active caspase 3 in G31R-treated cells (Figure 5.5B). However, despite this finding, neutralization of caspase 3 activity with the general caspase inhibitor Z-VAD-FMK at 100 uM only minimally reduced (<10%) G31 R-mediated apoptosis, even though this reduction was significant compared to RPMI control (p<0.05; paired t-test) (Figure 5.6). Because phosphatidylserine exposure was used in all cases as the marker for apoptosis, it would 121 be more accurate to describe phosphatidylserine exposure as being apparently independent of caspase activation in this system for the induction of apoptosis. As described in Chapter 3.2.5, 100 nM TSST-1 induced a near complete loss of pro-caspase 3 (Figure 5.5A), with no increase in active caspase 3 content (Figure 5.5B). 122 A. 32kD pro-caspase-3 R P M I G 3 1 R T S S T - 1 B. R P M I G 3 1 R T S S T - 1 Figure 5.5: A. Immunoblot for pro-caspase 3 content in human PBMC after treatment for 3 h with 100 nM G3 IR, 100 nM TSST-1 or RPMI medium control. Plot shows pro-caspase 3 content (average integrated density). B. Activated caspase 3 in human PBMC assessed by intracellular staining and flow cytometry (percentage positive) after treatment for 3 h with 100 nM G3 IR, 100 nM TSST-1 or RPMI medium control. 123 y DMSO carrier control (left) • General caspase inhibitor (right) RPMI G31R TSST-1 n=3 B DMSO carrier control (left) • General caspase inhibitor (right) RPMI G31R TSST-1 Figure 5.6: A. Activated caspase 3 in human PBMC assessed by intracellular staining and flow cytometry (percentage positive) after pre-incubation with the DMSO carrier control (left in each pair) or the general caspase inhibitor Z-VAD-FMK (100 uM) before treatment for 3 h with 100 nM G31R, 100 nM TSST-1, or RPMI medium control. B. Percentage apoptosis assessed by Annexin V-FITC staining and flow cytometry in human PBMC pre-incubated either with the DMSO carrier control (left in each pair) or the general caspase inhibitor Z-VAD-FMK (100 uM; right in each pair) before treatment for 3 h with 100 nM G3 IR, 100 nM TSST 1 or RPMI medium control. 124 5.3. Discussion and conclusions The involvement of the extrinsic pathway in G31 R-mediated apoptosis has been evaluated at both the extracellular (Figure 5.1 A & B) and intracellular (Figure 5.1C & D) levels. The lack of involvement of a secondary soluble factor in mediating apoptosis indicated that either G31R directly induced the apoptotic signal, or that G31R caused cells to acquire cytotoxic activity, thus killing neighbouring cells. In the latter case, this could still occur through the extrinsic pathway by the surface expression of Fas ligand on cytotoxic cells ligating to Fas receptor. However, because neutralization of Fas ligand could not prevent G31R-mediated apoptosis, it seems unlikely that this mechanism is involved. This does not, however, rule out G31R upregulating the surface expression of other death ligands such as the TRAIL on cytotoxic cells, or G31R directly binding to a death receptor, e.g. even to Fas receptor. However, because these classical death receptors all depend upon caspase 8 activity to initiate death signaling, the fact that caspase 8 inhibition was unable to reduce G31 R-mediated apoptosis suggested that the apoptotic mechanism is likely independent of any classical death receptor. This is further supported by the observation that the general caspase inhibitor Z-VAD-FMK, which potently inhibits most death receptor-mediated apoptosis, was only minimally effective in reducing G31R-mediated death. In other systems, the absence of caspase activity in the face of an apoptotic stimulus may still result in non-apoptotic forms of cell death that retain many of the features of apoptosis (Bidere et al., 2001; Deas et al., 1998; Ferraro-Peyret et al., 2002; Jaattela et al., 2003; Pettersen et al., 2001). However, while peptide inhibitors of caspases are widely used to demonstrate caspase dependency, criticisms have recently been levelled at this approach for being potentially misleading due to their propensity for incomplete caspase inhibition and lack of specificity, in the case of the sequence-specific inhibitors (Nicholson, 1999). Thus, the presence of apoptotic changes such as phosphatidylserine exposure even in the face of the inhibitor Z-VAD-FMK may merely reflect 125 activation of a small but sufficient fraction of caspases, below the detection sensitivity of the assays used (Figure 5.6A). Despite this possibility, the collective evidence suggests that G31 R-induced apoptosis is not primarily dependent upon the classical extrinsic pathway, and that the intrinsic pathway could be the major mechanism responsible. G31R might act on non-classical death receptors linked directly to the intrinsic pathway, such as CD47 and CD99 (Pettersen et al., 1999; Pettersen et al., 2001). It might also function as a cytotoxin that causes membrane or intracellular damage, although this would be considered unlikely, given that plasma membrane permeabilization is a late event (Chapter and that enzymatic activity that might contribute to cytotoxicity has not been demonstrated with TSST-1. Whatever the case, the hallmark event in the intrinsic pathway is mitochondrial outer membrane permeabilization (MOMP), that could occur through activation of BH3-only proteins or through elevation of intracellular calcium. These elements of the intrinsic pathway are considered in greater detail in the following Chapter. 126 Chapter 6 Intrinsic Pathway is the Predominant Pathway for G31R-Mediated Apoptosis "I have not failed. I've just found 10,000 ways that won't work. " -Thomas Edison 6.1. In t roduct ion The glycine 31 to arginine site-directed mutant of TSST-1 has been demonstrated to induce rapid apoptosis in a variety of cell types, including human PBMC, MOLT-16 and Jurkat E6-1 cell lines, as well as murine splenocytes (Chapter 4). The extrinsic pathway for apoptosis, classically involved in death receptor-mediated apoptosis such as Fas ligation, was not primarily responsible for G31R-mediated cell death (Chapter 5). Having excluded secondary soluble factors and the extrinsic pathway from involvement in the mechanism for G31R-induced apoptosis, a few possibilities remain. First, G31R could signal through a non-classical death receptor such as CD47, which links to the intrinsic pathway through recruitment and activation of a BH3-only pro-apoptotic protein (Figure 6.1-1). Second, G31R might function as a direct cytotoxin, for example by direct disruption of organelles such as the mitochondria and endoplasmic reticulum, or by promoting the generation of reactive oxygen species (Figure 6.1-II). However, there is no evidence for any enzymatic activity associated with TSST-1, rendering this hypothesis unfavourable. Lastly, G31R might promote direct membrane damage, leading to calcium dysregulation as the initiating event for apoptosis (Figure 6.1-III). Because the latter two possibilities are considered less likely, the specific aim of this chapter was to determine whether specific activation of the intrinsic pathway plays a role in G31R-mediated apoptosis. The involvement of the intrinsic pathway was assessed at five levels: 1) with the candidate death receptor CD47 (Figure 6.1 A); 2) upstream at the level of Bax translocation to the mitochondria (Figure 6. IB); 3) centrally in the intrinsic pathway in terms of mitochondrial outer 127 membrane permeabilization (MOMP) (Figure 6.1C); 4) with cytochrome c and AIF translocation out of the permeabilized mitochondria (Figure 6. ID & E); and 5) downstream with caspase 9 activation (Figure 6.1F). 128 Intrinsic Pathway G31R may bind to a yet unidentified non-classical death receptor like CD47 Membrane damage could lead to calcium1*" dysregulation C a 2 + G31R might function as a cytotoxin, promoting reactive oxygen species formation and mitochdonrial damage L Elevated calcium can cause caspase 12 and calpain activation Calpain cleaves Bax, .^-""promoting its aggregation M i t o c h o n d r i o n Caspase12 can cleave and activate caspase 9 directly Escaped cytochrome c binds to APAF-1 and promotes apoptosome assembly Bax/Bak oligomeric pores BH3-only protein frees Bax/Bak from Bcl-2/BcI-x, Bcl-xL ) Bax/Bak Freed Bax/Bak can translocate to the mitochondria Bax/Bak oligomerization forms pores Figure 6.1: Involvement of the Intrinsic Pathway. The approach to determining whether the intrinsic pathway is involved in G31 R-mediated apoptosis was to check for involvement of: A. a non-classical death receptor like CD47; B. Bax activation and translocation to the mitochondria; C. mitochondrial outer membrane permeabilization; D. & E. extravasation of cytochrome c and AIF from the mitochondria; F. caspase 9 activation. 129 6.2. Results 6.2.1. G31R does not primarily signal apoptosis through CD47 The CD47 receptor, also known as integrin-associated protein (IAP), is known to transmit a potent death signal to T lymphocytes when ligated at a specific epitope (Figure 6.1 A) (Pettersen et al., 1999; Lamy et al., 2003). The monoclonal anti-CD47 antibody B6H12 had previously been shown to inhibit interaction with the death-inducing epitope of the CD47 receptor. Therefore, to determine whether CD47 was involved in G31R-induced apoptosis, human PBMC were pre-incubated with B6H12 or IgGi isotype control between 0.1 to 10 jLig/ml for 15 minutes, as described in Chapter 2.8.1. G31R or TSST-1 was then administered at 100 nM, and apoptosis was assessed by Annexin V-FITC staining and flow cytometry. If G31R signals death through the CD47 receptor, then B6H12 pre-treatment should abrogate G31R-induced apoptosis. The results were expressed as percentage apoptosis, calculated by dividing the number of cells with increased surface phosphatidylserine exposure by the total number of cells. Pre-incubation of human PBMC with the highest dose of anti-CD47 neutralizing antibody, 10 jitg/ml, resulted in only a very modest but significant decrease in G31 R-induced (100 nM) apoptosis compared to the IgGi control (Figure 6.2A & B). However, the majority of G31R-induced apoptosis did not seem to depend on CD47 signaling. TSST-1 treatment, as usual, produced no significant change in apoptosis compared to RPMI medium control, with either antibody. 130 A. 0.1 1 10 A n t i - C D 4 7 Concentration (ug/ml) B. 0.1 1 1 0 IgGi Control Concentration (ug/ml) Figure 6.2: Percentage apoptosis in human PBMC assessed by Annexin V-FITC staining and flow cytometry, pre-incubated with either the B6H12 anti-CD47 antibody (A) or an IgGj isotype control (B) for 15 min before treatment for 3 h with 100 nM G3 IR, 100 nM TSST-1 or RPMI medium control. 131 6.2.2. G31R mediates Bax activation and translocation to the mitochondria Conformational changes in Bax following interaction with upstream BH3-only pro-apoptotic proteins cause Bax dissociation from inactive heterodimers with Bcl-2 and B C I - X L , and its subsequent translocation to the mitochondria. Mitochondrial membrane insertion of Bax results in the formation of oligomers which confer increased mitochondrial membrane permeability (Figure 6.IB). To measure these changes, the epitope-specific monoclonal antibody (6A7 clone) was used with intracellular staining to recognize only Bax protein that was present in its monomeric state, dissociated from Bcl-2 and/or Bcl-xL. When human PBMC treated with G31R or TSST-1 were stained with this antibody as described in Chapter 2.8.5, in conjunction with a polyclonal anti-Bax antibody to control for total cellular content of Bax, the cells could be analyzed by flow cytometry for Bax activation. If Bax activation is involved in G31R-induced apoptosis, individual cells undergoing apoptosis would be expected to initially show an increase in staining for the 6A7 epitope due to dissociation of Bax from Bcl-2 and/or Bcl-xL. As Bax accumulates on the mitochondrial membrane and oligomerizes, however, the staining would be expected to decrease again. Over a population of cells, with some at earlier stages and others at later stages of apoptosis, the net effect would be a broadening of the initial distribution of Bax staining. The results were therefore expressed as the coefficient of variation within each sample, which the analysis software, DakoCytomation Summit, defines as the standard deviation as a percentage of the population mean: The number of cells with changes in their content of the conformationally active Bax monomer was significantly increased in G31R-treated human PBMC within 1 hour after Coefficient of Variation (CV) = lOOx Standard deviation = 100x Population mean x 132 stimulation, plateauing by 2 hours (Figure 6.3A). In contrast, TSST-1 produced little change in the coefficient of variation compared to RPMI medium control (Figure 6.3A). Flow cytometry for Bax activation had the disadvantage of not being able to differentiate between Bax localized in the cytosol vs. the mitochondria. Therefore, cytosolic and mitochondrial fractions of cell lysates were isolated as described in Chapter 2.8.3, then immunoblotted with a rabbit polyclonal anti-Bax antibody. If Bax activation and translocation to the mitochondria occurs during G31R stimulation, then an increase in the mitochondrialxytosolic ratio of Bax protein should be seen in the cell lysates after immunoblotting. In contrast, since the cytochrome oxidase 4 (COX4) does not leave the mitochondria even during MOMP, the mitochondrialrcytosolic ratio of COX4 in human PBMC treated with G31R should remain unchanged and similar to cells treated with RPMI medium control. Thus, COX4 staining was used as a control for cross-contamination between mitochondrial and cytosolic fractions. As seen in the immunoblots in Figure 6.3B, a marked increase in Bax protein was found in the mitochondrial fraction of human PBMC after treatment for 30 minutes with 100 nM G31R, as compared to cells after treatment with 100 nM TSST-1 or RPMI medium control. Densitometric analysis revealed that the mitochondriahcytosolic ratio of Bax was considerably higher after treatment with G31R compared to TSST-1 or RPMI medium control, suggesting that G31R treatment resulted in Bax translocation into the cytosol. .. By contrast, staining for COX4 demonstrated consistently high mitonchondriahcytosolic ratio of COX4, indicating little cross-contamination of mitochondrial proteins into the cytosolic compartment (Figure 6.5B). Unfortunately, the patterns observed with immunoblot were not always consistent across experiments, likely due to technical difficulties in separating mitochondrial and cytosolic fractions prior to immunoblot analysis. However, results obtained by intracellular staining for 133 Bax and analyzed by flow cytometry were always consistent among all 3 donors studied (Figure 6.3A). Because neither flow cytometry nor immunoblot analysis could distinguish in situ intracellular localization of Bax, similar staining of toxin-treated human PBMC was performed as for flow cytometry above, but with final visualization by fluorescence microscopy (Chapter 2.8.4). Simultaneous staining with an antibody to COX4 permitted identification of mitochondria within cells. If Bax translocation to the mitochondria occurs following G31R stimulation, then the areas staining for Bax would be expected to co-localize to the areas staining for COX4. The Bax stained cells (red) showed a definite increase in punctate staining following G31R treatment for 30 minutes (Figure 6.4A). In contrast, the COX4 stain for mitochondria (green) indicated little change between RPMI, G31R and TSST-1 treatments (Figure 6.4B). An overlay between these two demonstrates close co-localization of Bax to the mitochondria following G31R treatment (Figure 6.4C). The initial stages of chromatin condensation could be seen using Hoechst 33258 (blue) to stain for nuclei (Figure 6.4D). An overlay of all three channels delineates the Bax and COX4 stain around the periphery from the centrally located nucleus (Figure 6.4E). 134 A. 80 70 C 60 o • — . re 50 > — o 40 • 4 - 1 C CD 30 o it 03 20 O o 10 0 B. Actin (42kD) Bax (20kD^ 1 ^ o 3 O 2 0.5 RPMI 1.5 Time (h) 2.5 G31R TSST-1 Mito Cyto Mito Cyto Mito Cyto oV Figure 6.3: A. Time-dependent Bax activation in human PBMC treated with 100 nM G3 IR as determined by intracellular staining with a conformation-specific anti-Bax antibody. Results were expressed as coefficient of variation (percentage) after flow cytometric analysis. Treatment with 100 nM TSST-1 and RPMI medium control had no effect. B. The immunoblot for Bax and actin in mitochondrial (Mito) and cytosolic (Cyto) fractions from human PBMC after treatment for 30 min with 100 nM G31R, 100 nM TSST-1 or RPMI medium control. Results were expressed as the ratio of mitochondrial to cytosolic protein content determined by densitometry from the immunoblot. 135 RPMI G31R TSST-1 Figure 6.4: Micrographs of human PBMC after treatment for 30 min with 100 nM G31R, 100 nM TSST-1 or RPMI medium control and examined by fluorescence microscopy. Results are representative from 3 separate donors. A. Red fluorescence from anti-Bax stain. Percentages in parentheses represent proportion of cells on slides exhibiting predominant Bax aggregation to mitochondria. B. Green fluorescence from anti-COX4 stain showing the localization of COX4 in the mitochondria. C. Overlay of red and green fluorescence showing that Bax is primarily colocalized (yellow) with COX4 in the mitochondria in G31R-treated cells. D. Blue fluorescence from Hoescht 33258 nuclear stain showing chromatin condensation with G3 IR. E. Overlay of red, green and blue fluorescence. 136 6.2.3. G31R causes loss of mitochondrial transmembrane potential The third stage in characterization of G31R's apoptotic activity involved direct determination of mitochondrial permeabilization (Figure 6.1C). To this end, human PBMC treated for 3 h with 100 nM G31R or TSST-1 were stained with the cationic-specific aggregating dye Mitosensor™, and the trans-mitochondrial membrane potential was examined by flow cytometry, as described in Chapter 2.8.2. If G31R induced apoptosis through the intrinsic pathway, an increase in green fluorescence associated with mitochondrial membrane disruption should have been seen. The results were expressed as the percentage of total cells exhibiting green fluorescence after toxin treatment. The mitochondrial transmembrane potential in many cells (as measured by the Mitosensor™ dye) was significantly reduced (as indicated by increased green fluorescence) following G31R treatment, implicating mitochondrial disruption in G31 R-mediated apoptosis of human PBMC (Figure 6.5A). In contrast, treatment with 100 nM TSST-1 produced no significant change in mitochondrial transmembrane potential compared with RPMI medium control. 137 a> 10 o Q- 0 B. Actin (42kD) COX4 (17kD) 3.5 l ] RPMI RPMI i G31R 100nM G31R TSST-1 100nM TSST-1 Mito Cyto Mito Cyto Mito Cyto mm * Figure 6.5: A. Human PBMC with mitochondrial transmembrane potential loss (percentage positive for green fluorescence) after treatment for 3 h with 100 nM G3 IR, 100 nM TSST-1 or RPMI medium control. B. Immunoblot for COX4 and actin in mitochondrial (Mito) and cytosolic (Cyto) fractions of human PBMC after treatment for 3 h with 100 nM G3 IR, 100 nM TSST-1 or RPMI medium control. Results are expressed as a ratio of mitochondrial to cytosolic protein content determined by densitometry from the immnoblot. 138 6.2.4. G31R causes cytochrome c and AIF translocation The next aspect of the intrinsic pathway examined logically follows MOMP: cytochrome c and AIF translocation (Figure 6. ID & E). Once MOMP is achieved, cytochrome c, AIF and a host of other mitochondrial constituents can leak out of the mitochondria into the cytosol. In order to determine whether leakage of mitochondrial constituents was responsible for activating further downstream events in the death program induced by G31R, toxin-treated human PBMC were lysed and separated into mitochondrial and cytosol fractions according to Chatper 2.8.3. Immunoblot analysis was performed on two key mitochondrial constituents, cytochrome c and AIF. The results were expressed as ratios of actin-normalized protein content between cytosolic and mitochondrial fractions. A moderate decrease in mitochondrialxytosolic cytochrome c ratio was observed in human PBMC after treatment with G31R as compared to the RPMI medium control, suggesting that cytochrome c had indeed translocated out of the mitochondria into the cytoplasm (Figure 6.6A). However, a decreased ratio was also observed with TSST-1, although this may be artifactual due to a rise in background to the right of the cytosolic cytochrome c blot (Figure 6.6A). Similarly, an apparent decrease in in mitochondriakcytosolic AIF ratio was observed in human PBMC after treatment with G31R as compared to cells treated with TSST-1 or the RPMI medium control, suggesting translocation of AIF rom the mitochondria into the cytoplasm after G31R treatment (Figure 6.6B). To confirm and extend these observations, immunocytochemical staining was performed as described in Chapter 2.8.4 to identify the intracellular localization of cytochrome c and AIF after toxin treatment. If G31R activates the intrinsic pathway, both cytochrome c and AIF would be expected to initially co-localize in a punctate fashion with the COX4 mitochondrial stain before subsequently becoming more diffuse. For AIF in particular, which translocates to the nucleus, some degree of nuclear co-localization would be expected. 139 Similar to the studies of Bax activation and translocation described earlier, far more specific and consistent information was obtained using immunocytochemical staining and fluorescence microscopy. Human PBMC treated for only 30 minutes with G31R had a notable loss of punctate staining with cytochrome c, compared with TSST-1 and RPMI medium control (Figure 6.7A). By contrast, the clustering of the COX4 stain was maintained, indicating that the mitochondria themselves were still present as intact structures (Figure 6.7B). An overlay of cytochrome c with COX4 staining areas shows that there was co-localization of these proteins in cells treated with TSST-1 or RPMI medium control, but not in cells treated with G31R, again confirming that cytochrome c has translocated following G31R treatment (Figure 6.7C). Early chromatin condensation can be seen with the Hoechst 33258 nuclei stain for G31R, while RPMI and TSST-1 treatments had normal euchromatin (Figure 6.7D). Finally, an overlay of all 3 stains confirms the central localization of the nuclei (Hoeschst stain), the peripheral distribution of mitochondria (Cox4 stain), and translocation of cytochrome c from the mitochondria after treatment with G31R, in contrast to TSST-1 or RPMI medium control (Figure 6.7E). Microscopic studies following immunocytochemical staining demonstrated little change in the AIF localization pattern after 30 minutes treatment with either G31R or TSST-1, compared with RPMI control (Figure 6.8). As expected, AIF co-localized with COX4 in the mitochondria (Figure 6.8A-C). However, after 3 hours, substantial diffusion of the AIF stain from the mitochondria into the cytosol was seen after G31R treatment (Figure 6.9A-C). Again, the COX4 stain remained punctate, defining the mitochondria on the periphery (Figure 6.9B). The overlay of AIF and COX4 showed a loss of AIF from the mitochondria and apparent translocation centrally in the cell (Figure 6.9C). The combination of the Hoechst 33258 nuclear stain (Figure 6.9D) with the AIF stain demonstrated co-localization of AIF to the nucleus, as expected (Figure 6.9E). 140 A. Actin (42kD) Cytochrome c (15kD) 8 -RPMI G31R TSST-1 o O % A -O b £ 2 H Cyto c i Actin Cyto c j. Ar.tin Actin B. RPMI G31R TSST-1 Mito Cyto Mito Cyto Mito Cyto AIF (57kD) Actin (42kD) mm 4 i i o (0 o o Figure 6.6: Immunoblots for cytochrome c, A IF and actin in mitochondrial (Mito) and cytosolic (Cyto) fractions of human P B M C after treatment for 3 h with 100 n M G3 IR, 100 n M TSST-1 or R P M I medium control. Results were expressed as a ratio of mitochondrial to cytosolic protein content determined by densitometry from the immunoblot. A . Cytochrome c vs. actin. B. A IF vs. actin 141 RPMI G31R TSST-1 Figure 6.7: Micrographs of human PBMC after treatment for 30 min with 100 nM G31R, 100 nM TSST-1 or RPMI medium control and examined by fluorescence microscopy. Results are representative from 3 separate donors. A. Red fluorescence from anti-cytochrome c stain. Percentages in parentheses represent proportion of cells on slides exhibiting loss of cytochrome c localization to the mitochondria. B. Green fluorescence from anti-COX4 stain showing the localization of COX4 in the mitochondria. C. Overlay of red and green fluorescence showing that relatively little cytochrome c is colocalized (yellow) with COX4 in the mitochondria after treatment with G3 IR, in contrast to cells treated with TSST-1 or RPMI. D. Blue fluorescence from Hoescht 33258 nuclear stain showing chromatin condensation with G31R. E. Overlay of red, green and blue fluorescence. 142 RPMI G31R TSST-1 Figure 6.8: Micrographs of human P B M C after treatment for 30 min with 100 n M G3 IR, 100 n M TSST-1 or RPMI medium control and examined by fluorescence microscopy. Results are representative from 3 separate donors. A. Red fluorescence from anti-ALF stain. B. Green fluorescence from anti-COX4 stain showing the localization of COX4 in the mitochondria. C. Overlay of red and green fluorescence showing that AIF is still colocalized (yellow) with COX4 in the mitochondria after treatment with G3 IR, similar to cells treated with TSST-1 or RPMI. D. Blue fluorescence from Hoescht 33258 nuclear stain. F.. Overlay of red and blue fluorescence. 143 RPMI (16%) G31R TSST-1 (24%) Figure 6.9: Micrographs of human PBMC after treatment for 3 h with 100 nM G3 IR, 100 nM TSST-1 or RPMI medium control. Results are representative from 3 separate donors. A. Red fluorescence from anti-AIF stain. Percentages in parentheses represent proportion of cells on slides exhibiting loss of AIF localization to the mitochondria. B. Green fluorescence from anti-COX4 stain showing the localization of COX4 in the mitochondria. C. Overlay of red and green fluorescence showing that relatively little AIF is colocalized (yellow) with COX4 in the mitochondria after treatment with G3 IR, in contrast to cells treated with TSST-1 or RPMI. D. Blue fluorescence from Hoescht 33258 nuclear. E. Overlay of red and blue fluorescence showing that AIF is colocalized (purple) with Hoescht 33258 stain in the nucleus after treatment with G3 IR, in contrast te cells treated with T.SST-3 cr RPMI. 144 6.2.5. G31 R-induced apoptosis is mostly independent of caspase 9 Finally, the caspase activation pathways, although only of modest importance based on studies with the general caspase inhibitor Z-VAD-FMK described in Chapter 5, were the last aspect of the intrinsic pathway considered (Figure 6. IF). Because cytochrome c, in conjunction with its adapter molecule APAF-1, activates the initiator caspase 9, the relatively caspase 9 specific inhibitor Z-LEHD-FMK was used to pre-treat human PBMC before exposure to toxin (Chapter 2.7.2). If G31R-induced apoptosis requires activation of the intrinsic pathway, the relatively caspase 9 specific inhibitor would be expected to prevent apoptois. The cells were analyzed by Annexin V-FITC staining and flow cytometry, and the results were expressed as percentage apoptosis. While a minimal but statistically significant reduction in G31 R-induced apoptosis was seen with the caspase 9 inhibitor (100 jitM), the majority of the cells were unaffected by caspase 9 inhibition (Figure 6.10). 145 Figure 6.10: Percentage apoptosis assessed by Annexin V-FITC staining and flow cytometry in human PBMC pre-incubated with either caspase 9 inhibitor (Z-LEHD-FMK, 100 uM) or DMSO carrier control for 15 min before exposure to 100 nM toxin for 3 h (*p<0.05, 2-tailed t-test compared to carrier control). 146 6.3. Discussion of data and conclusions Overall, the data obtained supports the notion that the intrinsic pathway is primarily involved as the major mechanism for G31R-mediated apoptosis. Data regarding the most apical part of this pathway, namely the possibility of CD47 as the death receptor involved, yielded inconclusive results. While a small but significant reduction in apoptosis was seen using a neutralizing antibody, the inability to demonstrate a dose-dependent effect with anti-CD47 and lack of a suitable positive control for CD47-mediated cell death meant that the effect of the neutralizing antibody itself could not be validated in this model system. Moreover, G31R might interact with a region of the CD47 receptor not sterically hindered by the B6H12 blocking antibody. This also cannot rule out other receptors being involved in G31R-mediated transmission of the death signal, nor the less likely direct cytotoxic and membrane disruptive hypotheses. The data from immunoblot analysis of mitochondrial compared with cytosolic fractions was inconclusive. This may be due to technical difficulties in obtaining pure mitochondrial and cytosolic fractions for immunoblot analysis. Thus, data obtained by immunocytochemical staining and fluorescence microscopy proved to be more reliable, not only in yielding more consistent results, but also in permitting detailed in situ localization of the proteins in the cells following toxin treatment. Concerning Bax, the combination of flow cytometry and fluorescence microscopy indicated that this protein dissociated from Bcl-2 and/or Bcl-xL following G31R stimulation, allowing it to translocate to the mitochondria to promote mitochondrial membrane disruption. It might seem surprising at first glance that little change in the average fluorescence per cell was seen with flow cytometry, while microscopy clearly demonstrates increased fluorescence at the mitochondria. However, this can be explained by noting that Bax translocation from the cytosol to the mitochondria results in the conversion from a diffuse stain to a punctate pattern; thus, the 147 averaged fluorescence over the entire cell may be relatively unchanged. Ultimately cellular localization would be best confirmed by using GFP-tagged Bax protein in conjunction with microscopy (Wolter et al., 1997). It is worthwhile to note that these visible changes with G31R were obtained after only 30 minutes exposure, the earliest time point assessed. While the mitochondrial membrane potential data shown was after 3 hours, preliminary assessment at earlier time points supports the very early induction of mitochondrial membrane destabilization by G31R. Similarly, cytochrome c translocation from the mitochondria was observed after only 30 minutes, both by immunoblot analysis and microscopy. In contrast, AIF translocation was delayed beyond 30 minutes, possibly due to its much greater size (57 kD vs. 15 kD for cytochrome c). Downstream activation of caspase 9 by cytochrome c likely provides the impetus for carrying out the remainder of the intrinsic death program. The incomplete inhibition of apoptosis by the moderately selective caspase 9 inhibitor Z-LEHD-FMK, combined with the very modest reduction in apoptosis obtained with the general caspase inhibitors Z-VAD-FMK (Chapter 5.2.4) and Boc-D-FMK (not shown), would suggest that G31 R-induced phosphatidylserine externalization and morphologic characteristics of cell death are mostly caspase-independent. While most authors have reported that apoptotic-specific changes such as the exposure of phosphatidylserine to be dependent on caspases, one group has demonstrated caspase-independent loss of phosphatidylserine exposure in human peripheral blood lymphocytes (PBLs) (Ferraro-Peyret et al., 2002). These investigators found that PBLs underwent loss of Ay m , phosphatidylserine exposure and DNA fragmentation in response to either CD95 ligation, IL-2 withdrawal, or exposure to the chemotherapeutic agents etoposide and staurosporine. While caspase inhibition prevented loss of A\|/m and phosphatidylserine externalization for CD95 ligation, it could not prevent loss of Av)/m and phosphatidylserine exposure after IL-2 withdrawal or treatment with the chemotherapeutic drugs. By contrast, DNA fragmentation was prevented 148 by caspase inhibition. Other mitochondrial constituents such as AIF released during MOMP have been hypothesized to be able to induce programmed cell death even in the absence of caspase activation, perhaps as backup mechanisms in vivo to guarantee the elimination of potentially dangerous cells such as activated T lymphocytes. This may well play a role with G31R-mediated apoptosis, where AIF translocation was demonstrated. Recent criticisms regarding the use of commonly employed caspase inhibitors have stemmed from observations that even small amounts of residual caspase activity may be sufficient to execute the apoptotic program (Nicholson, 1999), calling into question many studies which have drawn conclusions based on results using these peptide inhibitors of caspases. Several key questions remain unanswered regarding the hypothesis that G31 R-induced apoptosis is mediated by ligation of a death receptor leading to transmission of the death signal through the Bax-dependent intrinsic pathway. The first issue is whether G31R constitutes a unique mutation that confers an increased affinity for an unknown death receptor, or rather that it is merely one of a number of possible mutants of TSST-1 with such activity. The second major unknown is the identity of this putative death receptor. A large repertoire of surface antigens, usually associated with other functions than death signaling, have been shown to induce apoptosis or other forms of cell death, including: CD45 (Lesage et al., 1997), MHC class I (Pettersen et al., 1998), CD2 (Deas et al., 1998) and CD4 (Berndt et al, 1998). The next major gap in knowledge of this putative pathway is the link between the putative death receptor and Bax activation/oligomerization. Activation or redistribution of BH3-only proteins are normally responsible for Bax activation, so candidates would include Bid, Bad, Bim and the novel BNIP3, which couples CD47 to the intrinsic pathway (Lamy et al., 2003). 149 Chapter 7 G31 R-lnduced Apoptosis: A Working Model, Significance and Conclusions "When you know a thing, to hold that you know it; and when you do not know a thing, to allow that you do not know it - this is knowledge. "—Confucius 7.1. Introduction: the possibilities From the moment that the curious pro-apoptotic activity of G31R was confirmed by different methods including PI and 7-AAD staining, phosphatidylserine exposure by Annexin V, and DNA fragmentation by TdT incorporation (Chapter 4.2.1), clarification of the mechanism for apoptosis was of the highest priority both to understand what role this effect might have during TSS pathogenesis, and to explore possible applications of this form of apoptosis in immunological investigations and therapeutic strategies. The non-specificity of G31R-induced apoptosis involving different defined subpopulations within human PBMC argues against a mode of action that is confined to the known superantigenic activities of G31R involving only MHC class II-bearing APCs or V/32-specific T cells (Chapter 4.2.6). Therefore, our initial hypothesis that this form of apoptosis might be related to AICD cannot be supported. A systematic examination of the known apoptotic mechanisms has largely ruled out a significant contribution by the extrinsic apoptotic pathway (Chapter 5), favouring instead the intrinsic mitochondrial pathway (Chapter 6). However, the details linking G31R exposure to ultimate mitochondrial membrane de-stabilization and subsequent cell death remains unclear. A number of possibilities exist, and a working model is presented in Figure 7.1. Based on the rapid mode of action of G31R, requiring only momentary exposure to induce apoptosis (Chapter 4.2.5), the model of direct binding of G31R to a non-classical death receptor (i.e. not signaling through the extrinsic pathway) appears to be the most probable initiating event (Figure 7.1 A-1). The other possibilities include direct membrane disruption (Figure 7.1A-II), and cytotoxicity 150 through non-specific damage of intracellular targets following endocytosis of the toxin (Figure 7.1A-III). Superantigens are not known to be directly membrane-damaging, unlike other bacterial toxins such as staphylococcal alpha toxin (Jonas et al., 1994); nor is TSST-1 known to have any enzymatic activities that could account for its cytotoxicity in eukaryotic cells (Deresiewicz et al., 1994a). However, both epithelial and endothelial cells can facilitate the transport of intact TSST-1 by receptor-mediated transcytosis (Hamad et al., 1997) followed by apoptosis in response to the endocytosed toxin (Lee et al., 1991). Therefore, these non-specific cytotoxic mechanisms cannot be dismissed out of hand, in spite of the greater likelihood of an unknown death receptor being involved. 151 B P r o - S u r v i v a l ififtlaifc T X T 1 D o m i n a n t A p o o f o s f c Figure 7.1: Working Model. A. The most probable mechanism for G3 lR-induced apoptosis is direct ligation of a non-classical death receptor (I), that couples to the intrinsic mitochondrial pathway by a BH3-only pro-apoptotic protein. Other possibilities are direct membrane disruption (II) and cytotoxicity through non-specific damage (III). B. Based on this model, TSST-1 binds to separate high and low affinity receptors. Binding to high affinity receptors {left) leads to superantigenic activities, while binding to the low affinity receptor (right) leads to apoptosis. G3 IR binds predominantly to the low affinity receptor (right), leading to apoptosis and weak superantigenic activity. The ability of high dose TSST-1 to potentiate G31R-induced death during co-administration may be explained by the saturation of high affinity receptors, allowing the remaining TSST-1 and/or G31R to bmd to low affinity death receptors. 152 7.2. Evidence for a putative non-classical death receptor Proposing a ubiquitous death receptor (XDR) as the target of G31R might seem out of place at first, given the lower affinity of G31R for binding to MHC Class II in human PBMC compared to wild type TSST-1. However, the increase in potency of G31R for early apoptosis compared to TSST-1 might in fact be the net result of two properties: a decreased affinity for MHC class II or TCR (loss of function), an increased affinity for an XDR (gain of function), or both. Since other mutations in the MHC class II binding region and one in the TCR binding region of TSST-1 also resulted in enhanced pro-apoptotic activities (see details in Appendix A), a loss of function would be considered more likely. For example, TSST-1 has a relative high affinity for MHC class II (K<j = 440 nM) (Krakauer, 1999). The low affinity of G31R for MHC class II would result in relatively more unbound G31R that would be available to interact with the XDR at a given dose, compared with TSST-1 (Figure 7.1). This net result could account for the increased potency for triggering apoptosis by G31R compared with TSST-1. The pro-apoptotic activity of TSST-1 at high concentrations of 10 uM or above could be explained by the saturation of the high-affinity MHC class II binding sites, allowing the remaining TSST-1 molecules to bind to low-affinity XDR receptors. This hypothesis is also consistent with the finding that even though TSST-1 clearly activates anti-apoptotic pathways at moderate doses (<1 fiM), it was unable to reduce G31R-mediated apoptosis when human PBMC were co-stimulated by both toxins (Chapter 4.2.8). The critical aspect of this working model is that apoptosis signaling through the XDR must be dominant over any pro-survival pathways induced, since TSST-1 itself can promote survival through the degradation of caspase-3 in a wide range of cells (Chapter 3.2.5), as well as the activation of Akt in V82+ T cells (Chapter 3.2.4). This anti-apoptotic (pro-survival) activity of TSST-1 may be secondary to the activation of an inhibitor of apoptosis-inducing proteins (IAPs) (Figure 7.1B-IV). Once the signaling threshold for the activation of apoptosis through the XDR is achieved, the affected cell is committed to death 153 regardless of any attempted triggering of pro-survival pathways. This is evidently accomplished r through MOMP (Figure 7.1B-V), which could result in the translocation of the IAP-inhibitor Smac/Diablo from the mitochondria into the cytoplasm (Figure 7.1B-VI). This inhibition of IAP activity could explain why normal pro-caspase 3 processing and activation is observed following G31R, but not TSST-1 treatment (Chapter 5.2.4). 7.3. Precedence for important low-affinity interactions Since our XDR hypothesis for G31R mediated apoptosis implicates a low-affinity interaction between G31R and its target in various different cell types, it may be questioned whether there are other examples of biologically important low affinity receptors for staphylococcal superantigens. One such example is the low-affinity interaction between TSST-1 and the V/32+ TCR on T cells. It has only recently been possible to demonstrate direct binding of TSST-1 to V/32+ T cells (Kd = 2.3 uM) by surface plasmon resonance studies (McCormick et al., 2003), while previous evidence relied on inferences derived from TSST-1-driven expansion of V/32-specific T cells (Malchiodi et al., 1995; Fields et al., 1996; Li et al, 1998a). Additional receptors which interact with superantigens have been uncovered over time, including a p85 protein in the fibroblast cell line COS-1 which binds to SEB with low affinity but does not support TCR crosslinking (Rogers et al., 1995; Rogers et al., 1997); and a glycosphingolipid receptor (digalactosylceramide) on kidney proximal tubular cells that binds SEB but not SEA or TSST-1 (Chatterjee et al., 1995). Recent preliminary studies in our laboratory using fluorescent-labelled G31R and TSST-1 have also demonstrated low-affinity binding of these toxins to V/32" T cells that are significantly above background non-specific binding (data not shown), an observation that supports the notion that ubiquitous low-affinity receptors for staphylococcal superantigens, such as XDR, may exist in nature. 154 7.4. Properties of a putative death receptor for G31R The postulated XDR would need to fulfill several properties observed with G31R-induced apoptosis, including the ability to rapidly transduce an apoptotic signal within minutes of ligation. The extremely narrow timeframe required for G31R to induce apoptosis (<5 minutes' exposure) supports the concept of direct signaling through XDR (Chapter 4.2.5), while rendering the involvement of secondary mediators such as secretion or increased surface expression of pro-apoptotic factors less likely. A further property of this XDR would be that it can induce downstream caspase activation (Figure 7.1 A), likely through the induction of MOMP by the activation of BH3-only proteins such as Bim (as in the case for TCR ligation in thymocytes) (Puthalakath et al, 1999) or BNIP3 (for CD47 ligation on lymphocytes) (Lamy et al., 2003). While CD47 itself seems unlikely to be involved due to the relative ineffectiveness of anti-CD47 neutralizing antibodies in preventing G31R-induced death (Chapter 6.2.1), the kinetics and some of the characteristics of the cell death induced by CD47 seem similar to that induced by G31R. If a highly conserved XDR does prove to be the trigger for G31 R-mediated death, physiologic signaling through this XDR might account for some of the regulatory forms of apoptosis observed in the different hematopoietic cell lineages, including growth-factor withdrawal and T cell ACAD. Nonetheless, further work would be needed to definitively rule out the two categories of non-specific cytotoxic effects, membrane disruption and intracellular targeting, as being primarily responsible for G31 R-mediated apoptosis. 7.5. Concentration dependent effects of TSST-1 in the pathogenesis of staphylococcal toxic shock syndrome Revisiting the proliferation and cytokine dose response data available in the literature for the many known superantigens reveals a previously unappreciated but remarkably consistent pattern in which there appears to be a normal dose response within a relatively narrow concentration 155 range, followed by a paradoxical response of reduced biologic activity when the dose is further increased (Rink et al., 1997). No explanation has been provided for this phenomenon except to attribute such paradoxical responses to variable dependency on the T cell-APC bridge for the different superantigens. However, noting the striking paradoxical dose-effects of G31R on T cell proliferation and cytokine production at concentrations above 100 pM and its association with apoptosis (Kum et al., 1996), we propose that similar apoptotic activity may account for the observed decrease in proliferation with increasing doses of other superantigens including TSST-1, MAS, SEA, SED and SEE. In fact, the occurrence of spontaneous deletion of about 40-50% of superantigen-reactive T cells was noted with SEA, SED and SEE more than a decade ago, regardless of the presence of APCs (Kabelitz et al., 1992). In that context, the T cell death would appear to conform to the definition of activated T cell autonomous death (ACAD), discussed in Chapter 1.3. Should this type of rapid non-specific pro-apoptotic activity associated with G31R and high-dose TSST-1 be a general phenomenon with other superantigens, a revision of the current understanding of apoptosis in superantigen-mediated disease would be necessary. A possible scenario is depicted in Figure 9.1. At a localized site of infection or colonization by Staphylococcus aureus or Streptococcus pyogenes (termed the Zone of Destruction), high titers of superantigen could result in non-specific apoptosis of many first-line defender cells, including lymphocytes and monocytes/macrophages. This is supported by clinical observations that surgical wounds associated with nonmenstrual toxic shock syndrome demonstrate a remarkable lack of inflammation. This might confer an advantage to the micro-organism by allow it to establish a foothold past the initial innate immune response, but high concentrations of TSST-1 above 1 /xM are required to achieve this effect. At intermediate concentrations of TSST-1 in the local tissues (down to 50 pM), non-specific B cell apoptosis accumulates due to the regional production of IFN-7 (Figure 9.1; Zone ofB Cell Death) (Hofer et al, 1996). This might interfere 156 with the establishment of an effective humoral response to TSST-1 by condemning any B cells that might otherwise mount an antibody response. Finally, to assure long-term residence, the secreted superantigens may pack a secondary 'punch' by engaging in diversionary tactics (Figure 9.1; Zone of Distraction); that is, once diffused through the host circulation beyond the site of infection, superantigens such as TSST-1 would still be effective within the requisite dose range for conventional superantigenic activity by activating V/3-specific T cells and the induction of proinflammatory cytokines (down to fM levels). The ability of TSST-1 to delay AICD compared with SEB (Chapter 3.2.2) and to activate pro-survival pathways (Chapter 3.2.4 & 3.2.5) further confers an advantage by subverting and exacerbating an ineffective proinflammatory host response. Moreover, TSST-1 at these doses elicits polyclonal B cell activation and Ig production, again 'distracting' from a directed humoral response against the appropriate antigen(s). Should the host survive all three zones of superantigen-mediated immuno-dysregulation, eventual clearance of the superantigen follows (Figure 9.1; Clearance). A mutant such as G31R may be non-lethal in the murine model not because of its apoptotic activity, but because it is simply less potent at inducing lethal superantigenic effects (Kum et al., 1996). Wild type TSST-1, by contrast, would prolong its diversionary effects (Figure 9.1; Zone of Distraction) on the immune system in a susceptible host by maintaining dysfunctional T cell hyperactivity through the suppression of normal AICD and/or ACAD mechanisms for removing these cells. All this is conjecture at this point, since data on the actual concentrations of superantigens within local tissues and their correlation with immunobiologic consequences at the affected sites in vivo are lacking. 157 Apoptotic Bcell Apoptotic Tcell Toxin-producing S. aureus V Polyclonal B cell activation and proliferation VjB-specific T cell ^ activation and proliferation • • Figure 7.2: Apoptosis in Superantigen-Mediated Pathogenesis. Four zones of TSST-1 activity are defined by concentration ranges. Zone of Destruction: >1 uM, massive non-specific apoptosis occurs in immune cells that migrate into area near infection. Zone ofB Cell Death: >50pM, B cell apoptosis due to high local IFNy production. Zone of Distraction: >0.1fM, TSST-1 is superantigenic and promotes VP-specific T cell proliferation, proinflammatory cytokine release, as well as polyclonal B cell activation and non-sp-.cific Ig secretion. Clearance: eventually, TSST-1 is cleared. 158 7.6. Possible applications of G31R in cytotoxic therapies Superantigens are also potent inducers of cytotoxic T lymophocytes (CTLs), and several investigators have attempted to exploit this property for tumoricidal therapy by fusing SEA to a tumour marker-specific antibody in order to activate and direct CTLs at the tumour cells (Takemura et al., 2002). Although highly successful in vitro and in animal studies, clinical trials with this approach were unsuccessful due to unacceptable side-effects and toxicities associated with the administration of the SEA-fusion antibody. Since G31R and some of the other TSST-1 mutants have greatly reduced superantigenic activity, but exhibit potent pro-aptoptotic activity independent of CTLs, it may be possible to target tumor cells by using a similar approach with G31R. Additional fine-tuning of site-directed mutations might also minimize or eliminate any side effects associated with its superantigenic effects, while maintaining or even enhancing its cytotoxic activity. Further studies would be necessary to establish the range of G31R's cytotoxic activity on non-haematopoietic cell lines. If the toxin should prove specific only for cells of the haematopoietic lineage, possible applications would still include treatment of haematogenous malignancies including leukemias and lymphomas, as well as autoimmune conditions. If the toxin has wide-ranging cytotoxicity toward unrelated cell types, this approach might be extended to solid tumours as well. Further advantages of using superantigen or superantigen-derivatives in solid tumour therapy would include: the induction of an inflammatory state, which might aid in directing an immune response against the tumour; direct destruction of some portion of the cell mass; destruction of local endothelium established in the tumour which could complement anti-angiogenic strategies; and superantigen translocation across epithelial and endothelial barriers, which would allow superantigen entry into difiicult-to-access sites follwing infusion by catheterization. Lastly, identification of the putative death receptor triggered by G31R might lead to applications involving use of the physiologic ligand(s) of this receptor. 159 7.7. Potential application of G31R for research in cell biology Novel natural and synthetic compounds are routinely screened for cytotoxicity as a prelude to possible applications in therapeutic or research environments. Regardless of whether or not G31R ultimately shows promise for cytotoxic therapies, it will likely be a very useful reagent in apoptosis research, much like etopside, staurosporine, and camptothecin (Gorczyca et al., 1993b; Deas et al., 1998; Ferraro-Peyret et al., 2002). Novel, early and potent apoptotic effects similar to those associated with G31R have been of recent interest in the context of the regulation of activated T cells via ACAD (Hildeman et al., 2002b). The description of receptors such as CD47 provides a framework for how these processes may be relevant in vivo as either primary or backup systems for the removal of activated T cells (Pettersen et al., 1999; Lamy et al., 2003). Moreover, the variable susceptibility of different cell lines to G31R-mediated apoptosis (Chapter 4.2.7) might serve to underscore differences in their expression of anti-apoptotic molecules such as Bcl-2 or of pro-survival oncogenes among these tumour-derived cells. 7.8. Future directions The G31R mutant has been the preferred reagent for studying the early pro-apoptotic response for several reasons. Firstly, a substantial body of work has already been devoted to the characterization of this mutant, including the original data that sparked an interest in determining whether the precipitous decline in proliferative activity at high concentrations was attributable to anergy or apoptosis (Kum et al., 1996). The fact that G31R does still exhibit some proliferative activity indicates that it must still induce some T cell responses. Furthermore, since G31R is non-lethal in a murine model of toxic shock syndrome (Kum et al., 1996), it may have potential therapeutic applications in the treatment of human diseases such as tumours and autoimmunity. Therefore, future studies would aim to complete the gaps in understanding of the mechanism for G31R-mediated apoptosis. To identify candidates for the XDR, immuno-precipitation would 160 be performed using G31R and/or TSST-1 on membrane proteins of cells that do not express MHC class II or V/32-specific TCR. If such a receptor can be found, approaches such as the yeast two-hybrid system may be used to locate candidate signaling molecules such as BH3-only proteins to couple the XDR to MOMP. The actual dependency on individual molecules such as specific BH3-only proteins could be confirmed using knockouts in G31R-susceptible cell lines. Similarly, over-expression of Bcl-2 or BCI-XL could be attempted to prevent G31R-mediated death. Finally, the surprising lack of dependence on caspase activation could also be confirmed using caspase knockout cell lines. The range of cell types susceptible to the apoptotic activity of G31R might become clear with the identification of the XDR and downstream signaling partners, but concurrent studies could be performed to determine the scope of G31R's apoptotic activity on other cells of the haematopoietic and non-haematopoietic lineages, including granulocytes in the former category and epithelial cells in the latter. Similarly, correlations between the susceptibility of different tumour cell lines with their known assortment of under- or over-expressed proteins might provide useful information and complement the mechanistic studies summarized above. Besides the XDR, two other hypotheses explaining G31R-induced apoptosis were membrane disruption and direct cytotoxicity. The first could be tested by observing ionic fluxes and monitoring permeability of cells to varying sizes of cations (e.g. Na+, K + , Ca 2 + , Mg2 +) during the course of G31R treatment. These cations would be more sensitive to small pores or increases in membrane permeability that were not detectable using viability dyes such as trypan blue, PI or 7-AAD. The second hypothesis of direct cytoxicity requires that the toxin can translocate into cells. A combined approach of neutralizing externally bound G31R and intracellular staining of any internalized toxin could be used in concert with both flow cytometry and confocal microscopy to determine whether G31R can be endocytosed into cells. Cell-free systems containing only the cytosolic constituents and organelles without plasma membranes would lack 161 input from surface receptors, and allow testing to determine whether specific intracellular targets for G31R exist. Moreover, inhibitors of reactive oxygen species, including antioxidants and the superoxide dismutase mimetic Mn (III) tetrakis (5, 10, 15, 20-benzoic acid) porphyrin could be used to determine whether G31R relied upon non-specific oxidative mechanisms of cellular damage to induce death. Further structure-function analysis, including crystallography and epitope peptide mapping techniques, might permit the identification of death-inducing epitopes on the G31R and/or TSST-1 molecules themselves. These might prove fruitful for exploring its therapeutic applications in addition to clarification of its proinflammatory activity or immunogenicity. 7.9. Conclusion The study of apoptosis regulation by TSST-1 and several site-directed mutants have identified both anti-apoptotic and pro-apoptotic activities associated with the toxin. At moderate doses up to 1 [iM, TSST-1 specifically induces activation of the anti-apoptotic PBK/Akt pathway in V/32+ T cells, while simultaneously triggering rapid loss of cellular pro-caspase 3 content in diverse cell populations. Both of these biologic properties are suggested to play a role in the observed resistance of TSST-1-activated T cells to AICD and ACAD. However, these anti-apoptotic activities did not appear related to the potent and early pro-apoptotic activity associated with the G31R mutant and higher doses of TSST-1 (>1 /iM). Instead, this unique form of apoptosis appeared to utilize the intrinsic pathway to achieve mitochondrial dysfunction and cell death, but with minimal requirement for caspase activation. A weak interaction of TSST-1 and G31R with a death receptor capable of activating one or more BH3-only members of the Bcl-2 family is postulated to be responsible for the observed Bax activation leading to MOMP. 162 Appendix A Apoptotic Activity of TSST-1 Mutants: Structure and Function "The simplest explanation is that it doesn't make sense. " -Professor William Buechner 8.1. Introduction In the course of studying the structure-function correlations of TSST-1, our laboratory has previously generated a set of site-directed mutants other than G31R. Having demonstrated the pro-apoptotic activity of G31R, the question was raised as to whether these other TSST-1 mutants also have apoptogenic potential, and whether their apoptotic activity could be correlated with their known structure and superantigenic activities including T cell mitogenesis and cytokine production in human PBMC (Figure 8.1). The objective of the studies in this Chapter was therefore to determine whether any of this panel of site-directed TSST-1 mutants retained either the anti-apoptotic or pro-apoptotic activities of the wild type toxin, with the intent of identifying particular epitopes of the TSST-1 molecule that might be important for regulating apoptotic activity. To this end, the site directed mutants studied included those that are either defective in MHC class II binding (G31R, S49N, P50S, S53R, S53K, A55T and T57S) or TCR interaction (S14N and H135A), and one unrelated mutant, P95F. Among these mutants, G31R, S14N and H135A have been previously and described in the literature (Kum et al., 1996; Kum et al., 2001; Kum et al., 2002; Bonventre et al., 1995). All three bear some defects in superantigenic activity compared with the wild type toxin, such as a decreased potency in proliferation or cytokine production. 163 Figure 8.1: A. Three dimensional space-filling surface of TSST-1 showing colour coded regions corresponding to areas in which site-directed mutants were generated (turquoise: G31; gold: F47-E60; magenta: S14 and H135; green: P95. B. Reverse face of TSST-1 molecule in A. 164 8.1.1. Mutations around MHC class II binding sites Mutations generated in and around the MHC class II binding regions of TSST-1 included G31R, S49N, P50S, S53R, S53K, A55T and T57S. G31 is involved in the structure of the TSST-1 molecule by linking the Bl and 82 strands (Figure 8.2B). Thus, a change from glycine to the bulky, positively charged arginine residue at this location may well force some opening of the angle between the exit of the 81 strand and the entry into the 82 strand of the /3-barrel. During complex formation with HLA-DR1, the G31 residue provides sufficient flexibility within the region spanning residues 27-34, thus permitting optimal contact with the MHC class II molecule. The neighbouring residue L30 is involved in a hydrophobic interaction with nonpolar amino acids (M36, 163 and Y113) of HLA-DR1, and represents the most buried region of the TSST-1 molecule (Figure 8.3B) (Papageorgiou et al, 1996). Additionally, the adjacent polar residue D27 forms hydrogen bonds with residues on DR1 (Figure 8.3B), while the polar residue L67 of DR1 interacts with S29 and R34 on TSST-1 (not shown). Thus, it is not surprising that the mutant G3 IR loses most of its binding affinity for MHC class II molecules in human PBMC. Residues S49 to T57 all lie within a surface random coil portion of the TSST-1 polypeptide spanning from F47 to E60 within the N-terminal major domain of TSST-1 (Figure 8.1 A). Residue P50 permits a sharp turn that properly orients the subsequent Y51 to nestle its aromatic ring into a pocket formed by E173-P180 within the C-terminal domain (Figure 8.4C). P50 also forms a hydrogen bond interaction with DR1 in the TSST-1-HLA-DR1 complex (Figure 8.3B). Therefore, the initial portion of the turn formed by S49-Y52 seems intimately involved not only in the docking of the N-terminal TSST-1 domain onto the C-terminal domain, but also in MHC class II interaction. Most of the remainder of the random coil, from Y52 to E60, is surface exposed, framing the /3-barrel that constitutes most of the N-terminal domain of TSST-1 (Figure 8.1 A). With the majority of these R-groups oriented outward, the mutations from S53-T57 probably only minimally disrupt the secondary structure of the TSST-1 molecule. However, 165 both P50 and S53 are known to form hydrogen bonds with K39 of DR1, a residue that when mutated disrupts binding by TSST-1 (Kim et al., 1994). Moreover, the highly polar K58, along with D27, form hydrogen bonds to polar residues on DR1 (Figure 8.3B). 166 Figure 8.2: A . Three dimensional ribbon structure of TSST-1 showing colour coded structural regions (red: a helix; turquoise: P sheet; grey: random coil), with space-filling surface for F47-E60 and ball-and-stick model for S29-S32 (yellow: S29; orange: L30; turquoise: G31; magenta: S32). B. Detail of dotted line area from A showing G31 's involvement in the turn between the pi and P2 strands. 167 Figure 8.3: TSST-1 : H L A - D R Contact. A . Ribbon diagram of TSST-1 (light green) complexed with human H L A - D R 1 (spacefilling). TSST-1 residues involved in binding to H L A - D R 1 are shown with ball-and-stick models. TSST-1 contacts primarily the H L A - D R 1 a chain (yellow), with a small overlap on to the H L A - D R 1 p chain (orange) and the peptide presented in the antigen-presentation groove (magenta). B . Detail of dotted region from A . 168 Figure 8.4: A . Three dimensional ribbon structure of TSST-1 in usual orientation showing colour coded structural regions (red: a helix; turquoise: P sheet; grey: random coil), with space-filling surface for E173-P180 and ball-and-stick model for F47-E60 (purple: F47; turquoise: P50; orange: Y51). B. Rotation of molecule into orientation for C. C. Detail of dotted line area from B showing P50 (turquoise) producing a turn in the backbone that nestles Y51 (orange) in the pocket formed by the surface of E173-P180. 169 8.1.2. Mutations around TCR interaction site Mutations to the TCR interaction region of TSST-1 included S14N and H135A, with S14 exposed at the exit of the al helix, and HI35 exposed on the o2 helix adjacent to each other (Figure 8.5B). The crossed, central alpha helices constitute a second major stabilization point for the N-terminal and C-terminal domains of TSST-1. The amino group of S14 forms the last hydrogen bond in the backbone of the al helix with the carboxyl oxygen of L10, before the helix unravels with S15 and G16 (Figure 8.5B). S14 also forms a van der Waals interaction with H135 (Figure 8.5B) (Papageorgiou et al., 1996). H135 itself is solvent exposed, being situated in the middle of the cQ. helix spanning from S127-T138, and is involved in numerous intramolecular interactions such as hydrogen bonding to Y13 and SI5, and van der Waals interactions with Y13, S14, S15, Q136 and Q139 (Figure 8.5B). 8.1.3. Mutation not related to MHC class II or TCR interaction sites Finally, the mutant P95F has its mutation site in the C-terminal domain of TSST-1, on a surface-exposed region of the random coil linking the N-terminal and C-terminal domains (Figure 8. IB). The precise function of this region of the TSST-1 molecule is undetermined, but it does not appear to affect either TCR or MHC class II-dependent interactions. Therefore, it was used as a control to compare with mutant toxins known to affect either MHC class II or TCR interactions. 170 Figure 8.5: A. Three dimensional ribbon structure of TSST-1 showing colour coded residues and ball and stick models of pertinent amino acids in the two a helices (green: L10; orange: S14; purple: F131; indigo: H135; turquoise: Q139. B. Detail of dotted rectangle in A. 171 8.2. Experimental Approach The determination of the structure-function relationships between superantigens and their activities on T cells and APCs for proliferation and cytokine induction have depended upon both mutational studies and crystallography. Therefore, to examine the regulation of apoptosis by TSST-1, it would also seem reasonable to examine and correlate the structure and function relationships of these site-directed TSST-1 mutants with their apoptotogenic activity. Given that TSST-1 possesses a dual personality in exhibiting both anti-apoptotic properties at low doses (<1 /xM) and pro-apoptotic effects at high doses (>1 fiM), the question arises as to whether mutants with specific defects in either TCR or MHC class II interaction could affect the toxin's ability to induce or suppress apoptosis. All of the mutants were initially screened for early pro-apoptotic activity by treating human PBMC with either 100 nM or 1 /iM of the toxin for 3 hours, prior to staining for phosphatidylserine exposure with Annexin V-FITC and flow cytometric analysis. Mutants demonstrating no significant elevation in percentage apoptosis compared with the RPMI medium control were re-assessed at a dose of 10 itM. The results were expressed as percentage apoptosis, calculated by dividing the number of cells staining with Annexin V-FITC by the total number of cells. 8.3. Results 8.3.7. All TSST-1 mutants induce apoptosis to varying extents, and there is no correlation between apoptotic and superantigenic activities G31R and S53R were found to be particularly potent at inducing apoptosis, with significant activity at 100 nM (Figure 8.6A). In contrast, the majority of mutants, including S49N, P50S, S53K, A55T and HI35A only induced an early pro-apoptotic response at 1 /xM but not 100 nM (Figure 8.6A). Three mutants including S14N, T57S and P95F did not induce any apoptosis above the RPMI medium control even at 1 iiM (Figure 8.6A). When these three mutants and the 172 low responder S49N were assayed for apoptotic activity at 10 juM, all induced an apoptotic response (Figure 8.6B). Wild type TSST-1 assessed in parallel also required 10 /xM to induce a notable level of apoptosis compared to the RPMI medium control (Figure 8.6B). The correlation between the potency of the early apoptotic response and their known superantigenic properties (mitogenic activity and proinflammatory cytokine production) for all the TSST-1 mutants are shown in Figure 8.7 and summarized in Table 7.1. All of the mutants, like wtTSST-1, were capable of inducing rapid, non-specific apoptosis in human PBMC when sufficiently high doses (10 uM) were used. There was no correlation between apoptotic and superantigenic activities. 173 RPMI S U N G31R S49N P50S S53K S53R A55T T57S P95F H135A TSST-1 RPMI S14N S49N T57S P95F TSST-1 Figure 8.6: A . Percentage apoptosis by Annexin V - F I T C staining in human P B M C from three donors after toxin treatment. B . TSST-1 and mutants toxins that induced only low levels of apoptosis at 1 u M ( S U N , S49N, T57S, and P95F) were re-studied at a higher concentration of 10 u M in one additional donor. 174 A. T57S S53R/K B. T57S Figure 8.7: Three dimensional space-filling surfaces of TSST-1 showing the location of mutations, with colour coding for the mutants' apoptotic activities (apoptotic activity at: 10 uM, red; 1 uM, purple; 100 nM, blue). A. Dotted line denotes the known MHC class II interaction region. B. Dotted line denotes the TCR interaction region. 175 Table 8.1: Summary of TSST-1 mutants and their apoptotic vs. superantigenic activities Mutant TCR/MHCII Mutation Region MHCII Binding Affinity Mitogenic Activity Cytokine Induction Apoptotic Activity wt TSST-1 N/A +++ ++++ ++++ + S14N TCR +++ ++ - + G31R MHCII + + + +++ S49N MHCII unknown +++ +++ ++ P50S MHCII unknown +++ +++ ++ S53R MHCII +++ +++ +++ +++ S53K MHCII unknown unknown unknown ++ A55T MHCII +++ ++++ ++++ ++ T57S MHCII +++ ++++ ++++ + P95F N/A unknown ++++ ++++ + H135A TCR +++ _* - ++ Mutation region refers to either TCR or MHC class II interaction sites. Binding affinity was determined by the mutant's ability to compete with 125I-labeled wtTSST-1 for binding to human PBMC (Kum et al., 1996). The mitogenic activity was determined using 3H-thymidine incorporation over 24 hours as a marker of proliferation. Cytokine induction refers to TNFa levels detected by ELISA in human PBMC supernatants after treatement for 24 hours. 'Unknown' means that the mutant has not been tested. Apoptotic activity was measured by Annexin V-FITC staining, with a '+' referring to significant apoptosis above baseline only at 10 /xM and above; '++' at 1 /xM and above, and '+++' at 100 nM and above. * While originally described as a mutation abolishing mitogenicity (Blanco et al, 1990; Cullen et al., 1995; Drynda et al., 1995), H135A was more recently shown to induce slight mitogenesis in rabbit splenocytes at 0.01 /xg/well, but not at higher or lower doses (Prasad et al., 1997). 176 8.4. Conclusions and discussion of data At face value, the apoptotic activity of the TSST-1 mutants would seem to suggest the existence of regulatory regions for apoptosis. With the notable exception of H135A, most of the TSST-1 mutants that exhibit potent apoptotic activity (requiring 1 uM or higher) have alterations within the MHC class II binding region (Table 7.1). On the other hand, the four toxins that exhibit low-level apoptotic activity (requiring 10 uM or higher) include the wtTSST-1, or mutations in either the MHC class II binding site (T57S), the TCR interaction site (S14N), or an unrelated region (P95F). However, the various MHC Class II binding mutants likely differ from each other in their relative binding kinetics and affinities to multiple receptors. For example, while TSST-1 may bind strongly to both MHC class II and the TCR, G31R exhibits a weaker MHC class II binding interaction. This may have permitted G3 IR to trigger some unknown death receptor rather than cognate interactions with the MHC class II-TCR complex. Furthermore, the lower affinity for MHCII and/or TCR may allow more G31R to be available for binding to a death receptor. S53R and A55T would seemingly violate this possibility by promoting apoptosis without detectable loss of MHC class II binding activity, or marked impairment in mitogenesis and cytokine production. However, the proximity of S53 and A55 (Figure 8.7A) suggests that this random coil region may perhaps function as a unique anti-apoptotic domain or an interaction region for a putative death receptor, such that the mutations in this region have resulted in an enhancement of apoptosis. Moreover, G31 is situated immediately adjacent to the F47-E60 random coil containing S53 and A55 (Figure 8.1 A & 8.7A). The more potent apoptotic activity exhibited by S53R (about 10-fold higher than G31R) might have resulted from its interaction with the putative death receptor, in contrast to the closely related S53K which is more similar to G31R in activity. 177 The apoptotic activity of the HI35A mutant with a known defect in the TCR interaction region could also be consistent with this hypothesis. The reduction in cognate interactions between the TCR and MHC Class II following HI35A stimulation might be sufficient to affect the on- and off-rate kinetics, and permit triggering of the putative death receptor. The closely related S14N, by contrast, retains some mitogenic activity, suggesting only a partial defect in T cell interaction that might not be sufficient to indirectly promote death receptor binding or triggering. Further structure-function correlations would require accurate structural determinations and additional studies by site-directed mutagenesis. While circular dichroism on this set of mutants was attempted to identify gross changes in secondary and tertiary structure, difficulties in obtaining reliable component structural data for a-helical, /3-sheet and random coil content prevented comparisons at this general level. However, more accurate structural data from X-ray crystallography or nuclear magnetic resonance, with or without complexing with target proteins, might prove valuable toward uncovering the differences between these mutants. Moreover, new mutations could be generated, either by modifying the R- group for existing mutants or creating new mutations at adjacent residues, to probe the nature of the interactions responsible for apoptosis. 178 Appendix B Publications 9.1. Papers Hung R W Y , Chow AW. Apoptosis: Molecular mechanisms, regulation and role in pathogenesis. Can J Infect Dis 1997;8(2):103-109. Kum WW, Cameron SB, Hung R W , Kalyan S, Chow AW. Temporal sequence and kinetics of proinflammatory and anti-inflammatory cytokine secretion induced by toxic shock syndrome toxin 1 in human peripheral blood mononuclear cells. Infect Immun 2001 Dec;69(12):7544-9 Kum WW, Hung R W , Cameron SB, Chow AW. Temporal sequence and functional implications of V beta-specific T cell receptor down-regulation and costimulatory molecule expression following in vitro stimulation with the staphylococcal superantigen Toxic shock syndrome toxin-1. J Infect Dis 2002 Feb 15;185(4):555-60. 9.2. Abstracts Hung R, Chow AW: Toxic shock syndrome toxin-1 (TSST-1) induced apoptosis of T cells in vitro. Abstract 369, Royal College/CSCI Annual Meeting, Vancouver, Sept 26, 1997. Hung RW, Chow AW: Toxic shock syndrome toxin-1 induced apoptosis - possible mechanisms. Western AFCR, Carmel, Feb 1998. J Invest Med, 46:161 A, 1998. Hung R W , Chow AW: Toxic shock syndrome toxin-1: apoptosis in vitro. Royal College/CSCI Annual Meeting, Toronto, September, 1998. Hung RW, Chow AW: Toxic shock syndrome toxin-1 induced apoptosis in thymocytes in vivo. Western AFCR, Carmel, January 1999. Hung R, Chow AW: Toxic shock syndrome toxin-1 induction of apoptosis in naive and mature immune cell populations. Western CSCI Regional Meeting: Young Investigators' Forum. Vancouver, June 26, 1999. Hung R, Chow AW: Toxic shock syndrome toxin-1 induction of apoptosis in naive and mature immune cell populations. CSCI/MRI Joint Program, Montreal, September 23, 1999 (CSCI trainee award). Hung RW, Chow AW: Anti-apoptotic regulation by toxic shock syndrome toxin-1 - role of Fas receptor, Western AFCR, Carmel, February 2000. Hung RW, Chow AW: An ode to the dead T lymphocyte - a tale of apoptotic proportions. MRC National Health Research Day, Winnipeg, June, 2000. 179 Hung RW, Chow AW: A single-amino-acid substitution mutant of toxic shock syndrome toxin-1 induces apoptosis independent of Fas receptor and CD8+ cytotoxic T lymphocytes. CSCI-MRC Program, Edmonton, AB, Sept 21, 2000. Hung RW, Kum WW, Chow AW: A single-amino-acid substitution mutant of toxic shock syndrome toxin-1 induces apoptosis independent of Fas receptor and CD8+ cytotoxic T lymphocytes. UBC Experimental Medicine Student Research Day, Vancouver, October, 2000 (best poster presentation award). Hung RW, Kum WW, Chow AW: A mutant of toxic shock syndrome toxin-1 exerts apoptotic activity by a Fas receptor and perforin independent pathway. Western AFCR, Carmel, CA, Feb 2001. Hung RW, Kum WWS, Chow AW: Identification of regulatory domains for apoptosis in the toxic shock syndrome toxin-1 molecule. CSCI-CIHR Program, Ottawa ON, Sept 20, 2001. Clin Invest Med 24:220, 2001. Hung RW, Kum WWS, Chow AW: Regulation of apoptosis by the toxic shock syndrome toxin-1 (TSST-1) and mutant toxin G31R. UBC Experimental Medicine Student Research Day, Vancouver, Nov 2, 2001 (best poster presentation award). Hung RW, Kum WWS, Chow AW: Mechanisms for induction of apoptosis by the toxic shock syndrome toxin-1 (TSST-1) mutant G31R. Western AFCR, Carmel, CA, Feb 2001; J Invest Med 50:48A, 2002. Hung RW, Kum WWS, Chow AW: Regulation of apoptosis by toxic shock syndrome toxin-1 (TSST-1) and the mutant G31R. Exp Biol Annual Meeting, New Orleans, Apr 2002 (AFMR Trainee Travel Award). Hung RW, Kum WWS, Chow AW: The G31R mutant of toxic shock syndrome toxin-1: intrinsic and extrinsic pathways for apoptosis. CSCI Young Investigators Forum, Ottawa, September, 2002. Hung RW, Chow AW: Mitochondrial Involvement During Apoptosis Triggered by the G31R Mutant of Toxic Shock Syndrome Toxin-1. CSCI Young Investigators Forum, Ottawa, September, 2003. 180 Bibliography Acehan, D., Jiang, X., Morgan, D. G., Heuser, J. E., Wang, X., and Akey, C. W. (2002). Three-dimensional structure of the apoptosome: implications for assembly, procaspase-9 binding, and activation. Mol.Cell 9 , 423-432. Acharya, K. R., Passalacqua, E. F., Jones, E. Y., Harlos, K., Stuart, D. I., Brehm, R. D., and Tranter, H. S. (1994). Structural basis of superantigen action inferred from crystal structure of toxic-shock syndrome toxin-1. Nature 3 6 7 , 94-97. Adrain, C , Slee, E. A., Harte, M. T., and Martin, S. J. (1999). Regulation of apoptotic protease activating factor-1 oligomerization and apoptosis by the WD-40 repeat region. J.Biol. Chem. 214, 20855-20860. Akatsuka, H., Imanishi, K., Inada, K., Yamashita, H., Yoshida, M., and Uchiyama, T. (1994). Production of tumour necrosis factors by human T cells stimulated by a superantigen, toxic shock syndrome toxin-1. Clin.Exp.Immunol. 9 6 , 422-426. Al Madani, O., Gordon, A. E., Weir, D. M., Raza, M. W., Busuttil, A., and Blackwell, C. (1999). Pyrogenic toxins of Staphylococcus aureus in sudden unexpected nocturnal deaths in adults and older children: factors influencing the control of inflammatory responses to toxic shock syndrome toxins. FEMS Immunol.Med.Microbiol. 2 5 , 207-219. Alam, A., Cohen, L. Y., Aouad, S., and Sekaly, R. P. (1999). Early activation of caspases during T lymphocyte stimulation results in selective substrate cleavage in nonapoptotic cells. J.Exp.Med. 1 9 0 , 1879-1890. Algeciras-Schimnich, A., Griffith, T. S., Lynch, D. H., and Paya, C. V. (1999). Cell cycle-dependent regulation of FLIP levels and susceptibility to Fas-mediated apoptosis. J.Immunol. 1 6 2 , 5205-5211. Alouf J. E. and Muller-Alouf, H. (2003). Staphylococcal and streptococcal superantigens: molecular, biological and clinical aspects. Int.J.Med.Microbiol. 2 9 2 , 429-440. Andrews, M. M., Parent, E. M., Barry, M., and Parsonnet, J. (2001). Recurrent nonmenstrual toxic shock syndrome: clinical manifestations, diagnosis, and treatment. Clin.Infect.Dis. 3 2 , 1470-1479. Antonsson, B. (2001). Bax and other pro-apoptotic Bcl-2 family "killer-proteins" and their victim the mitochondrion. Cell Tissue Res. 3 0 6 , 347-361. Ashkenazi, A. and Dixit, V. M. (1998). Death receptors: signaling and modulation. Science 2 8 1 , 1305-1308. Aubert, V., Schneeberger, D., Sauty, A., Winter, J., Sperisen, P., Aubert, J. D., and Spertini, F. (2000). Induction of tumor necrosis factor alpha and interleukin-8 gene expression in bronchial epithelial cells by toxic shock syndrome toxin 1. Infect.Immun. 6 8 , 120-124. 181 Bauer, B., Jenny, M., Fresser, F., Uberall, F., and Baier, G. (2003). AKTl/PKBalpha is recruited to lipid rafts and activated downstream of PKC isotypes in CD3-induced T cell signaling. FEBS Lett. 541, 155-162. Bergdoll, M. S., Crass, B. A., Reiser, R. F., Robbins, R. N., and Davis, J. P. (1981). A new staphylococcal enterotoxin, enterotoxin F, associated with toxic-shock-syndrome Staphylococcus aureus isolates. Lancet 1, 1017-1021. Berndt, C , Mopps, B., Angermuller, S., Gierschik, P., and Krammer, P. H. (1998). CXCR4 and CD4 mediate a rapid CD95-independent cell death in CD4(+) T cells. Proc.Natl.Acad.Sci. U.S.A 95, 12556-12561. Bidere, N. and Senik, A. (2001). Caspase-independent apoptotic pathways in T lymphocytes: a minireview. Apoptosis. 6, 371-375. Blanco, L., Choi, E. M., Connolly, K., Thompson, M. R., and Bonventre, P. F. (1990). Mutants of staphylococcal toxic shock syndrome toxin 1: mitogenicity and recognition by a neutralizing monoclonal antibody. Infect.Immun. 58, 3020-3028. Blume-Jensen, P., Janknecht, R., and Hunter, T. (1998). The kit receptor promotes cell survival via activation of PI 3-kinase and subsequent Akt-mediated phosphorylation of Bad on Serl36. Curr.Biol. 8, 779-782. Bohach, G. A., Fast, D. J., Nelson, R. D., and Schlievert, P. M. (1990). Staphylococcal and streptococcal pyrogenic toxins involved in toxic shock syndrome and related illnesses. Crit Rev.Microbiol. 17, 251-272. Boldin, M. P., Goncharov, T. M., Goltsev, Y. V., and Wallach, D. (1996). Involvement of MACH, a novel MORTl/FADD-interacting protease, in Fas/APO-1- and TNF receptor-induced cell death. Cell $5, 803-815. Boldin, M. P., Varfolomeev, E. E., Pancer, Z., Mett, I. L., Camonis, J. H., and Wallach, D. (1995). A novel protein that interacts with the death domain of Fas/APOl contains a sequence motif related to the death domain. J.Biol.Chem. 270, 7795-7798. Bonventre, P. F., Heeg, H., Edwards, C. K., Ill, and Cullen, C. M. (1995). A mutation at histidine residue 135 of toxic shock syndrome toxin yields an immunogenic protein with minimal toxicity. Infect.Immun. 63, 509-515. Borutaite, V. and Brown, G. C. (2003). Mitochondria in apoptosis of ischemic heart. FEBS Lett. 541, 1-5. Bouillet, P., Metcalf, D., Huang, D. C , Tarlinton, D. M., Kay, T. W., Kontgen, F., Adams, J. M., and Strasser, A. (1999). Proapoptotic Bcl-2 relative Bim required for certain apoptotic responses, leukocyte homeostasis, and to preclude autoimmunity. Science 286, 1735-1738. Bouillet, P., Purton, J. F., Godfrey, D. I., Zhang, L. C , Coultas, L., Puthalakath, H., Pellegrini, M., Cory, S., Adams, J. M., and Strasser, A. (2002). BH3-only Bcl-2 family member Bim is required for apoptosis of autoreactive thymocytes. Nature 415, 922-926. 182 Brandt, K., van der, B. J., Fliegert, R., and Gehring, S. (2002). TSST-1 induces Thl or Th2 differentiation in naive CD4+ T cells in a dose- and APC-dependent manner. Scand J.Immunol. 56, 572-579. Bratton, D. L., May, K. R., Kailey, J. M., Doherty, D. E., and Leung, D. Y. (1999). Staphylococcal toxic shock syndrome toxin-1 inhibits monocyte apoptosis. J.Allergy Clin.Immunol. 103, 895-900. Brunner, T., Mogil, R. J., LaFace, D., Yoo, N. J., Mahboubi, A., Echeverri, F., Martin, S. J., Force, W. R., Lynch, D. FL, Ware, C. F., and . (1995). Cell-autonomous Fas (CD95)/Fas-ligand interaction mediates activation-induced apoptosis in T-cell hybridomas. Nature 373, 441-444. Budihardjo, I., Oliver, PL, Lutter, M., Luo, X., and Wang, X. (1999). Biochemical pathways of caspase activation during apoptosis. Annu.Rev. Cell Dev.Biol. 15, 269-290. Burgering, B. M. and Medema, R. H. (2003). Decisions on life and death: FOXO Forkhead transcription factors are in command when PKB/Akt is off duty. J.Leukoc.Biol. 73, 689-701. Cameron, S. B., Nawijn, M. C , Kum, W. W., Savelkoul, H. F., and Chow, A. W. (2001). Regulation of helper T cell responses to staphylococcal superantigens. Eur. Cytokine Netw. 12, 210-222. Cappello, F., Bellafiore, M., Palma, A., and Bucchieri, F. (2002). Defective apoptosis and tumorigenesis: role of p53 mutation and Fas/FasL system dysregulation. Eur.J.Histochem. 46, 199-208. Chan, W. Y., Lorke, D. E., Tiu, S. C , and Yew, D. T. (2002). Proliferation and apoptosis in the developing human neocortex. Anat.Rec. 267, 261-276. Chatterjee, S., Khullar, M., and Shi, W. Y. (1995). Digalactosylceramide is the receptor for staphylococcal enterotoxin-B in human kidney proximal tubular cells. Glycobiology 5, 327-333. Check, E. (2002). Worm cast in starring role for Nobel prize. Nature 419, 548-549. Chen, J., Amasaki, Y., Kamogawa, Y., Nagoya, M., Arai, N., Arai, K., and Miyatake, S. (2003). Role of NFATx (NFAT4/NFATc3) in expression of immunoregulatory genes in murine peripheral CD4+ T cells. J.Immunol. 170, 3109-3117. Cheng, E. FL, Nicholas, J., Bellows, D. S., Hayward, G. S., Guo, H. G., Reitz, M. S., and Hardwick, J. M. (1997). A Bcl-2 homolog encoded by Kaposi sarcoma-associated virus, human herpesvirus 8, inhibits apoptosis but does not heterodimerize with Bax or Bak. Proc.Natl.Acad.Sci.U.S.A 94, 690-694. Chesney, P. J., Davis, J. P., Purdy, W. K., Wand, P. J., and Chesney, R. W. (1981). Clinical manifestations of toxic shock syndrome. JAMA 246, 741-748. Chinnaiyan, A. M., O'Rourke, K., Tewari, M., and Dixit, V. M. (1995). FADD, a novel death domain-containing protein, interacts with the death domain of Fas and initiates apoptosis. Cell 81, 505-512. 183 Choi, Y., Lafferty, J. A., Clements, J. R., Todd, J. K., Gelfand, E. W., Kappler, J., Marrack, P., and Kotzin, B. L. (1990). Selective expansion of T cells expressing V beta 2 in toxic shock syndrome. J.Exp.Med. 172, 981-984. Chow, A. W., Wong, C. K., MacFarlane, A. M., and Bartlett, K. H. (1984). Toxic shock syndrome: clinical and laboratory findings in 30 patients. Can.Med.AssocJ. 130, 425-430. Chun, H. J., Zheng, L., Ahmad, M., Wang, J., Speirs, C. K., Siegel, R. M., Dale, J. K., Puck, J., Davis, J., Hall, C. G., Skoda-Smith, S., Atkinson, T. P., Straus, S. E., and Lenardo, M. J. (2002). Pleiotropic defects in lymphocyte activation caused by caspase-8 mutations lead to human immunodeficiency. Nature 419, 395-399. Cikala, M., Wilm, B., Hobmayer, E., Bottger, A., and David, C. N. (1999). Identification of caspases and apoptosis in the simple metazoan Hydra. Curr.Biol. 9, 959-962. Creagh, E. M., Conroy, H., and Martin, S. J. (2003). Caspase-activation pathways in apoptosis and immunity. Immunol Rev 193, 10-21. Cullen, C. M., Blanco, L. R., Bonventre, P. F., and Choi, E. (1995). A toxic shock syndrome toxin 1 mutant that defines a functional site critical for T-cell activation. Infect.Immun. 63, 2141-2146. Dao, T., Holan, V., and Minowada, J. (1993). Interleukin-2 production by T cells: a study of the immunoregulatory actions of interferon-alpha, interferon-gamma, and tumor necrosis factor-alpha in phenotypically different T cell clones. Cell Immunol. 151, 451-459. Datta, S. R., Dudek, H., Tao, X., Masters, S., Fu, H., Gotoh, Y., and Greenberg, M. E. (1997). Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 91,231-241. Deas, O., Dumont, C , MacFarlane, M., Rouleau, M., Hebib, C , Harper, F., Hirsch, F., Charpentier, B., Cohen, G. M., and Senik, A. (1998). Caspase-independent cell death induced by anti-CD2 or staurosporine in activated human peripheral T lymphocytes. J.Immunol. 161, 3375-3383. del Peso, L., Gonzalez-Garcia, M., Page, C , Herrera, R., and Nunez, G. (1997). Interleukin-3-induced phosphorylation of BAD through the protein kinase Akt. Science 278, 687-689. Demers, B., Simor, A. E., Vellend, H., Schlievert, P. M., Byrne, S., Jamieson, F., Walmsley, S., and Low, D. E. (1993). Severe invasive group A streptococcal infections in Ontario, Canada: 1987-1991. Clin.Infect.Dis. 16, 792-800. Dennig, D., Lacerda, J., Yan, Y., Gasparetto, C , and O'Reilly, R. J. (1994). ICAM-1 (CD54) expression on B lymphocytes is associated with their costimulatory function and can be increased by coactivation with IL-1 and IL-7. Cell Immunol. 156, 414-423. Deresiewicz, R. L., Flaxenburg, J. A., Chan, M., Finberg, R. W., and Kasper, D. L. (1994a). Intracellular expression of toxic shock syndrome toxin 1 in Saccharomyces cerevisiae. Infect.Immun. 62, 2202-2207. 184 Deresiewicz, R. L., Woo, J., Chan, M., Finberg, R. W., and Kasper, D. L. (1994b). Mutations affecting the activity of toxic shock syndrome toxin-1. Biochemistry 33, 12844-12851. Desagher, S., Osen-Sand, A., Nichols, A., Eskes, R., Montessuit, S., Lauper, S., Maundrell, K., Antonsson, B., and Martinou, J. C. (1999). Bid-induced conformational change of Bax is responsible for mitochondrial cytochrome c release during apoptosis. J. Cell Biol. 144, 891-901. Dijkers, P. F., Medema, R. PL, Lammers, J. W., Koenderman, L., and Coffer, P. J. (2000). Expression of the pro-apoptotic Bcl-2 family member Bim is regulated by the forkhead transcription factor FKHR-L1. Curr.Biol. 10, 1201-1204. Drynda, A., Konig, B., Bonventre, P. F., and Konig, W. (1995). Role of a carboxy-terminal site of toxic shock syndrome toxin 1 in eliciting immune responses of human peripheral blood mononuclear cells. Infect.Immun. 63, 1095-1101. Du, C , Fang, M., Li, Y., Li, L., and Wang, X. (2000). Smac, a mitochondrial protein that promotes cytochrome c-dependent caspase activation by eliminating IAP inhibition. Cell 102, 33-42. Earhart, C. A., Mitchell, D. T., Murray, D. L., Pinheiro, D. M., Matsumura, M., Schlievert, P. M., and Ohlendorf, D. H. (1998). Structures of five mutants of toxic shock syndrome toxin-1 with reduced biological activity. Biochemistry 37, 7194-7202. Earnshaw, W. C , Martins, L. M., and Kaufmann, S. H. (1999). Mammalian caspases: structure, activation, substrates, and functions during apoptosis. Annu.Rev Biochem. 68, 383-424. Enari, M., Sakahira, FL, Yokoyama, FL, Okawa, K., Iwamatsu, A., and Nagata, S. (1998). A caspase-activated DNase that degrades DNA during apoptosis, and its inhibitor ICAD. Nature 391, 43-50. Eskes, R., Desagher, S., Antonsson, B., and Martinou, J. C. (2000). Bid induces the oligomerization and insertion of Bax into the outer mitochondrial membrane. Mol. Cell Biol. 20, 929-935. Ettinger, R., Panka, D. J., Wang, J. K., Stanger, B. Z., Ju, S. T., and Marshak-Rothstein, A. (1995). Fas ligand-mediated cytotoxicity is directly responsible for apoptosis of normal CD4+ T cells responding to a bacterial superantigen. J.Immunol. 154, 4302-4308. Fadok, V. A., Savill, J. S., Haslett, C , Bratton, D. L., Doherty, D. E., Campbell, P. A., and Henson, P. M. (1992a). Different populations of macrophages use either the vitronectin receptor or the phosphatidylserine receptor to recognize and remove apoptotic cells. J.Immunol 149, 4029-4035. Fadok, V. A., Voelker, D. R., Campbell, P. A., Cohen, J. J., Bratton, D. L., and Henson, P. M. (1992b). Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J.Immunol 148, 2207-2216. Ferraro-Peyret, C , Quemeneur, L., Flacher, M., Revillard, J. P., and Genestier, L. (2002). Caspase-independent phosphatidylserine exposure during apoptosis of primary T lymphocytes. J.Immunol. 169, 4805-4810. 185 Fields, B. A., Malchiodi, E. L., Li, FL, Ysern, X., Stauffacher, C. V., Schlievert, P. M., Karjalainen, K., and Mariuzza, R. A. (1996). Crystal structure of a T-cell receptor beta-chain complex ed with a superantigen. Nature 384, 188-192. Froelich, C. J., Dixit, V. M., and Yang, X. (1998). Lymphocyte granule-mediated apoptosis: matters of viral mimicry and deadly proteases. Immunol Today 19, 30-36. Gollapudi, S., McCormick, M. J., and Gupta, S. (2003). Changes in mitochondrial membrane potential and mitochondrial mass occur independent of the activation of caspase-8 and caspase-3 during CD95-mediated apoptosis in peripheral blood T cells. Int.J.Oncol. 22, 597-600. Gonzalo, J. A., Moreno, d. A., I, Ales-Martinez, J. E., Martinez, C , and Kroemer, G. (1992). Expansion and clonal deletion of peripheral T cells induced by bacterial superantigen is independent of the interleukin-2 pathway. Eur. J.Immunol. 22, 1007-1011. Goping, I. S., Barry, M., Liston, P., Sawchuk, T., Constantinescu, G., Michalak, K. M., Shostak, I., Roberts, D. L., Hunter, A. M., Korneluk, R., and Bleackley, R. C. (2003). Granzyme B-induced apoptosis requires both direct caspase activation and relief of caspase inhibition. Immunity. 18, 355-365. Gorak-Stolinska, P., Kemeny, D. M., and Noble, A. (2002). Activation-induced cell death in human T cells is a suicidal process regulated by cell density but superantigen induces T cell fratricide. Cell Immunol. 219, 98-107. Gorczyca, W., Bigman, K., Mittelman, A., Ahmed, T., Gong, J., Melamed, M. R., and Darzynkiewicz, Z. (1993 a). Induction of DNA strand breaks associated with apoptosis during treatment of leukemias. Leukemia 7, 659-670. Gorczyca, W., Gong, J., and Darzynkiewicz, Z. (1993b). Detection of DNA strand breaks in individual apoptotic cells by the in situ terminal deoxynucleotidyl transferase and nick translation assays. Cancer Res. 53, 1945-1951. Gougeon, M. L. (2003). Apoptosis as an HIV strategy to escape immune attack. Nat.Rev.Immunol. 3, 392-404. Grasl-Kraupp, B., Ruttkay-Nedecky, B., Koudelka, H., Bukowska, K., Bursch, W., and Schulte-Hermann, R. (1995). In situ detection of fragmented DNA (TUNEL assay) fails to discriminate among apoptosis, necrosis, and autolytic cell death: a cautionary note. Hepatology 21, 1465-1468. Guo, Y., Srinivasula, S. M., Druilhe, A., Fernandes-Alnemri, T., and Alnemri, E. S. (2002). Caspase-2 induces apoptosis by releasing proapoptotic proteins from mitochondria. J.Biol. Chem. 277, 13430-13437. Hajjeh, R. A., Reingold, A., Weil, A., Shutt, K., Schuchat, A., and Perkins, B. A. (1999). Toxic shock syndrome in the United States: surveillance update, 1979 1996. Emerg.Infect.Dis. 5, 807-810. Hamad, A. R., Marrack, P., and Kappler, J. W. (1997). Transcytosis of staphylococcal superantigen toxins. J.Exp.Med. 185, 1447-1454. 186 Henderson, S., Huen, D., Rowe, M., Dawson, C , Johnson, G., and Rickinson, A. (1993). Epstein-Barr virus-coded BHRF1 protein, a viral homologue of Bcl-2, protects human B cells from programmed cell death. Proc.Natl.Acad.Sci.U.S.A 90, 8479-8483. Hengartner, M. O. (2000). The biochemistry of apoptosis. Nature 407, 770-776. Hildeman, D. A., Mitchell, T., Teague, T. K., Henson, P., Day, B. J., Kappler, J., and Marrack, P. C. (1999). Reactive oxygen species regulate activation-induced T cell apoptosis. Immunity. 10, 735-744. Hildeman, D. A., Zhu, Y., Mitchell, T. C , Bouillet, P., Strasser, A., Kappler, J., and Marrack, P. (2002a). Activated T cell death in vivo mediated by proapoptotic bcl-2 family member bim. Immunity. 16, 759-767'. Hildeman, D. A., Zhu, Y., Mitchell, T. C., Kappler, J., and Marrack, P. (2002b). Molecular mechanisms of activated T cell death in vivo. Curr.Opin.Immunol. 14, 354-359. Hinton, H. J. and Welham, M. J. (1999). Cytokine-induced protein kinase B activation and Bad phosphorylation do not correlate with cell survival of hemopoietic cells. J.Immunol 162, 7002-7009. Hirata, H., Takahashi, A., Kobayashi, S., Yonehara, S., Sawai, H., Okazaki, T., Yamamoto, K., and Sasada, M. (1998). Caspases are activated in a branched protease cascade and control distinct downstream processes in Fas-induced apoptosis. J.Exp.Med. 187, 587-600. Hofer, M. F., Newell, K., Duke, R. C , Schlievert, P. M., Freed, J. H., and Leung, D. Y. (1996). Differential effects of staphylococcal toxic shock syndrome toxin-1 on B cell apoptosis. Proc.Natl.Acad.Sci.U.S.A 93, 5425-5430. Hsu, Y. T. and Youle, R. J. (1997). Nonionic detergents induce dimerization among members of the Bcl-2 family. J.Biol.Chem. 272, 13829-13834. Hsu, Y. T. and Youle, R. J. (1998). Bax in murine thymus is a soluble monomeric protein that displays differential detergent-induced conformations. J.Biol.Chem. 273, 10777-10783. Hu, W. G. and Zhu, X. H. (1996). V beta specificity of superantigen TSST-1 plus CD28 costimulation without APCs. Immunol.Invest 25, 405-411. Hu, W. G., Zhu, X. H., Wu, Y. Z., and Jia, Z. C. (1998). Localization of a T-cell epitope of superantigen toxic shock syndrome toxin 1 to residues 125 to 158. Infect.Immun. 66, 4971-4975. Huang, H., Joazeiro, C. A., Bonfoco, E., Kamada, S., Leverson, J. D., and Hunter, T. (2000). The inhibitor of apoptosis, cIAP2, functions as a ubiquitin-protein ligase and promotes in vitro monoubiquitination of caspases 3 and 7. J.Biol.Chem. 275, 26661-26664. Irmler, M., Thome, M., Hahne, M., Schneider, P., Hofmann, K., Steiner, V., Bodmer, J. L., Schroter, M., Burns, K., Mattmann, C , Rimoldi, D., French, L. E., and Tschopp, J. (1997). Inhibition of death receptor signals by cellular FLIP. Nature 388, 190-195. Jaattela, M. and Tschopp, J. (2003). Caspase-independent cell death in T lymphocytes. Nat.Immunol. 4, 416-423. 187 Jardetzky, T. S., Brown, J. E L , Gorga, J. C , Stern, L. J., Urban, R. G., Chi, Y. L, Stauffacher, C , Strominger, J. L., and Wiley, D. C. (1994). Three-dimensional structure of a human class II histocompatibility molecule complex ed with superantigen. Nature 368, 711-718. Jellinger, K. A. (2001). Cell death mechanisms in neurodegeneration. J.Cell Mol.Med. 5 , 1-17. Jonas, D., Walev, I., Berger, T., Liebetrau, M., Palmer, M., and Bhakdi, S. (1994). Novel path to apoptosis: small transmembrane pores created by staphylococcal alpha-toxin in T lymphocytes evoke internucleosomal DNA degradation. Infect.Immun. 62, 1304-1312. Kabelitz, D. and Wesselborg, S. (1992). Life and death of a superantigen-reactive human CD4+ T cell clone: staphylococcal enterotoxins induce death by apoptosis but simultaneously trigger a proliferative response in the presence of HLA- DR+ antigen-presenting cells. Int.Immunol. 4, 1381-1388. Kain, K. C , Schulzer, M., and Chow, A. W. (1993). Clinical spectrum of nonmenstrual toxic shock syndrome (TSS): comparison with menstrual TSS by multivariate discriminant analyses. Clin.Infect.Dis. 16, 100-106. Kappler, J., Kotzin, B., Herron, L., Gelfand, E. W., Bigler, R. D., Boylston, A., Carrel, S., Posnett, D. N., Choi, Y., and Marrack, P. (1989). V beta-specific stimulation of human T cells by staphylococcal toxins. Science 244, 811-813. Kayagaki, N., Kawasaki, A., Ebata, T., Ohmoto, FL, Ikeda, S., Inoue, S., Yoshino, K., Okumura, K., and Yagita, H. (1995). Metalloproteinase-mediated release of human Fas ligand. J.Exp.Med. 182, 1777-1783. Kerr, J. F. (2002). History of the events leading to the formulation of the apoptosis concept. Toxicology 181-182, 471-474. Kerr, J. F., Wyllie, A. H., and Currie, A. R. (1972). Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br.J.Cancer 26, 239-257. Kim, J., Urban, R. G., Strominger, J. L., and Wiley, D. C. (1994). Toxic shock syndrome toxin-1 complexed with a class II major histocompatibility molecule HLA-DR1. Science 266, 1870-1874. Kischkel, F. C , Hellbardt, S., Behrmann, I., Germer, M., Pawlita, M., Krammer, P. H., and Peter, M. E. (1995). Cytotoxicity-dependent APO-1 (Fas/CD95)-associated proteins form a death-inducing signaling complex (DISC) with the receptor. EMBO J. 14, 5579-5588. Kischkel, F. C , Lawrence, D. A., Tinel, A., LeBlanc, H., Virmani, A., Schow, P., Gazdar, A., Blenis, J., Arnott, D., and Ashkenazi, A. (2001). Death receptor recruitment of endogenous caspase-10 and apoptosis initiation in the absence of caspase-8. J.Biol. Chem. 276, 46639-46646. Krakauer, T. (1996). Detection of adhesion of superantigen-activated T lymphocytes to human endothelial cells by ELISA. J.Immunoassay 17, 1-12. Krakauer, T. (1998). Interleukin-8 production by human monocytic cells in response to staphylococcal exotoxins is direct and independent of interleukin-1 and tumor necrosis factor-alpha. J.Infect.Dis. 178, 573-577. 188 Krakauer, T. (1999). Immune response to staphylococcal superantigens. Immunol.Res. 2 0 , 163-173. Kreisel, D., Sankaran, D., Wells, A. D., and Turka, L. A. (2002). Interleukin-2-mediated survival and proliferative signals are uncoupled in T lymphocytes that fail to divide after activation. Am.J. Transplant. 2 , 120-128. Krueger, A., Fas, S. C , Baumann, S., and Krammer, P. H. (2003). The role of CD95 in the regulation of peripheral T-cell apoptosis. Immunol.Rev. 1 9 3 , 58-69. Krueger, A., Schmitz, I., Baumann, S., Krammer, P. PL, and Kirchhoff, S. (2001). Cellular FLICE-inhibitory protein splice variants inhibit different steps of caspase-8 activation at the CD95 death-inducing signaling complex. J.Biol.Chem. 2 7 6 , 20633-20640. Kum, W. W., Cameron, S. B., Hung, R. W., Kalyan, S., and Chow, A. W. (2001). Temporal sequence and kinetics of proinflammatory and anti-inflammatory cytokine secretion induced by toxic shock syndrome toxin 1 in human peripheral blood mononuclear cells. Infect.Immun. 6 9 , 7544-7549. Kum, W. W., Hung, R. W., Cameron, S. B., and Chow, A. W. (2002). Temporal sequence and functional implications of V beta-specific T cell receptor down-regulation and costimulatory molecule expression following in vitro stimulation with the staphylococcal superantigen Toxic shock syndrome toxin-1. J.Infect.Dis. 1 8 5 , 555-560. Kum, W. W., Laupland, K. B., See, R. H., and Chow, A. W. (1993). Improved purification and biologic activities of staphylococcal toxic shock syndrome toxin 1. J.Clin.Microbiol. 3 1 , 2654-2660. Kum, W. W., Wood, J. A., and Chow, A. W. (1996). A mutation at glycine residue 31 of toxic shock syndrome toxin-1 defines a functional site critical for major histocompatibility complex class II binding and superantigenic activity. J.Infect.Dis. 1 7 4 , 1261-1270. Kushnaryov, V. M., MacDonald, H. S., Reiser, R., and Bergdoll, M. S. (1984). Staphylococcal toxic shock toxin specifically binds to cultured human epithelial cells and is rapidly internalized. Infect.Immun. 4 5 , 566-571. Lamy, L., Ticchioni, M., Rouquette-Jazdanian, A. K., Samson, M., Deckert, M., Greenberg, A. H., and Bernard, A. (2003). CD47 and the 19 kDa Interacting Protein-3 (BNIP3) in T Cell Apoptosis. J.Biol.Chem. 2 7 8 , 23915-23921. Lassus, P., Opitz-Araya, X., and Lazebnik, Y. (2002). Requirement for caspase-2 in stress-induced apoptosis before mitochondrial permeabilization. Science 2 9 7 , 1352-1354. Lecoeur, H., Ledru, E., Prevost, M. C , and Gougeon, M. L. (1997). Strategies for phenotyping apoptotic peripheral human lymphocytes comparing ISNT, annexin-V and 7-AAD cytofluorometric staining methods. J.Immunol.Methods 2 0 9 , 111-123. Lee, P. K., Vercellotti, G. M., Deringer, J. R., and Schlievert, P. M. (1991). Effects of staphylococcal toxic shock syndrome toxin 1 on aortic endothelial cells. J.Infect.Dis. 1 6 4 , 711-719. 189 Leist, M. and Jaattela, M. (2001). Four deaths and a funeral: from caspases to alternative mechanisms. Nat.Rev.Mol. Cell Biol. 2, 589-598. Lendahl, U. and Orrenius, S. (2002). [Sydney Brenner, Robert Horvitz and John Sulston. Winners of the 2002 Nobel Prize in medicine or physiology. Genetic regulation of organ development and programmed cell death]. Lakartidningen 99, 4026-4032. Lens, S. M., Kataoka, T., Former, K. A., Tinel, A., Ferrero, I., MacDonald, R. FL, Hahne, M., Beermann, F., Attinger, A., Orbea, H. A., Budd, R. C., and Tschopp, J. (2002). The caspase 8 inhibitor c-FLIP(L) modulates T-cell receptor-induced proliferation but not activation-induced cell death of lymphocytes. Mol. Cell Biol. 22, 5419-5433. Lesage, S., Steffi A. M., Philippoussis, F., Page, M., Trop, S., Mateo, V., and Hugo, P. (1997). CD4+ CD8+ thymocytes are preferentially induced to die following CD45 cross-linking, through a novel apoptotic pathway. J.Immunol. 159, 4762-4771. Lester, M. R., Hofer, M. F., Gately, M., Trumble, A., and Leung, D. Y. (1995). Down-regulating effects of IL-4 and IL-10 on the IFN-garnma response in atopic dermatitis. J.Immunol. 154, 6174-6181. Letai, A., Bassik, M. C., Walensky, L. D., Sorcinelli, M. D., Weiler, S., and Korsmeyer, S. J. (2002). Distinct BH3 domains either sensitize or activate mitochondrial apoptosis, serving as prototype cancer therapeutics. Cancer Cell 2, 183-192. Leung, D. Y., Meissner, C., Fulton, D., and Schlievert, P. M. (1995). The potential role of bacterial superantigens in the pathogenesis of Kawasaki syndrome. J.Clin.Immunol. 15, 11S-17S. Leung, D. Y., Meissner, H. C., Shulman, S. T., Mason, W. H., Gerber, M. A., Glode, M. P., Myones, B. L., Wheeler, J. G., Ruthazer, R., and Schlievert, P. M. (2002). Prevalence of superantigen-secreting bacteria in patients with Kawasaki disease. J.Pediatr. 140, 742-746. Li, H., Llera, A., Tsuchiya, D., Leder, L., Ysern, X., Schlievert, P. M., Karjalainen, K., and Mariuzza, R. A. (1998a). Three-dimensional structure of the complex between a T cell receptor beta chain and the superantigen staphylococcal enterotoxin B. Immunity. 9, 807-816. Li, H., Zhu, H., Xu, C. J., and Yuan, J. (1998b). Cleavage of BID by caspase 8 mediates the mitochondrial damage in the Fas pathway of apoptosis. Cell 94, 491-501. Li, L. Y., Luo, X., and Wang, X. (2001a). Endonuclease G is an apoptotic DNase when released from mitochondria. Nature 412, 95-99. Li, X. C , Demirci, G., Ferrari-Lacraz, S., Groves, C , Coyle, A., Malek, T. R., and Strom, T. B. (2001b). IL-15 and IL-2: a matter of life and death for T cells in vivo. Nat.Med. 7, 114-118. Lindsten, T., Ross, A. J., King, A., Zong, W. X., Rathmell, J. C , Shiels, H. A., Ulrich, E., Waymire, K. G., Mahar, P., Frauwirth, K., Chen, Y., Wei, M., Eng, V. M., Adelman, D. M., Simon, M. C , Ma, A., Golden, J. A., Evan, G., Korsmeyer, S. J., MacGregor, G. R., and Thompson, C. B. (2000). The combined functions of proapoptotic Bcl-2 family members bak and bax are essential for normal development of multiple tissues. Mol.Cell 6, 1389-1399. 190 Liu, X., Kim, C. N., Yang, J., Jemmerson, R., and Wang, X. (1996). Induction of apoptotic program in cell-free extracts: requirement for dATP and cytochrome c. Cell 86, 147-157. Luo, X., Budihardjo, I., Zou, FL, Slaughter, C , and Wang, X. (1998). Bid, a Bcl2 interacting protein, mediates cytochrome c release from mitochondria in response to activation of cell surface death receptors. Cell 94, 481-490. Makida, R., Hofer, M. F., Takase, K., Cambier, J. C , and Leung, D. Y. (1996). Bacterial superantigens induce V beta-specific T cell receptor internalization. Mol.Immunol. 33, 891-900. Malchiodi, E. L., Eisenstein, E., Fields, B. A., Ohlendorf, D. PL, Schlievert, P. M., Karjalainen, K., and Mariuzza, R. A. (1995). Superantigen binding to a T cell receptor beta chain of known three-dimensional structure. J.Exp.Med. 182, 1833-1845. Marrack, P. and Kappler, J. (1990). The staphylococcal enterotoxins and their relatives. Science 248, 705-711. Marsden, V. S. and Strasser, A. (2003). CONTROL OF APOPTOSIS IN THE IMMUNE SYSTEM: Bcl-2, BH3-Only Proteins and More. Annu.Rev Immunol 21, 71-105. Marx, J. (2002). Nobel Prize in Physiology or Medicine. Tiny worm takes a star turn. Science 298, 526. McCormick, J. K., Tripp, T. J., Llera, A. S., Sundberg, E. J., Dinges, M. M., Mariuzza, R. A., and Schlievert, P. M. (2003). Functional Analysis of the TCR Binding Domain of Toxic Shock Syndrome Toxin-1 Predicts Further Diversity in MHC Class II/Superantigen/TCR Ternary Complexes. J.Immunol. 171, 1385-1392. McCormick, J. K., Yarwood, J. M., and Schlievert, P. M. (2001). Toxic shock syndrome and bacterial superantigens: an update. Annu.Rev.Microbiol. 5 5 , 77-104. McKallip, R. J., Do, Y., Fisher, M. T., Robertson, J. L., Nagarkatti, P. S., and Nagarkatti, M. (2002). Role of CD44 in activation-induced cell death: CD44-deficient mice exhibit enhanced T cell response to conventional and superantigens. Int.Immunol. 14, 1015-1026. Mehindate, K., al Daccak, R., Damdoumi, F., and Mourad, W. (1996). Synergistic effect between CD40 and class II signals overcome the requirement for class II dimerization in superantigen-induced cytokine gene expression. Eur. J.Immunol. 26, 2075-2080. Mogil, R. J., Radvanyi, L., Gonzalez-Quintial, R., Miller, R., Mills, G., Theofilopoulos, A. N., and Green, D. R. (1995). Fas (CD95) participates in peripheral T cell deletion and associated apoptosis in vivo. Int.Immunol. 7, 1451-1458. Monday, S. R., Vath, G. M., Ferens, W. A., Deobald, C , Rago, J. V., Gahr, P. J., Monie, D. D., Iandolo, J. J., Chapes, S. K., Davis, W. C , Ohlendorf, D. H., Schlievert, P. M., and Bohach, G. A. (1999). Unique superantigen activity of staphylococcal exfoliative toxins. J.Immunol. 162, 4550-4559. Morishima, N., Nakanishi, K., Takenouchi, H., Shibata, T., and Yasuhiko, Y. (2002). An endoplasmic reticulum stress-specific caspase cascade in apoptosis. Cytochrome c-independent activation of caspase-9 by caspase-12. J.Biol.Chem. 277, 34287-34294. 191 Moulding, D. A., Walter, C , Hart, C. A., and Edwards, S. W. (1999). Effects of staphylococcal enterotoxins on human neutrophil functions and apoptosis. Infect.Immun. 67, 2312-2318. Mountz, J. D., Baker, T. J., Borcherding, D. R., Bluethmann, H., Zhou, T., and Edwards, C. K., Ill (1995). Increased susceptibility of fas mutant MRL-lpr/lpr mice to staphylococcal enterotoxin B-induced septic shock. J.Immunol. 155, 4829-4837. Muraille, E., De Smedt, T., Thielemans, K., Urbain, J., Moser, M., and Leo, O. (1995). Activation of murine T cells by bacterial superantigens requires B7-mediated costimulation. Cell Immunol. 162, 315-320. Murphy, L. L., Mazanet, M. M., Taylor, A. C , Mestas, J., and Hughes, C. C. (1999). Single-cell analysis of costimulation by B cells, endothelial cells, and fibroblasts demonstrates heterogeneity in responses of CD4(+) memory T cells. Cell Immunol. 194, 150-161. Murzin, A. G. (1993). OB(oligonucleotide/oligosaccharide binding)-fold: common structural and functional solution for non-homologous sequences. EMBO J. 12, 861-867. Muzio, M., Chinnaiyan, A. M., Kischkel, F. C., O'Rourke, K., Shevchenko, A., Ni, J., Scaffidi, C., Bretz, J. D., Zhang, M., Gentz, R., Mann, M., Krammer, P. H., Peter, M. E., and Dixit, V. M. (1996). FLICE, a novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95 (Fas/APO-1) death-inducing signaling complex. Cell 85, 817-827. Nakano, K. and Vousden, K. H. (2001). PUMA, a novel proapoptotic gene, is induced by p53. Mol.Celll, 683-694. Nicholson, D. W. (1999). Caspase structure, proteolytic substrates, and function during apoptotic cell death. Cell Death.Differ. 6, 1028-1042. Nicoletti, I., Migliorati, G., Pagliacci, M. C , Grignani, F., and Riccardi, C. (1991). A rapid and simple method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry. J.Immunol.Methods 139, 271-279. Ochi, A., Yuh, K., and Migita, K. (1993). Not every superantigen induces tolerance in vivo. Semin.Immunol. 5, 57-63. Oda, E., Ohki, R., Murasawa, H., Nemoto, J., Shibue, T., Yamashita, T., Tokino, T., Taniguchi, T., and Tanaka, N. (2000). Noxa, a BH3-only member of the Bcl-2 family and candidate mediator of p53-induced apoptosis. Science 288, 1053-1058. Okkenhaug, K. and Vanhaesebroeck, B. (2003). PI3K in lymphocyte development, differentiation and activation. ISS at.Rev.Immunol. 3, 317-330. Ozes, O. N., Mayo, L. D., Gustin, J. A., Pfeffer, S. R., Pfeffer, L. M., and Dormer, D. B. (1999). NF-kappaB activation by tumour necrosis factor requires the Akt serine-threonine kinase. Nature 401, 82-85. Panka, D. J., Mano, T., Suhara, T., Walsh, K., and Mier, J. W. (2001). Phosphatidylinositol 3-kinase/Akt activity regulates c-FLIP expression in tumor cells. J.Biol.Chem. 276, 6893-6896. 192 Papageorgiou, A. C , Brehm, R. D., Leonidas, D. D., Tranter, H. S., and Acharya, K. R. (1996). The refined crystal structure of toxic shock syndrome toxin-1 at 2.07 A resolution. J.Mol.Biol. 260, 553-569. Perlman, H., Pagliari, L. J., Nguyen, N., Bradley, K., Liu, H., and Pope, R. M. (2001). The Fas-FasL death receptor and PI3K pathways independently regulate monocyte homeostasis. Eur. J.Immunol. 31, 2421-2430. Pettersen, R. D., Bernard, G., Olafsen, M. K., Pourtein, M., and Lie, S. O. (2001). CD99 signals caspase-independent T cell death. J.Immunol. 166, 4931-4942. Pettersen, R. D., Gaudernack, G., Olafsen, M. K., Lie, S. O., and Hestdal, K. (1998). The TCR-binding region of the HLA class I alpha2 domain signals rapid Fas-independent cell death: a direct pathway for T cell-mediated killing of target cells? J.Immunol. 160, 4343-4352. Pettersen, R. D., Hestdal, K., Olafsen, M. K., Lie, S. O., and Lindberg, F. P. (1999). CD47 signals T cell death. J.Immunol. 162, 7031-7040. Pinkoski, M. J., Droin, N. M., and Green, D. R. (2002). Tumor necrosis factor alpha up-regulates non-lymphoid Fas-ligand following superantigen-induced peripheral lymphocyte activation. J.Biol.Chem. 277, 42380-42385. Pinkoski, M. J., Waterhouse, N. J., Heibein, J. A., Wolf, B. B., Kuwana, T., Goldstein, J. C., Newmeyer, D. D., Bleackley, R. C., and Green, D. R. (2001). Granzyme B-mediated apoptosis proceeds predominantly through a Bcl-2-inhibitable mitochondrial pathway. J.Biol. Chem. 276, 12060-12067. Piano, L. R., Gutman, D. M., Woischnik, M., and Collins, C. M. (2000). Recombinant Staphylococcus aureus exfoliative toxins are not bacterial superantigens. Infect.Immun. 68, 3048-3052. Prasad, G. S., Earhart, C. A., Murray, D. L., Novick, R. P., Schlievert, P. M., and Ohlendorf, D. H. (1993). Structure of toxic shock syndrome toxin 1. Biochemistry 32, 13761-13766. Prasad, G. S., Radhakrishnan, R., Mitchell, D. T., Earhart, C. A., Dinges, M. M., Cook, W. J., Schlievert, P. M., and Ohlendorf, D. H. (1997). Refined structures of three crystal forms of toxic shock syndrome toxin-1 and of a tetramutant with reduced activity. Protein Sci. 6, 1220-1227. Putcha, G. V., Harris, C. A., Moulder, K. L., Easton, R. M., Thompson, C. B., and Johnson, E. M., Jr. (2002). Intrinsic and extrinsic pathway signaling during neuronal apoptosis: lessons from the analysis of mutant mice. J.Cell Biol. 157, 441-453. Puthalakath, H., Huang, D. C , O'Reilly, L. A., King, S. M., and Strasser, A. (1999). The proapoptotic activity of the Bcl-2 family member Bim is regulated by interaction with the dynein motor complex. Mol. Cell 3, 287-296. Puthalakath, H. and Strasser, A. (2002). Keeping killers on a tight leash: transcriptional and post-translational control of the pro-apoptotic activity of BH3-only proteins. Cell Death.Differ. 9, 505-512. 193 Rami, A. (2003). Ischemic neuronal death in the rat hippocampus: the calpain-calpastatin-caspase hypothesis. Neurobiol.Dis. 13, 75-88. Ramirez, M., Fernandez-Troy, N., Buxade, M., Casaroli-Marano, R. P., Benitez, D., Perez-Maldonado, C , and Espel, E. (1999). Wortmannin inhibits translation of tumor necrosis factor-alpha in superantigen-activated T cells. Int.Immunol. 11, 1479-1489. Rao, R. V., Castro-Obregon, S., Frankowski, H., Schuler, M., Stoka, V., del Rio, G., Bredesen, D. E., and Ellerby, H. M. (2002). Coupling endoplasmic reticulum stress to the cell death program. An Apaf-1 -independent intrinsic pathway. J.Biol.Chem. 277, 21836-21842. Rathmell, J. C , Lindsten, T., Zong, W. X., Cinalli, R. M., and Thompson, C. B. (2002). Deficiency in Bak and Bax perturbs thymic selection and lymphoid homeostasis. Nat.Immunol. 3, 932-939. Refaeli, Y., Van Parijs, L., London, C. A., Tschopp, J., and Abbas, A. K. (1998). Biochemical mechanisms of IL-2-regulated Fas-mediated T cell apoptosis. Immunity. 8, 615-623. Renno, T., Attinger, A., Locatelli, S., Bakker, T., Vacheron, S., and MacDonald, H. R. (1999). Cutting edge: apoptosis of superantigen-activated T cells occurs preferentially after a discrete number of cell divisions in vivo. J.Immunol. 162, 6312-6315. Rieux-Laucat, F., Fischer, A., and Deist, F. L. (2003). Cell-death signaling and human disease. Curr. Opin.Immunol 15, 325-331. Rink, L., Nicklas, W., Alvarez-Ossorio, L., Fagin, U., and Kirchner, H. (1997). Microbial superantigens stimulate T cells by the superantigen bridge and independently by a cytokine pathway. J.Interferon Cytokine Res. 17, 489-499. Rogers, T. J., Guan, L., and Zhang, L. (1995). Characterization of an alternative superantigen binding site expressed on a renal fibroblast cell line. Int.Immunol. 7, 1721-1727. Rogers, T. J. and Zhang, L. (1997). Structural basis for the interaction of superantigen with the alternative superantigen-binding receptor p85. Mol.Immunol. 34, 263-272. Rosten, P. M., Bartlett, K. FL, and Chow, A. W. (1987). Detection and quantitation of toxic shock syndrome toxin 1 in vitro and in vivo by noncompetitive enzyme-linked immunosorbent assay. J.Clin.Microbiol. 25, 327-332. Rothenberg, F., Fisher, S. A., and Watanabe, M. (2003). Sculpting the cardiac outflow tract. Birth Defects Res.Part C.Embryo.Today 69, 38-45. Roucou, X., Montessuit, S., Antonsson, B., and Martinou, J. C. (2002a). Bax oligomerization in mitochondrial membranes requires tBid (caspase-8-cleaved Bid) and a mitochondrial protein. BiochemJ. 368, 915-921. Roucou, X., Rostovtseva, T., Montessuit, S., Martinou, J. C , and Antonsson, B. (2002b). Bid induces cytochrome c-impermeable Bax channels in liposomes. BiochemJ. 363, 547-552. 194 Ruchaud, S., Korfali, N., Villa, P., Kottke, T. J., Dingwall, C , Kaufrnann, S. H., and Earnshaw, W. C. (2002). Caspase-6 gene disruption reveals a requirement for lamin A cleavage in apoptotic chromatin condensation. EMBO J. 21, 1967-1977. Sade, H. and Sarin, A. (2003). IL-7 inhibits dexamethasone-induced apoptosis via Akt/PKB in mature, peripheral T cells. Eur. J.Immunol 33, 913-919. Saha, B., Harlan, D. M., Lee, K. P., June, C. H., and Abe, R. (1996a). Protection against lethal toxic shock by targeted disruption of the CD28 gene. J.Exp.Med. 183, 2675-2680. Saha, B., Jaklic, B., Harlan, D. M., Gray, G. S., June, C. H., and Abe, R. (1996b). Toxic shock syndrome toxin-1-induced death is prevented by CTLA4Ig. J.Immunol. 157, 3869-3875. Salvesen, G. S. and Dixit, V. M. (1999). Caspase activation: the induced-proximity model. Proc.Natl.Acad.Sci.U.S.A 96, 10964-10967. Samejima, K., Tone, S., and Earnshaw, W. C. (2001). CAD/DFF40 nuclease is dispensable for high molecular weight DNA cleavage and stage I chromatin condensation in apoptosis. J.Biol.Chem. 276, 45427-45432. Savill, J. and Fadok, V. (2000). Corpse clearance defines the meaning of cell death. Nature 407, 784-788. Scaffidi, C , Fulda, S., Srinivasan, A., Friesen, C , Li, F., Tomaselli, K. J., Debatin, K. M., Krammer, P. H., and Peter, M. E. (1998). Two CD95 (APO-l/Fas) signaling pathways. EMBO J. 17, 1675-1687. Schaeffer, E. M. and Schwartzberg, P. L. (2000). Tec family kinases in lymphocyte signaling and function. Curr.Opin.Immunol. 12, 282-288. Schlievert, P. M. (1981). Staphylococcal scarlet fever: role of pyrogenic exotoxins. Infect.Immun. 31, 732-736. Schlievert, P. M. (1986). Staphylococcal enterotoxin B and toxic-shock syndrome toxin-1 are significantly associated with non-menstrual TSS. Lancet 1, 1149-1150. Schlievert, P. M., Jablonski, L. M., Roggiani, M., Sadler, I., Callantine, S., Mitchell, D. T., Ohlendorf, D. H., and Bohach, G. A. (2000). Pyrogenic toxin superantigen site specificity in toxic shock syndrome and food poisoning in animals. Infect.Immun. 68, 3630-3634. Schlievert, P. M., Shands, K. N., Dan, B. B., Schmid, G. P., and Nishimura, R. D. (1981). Identification and characterization of an exotoxin from Staphylococcus aureus associated with toxic-shock syndrome. J.Infect.Dis. 143, 509-516. Schmid, I., Krall, W. J., Uittenbogaart, C. H., Braun, J., and Giorgi, J. V. (1992). Dead cell discrimination with 7-amino-actinomycin D in combination with dual color immunofluorescence in single laser flow cytometry. Cytometry 13, 204-208. Schmidt, M., Fernandez, d. M., van der, H. A., Klompmaker, R., Kops, G. J., Lam, E. W., Burgering, B. M., and Medema, R. H. (2002). Cell cycle inhibition by FoxO forkhead transcription factors involves downregulation of cyclin D. Mol. Cell Biol. 22, 7842-7852. 195 Scholl, P. R., Trede, N., Chatila, T. A., and Geha, R. S. (1992). Role of protein tyrosine phosphorylation in monokine induction by the staphylococcal superantigen toxic shock syndrome toxin-1. J.Immunol. 148, 2237-2241. Schultz, D. R. and Harringto, W. J., Jr. (2003). Apoptosis: Programmed cell death at a molecular level. Semin.Arthritis Rheum. 32, 345-369. See, R. FL, Kum, W. W., Chang, A. PL, Goh, S. PL, and Chow, A. W. (1992). Induction of tumor necrosis factor and interleukin-1 by purified staphylococcal toxic shock syndrome toxin 1 requires the presence of both monocytes and T lymphocytes. Infect.Immun. 60, 2612-2618. Seward, R. J., von Haller, P. D., Aebersold, R., and Huber, B. T. (2003). Phosphorylation of the pro-apoptotic protein Bim in lymphocytes is associated with protection from apoptosis. Mol.Immunol 39, 983-993. Shi, Y. F., Sahai, B. M., and Green, D. R. (1989). Cyclosporin A inhibits activation-induced cell death in T-cell hybridomas and thymocytes. Nature 339, 625-626. Shupp, J. W., Jett, M., and Pontzer, C. H. (2002). Identification of a transcytosis epitope on staphylococcal enterotoxins. Infect.Immun. 70, 2178-2186. Siegel, R. M., Muppidi, J., Roberts, M., Porter, M., and Wu, Z. (2003). Death receptor signaling and autoimmunity. Immunol Res. 27, 499-512. Singh, B. R., Evenson, M. L., and Bergdoll, M. S. (1988a). Structural analysis of staphylococcal enterotoxins B and CI using circular dichroism and fluorescence spectroscopy. Biochemistry 27, 8735-8741. Singh, B. R., Kokan-Moore, N. P., and Bergdoll, M. S. (1988b). Molecular topography of toxic shock syndrome toxin 1 as revealed by spectroscopic studies. Biochemistry 27, 8730-8735. Slee, E. A., Adrain, C , and Martin, S. J. (1999a). Serial killers: ordering caspase activation events in apoptosis. Cell Death.Differ. 6, 1067-1074. Slee, E. A., Harte, M. T., Kluck, R. M., Wolf, B. B., Casiano, C. A., Newmeyer, D. D., Wang, H. G., Reed, J. C , Nicholson, D. W., Alnemri, E. S., Green, D. R., and Martin, S. J. (1999b). Ordering the cytochrome c-initiated caspase cascade: hierarchical activation of caspases-2, -3, -6, -7, -8, and -10 in a caspase-9-dependent manner. J.CellBiol. 144, 281-292. Sommer, K. W., Schamberger, C. J., Schmidt, G. E., Sasgary, S., and Cerni, C. (2003). Inhibitor of apoptosis protein (IAP) survivin is upregulated by oncogenic c-H-Ras. Oncogene 22, 4266-4280. Stahl, M., Dijkers, P. F., Kops, G. J., Lens, S. M., Coffer, P. J., Burgering, B. M., and Medema, R. H. (2002). The forkhead transcription factor FoxO regulates transcription of p27Kipl and Bim in response to IL-2. J.Immunol 168, 5024-5031. Strasser, A., Harris, A. W., Huang, D. C , Krammer, P. H., and Cory, S. (1995). Bcl-2 and Fas/APO-1 regulate distinct pathways to lymphocyte apoptosis. EMBO J. 14, 6136-6147. 196 Strasser, A., O'Connor, L., and Dixit, V. M. (2000). Apoptosis signaling. Annu.Rev Biochem. 69, 217-245. Strausbaugh, L. J. (1993). Toxic shock syndrome. Are you recognizing its changing presentations? Postgrad.Med. 94, 107-3, 117. Susin, S. A., Daugas, E., Ravagnan, L., Samejima, K., Zamzami, N., Loeffler, M., Costantini, P., Ferri, K. F., Irinopoulou, T., Prevost, M. C , Brothers, G., Mak, T. W., Penninger, J., Earnshaw, W. C , and Kroemer, G. (2000). Two distinct pathways leading to nuclear apoptosis. J.Exp.Med. 192, 571-580. Susin, S. A., Lorenzo, H. K., Zamzami, N., Marzo, I., Snow, B. E., Brothers, G. M., Mangion, J., Jacotot, E., Costantini, P., Loeffler, M., Larochette, N., Goodlett, D. R., Aebersold, R., Siderovski, D. P., Penninger, J. M., and Kroemer, G. (1999). Molecular characterization of mitochondrial apoptosis-inducing factor. Nature 397, 441-446. Sutton, V. R., Davis, J. E., Cancilla, M., Johnstone, R. W., Ruefli, A. A., Sedelies, K., Browne, K. A., and Trapani, J. A. (2000). Initiation of apoptosis by granzyme B requires direct cleavage of bid, but not direct granzyme B-mediated caspase activation. J.Exp.Med. 192, 1403-1414. Sutton, V. R., Wowk, M. E., Cancilla, M., and Trapani, J. A. (2003). Caspase activation by granzyme B is indirect, and caspase autoprocessing requires the release of proapoptotic mitochondrial factors. Immunity. 18, 319-329. Taimen, P. and Kallajoki, M. (2003). NuMA and nuclear lamins behave differently in Fas-mediated apoptosis. J.Cell Sci. 116, 571-583. Takemura, S., Kudo, T., Asano, R., Suzuki, M., Tsumoto, K., Sakurai, N., Katayose, Y., Kodama, H., Yoshida, FL, Ebara, S., Saeki, FL, Imai, K., Matsuno, S., and Kumagai, I. (2002). A mutated superantigen SEA D227A fusion diabody specific to MTJC1 and CD3 in targeted cancer immunotherapy for bile duct carcinoma. Cancer Immunol.Immunother. 51, 33-44. Tang, T. T., Dowbenko, D., Jackson, A., Toney, L., Lewin, D. A., Dent, A. L., and Lasky, L. A. (2002). The forkhead transcription factor AFX activates apoptosis by induction of the BCL-6 transcriptional repressor. J.Biol.Chem. 277, 14255-14265. Taskapan, M. O. and Kumar, P. (2000). Role of staphylococcal superantigens in atopic dermatitis: from colonization to inflammation. Ann.Allergy Asthma Immunol. 84, 3-10. Tewari, M., Quan, L. T., O'Rourke, K., Desnoyers, S., Zeng, Z., Beidler, D. R., Poirier, G. G., Salvesen, G. S., and Dixit, V. M. (1995). Yama/CPP32 beta, a mammalian homolog of CED-3, is a CrmA-inhibitable protease that cleaves the death substrate poly(ADP-ribose) polymerase. Cell 81, 801-809. Thome, M., Schneider, P., Hofmann, K., Fickenscher, H., Meinl, E., Neipel, F., Mattmann, C , Burns, K., Bodmer, J. L., Schroter, M., Scafiidi, C , Krammer, P. H., Peter, M. E., and Tschopp, J. (1997). Viral FLICE-inhibitory proteins (FLIPs) prevent apoptosis induced by death receptors. Nature 517-521. Thome, M. and Tschopp, J. (2001). Regulation of lymphocyte proliferation and death by FLIP. Nat.Rev Immunol 1, 50-58. 197 Thornberry, N. A. and Lazebnik, Y. (1998). Caspases: enemies within. Science 281, 1312-1316. Thornberry, N. A., Rano, T. A., Peterson, E. P., Rasper, D. M., Timkey, T., Garcia-Calvo, M., Houtzager, V. M., Nordstrom, P. A., Roy, S., Vaillancourt, J. P., Chapman, K. T., and Nicholson, D. W. (1997). A combinatorial approach defines specificities of members of the caspase family and granzyme B. Functional relationships established for key mediators of apoptosis. J.Biol.Chem. 272, 17907-17911. Tibbetts, M. D., Zheng, L., and Lenardo, M. J. (2003). The death effector domain protein family: regulators of cellular homeostasis. Nat.Immunol 4, 404-409. Toyota, PL, Yanase, N., Yoshimoto, T., Moriyama, M., Sudo, T., and Mizuguchi, J. (2003). Calpain-induced Bax-cleavage product is a more potent inducer of apoptotic cell death than wild-type Bax. Cancer Lett. 189, 221-230. Trede, N. S., Castigli, E., Geha, R. S., and Chatila, T. (1993a). Microbial superantigens induce NF-kappa B in the human monocytic cell line THP-1. J.Immunol. 150, 5604-5613. Trede, N. S., Chatila, T., and Geha, R. S. (1993b). Activator protein-1 (AP-1) is stimulated by microbial superantigens in human monocytic cells. Eur. J.Immunol. 23, 2129-2135. Trede, N. S., Morio, T., Scholl, P. R., Geha, R. S., and Chatila, T. (1994). Early activation events induced by the staphylococcal superantigen toxic shock syndrome toxin-1 in human peripheral blood monocytes. Clin.Immunol.Immunopathol. 70, 137-144. Vanhaesebroeck, B. and Alessi, D. R. (2000). The PI3K-PDK1 connection: more than just a road to PKB. BiochemJ. 346 Pt 3, 561-576. Vaux, D. L., Cory, S., and Adams, J. M. (1988). Bcl-2 gene promotes haemopoietic cell survival and cooperates with c-myc to immortalize pre-B cells. Nature 335, 440-442. Vella, A. T., Dow, S., Potter, T. A., Kappler, J., and Marrack, P. (1998). Cytokine-induced survival of activated T cells in vitro and in vivo. Proc.Natl.Acad.Sci.U.S.A 95, 3810-3815. Verhagen, A. M., Coulson, E. J., and Vaux, D. L. (2001). Inhibitor of apoptosis proteins and their relatives: IAPs and other BIRPs. Genome Biol. 2, REVIEWS3009. Vermes, I., Haanen, C , Steffens-Nakken, PL, and Reutelingsperger, C. (1995). A novel assay for apoptosis. Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled Annexin V. J.Immunol.Methods 184, 39-51. Wahl, C , Miethke, T., Heeg, K., and Wagner, H. (1993). Clonal deletion as direct consequence of an in vivo T cell response to bacterial superantigen. Eur. J.Immunol. 23, 1197-1200. Walsh, G. M., Sexton, D. W., and Blaylock, M. G. (2003). Corticosteroids, eosinophils and bronchial epithelial cells: new insights into the resolution of inflammation in asthma. J.Endocrinol. 178, 37-43. Wang, C. Y., Mayo, M. W., Korneluk, R. G., Goeddel, D. V., and Baldwin, A. S., Jr. (1998). NF-kappaB antiapoptosis: induction of TRAF1 and TRAF2 and c-IAPl and C-IAP2 to suppress caspase-8 activation. Science 281, 1680-1683. 198 Wang, K. K. (2000). Calpain and caspase: can you tell the difference? Trends Neurosci. 23, 20-26. Ward, S. G. and Cantrell, D. A. (2001). Phosphoinositide 3-kinases in T lymphocyte activation. Curr.Opin.Immunol. 13, 332-338. Weber, A. K., Wahn, U., and Renz, H. (2000). Superantigen-induced T cell death by apoptosis: analysis on a single cell level and effect of EFN-gamma and IL-4 treatment. Int.Arch.Allergy Immunol. 121, 215-223. Wehner, J. and Neuber, K. (2001). Staphylococcus aureus enterotoxins induce histamine and leukotriene release in patients with atopic eczema. Br. J.Dermatol. 145, 302-305. Wolter, K. G., Hsu, Y. T., Smith, C. L., Nechushtan, A., Xi, X. G., and Youle, R. J. (1997). Movement of Bax from the cytosol to mitochondria during apoptosis. J.Cell Biol. 139, 1281-1292. Wyllie, A. H., Kerr, J. F., and Currie, A. R. (1980). Cell death: the significance of apoptosis. Int.Rev.Cytol. 68, 251-306. Xiao, C. W., Asselin, E., and Tsang, B. K. (2002). Nuclear factor kappaB-mediated induction of Flice-like inhibitory protein prevents tumor necrosis factor alpha-induced apoptosis in rat granulosa cells. Biol.Reprod. 67, 436-441. Yamaguchi, H. and Wang, H. G. (2001). The protein kinase PKB/Akt regulates cell survival and apoptosis by inhibiting Bax conformational change. Oncogene 20, 7779-7786. Yang, Q. H., Church-Hajduk, R., Ren, J., Newton, M. L., and Du, C. (2003). Omi/HtrA2 catalytic cleavage of inhibitor of apoptosis (IAP) irreversibly inactivates IAPs and facilitates caspase activity in apoptosis. Genes Dev. 17, 1487-1496. Yang, Y., Kim, D., and Fathman, C. G. (1998). Regulation of programmed cell death following T cell activation in vivo. Int.Immunol. 10, 175-183. Yarwood, J. M., Leung, D. Y., and Schlievert, P. M. (2000). Evidence for the involvement of bacterial superantigens in psoriasis, atopic dermatitis, and Kawasaki syndrome. FEMS Microbiol.Lett. 192, 1-7. Yasuda, M., Umemura, S., Osamura, R. Y., Kenjo, T., and Tsutsumi, Y. (1995). Apoptotic cells in the human endometrium and placental villi: pitfalls in applying the TUNEL method. Arch.Histol.Cytol. 58, 185-190. Yuan, J., Shaham, S., Ledoux, S., Ellis, H. M., and Horvitz, H. R. (1993). The C. elegans cell death gene ced-3 encodes a protein similar to mammalian interleukin-1 beta-converting enzyme. Cell 75, 641-652. Zha, J., Harada, H., Yang, E., Jockel, J., and Korsmeyer, S. J. (1996). Serine phosphorylation of death agonist BAD in response to survival factor results in binding to 14-3-3 not BCL-X(L). Cell 87, 619-628. 199 Zhao, Y. X., Holmdahl, R., and Tarkowski, A. (1997). Induction of selective anergy by toxic shock syndrome toxin-1 in CD8+ T cells. Immunology 92, 188-193. Zheng, Z., Lee, J. E., and Yenari, M. A. (2003). Stroke: molecular mechanisms and potential targets for treatment. Curr.Mol.Med. 3, 361-372. Zollner, T. M., Munk, M. E., Keller, T., Nuber, V., Boehncke, W. PL, Kaufmann, S. H., Duijvestijn, A. M., Sterry, W., and Kaufmann, R. (1996). The superantigen exfoliative toxin induces cutaneous lymphocyte-associated antigen expression in peripheral human T lymphocytes. Immunol.Lett. 49 , 111-116. Zorgani, A., Essery, S. D., Madani, O. A., Bentley, A. J., James, V. S., MacKenzie, D. A., Keeling, J. W., Rambaud, C , Hilton, J., Blackwell, C. C , Weir, D. M., and Busuttil, A. (1999). Detection of pyrogenic toxins of Staphylococcus aureus in sudden infant death syndrome. FEMS Immunol.Med.Microbiol. 25 , 103-108. Zou, H., Li, Y., Liu, X., and Wang, X. (1999). An APAF-1.cytochrome c multimeric complex is a functional apoptosome that activates procaspase-9. J.Biol. Chem. 274, 11549-11556. 200 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items