UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Caspases in olfactory neuron aptosis Cowan, Catherine M. 2003

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

Item Metadata

Download

Media
831-ubc_2003-859517.pdf [ 8.59MB ]
Metadata
JSON: 831-1.0091148.json
JSON-LD: 831-1.0091148-ld.json
RDF/XML (Pretty): 831-1.0091148-rdf.xml
RDF/JSON: 831-1.0091148-rdf.json
Turtle: 831-1.0091148-turtle.txt
N-Triples: 831-1.0091148-rdf-ntriples.txt
Original Record: 831-1.0091148-source.json
Full Text
831-1.0091148-fulltext.txt
Citation
831-1.0091148.ris

Full Text

CASPASES IN O L F A C T O R Y NEURON APOPTOSIS by CATHERINE M . C O W A N B.Sc. Hons. (Neurosci.), University of Sheffield, U.K. , 1997 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Zoology, Neuroscience Program) We accept^this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A 2003 © Catherine M . Cowan, 2003 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Z o o l p ^ q The University of British Columbia Vancouver, Canada Date H "Time. ZOOl DE-6 (2/88) ABSTRACT The process of neuronal apoptosis is essential in brain development, and is an important component of many neurodegenerative diseases. Protease enzymes of the caspase family are key elements in the molecular events occurring during apoptosis. The combination of caspases utilized depends upon the death stimulus and cell type. Caspases-3 and -9 have been implicated in developmental neuronal apoptosis, as both caspase-3 and caspase-9 null mice display attenuated apoptosis during embryonic development resulting in excess neurons. The olfactory system (OS) is an excellent system for studying apoptosis of mature neurons in vivo, because the mature olfactory receptor neurons (ORNs) are physically distinct from other neuronal populations, are easily manipulated, and project to one target, the olfactory bulb. I show that ORNs undergoing apoptosis after removal of the olfactory bulb, undergo a retrograde wave of activation of caspase-9 and caspase-3, proceeding from synapse to cell body. This is the first time such a wave has been demonstrated in any neuronal system. Caspase-3 null mice demonstrate that caspase-3 is required for acute bulbectomy-induced apoptosis. Nothing was previously known about the role of any apoptotic molecules in OS development. I show that caspase-3 is necessary for correct OS development: caspase-3 null adult mice have OS abnormalities including an expanded ORN population, and post-natal onset olfactory bulb abnormalities. An expanded olfactory epithelium (OE) is evident as early as embryonic day (E)13 in caspase-3 and caspase-9 nulls, coinciding with wildtype peak of OE apoptosis and onset of OE caspase-3 expression. OE apoptosis is decreased five-fold at E l 3 in caspase-3 and caspase-9 nulls, but normal low levels are seen postnatally. Caspase-3 and -9 are crucial for E l 3-specific presynaptic wave of developmental death, the absence of which results in life-long neuronal abnormalities. However, ORNs can utilize a caspase-3/9-independent death mechanism depending upon maturational state and possibly signals received. Taken together, these results suggest that neurons at different stages of the same lineage can die by different mechanisms. TABLE OF CONTENTS Title page i Abstract i i Table of contents i i i List of tables v List of figures vi List of abbreviations vii i Acknowledgements ix CHAPTER 1: INTRODUCTION 1 CHAPTER 2: LITERATURE REVIEW 2.1 Apoptosis in the mature and developing olfactory neuroepithelium 9 2.2 Caspases .21 CHAPTER 3: AIMS 3.1 Caspase-3 and caspase-9 in mature ORN apoptosis in vivo 27 3.2 Caspase-3 and caspase-9 in olfactory system development 29 CHAPTER 4: METHODS Mice and genotyping 31 Bulbectomy 32 Tissue preparation 32 Immunoblotting 33 Immunoprecipitation 33 Immunohistochemistry 34 Terminal dUTP nick-end labelling (TUNEL) 35 In situ hybridization 36 Histological stains 36 Image capture 36 Methods of sampling, counting and measurement 37 RESULTS CHAPTER 5: Apoptosis and caspases in mature olfactory neurons of the adult mouse 5.1 Apoptosis of mature ORNs after olfactory bulbectomy 40 5.2 Axonal activation of caspase-3 during ORN apoptosis 41 5.3 Caspase-3-dependent cleavage of axonal APLP2 42 5.4 Evidence that proteolysis starts at the presynaptic complex 43 5.5 Activation of caspase-9 during ORN apoptosis 44 5.6 Expression of other caspases in the mature OE during O R N apoptosis 45 5.7 Mature ORNs do not undergo acute bulbectomy-induced apoptosis in caspase-3 null mice 47 DISCUSSION 48 CHAPTER 6: Caspase-3 and caspase-9 in development of the mouse olfactory system 6.1 The caspase-3 and caspase-9 null mice 72 6.2 Caspase-3 and caspase-9 expression in the developing olfactory system 73 6.3 Decreased apoptosis in the olfactory system of caspase-3 and caspase-9 null E13 embryos 73 6.4 Caspase-3 and caspase-9 null mice have a thicker OE neuronal layer at E l 3 74 6.5 Identity of OE cells which undergo caspase-dependent or -independent apoptosis . .75 6.6 Abnormal shape and size of olfactory bulbs in caspase-3 null adults : 76 6.7 Normal size and shape of olfactory bulbs in caspase-3 null neonates 77 6.8 Increased caspase-9 expression in the caspase-3 null mouse 77 DISCUSSION 77 CHAPTER 7: CONCLUSIONS 7.1 Summary of results and conclusions 109 7.2 General discussion I l l 7.3 Future directions 118 REFERENCES 122 LIST OF TABLES Table # page 4.1 Primary antibodies used in immunoblotting 39 4.2 Primary antibodies used in immunohistochemistry 39 6.1 Genotype frequencies and lifespan of caspase null mice 84 LIST OF FIGURES fig# title page 1.3.1 Diagram to illustrate the location of the olfactory epithelium 6 1.3.2 Diagram to illustrate the cell types of the neuronal lineage in the olfactory epithelium . . . 7 1.3.3 Schematic diagram of the olfactory bulbectomy 8 2.1.1 Features of apoptosis 23 2.1.2 Experimental models to manipulate ORN death 24 2.1.3 Pathways implicated in ORN apoptosis 25 2.1.4 Apoptotic molecules within Olfactory Receptor Neurons 26 5.1.1 Apoptosis after bulbectomy: D N A laddering pattern 51 5.1.2 Apoptosis after bulbectomy: T U N E L labelling of ORNs proceeds away from the bulb . .52 5.1.3 Spatial evaluation of T U N E L at 24 hours after bulbectomy 53 5.1.4 Apoptosis after bulbectomy: T U N E L labelling of ORNs indicates progression of apoptosis over time 54 5.1.5 O M P decrease after bulbectomy confirms loss of ORNs 55 5.1.6 Temporal analysis of ORN apoptosis by T U N E L and in situ hybridization 56 5.1.7 Single-stranded D N A immunodetection shows the same pattern of apoptotic nuclei postbulbectomy as T U N E L 57 5.2.1 Caspase-3 proenzyme expression and activation during ORN apoptosis 58 5.2.2 Caspase-3 activation in ORN axons and cytoplasm during apoptosis 60 5.3.1 APLP2 is cleaved in the axons and cell bodies of apoptotic ORNs 61 5.4.1 Caspase-3-mediated APLP2 cleavage occurs first in the synaptic complexes, then axons and cell bodies of ORNs 63 5.5.1 Schematic diagram of caspase-9 protein 64 5.5.2 Caspase-9 proenzyme expression and activation in the OE during ORN apoptosis 65 5.5.3 Caspase-9 expression in ORN axons and synapses 66 5.6.1 Expression of other caspases in the mouse OE 67 5.6.2 Caspases-8 and -10 in the olfactory epithelium after bulbectomy 68 5.6.3 Caspase-8 is activated on the lesioned side of the OE after bulbectomy 69 5.7.1 Caspase-3 is required for ORN apoptosis after bulbectomy in the adult mouse 70 5.7.2 OE of caspase-3 null mice retains normal thickness 72 hours after bulbectomy 71 6.1.1 Caspase-3 null mice are often small with hydrocephaly : . . 85 6.2.1 Caspase-9 is strongly expressed in ORN axons throughout development 86 6.2.2 Caspase-3 is strongly expressed in ORN axons and cell bodies throughout development from E l 3 87 6.3.1 Apoptotic neurons in the olfactory epithelium of wildtype developing mice 88 6.3.2 E l 3 is the peak of apoptosis in the olfactory epithelium 89 6.3.3 Apoptotic neurons in the OE of caspase-3 and caspase-9 null E13 embryos 90 6.3.4 Caspase-3 and caspase-9 null E13 embryos have fewer apoptotic cells in the OE than littermates 91 6.3.5 GAP-43 expression in the same population as T U N E L provides proof of OE identity . . .92 6.3.6 Apoptosis in caspase-3 and caspase-9 null E13 embryos 93 6.3.7 Other notable features of apoptosis in E l 3 embryos 94 6.4.1 Caspase-3 and caspase-9 null E l 3 mice have a thicker OE than littermates 95 6.4.2 Caspase-3 and caspase-9 null E l 3 mice have an OE of more varied thickness 96 vi 6.5.1 N C A M expression provides proof of ORN identity of caspase-dependent cells at E l 3 . . 97 6.5.2 Caspase-3 is not required for all ORN apoptosis in the adult mouse 98 6.6.1 Olfactory bulbs of caspase-3 null adult mice are of more varied sizes 100 6.6.2 Abnormally shaped olfactory bulbs with ectopic bulb tissue in caspase-3 null adults . . .101 6.6.3 Caspase-3 null mice with hydrocephaly have abnormally shaped olfactory bulbs 102 6.6.4 Enlarged glomeruli and ORN axon bundles in caspase-3 null adults 103 6.6.5 Caspase-3 null adults have increased ORN axon bundle area 104 6.7.1 Olfactory bulbs of P4 caspase-3 null mice are the same size and shape as littermates . . 105 6.7.2 Tyrosine hydroxylase immunolabelling of periglomerular cells at P4 in caspase-3 nulls 106 6.7.3 No difference in number of apoptotic ORNs at P4 between caspase-3 nulls and littermates 107 6.8.1 Caspase-9 expression is increased in caspase-3 null mice 108 7.2.1 Pathways implicated in ORN apoptosis 116 7.2.2 Apoptotic molecules within olfactory receptor neurons 117 vii LIST OF ABBREVIATIONS Apaf-1 apoptotic protease activating factor-1 APLP2 amyloid precursor-like protein 2 APP amyloid precursor protein A x axon , C3 caspase-3 C9 caspase-9 CNS central nervous system CP cribriform plate E embryonic day EPL external plexiform layer G glomerulus GAP-43 growth-associated protein-43 G C granule cell G F A P glial fibrillary acidic protein het heterozygote HRP horseradish peroxidase kDa kilo Dalton M C mitral cell N C nasal cavity N C A M neural cell adhesion molecule N F L nerve fibre layer N G F nerve growth factor NST neuron-specific tubulin (pill tubulin) OB olfactory bulb OE olfactory epithelium OEG olfactory ensheathing glia OMP olfactory marker protein OP olfactory placode ORN olfactory receptor neuron P postnatal day PBS phosphate-buffered saline PCR polymerase chain reaction PFA paraformaldehyde pOE presumptive olfactory epithelium T H tyrosine hydroxylase T U N E L terminal dUTP nick-end labelling viii ACKNOWLEDGEMENTS I would like to thank all of the Roskams' laboratory that I have known over the years March 1998-March 2003 for moral support and a pleasant work environment. In particular, I would like to sincerely thank the following people for technical assistance: Teresa Wang for mouse colony management and genotyping, as well as assistance with mouse sacrifices Jimmy Thai for optimizing and teaching me alkaline phosphatase T U N E L Harinder Janjua for assistance with immunochemistry Joellen Fung for in situ hybridization Dr Murry Gilbert for advice on confocal microscopy General assistance over the years from Rhonda Oshanek, Sheila Ayres, Kathryn Clark, Lisa Garcia, and Andrea Griffiths. And, of course, Dr A. Jane Roskams for assistance technical, intellectual, moral and otherwise. I would like to acknowledge receipt of gift antibodies and mice from: Drs Stanislaw Krajewski and John Reed (The Burnham Institute, La Jolla, C A ) Dr Donald Nicholson (Merck Frosst Centre for Therapeutics, Montreal, Quebec) Dr Scott Kaufmann (Mayo Graduate School, Rochester) Dr Frank Margolis (University of Maryland) Dr Gopal Thinakaran (University of Chicago) Dr Keisuke Kuida (Vertex Pharmaceuticals) Financial Support for my project or myself was received from: 1998-1999 British Columbia Health Research Foundation, and C M M T operating support from Merck Frosst B C Health Research Foundation National Institutes of Health 2001-2002 Rick Hansen Neurotrauma Initiative Studentship ix CHAPTER 1: INTRODUCTION 1.1 NEURONAL APOPTOSIS Neuronal apoptosis is an important process in normal development, as well as being a major feature of a number of pathological states including neurodegenerative diseases, stroke and trauma (Choi 1996, Rink et al 1995, Stefanis et al 1997). In the developing nervous system, embryonic or early postnatal neuronal apoptosis is required for normal development (Oppenheim 1991, Pettmann and Henderson 1998). Often, it occurs because an excess number of neurons are born and compete for limiting quantities of neurotrophic factors in their immediate and target environment (Burek and Oppenheim 1999, Barde 1989). Trophic support is required from various sources throughout the neuraxis to ensure neuronal survival both before and after the critical period of naturally occurring cell death (Davies 1997, Wang and Tessier-Lavigne 1999). Neuronal apoptosis has also been shown to have a role in Huntington's disease (Thomas et al 1995, Portera-Cailliau et al 1995), Alzheimer's disease (Loo et al 1993, Su et al 1994, Lassmann et al 1995), Parkinson's disease (Olson 1997, Anglade et al 1997), Amyotrophic Lateral Sclerosis (Martin 1999), traumatic injury (Yakovlev et al 1997) and in secondary damage following stroke (Hara et al 1995, Mehmet and Edwards 1996, Choi 1996). 1.2 CASPASES Various members of the caspase family of proteases are involved in the initiation and execution of the apoptotic process in different cell types. The activation of caspases (Cysteine-containing, ASPartate-specific proteASES) is a crucial stage in the series of molecular events that take place during apoptotic cell death. The caspases that have been implicated in cell death exist as proenzymes and themselves require cleavage prior to activation and heterotetramerization. Although at least 14 members of the highly conserved caspase family (all homologues of the C. elegans CED-3) are now known in mammals, at least six of which have clearly defined role in apoptosis (reviewed in Earnshaw et al 1999), it is not clear that all caspases are involved in the apoptotic process. What is clear from a number of different experiments in cell lines and in situ experiments is that there is a close and direct interaction between different members of the caspase family and that cell death results from an enzyme cascade of mutual activation (Slee et al 1999, McCall and Steller 1997, Nicholson and Thoraberry 1997, Salvesen and Dixit 1997). The particular caspases involved in a given apoptotic situation are likely to be cell type and stimulus specific. A more detailed review of the caspase literature is provided in section 2.2. This thesis makes a contribution to the caspase field by using a system in which the molecular mechanisms of in vivo neuronal apoptosis can be studied. Although a role for caspase family members in neuronal apoptosis is clearly evident (especially caspases-3 and -9, see section 1.4), it is not clear which caspase family members co-exist within neurons and what the sequential hierarchy of caspase cleavage is that results in the end-stage of D N A fragmentation. This thesis is novel in presenting in vivo data implicating co-activation of different members of the caspase family in neuronal cell death in a genetically normal animal model. Caspases-3 and -9 have been implicated in distant neuronal and glial apoptosis following ischemia and spinal cord injury, although the manner in which they transmit apoptotic signals in these in vivo models has yet to be revealed (Clark et al 1999, Endres et al 1998, Springer et al 1999); This thesis offers 1 evidence that caspases-3 and -9 can propagate their proapoptotic signal from the synapse to the cell body, and provides other insights into possible mechanisms of action in developing and mature neurons. 1.3 THE OLFACTORY SYSTEM AS A MODEL OF NEURONAL APOPTOSIS The olfactory system provides an ideal model in which to study neuronal apoptosis in vivo. The olfactory receptor neurons (ORNs) of the mammalian olfactory epithelium (OE) are convenient to study because: they are directly exposed to the external environment of the nasal cavity and thus easily accessible; they project their axons back to a single target, the glomeruli of the olfactory bulb; and the OE is separated from the olfactory bulb and the rest of the brain by the cribriform plate, and thus the ORN cell bodies are physically distant from all other neuronal cell types (figure 1.3.1). ORNs die by apoptosis normally in vivo in both human and rodent. This is part of a regular turnover which continues throughout adult life (Moulton 1974, Graziadei et al 1979, Graziadei and Graziadei 1979) and in fact ORNs are the only neurons in the mature nervous system which do this. In the rat and mouse, each ORN normally degenerates every 4-6 weeks (although this time is variable, as discussed in the Literature Review), and is replaced by a newly differentiated precursor neuron from the base of the neuronal layer, which assumes a mature state within the upper layers of the OE and projects an axon directly back into the olfactory bulb (Calof et al 1996, Carr and Farbman 1992, Huard et al 1998, Roskams et al 1996, Schwartz Levey et al 1991). Consequently, at any given moment the OE consists of cells at all stages of the neuronal lineage, from neuronal precursor to mature neuron to apoptotic neuron (figure 1.3.2). ORNs can readily be studied as they undergo apoptosis by inducing the synchronized death of a large number of ORNs in the rodent by one of a number of paradigms. These include: physical removal of their target, the olfactory bulb; chemical lesion of the bulb; or direct chemical ablation of the OE. These models are discussed in more detail in the Literature Review, section 2.1. In our laboratory we most commonly use the olfactory bulbectomy paradigm to induce a synchronous wave of apoptosis in ORNs (figure 1.3.3). In this paradigm, all of the mature ORNs (an estimated 8-12 million in the mouse) die within 72 hours of the target ablation (Roskams et al 1996, Calof et al 1996). There is evidence that this death is apoptotic (Morrison and Costanzo 1989, Michel et al 1994, Holcomb et al 1995), and we were also able to confirm this (see Results section 5.1). The approach of monitoring the activation of cell death pathways and the timecourse of apoptosis in the olfactory neuroepithelium following bulbectomy has already been proven to be a valuable one in establishing a neuroprotective role for Bcl-2 in the neuronal apoptotic cascade in mice overexpressing human Bcl-2 (Jourdan et al 1998). Given the large number of neurons undergoing a synchronous wave of apoptosis and the accessibility of this population for histological, biochemical and molecular analysis, the OE is thus a useful in vivo model for studying the activation and pharmacological manipulation of proteins that lead to the apoptosis. One of the major aims of this thesis is to determine which pathways act in ORNs in vivo as the neurons undergo apoptosis. The olfactory bulbectomy model gives us a means to study and analyse neuronal cell death in real neurons as they are dying. Despite significant progress in 2 understanding the enzymological process of neuronal apoptosis, essentially nothing is known about how a neuron, which exists in a complex three-dimensional environment in vivo, integrates the pro- and anti-apoptotic signals received from different parts of the neuraxis. I have taken advantage of the unique structural and organizational features of the OE to examine where caspases are activated in mature ORNs after deafferentation, and how this activity is propagated in space and time. Considering the number of developmental mechanisms that are also conserved between olfactory neurons and other peripheral and central neurons (reviewed in Roskams et al 1996), I believe that studying the signal transduction mechanisms that contribute to apoptosis in the OE will provide insight not only into how olfactory neurons die, but also into how this process may be regulated in other neuronal groups. Aside from inferences that may be made about neuronal apoptosis in general, studying how olfactory neurons die in vivo will also be of importance in understanding the cell biology of these particular neurons. It would be interesting to understand the molecular players that are involved in the turnover of ORNs. It is known that Bcl-2 is involved (Jourdan et al 1998), but the involvement of caspases in this process has not been investigated. The olfactory bulbectomy model also has the added advantage of allowing us to compare mature neuronal cell death and developmental neuronal cell death in the same lineage. Despite considerable progress in elucidating the cellular pathways of apoptosis, it is not yet known whether immature and mature neurons from the same lineage utilize identical cellular pathways to mediate apoptosis. 1.4 CASPASE-3 AND CASPASE-9 NULL MICE Caspase-3 and caspase-9 null mice both show similarly dramatically abnormal brain development, with no gross abnormalities in any other organ, which has led to the idea that caspases-3 and -9 in particular are the important caspases in the nervous system. The major features are ectopic neuronal tissue protruding from the skull, which occurs as a result of decreased neuronal apoptosis; disruption of cortical structure; expansion of the ventricular zone; and sometimes hydrocephaly (Kuida et al 1996, 1998, Hakem et al 1998). The caspase-9 null phenotype is more severe than the caspase-3 null phenotype in that caspase-9 null mice generally do not live beyond postnatal day (P)3, whereas caspase-3 null mice are reported to live 3-6 weeks postnatally (Kuida et al 1996). In addition, it has been shown that there is reduced terminal dUTP nick-end labelling (TUNEL) in caspase-3 and caspase-9 null mouse embryonic brains (Srinivasan et al 1998 and Hakem et al 1998 respectively). The caspase-3 and caspase-9 null phenotypes have therefore been cited as evidence that these two caspases play a crucial role in apoptosis in the developing brain. This thesis will address for the first time whether there is an olfactory system phenotype in caspase-3 and caspase-9 null mice. 1.5 APOPTOSIS AND CASPASES IN OLFACTORY SYSTEM DEVELOPMENT There has been some work on programmed cell death in the OE during development. A massive wave of cell death has been described in the developing rat OE at embryonic day (E)12 and E l 3 (Pellier and Astic 1994a,b), which is the time of olfactory pit involution and establishment of 3 early olfactory turbinates. Particularly dense apoptotic nuclei are seen in the posterior region of the olfactory pit at E13 and E14, immediately prior to and during the time when the two nasal swellings contact each other posteriorly and fuse to form the epithelial seam (Pellier and Astic 1994a). This process is a part of palate formation. Another function of apoptosis at this time is that the extracellular space created after E14 is suggested to provide a path of least resistance for the newly outgrowing olfactory axons and their accompanying migrating cells (Pellier and Astic 1994a). After E14, very few apoptotic nuclei are seen in the developing OE (Pellier and Astic 1994b). Developmental apoptosis was quantified more carefully, and cell types were identified, by Voyron et al 1999. They found two peaks of apoptosis at the septum of the developing mouse OE: one at E12 (corresponding to the death seen by Pellier and Astic), and one at E16 (at the time of initial olfactory axon synaptogenesis). Very little is known about expression of caspases in the developing olfactory system, except for the isolated facts that caspase-3 mRNA is present in the proliferative region of the olfactory bulb throughout life (Yan et al 2001), and active caspase-3 protein is present in this region in the first 12 postnatal days (de Bilbao et al 1999). This is not believed to be related to apoptotic death of the proliferating cells. Essentially nothing is known about the role of caspases in the programmed cell death events described above. These issues will be addressed in A I M 2. 1.6 OVERVIEW OF AIMS AIM 1: To test the hypothesis that caspase-3 and caspase-9 are involved in mature ORN apoptosis in vivo. These two caspases are the most strongly implicated in neuronal apoptosis out of those thus far detected in the OE. I use a combination of western blotting and immunohistochemical approaches to assess changes in amount and distribution of proenzyme, active forms, and downstream cleavage products of the caspases over the timecourse of apoptosis. I analyse bulbectomized tissue from caspase-3 null mice to confirm whether caspase-3 is necessary for mature ORN apoptosis. AIM 2: To investigate the role of caspase-3 and caspase-9 in olfactory system development using null mice. Both caspase-3 and caspase-9 null mice are known to have grossly abnormal brain development. I will assess whether they have abnormal olfactory system development. Almost nothing is known about the involvement of caspases in programmed cell death in the developing olfactory system. The main questions asked in this section are: • What is the expression pattern of caspase-3 and caspase-9 in the normal mouse embryo OE? • Is apoptosis abnormal in the olfactory system of caspase-3 and caspase-9 null mice? • Does the caspase-3 null mouse have gross olfactory structural abnormalities? 1.7 OVERVIEW OF RESULTS AND CONCLUSIONS CHAPTER 5 I have confirmed that ORNs lesioned by olfactory bulbectomy do indeed die by apoptosis, as revealed by D N A laddering, T U N E L labelling, and single-stranded D N A immunohistochemistry. I determined spatial and temporal parameters of this bulbectomy-induced O R N apoptosis, as a framework in which to examine the role of caspases. Using a 4 combination of western blotting, immunohistochemistry and TUNEL, I determined that caspase-3 and caspase-9 proenzyme expression, activation, and cleavage of downstream targets, all change in an apoptosis stage-specific manner. Maximal caspase-9 cleavage occurs earlier than maximal caspase-3 cleavage. Caspases are expressed throughout the O R N from synapse to cell body, and caspase-dependent cleavage of downstream targets occurs in the axon earlier than in the cell body. I conclude that caspases-9 and -3 are situated to propagate the apoptotic signal from synapse to cell body. ORNs of caspase-3 null mice do not undergo acute bulbectomy-induced apoptosis. CHAPTER 6 Caspases-3 and -9 are strongly expressed in ORNs throughout development. Caspase-3 is strongly expressed in many layers of the olfactory bulb in postnatal development. The peak of developmental apoptosis in the OE is at E l 3 . The OE of caspase-3 and caspase-9 null mice at E l 3 contains 4-fold fewer apoptotic nuclei, and is thicker, than wildtype. Neonatal and adult caspase-3 nulls on the other hand have too many ORNs, but they do not have thicker OE or less apoptosis. Therefore an isolated period of requirement for caspase-3 leads to a lifelong expanded ORN population, but the OE is eventually able to regulate the local organization of its layers. Caspase-3 null adults have too many ORNs with abnormally organized axon bundles, and olfactory bulbs of an abnormal shape and size. Neonatal caspase-3 nulls, however, do not yet have abnormal olfactory bulbs. Postnatal onset of bulb abnormalities may be because of postnatal functional input from an already abnormal ORN population, and/or postnatal onset of caspase-3 expression in cells of the bulb, including postnatal birth of caspase-3 expressing periglomerular cells. The requirement for caspase-3 in ORN apoptosis varies according to maturity of the olfactory neuron: in the adult, mature ORNs require caspase-3 but immature ORNs might not. The requirement for caspase-3 in ORN apoptosis varies according to maturity of the olfactory system: immature ORNs in the E l 3 embryo require caspase-3 (and caspase-9) for apoptosis, but immature ORNs in the adult do not require caspase-3 for apoptosis. 5 FIGURE 1.3.1 a) Diagram to illustrate the location of the olfactory epithelium The olfactory epithelium (OE) is a convenient tissue in which to study neuronal apoptosis because it is directly exposed to the external environment of the nasal cavity; it projects to one target, the olfactory bulb (OB); and it is separated from the brain by the cribriform plate (CP). b) The primary olfactory neuraxis Olfactory receptor neurons (ORNs) within the OE send their axons through the cribriform plate to synapse with mitral cells within the glomeruli of the olfactory bulb (from Margolis et al 1991). 6 FIGURE 1.3.2 Diagram to illustrate the cell types of the neuronal lineage in the olfactory epithelium Due to ongoing turnover of olfactory receptor neurons, all of these cells types are present in the mature OE. Depicted (from left to right) are a basal precursor cell, a mitotic basal cell, a neuronal precursor, an immature neuron, three healthy mature neurons, and an apoptotic mature neuron (from Halasz 1990). 7 Adapted from Calof et al 1996 FIGURE 1.3.3 Schematic diagram of the olfactory bulbectomy Following a unilateral lesion, the mature olfactory receptor neurons (ORNs) which project to the olfactory bulb are induced to die in a synchronous wave of degeneration. 8 CHAPTER 2: LITERATURE REVIEW 2.1 Apoptosis in the mature and developing olfactory neuroepithelium NOTE: This review was published as Cowan and Roskams (2002), Microsc. Res. Tech. 58:204-215. Section 2.1 of this thesis consists of this paper almost in its entirety, except that some data that I generated has been removed as it will be presented later in this thesis. ABSTRACT Neuronal apoptosis is important in the developmental sculpting of a normal nervous system and also in the loss of neurons caused by neurodegenerative disease, ischemia or trauma. In a developing embryo, exquisite mechanisms of regulation exist to balance factors that control neuronal birth and death within a given neuronal group, so that sufficient neurons develop and survive to elicit normal function. Postnatally, the only part of the mammalian nervous system where many of these regulatory balance mechanisms are retained is the OE. During the last thirty years, researchers investigating ORN cellular and developmental biology have focussed on the regeneration of the neuronal population within the olfactory neuroepithelium, following the induced death of the mature neuronal population. This body of work has thus far overshadowed the equally important and intrinsically linked phenomenon of death of mature ORNs which is required to initiate regeneration. The purpose of this review is to reveal what has been established about the different forms of cell death that can occur in neurons of the OE, and highlight the identified pro- and anti-apoptotic pathways that control the normal and induced turnover of ORNs. INTRODUCTION The last thirty years (since Graziadei 1973) have provided us with a plethora of data aimed at understanding the mechanisms driving the repopulation of the OE by regenerating ORNs. Little, however, is known about the pathways that drive the apoptotic death of these same ORNs, despite the fact that there is recognized signalling between regenerating ORNs and their apoptotic counterparts to ensure tightly regulated repopulation (Calof et al 1996a, Carr and Farbman 1992, 1993). Now that model organisms have granted us access to the identification of molecular pathways that mediate apoptosis, apoptotic mechanisms can now be explored during both normal and induced turnover of neurons within the olfactory neuroepithelium. Additional analysis of the human genome sequence has targeted the apoptotic machinery as a model for understanding both the conservation and complexity that arises in a given cellular process, as a diverse number of organisms, dependent upon that process for survival, evolve (Aravind et al 2001). WHAT IS APOPTOSIS? At the same time that Pasquale Graziadei was establishing the regenerative response of ORNs to lesion, the term apoptosis (from the Greek for falling leaves) was coined (Kerr et al 1972). 9 Apoptosis was first defined as a morphologically distinctive form of cell death associated with normal physiology, whereby a cell is eliminated by an endogenous suicide program, leaving no trace of its existence. This seminal paper demonstrated how widespread apoptosis is in nature (Kerr et al 1972). The same group distinguished the active process of apoptosis from passive necrosis (Wyllie et al 1980) and defined the distinctive morphology and biochemical features of apoptosis (Wyllie et al 1984) (figure 2.1.1). It has now been established that the genetically regulated apoptotic apparatus has been highly conserved in evolution (Vaux et al 1994, Aravind et al 2001). Apoptosis is essential for processes as diverse as removal of vestigial organs (e.g. tail resorption in the tadpole), normal tissue turnover (e.g. skin, intestinal crypts), organ involution after trophic factor withdrawal (e.g. prostate, endometrial sloughing during menstruation, mammary remodelling during pregnancy and nursing), tissue sculpting during embryogenesis (e.g. formation of digits, canalization of ducts and tubes, fusion of palate, closure of neural tube), and removal of damaged cells (e.g. sunburn, response of host cell to virus) (reviewed in Haanen and Vermes 1996, Milligan and Schwartz 1997). Signals that trigger apoptosis include damage due to ionizing radiation or viral infection, or extracellular signals (Milligan and Schwartz 1997). Extrinsic signals may either suppress or promote apoptosis, and the same signals can promote survival in one cell type and invoke the suicide program in others (Steller 1995). Apoptosis is therefore a complex phenomenon of related morphological and biochemical processes that can vary with tissue and cell type. CELLULAR HALLMARKS OF APOPTOSIS Apoptosis has traditionally been demonstrated using the morphology typical of end-stage apoptosis, which can be determined by light and electron microscopy. Features of apoptosis include condensed chromatin, electron-dense nuclei, reduction of cytoplasmic volume, intact organelles and plasma membrane initially, and membrane blebbing, leading to fragmentation of nuclear contents, and finally encapsulation of fragments into apoptotic bodies and rapid clearance by phagocytosis (summarized in figure 2.1.1). Biochemical hallmarks accompanying morphological changes include a decreased mitochondrial membrane potential, decreased intracellular pH, externalization of phosphatidylserine on the surface of the apoptotic bodies, selective proteolysis, and the degradation of D N A in 180 base pair fragments, reflecting the spacing between individual nucleosomes (Wyllie et al 1984). This pattern can be identified using genomic D N A agarose electrophoresis, and presents as a laddering pattern which has been used to definitively identify apoptosis. This D N A fragmentation pattern has now been developed into histological techniques to detect D N A fragmentation, such as T U N E L (Gavrieli et al 1992). The limitations of T U N E L are that it can sometimes identify necrotic cells (Rink et al 1995), that some apoptotic cells are TUNEL-negative (Cohen et al 1992), and it could potentially also detect actively dividing cells. Now that some of the apoptotic pathways have been defined, fluorogenic peptides are widely utilized to demonstrate cytoplasmic cleavage events upstream of D N A fragmentation that are indicative of the apoptotic machinery in action. A limitation of all these methods is that the execution phase of apoptosis is relatively short, so that the number of cells in any given tissue or state undergoing apoptosis may be underestimated. 10 DEVELOPMENTAL APOPTOSIS IN THE NERVOUS SYSTEM Apoptosis is especially crucial within the developing nervous system, where the embryonic or early postnatal apoptotic death of neurons is a prerequisite for normal development (reviewed by Oppenheim 1991, Pettmann and Henderson 1998). Vertebrate neurons undergo developmental cell death because they are produced in excess and compete for access to limiting quantities of neurotrophic factors produced by their target and neighbouring cells (Barde 1989). By this edict, neurons that do not make an appropriate connection, or that are not surrounded by the optimal supportive environment, then die (the Neurotrophic Factor Hypothesis) (Burek and Oppenheim 1998). The mechanisms that link trophic support directly with survival or apoptosis in vivo are only just beginning to be revealed. There is now genetic and in vitro evidence that in many organisms both neurons and their precursors require trophic support from different sources throughout their neuraxis for survival at developmental stages both before and after the critical period of naturally occurring cell death (Davies 1997, ElShamy and Ernfors 1996, Farinas et al 1996). There are therefore exquisite mechanisms of regulation to balance factors that control neuronal birth and death within a given neuronal group in an embryo, so that sufficient neurons survive to elicit normal function. Postnatally, the only part of the mammalian nervous system where this balance of regulatory mechanisms is known to be retained is the OE (Calof et al 1996a, 1998, Roskams et al 1996). ABERRANT APOPTOSIS IN THE MATURE NERVOUS SYSTEM Although neuronal cell death is critical to ensure normal central nervous system (CNS) development, there are times in the mature nervous system when it is inappropriate for neuronal cell death to occur. Extensive neuronal death is observed in neurodegenerative diseases and also after stroke and trauma (Choi 1996, Rink et al 1995, Stefanis et al 1997). Although many neurodegenerative diseases have an established genetic basis, the majority of cases of Amyotrophic Lateral Sclerosis, Parkinson's disease, Multiple Sclerosis and Alzheimer's disease are not familial, but sporadic. Many key causal cellular events that result in the loss of neurons in these diseases have not yet been delineated, but they may share many common mediators (Bredesen 1995, Mattson et al 1998, Rink et al 1995, Stefanis et al 1997). Monitoring neuronal degeneration at the cellular level is therefore fundamental in order for us to understand the mechanisms that drive a neuron towards irreversible cell death, and to devise effective ways to intervene early in the pathways that mediate aberrant neuronal apoptosis. APOPTOSIS IN THE OLFACTORY EPITHELIUM Electron microscopy was initially used to demonstrate ORN degeneration after nerve transection in the frog (Graziadei 1973), the salamander (Simmons et al 1981) and the rodent (Graziadei and Monti Graziadei 1978), and after bulbectomy in the rodent (Graziadei et al 1979). The morphology of dying ORNs is consistent with apoptosis, after bulbectomy in the rodent (Morrison and Costanzo 1989), after nerve transection (Doucette et al 1983), in normal turnover (Magrassi and Graziadei 1995), and in development (Pellier and Astic 1994). D N A laddering has subsequently been demonstrated in the rodent OE (Michel et al 1994). T U N E L has also been used to label apoptotic ORNs in development (Voyron et al 1999), in normal turnover 11 (Magrassi and Graziadei 1995), after olfactory bulbectomy (Holcomb et al 1995, Cowan et al 2001), and after olfactory nerve transection (Deckner et al 1997). Apoptosis in olfactory receptor neuron development Apoptosis is as widespread in the formation of the olfactory system during embryonic development, as it is in the rest of the nervous system. As early as E10 of mouse development, small numbers of degenerating cells (as evidenced by swelling and disrupted cytoplasm) can be detected in the developing mouse olfactory neuroepithelium, during formation of the olfactory pit (Cuschieri and Bannister 1975). A massive wave of cell death has been described in the developing rat OE at E12 (concomitant with olfactory pit involution and the establishment of the early olfactory turbinates), with only occasional apoptotic cells from E14 onwards (Pellier and Astic 1994). Two types of cell death are proposed to occur at these early stages of OE development: apoptosis (as determined by electron microscopy); and a non-lysosomal type of cell death (distinguished from apoptosis by cytoplasmic swelling and no phagocytosis, and from necrosis by chromatin condensation and lack of mitochondrial pathology). In this study, cell density was highest during the peak time of apoptosis, and more diffuse later, and the extracellular space created after E14 was suggested to provide a path of least resistance for the newly outgrowing olfactory axons and their accompanying migrating cells. The more sensitive T U N E L technique has subsequently been used, in conjunction with double-labelling with olfactory marker protein (OMP) and growth-associated protein-43 (GAP-43) immunochemistry, to determine the cell types in the developing OE undergoing apoptosis (Voyron et al 1999). Peak phases of TUNEL-positivity in the developing mouse OE occur at E12 (corresponding to the formation of the olfactory pit, as seen by Pellier and Astic 1994) and at E l 6 (at the time of initial olfactory axon synaptogenesis). Apoptosis was shown to occur in both immature and mature ORNs, with a dramatic decrease in apoptosis at E l 8 , and by E l 9 very low levels are reached which are maintained through to adulthood. Apoptosis in mature olfactory receptor neurons Like other epithelia, but unlike other parts of the nervous system, olfactory sensory neurons undergo both apoptosis and genesis as a part of normal turnover throughout adult life. The ORN population density is tightly controlled by local signals, whereby mature sensory neurons either stimulate or inhibit their replacement signalling to progenitor populations within the lower layers of the neuroepithelium (Calof et al 1996a). Indeed, dying ORNs are always present in the OE (Graziadei et al 1978). A combination of TUNEL, electron microscopy (i.e. to visualize chromatin condensation), and light microscopy (Feulgen staining to label dying nuclei; acridine orange to label apoptotic bodies) has demonstrated that the cell death component of normal turnover in the normal adult rat OE is apoptotic (Magrassi and Graziadei 1995). T U N E L -positivity has also been demonstrated in all stages of the ORN lineage in the normal adult mouse (Holcomb et al 1995) and rat (Deckner et al 1997). The extracellular or intracellular stimuli that initiate the apoptotic response in normal ORN turnover have yet to be identified. Phenomena that influence the turnover of normal olfactory neurons It was once thought that ORNs have a fixed, finite lifespan that is intrinsically determined, as ORNs in an untreated rodent were shown to live for 1 month after differentiation (Graziadei and Monti Graziadei 1978, Moulton 1974). It is now known that the environment to which the OE is exposed can affect the longevity of the ORNs (reviewed in Farbman 1990). It was subsequently 12 thought that mammalian ORNs could live up to 12 months in a clean environment (Hinds et al 1984), although these data are open to a different interpretation as the cells labelled in this study could have been quiescent basal cells (Magrassi and Graziadei 1995). Some ORNs do have a lifespan exceeding 3 months (found using retrogradely transported markers; Mackay-Sim and Kittel 1991), but normal ORNs usually live 30-40 days (following ORN clones from labelled progenitors; Caggiano et al 1994). As an animal ages, its ORNs spend a longer time in the differentiation phase of their life cycle (Breipohl et al 1986), but a relationship between age and O R N apoptosis has yet to be established. The rate of apoptosis (and therefore the longevity of the ORN) and the rate of neurogenesis are intrinsically linked, and manipulating one affects the other (Farbman et al 1988). It has been suggested that mechanisms to regulate cell division and apoptosis to maintain a constant number of ORNs are locally controlled, because both the thickness and the rate of proliferation of the OE within an individual animal may vary (Farbman 1990). In lobster, ORNs are added proximally and lost distally within the flagellum, and a spatial gradient of ORNs of different ages exists (Steullet et al 2000). The lifespan of lobster ORNs is much longer than in mammals (6-12 months), but longevity is also affected by mechanical damage. Developing lobster ORNs also die within a defined period if they fail to make a connection with their target. The parallels with ORN turnover in mammals are important for mechanistic considerations, but this is a situation in which asthetacs (sensory units containing ORNs) are being physically shed, rather than cells undergoing apoptosis. Experimental manipulations that influence the turnover of normal olfactory neurons Specific lesions which target different sectors of the olfactory pathway are becoming increasingly important in studying the molecular signals that drive apoptosis in ORNs (figure 2.1.2). Three experimental methods for causing mass degeneration of ORNs have been used extensively: direct chemical lesion (introduction of zinc sulphate, Triton X-100 or methyl bromide into the nasal cavity, figure 2.1.2c), olfactory bulbectomy (figure 2.1.2a), and olfactory nerve transection (figure 2.1.2b). The latter two paradigms result in O R N apoptosis. In contrast, most chemical insults to the OE, although they may be considered in some respects more natural causes of death, potentially mimicking the effects of toxic environmental stimuli, actually induce necrosis in ORN and non-ORN cells within the OE. Sensory deprivation (e.g. by naris occlusion), a model extensively used in other developing sensory systems, has also been used to manipulate O R N development and to examine subsequent cellular remodelling (including apoptosis) within the olfactory bulb (figure 2.1.2d). Sensory deprivation Naris occlusion (first used by Meisami 1976) causes a decrease in O R N density within the OE in parallel with a decrease in size of the olfactory bulb (figure 2.12d). In neonatal rats, the OE became thinner at 20 days post-naris occlusion (before any change in cell number), with a significant reduction in ORN number reported by 30 days (Farbman et al 1988). This reduction in O R N number did not change the number of cells immunoreactive for OMP, but did report a 40% reduction in the rate of neurogenesis. The fact that there is a reduction in the rate of neurogenesis but the same number of mature ORNs, leads to the conclusion that the mature ORNs must have an increased lifespan. Thus, a delay in ORN apoptosis caused by naris occlusion was attributed to protection from infectious or toxic agents in the environment, rather than a change in the functional activity of the ORNs due to the sensory deprivation itself. This 13 conclusion is supported by the fact that similar changes are observed in the non-sensory respiratory epithelium. Loss of mature ORNs following bulbectomy Surgical removal of the olfactory bulb (bulbectomy) is both a target-derived trophic factor deprivation and a deafferentation, and causes a specific and somewhat synchronous retrograde degeneration of mature ORNs (Graziadei et al 1979, Costanzo and Graziadei 1983, Le Gros Clark 1951) (figure 2.1.2a). The olfactory bulbectomy model has thus far been utilized chiefly to examine neurogenesis and regeneration of ORNs following loss of the mature O R N population. Only in the last decade has the acute degeneration seen in this model been proved to be apoptotic (Holcomb et al 1995, Morrison and Constanzo 1989, Cowan et al 2001). In the hamster, at 4 days post-bulbectomy the number of ORNs (assessed using morphological criteria) is reduced to 39% of normal, and OE thickness is reduced to 60-70% (Costanzo and Graziadei 1983). A subsequent study demonstrated an almost total absence of ORNs at 4 days after bulbectomy in the adult hamster, and showed by scanning electron microscopy shrunken O R N cell bodies with pocked cell surfaces (Morrison and Costanzo 1989). O R N degeneration is also evident within the first 24 hours post-bulbectomy in the mouse, with significantly fewer mature ORNs by 72 hours. Post-bulbectomy, the peak of apoptosis (2 days) precedes the peak of cell loss (5 days) as measured by epithelial thickness using septal OE as a reference point (Calof et al 1996a). This thesis will attempt to take into account the convoluted nature of the OE turbinates in assessing rates of apoptosis following bulbectomy, and to establish a relationship between the distance of the ORN cell body from the site of lesion and the timecourse of apoptosis (Cowan et al 2001). Although specifically a lesion model, bulbectomy may also represent a more physiologically relevant model of normal ORN turnover than it initially seems. Even though wear and tear through exposure to the environment does influence O R N lifespan, the signal for ORNs to die could also come from the bulb withdrawing trophic support from ORNs that it senses to be functioning at a sub-threshold level. Several target growth factors, including neurotrophins, are expressed within and around the olfactory bulb and may be at least partially responsible for the control of survival of the mature ORN population (Carter and Roskams 2002). Role of the olfactory bulb in ORN longevity The bulbectomy model was utilized to demonstrate that the olfactory bulb provides trophic support to maintain the mature ORN and prevent it from undergoing apoptosis (Schwob et al 1992). In adult rats, 1 month following unilateral bulbectomy (after acute retrograde degeneration has ceased), 90% of the newly generated cells of the O R N lineage were lost between 5 and 14 days after neurogenesis, as opposed to 40% in the normal situation. Thus, in the absence of the olfactory bulb, the average ORN lifespan becomes 1/5 to 1/2 of normal (2 weeks maximum), and regenerating ORNs die during the transition from GAP-43-positive to OMP-positive. At chronic timepoints following bulbectomy, increased O R N death (a 2-fold increase between 1 and 7 weeks post-bulbectomy) has been reported in newly generated cells at all stages in ORN post-lesion development (Carr and Farbman 1992,1993). An increase in cell death among newly generated immature neurons, 6-7 days old, corresponds with the stage at which the axons of these immature ORNs would have reached the bulb. These papers, collectively, provided compelling evidence of a link between olfactory bulb-drive survival of 14 ORNs, and between O R N death and precursor cell proliferation within the OE (Carr and Farbman 1992, 1993, Schwob et al 1992). Loss of mature ORNs following nerve transection A number of model organisms have been employed to demonstrate the loss of the mature O R N population following olfactory nerve axotomy (figure 2.1.2b). In the frog and rodent, olfactory nerve axotomy, like bulbectomy, leads to the rapid retrograde degeneration of the entire mature neuron population by 6-8 days post-axotomy, with loss of all OMP-positive ORNs by iO days (Graziadei 1973, Graziadei and Monti Graziadei 1978, 1980). In the axotomized frog OE, ORNs progressively degenerate from 4 to 15 days post-axotomy and then phagocytic cells appear within the OE (Graziadei 1973). The restoration of OMP-positivity has also been reported at 9 days post-axotomy in the rat (Huang et al 1995). Condensed chromatin (indicative of pyknosis) was observed in salamander ORNs following axotomy (Simmons et al 1981). Functionally, odour-evoked voltage transients decrease in amplitude over the first 7 days, and are lost by 10 days following axotomy (Simmons and Getchell 1981). Unlike a bulbectomy, the axotomized and regenerating olfactory nerve has a chance to reconnect with the olfactory bulb, therefore in this model the OE is not thrown into a prolonged cycle of degeneration and regeneration. Proof of apoptosis in ORNs D N A laddering of genomic D N A prepared from the mouse OE at 8-72 hours post-bulbectomy provided the first molecular evidence that cell death of ORNs following lesion is, indeed, by apoptosis (Michel et al 1994). Subsequently, TUNEL-positive ORNs were demonstrated in the mouse and rat OE in both normal adult and post-bulbectomy states (Holcomb et al 1995, Magrassi and Graziadei 1995). In accordance with the morphological evidence (above), TUNEL-positive cells were seen at all stages within the ORN lineage (Holcomb et al 1995), with maximal TUNEL-positivity at 48 hours following bulbectomy. A n increase in T U N E L -positivity within the OE, concomitant with the increased turnover rate in the absence of the olfactory bulb (reported above), is still seen at 84 days after bulbectomy. The shift in ratio of TUNEL+ immature ORNs: mature ORNs reported following the initial wave of regeneration (immature and mature ORNs in the first regenerative wave, followed by mature ORNs only in the chronic state) suggests that different factors are responsible for survival of cells at different stages in the ORN lineage (Calof et al 1996a). The differences in assessment of numbers of dying (Carr and Farbman 1992, 1993) versus TUNEL-positive apoptotic ORNs (Holcomb et al 1995) can be attributed in large part to better sensitivity of T U N E L over assessment of pyknotic morphology. This thesis uses T U N E L to demonstrate a three-dimensional pattern of retrograde apoptosis (proportional to length of axon) throughout different turbinates of the OE following unilateral bulbectomy. Macrophage infiltration, indicative of apoptosis (see figure 2.1.1), also occurs in the rat OE following bulbectomy (Suzuki et al 1995). Macrophages were seen in the lesioned OE as early a 1 day post-bulbectomy, with the number of macrophages peaking at 3-6 days post-bulbectomy (immediately following the apoptotic death of the entire mature ORN population). Macrophages were concluded to be the most important phagocyte acutely, whereas the supporting (sustentacular) cells of the OE assumed the more chronic phagocytic role, filling phagosomes with cell debris peaking at 30-90 days post-bulbectomy. T U N E L has also been used to confirm that the ORNs lost following unilateral olfactory nerve 15 transection are indeed, apoptotic. Following axotomy, the number of TUNEL-positive ORNs increased dramatically (8-fold) 36 hours after transection of the olfactory nerve, decreasing back to normal levels by 4 days post-axotomy (Deckner et al 1997). It should be noted, however, that the widely held view that axonal injury always causes rapid death of ORNs was contested in a study that found that the site of axonal transection profoundly affects the timecourse of ORN death (Doucette et al 1983). ORN degeneration (as determined by morphological criteria and epithelial thickness) is significantly delayed when the ORN axon transection is made on the dorsal surface of the bulb as opposed to at the cribriform plate. This form of delayed degeneration (when the lesion is closer to the terminal) is similar to that seen in other parts of the nervous system following axotomy. MOLECULAR SIGNALS THAT REGULATE ORN APOPTOSIS The last 5 years have witnessed a revolution in our understanding of cellular apoptotic machinery (Aravind et al 2001). From the initial identification of the CED genes which mediate developmental cell death in C. elegans, there are now large families of CED homologues that participate in the apoptotic pathways of mammalian cells (reviewed by Earnshaw et al 1999). For mechanistic purposes, apoptotic signalling pathways can be conveniently divided into initiation (either extracellular or intracellular), propagation (amplification) and termination. The apoptotic changes indicative of termination (figure 2.1.1) only result from a summated signalling process that can involve a variety of stimuli and parallel or intersecting pro-apoptotic pathways (figure 2.1.3). Initiators of ORN apoptosis The Tumour Necrosis Factor Receptor (TNFR) and Fas (or Apol/CD95) are two members of a family of cell surface death receptors capable of inducing intracellular proapoptotic pathways (Baglioni 1992, Itoh et al 1991). Both have already been implicated in neuronal apoptosis (MacEwan 1996, Rabizadeh et al 1993). The mRNA for Fas, Fas ligand, TNFR1 and its ligand TNF-a, are all detected in the normal adult rat OE (by reverse transcription polymerase chain reaction (RT-PCR)) (Farbman et al 1999). The Fas ligand is localized within the ORNs to the olfactory dendrites and dendritic knobs, and some axon bundles. Addition of either Fas ligand or TNF-a to organotypic cultures of foetal E19 rat OE causes an increase in the number of apoptotic bodies (determined by Feulgen staining) after 4-6 hours. The Low Affinity Nerve Growth Factor Receptor (p75-NTR) can have different pro- or anti-apoptotic activities dependent upon its cellular environment (and co-expression with other neurotrophin receptors) (see Carter and Roskams 2002, Kaplan and Miller 2000). Most recently, p75-NTR (which has close homology to the p75 TNFR, which has been implicated in neuronal apoptosis) has gained favour as a neuronal pro-apoptotic receptor (Barrett and Bartlett 1994). p75-NTR is up-regulated after seizure in susceptible (TUNEL+) brain regions, where T U N E L -positivity correlated highly with p75-NTR expression within individual neurons (Roux et al 1999). p75-NTR mRNA levels are also increased in striatal cholinergic neurons in a model of focal cerebral ischemia (Kokaia et al 1998), in motoneurons after sciatic nerve crush (Ernfors et al 1989), and in Purkinje cells after axotomy (Armstrong et al 1991). Genetic evidence for a proapoptotic function of p75-NTR comes from the p75-NTR null mouse, in which reduced developmental apoptosis is detected in the retina and spinal cord (Trade and Barde 1999). p75-16 NTR can drive apoptosis via activation of sphingomyelinase, production of ceramide, activation of c-jun N-terminal kinase (INK) and accumulation of p53 (Jarvis et al 1994, figure 2.1.3). There is recent evidence that caspases-9, -3, and -6 are required for p75-mediated apoptosis of hippocampal neurons (Troy et al 2002). However, there is little evidence thus far to support the activation of this pathway in ORNs. However, JNK is induced after p75 activation in sympathetic neurons (Bamji et al 1998), and p53 is implicated in p75-dependent apoptosis in sympathetic neurons induced by nerve growth factor (NGF) withdrawal (Aloyz et al 1998), so it is not unlikely that a similar pathway to this could drive ORN apoptosis. p75-NTR is widely expressed in the nervous system during development, but in the adult CNS is confined to neurons of the basal forebrain, caudate, putamen and Purkinje cells (Barker 1998). Within the olfactory system, the p75-NTR is expressed chiefly by olfactory ensheathing glia (OEG) (Ramon-Cueto and Avila 1998), where its glial expression is maximal during development or after lesion in the olfactory nerve layer of the bulb (Gong et al 1994, Turner and Perez-Polo 1993). Neuronal p75-NTR immunoreactivity has also been demonstrated in ORNs of the neonatal and adult rat (Turner and Perez-Polo 1992) and the synaptic terminals of ORNs (Turner and Perez-Polo 1992, Roskams et al 1996). Dissociated immature ORNs in primary culture also express p75-NTR, as do placodally-derived ORN cell lines (Cunningham et al 1999, Illing et al 2001). Deafferentation down-regulates glomerular p75 expression, concomitant with an increase in p75-NTR in the olfactory nerve layer (Turner and Perez-Polo 1993, 1994). A n immunotoxin that targets p75-NTR-expressing neurons causes a dramatic loss of synapses in the olfactory bulb (Pallera et al 1994). Thus, p75-NTR is ideally situated to mediate survival/death signalling between the ORN and its target, the mitral cell of the olfactory bulb. Clusterin/ApoJ is up-fegulated during apoptosis in a number of tissues (Buttyan et al 1989). During post-bulbectomy ORN apoptosis, D N A fragmentation has been found to correlate closely with an increase in clusterin mRNA (by RT-PCR) in the lesioned OE. The induction of clusterin expression, however, was not found within ORNs but was shown to be in OEGs (Michel et al 1997). Interestingly, while clusterin and glial fibrillary acidic protein (GFAP) co-localize in the OEGs in the lamina propria as expected, clusterin-positive /GFAP-negative cells were seen above the basal lamina in the post-bulbectomy OE. The authors suggest that clusterin could therefore be made in and secreted by the OEGs before being internalized by cells of the neuroepithelium. Thus, although clusterin is not directly involved in the apoptotic program of ORNs, a link has been made between clusterin and ORN apoptosis. Intracellular early pro-apoptotic pathways in ORNs Bcl-2 is an anti-apoptotic proto-oncogene first identified in B-Cell Lymphoma that when mutated can prevent the normal apoptotic response (reviewed in Gross et al 1999, Pellegrini and Strasser 1999). Bcl-2 can block apoptosis in many neurons when overexpressed, and has been shown in vitro to rescue neurons from neurotrophic factor deprivation-induced apoptosis (Garcia et al 1992, Allsopp et al 1993). Although Bcl-2 expression has not been reported in ORNs, Bcl-2 is expressed normally in the mitral cells of the olfactory bulb (Castren et al 1994). Bcl-2 knockout mice have the full complement of neurons at birth, but exhibit progressive degeneration of several neuronal types postnatally, including sensory neurons of the trigeminal ganglion (Michaelidis et al 1996, Pinon et al 1997). Bcl-2 therefore appears to be necessary for the survival of specific sensory neuron populations in the early postnatal period, a time when ORNs are making many new connections with the olfactory bulb. Bcl-2 overexpression does 17 protect ORNs from bulbectomy-induced apoptosis (Jourdan et al 1998). This protection from apoptosis is long-term: ORNs remain TUNEL-negative at 5 days post-bulbectomy. p53 is a tumour suppressor gene (Michalovitz et al 1990) involved in negative regulation of the cell cycle (Kastan et al 1992). p53 exerts a proapoptotic effect by inducing the Bax promoter, leading to an increase in the Bax/Bcl-2 ratio (Miyashita and Reed 1995). This change in Bax:Bcl-2 ratio can then lead to the activation of mitochondrial pro-apoptotic pathways (figure 2.1.3). p53 null mice do not exhibit a significant change in the apoptotic profile (TUNEL-positivity) of ORNs (compared with wildtypes) up to 5 days post-bulbectomy, however, suggesting that apoptosis of mature ORNs following bulbectomy is p53-independent (Calof et al 1996b). As p53 is more associated with apoptosis in proliferating cell populations (Lowe et al 1993) the O R N precursor population may be more likely to undergo p53-dependent apoptosis than the mature ORNs. However, p53 is present in ORNs in more mature regions of the OE, and its expression is more widespread following axotomy (Huang et al 1995). p53 is also expressed in the olfactory bulb (Wong et al 2000). Transcription factors implicated in ORN apoptosis There is an absolute requirement for R N A and protein synthesis in the early phases of apoptotic induction (Wyllie et al 1984, Martin et al 1988). Therefore, the early transcriptional events that occur during apoptotic induction are of particular importance in understanding the pathways used by a given neuronal group to achieve terminal apoptosis. Within ORNs, these same transcriptional changes may also drive the events needed for ORNs to signal their replacement to localized progenitors. A transient increase in c-fos mRNA expression (indicative of the early stages of apoptosis) is seen in apoptotic ORNs, peaking at 16 hours after bulbectomy in adult mice (Michel et al 1994). C-jun is necessary for neuronal apoptosis in vitro (Estus et al 1994, Ham et al 1995), is expressed during neuronal apoptosis in development (Ferrer et al 1996) and in a wide variety of experimental models of apoptosis. In the olfactory system, C-jun is expressed in ORNs close to the basal cell layer, and in the periglomerular, tufted, mitral and granule cells of the olfactory bulb (Baba et al 1997). 1-3 days following olfactory nerve transection c-jun immunoreactivity disappears from the OE, and by day 9 has returned to neurons in the basal area (Huang et al 1995). As C-jun is also implicated in a variety of growth regulation pathways (both mitogenic and differentiation-promoting), it could also be serving this role (and have nothing to do with apoptosis) in progenitors of the ORN lineage (Vogt and Bos 1990). Terminators of ORN apoptosis A variety of in vitro and in situ systems have been employed to establish the seminal interactions between facilitators of the apoptotic cascade (figure 2.1.3). The final stage of apoptosis, called execution or termination, occurs through the activation of caspases, a highly conserved family of cysteine proteases with specificity for aspartic acid residues in their substrate, which are the mammalian homologues of C. elegans CED-3 (Xue et al 1996). This highly conserved protease family now contains 14 mammalian family members, at least six of which have a clearly defined role in apoptosis (Earnshaw et al 1999). The cleavage of caspase substrates leads a cell to malfunction, and ultimately orchestrates the final death-blow for a cell to die and be cleared. Caspases have been shown to be involved in apoptosis in many cell types including motoneurons, sympathetic ganglion cells, fibroblasts, interdigital cells, mammary epithelium, lymphocytes and fly ommatidia (reviewed in Milligan and Schwartz 1997). 18 The mRNA for caspases-1, -2 and -3 has been detected by RT-PCR in the OE of both adult and foetal rats (Suzuki and Farbman 2000). Caspase-2 mRNA is absent at 3 and 5 days post-bulbectomy, but reappears by 21 days, suggesting that caspase-2 may be expressed in ORNs. Pan-caspase inhibitors also blocked apoptosis (in a dose-dependent manner) induced by TNF-a in organotypic cultures of OE, which normally causes a 2.5-fold increase in apoptotic bodies in the OE. The pan-caspase inhibitors also reduce apoptosis under control conditions, i.e. in the absence of exogenous TNF-a, suggesting that normal cell death in the OE also involves caspases. Knockout mice have been utilized to highlight the critical importance of caspase-9, an initiator caspase, and caspase-3, an executioner caspase, in neuronal apoptosis during development (Hakem et al 1998, Kuida et al 1996, 1998). This thesis will add to the literature by attempting to prove that caspases-3 and -9 are a major pathway to drive ORN apoptosis, both in development and in adulthood, as has been suggested for other neuronal populations. Caspase cleavage targets Final execution of the apoptotic program is carried out by a variety of caspase substrate proteins, which themselves are responsible for the targeted shutdown of cellular machinery (reviewed in Cryns and Yuan 1998). ORNs do possess a number of the classic caspase cleavage targets found in non-neuronal cells such as poly A D P ribose polymerase and inhibitor of caspase-activated DNAse (which enzymatically produces the D N A fragments characteristic of apoptosis). They do, however, also contain a number of alternative caspase target proteins which could be more neuron-specific in their action. For example, the Amyloid Precursor Protein (APP) homologue, amyloid precursor-like protein 2 (APLP2), which as I will show in this thesis is not only present in the axons of ORNs, but is also cleaved in vivo in a caspase 3-dependent manner. These results are presented in section 5.4, and discussed further in the general discussion, section 7.2. LOSS OF OLFACTION AND NEURODEGENERATIVE DISEASE A gradual loss of smell perception occurs with aging, beginning on average around 60 years and becoming markedly more severe after 70 years (Doty et al 1984, Schiffman 1993, Schiffman and Gatlin 1993). A loss of olfaction is also a central component of some neurodegenerative diseases, such as Alzheimer's disease (Doty et al 1987, Eichenbaum et al 1983, Serby et al 1985a,b), Parkinson's disease (Ansari and Johnson 1975, Ward et al 1983, Doty et al 1988), Huntington's disease (Moberg et al 1987, Bylsma et al 1997, Moberg and Doty 1997) and Multiple Sclerosis (Pinching 1977, Constantinescu et al 1994, Doty et al 1999). Indeed, olfactory deficit is often one of the first symptoms of some neurodegenerative diseases (Doty et al 1991). Patients with Alzheimer's disease have morphological and biochemical abnormalities throughout the olfactory pathway, including at the level of the ORN (Talamo et al 1989). Whether these changes reflect a corresponding change in the rate of ORN apoptosis or their replacement, which could represent neurodegenerative events occurring deeper in the CNS, has yet to be determined. 19 FUTURE DIRECTIONS Despite the progress made in identifying components of the apoptotic machinery utilized by ORNs, we still have a long way to go to understand how apoptotic signals can be coordinated in a cell that exists in such a complex three-dimensional environment. Even within the OE, given that ORNs can die at any stage in their development, we have yet to establish the signals (secreted or cell-cell) that initiate this response, or determine whether their underlying cytoplasmic apoptotic pathways can change from one stage of development to the next. Is loss of olfaction in aging a consequence of a change in the efficiency of O R N signalling, an increase in the apoptotic rate of mature ORNs, or due to a loss in the ability of ORNs to be replaced by their progenitors? Although there are long-established links between neurodegenerative disease and loss of olfaction, we are still a long way from understanding which part of the olfactory system is specifically impacted in these disorders, or from establishing the cellular basis for this change. Thankfully, technology is now on our side. With the identification of more neurodegenerative disease-related genes and their modifiers (via the human genome sequence), the apoptotic pathways that impact them (courtesy of model organism genetics), the production of viable animal models for disease (improved techniques in transgenic production), and advances in diagnostic approaches to examine changes in state (Array technology), this will be an area that will witness considerable progress in the coming years. 20 2.2 Caspases Various members of the caspase family of proteases (Alnernri et al 1996) are crucial in the series of molecular events that take place during apoptosis in different cell types. Caspases are all homologues of C. elegans CED-3. CED-3 was discovered as a gene in the absence of which developmental programmed cell death does not occur (Hengartner and Horvitz 1994), and homology to a family of mammalian proteins was soon discovered (Yuan et al 1993). Although at least 14 members of the highly conserved caspase family are now known in mammals, many of which have a clearly defined role in apoptosis (reviewed in Earnshaw et al 1999), it is not clear that all caspases are involved in the apoptotic process. What is clear from a number of different experiments in cell lines and in situ experiments is that there is a close and direct interaction between different members of the caspase family and that cell death results from an enzyme cascade of mutual activation (Slee et al 1999, McCall and Steller 1997, Nicholson and Thornberry 1997, Salvesen and Dixit 1997). The particular caspases involved in a given apoptotic situation are likely to be cell type and stimulus specific. Caspases exist as proenzymes, consisting of an N-terminal prodomain and a large and small subunit. The cysteine-containing catalytic site is located within the large subunit. In general, the proenzyme form has little or no catalytic activity itself. The active form consists of a heterotetramer of two large and two small subunits. There is more than one way of achieving activation and the formation of the heterotetramer, depending upon the type of caspase. Some caspases, in particular the "effector" caspases-3, -6, and -7, normally exist as proenzyme dimers. This is generally thought of as the inactive form, although a very, low level of activity has been detected in the caspase-3 proenzyme (Nicholson and Thornberry 1997). For full activation, enzymatic cleavage of the prodomain is required, allowing the heterotetramer to form. In contrast, some upstream "initiator" caspases such as -8, -9 and -10 normally exist in their inactive form as monomers, and do not require N-terminal prodomain cleavage for activation. This scheme of retaining the prodomain on the large subunit as part of the active heterotetramer is illustrated in figure 5.5.1. Because of the way in which these upstream caspases are recruited to adaptor proteins, many molecules of the caspase in question are clustered together prior to activation. The so-called "induced proximity" model of autoactivation proposed that any low levels of proenzyme activity would thus be sufficient to cleave a neighbouring molecule and activate it (Martin et al 1998, Muzio et al 1998, Srinivasula et al 1998, Yang et al 1998). While this may indeed occur, it has also subsequently been discovered that caspases-8, -9 and -10 are activated by forming dimers, and that proteolytic cleavage is not necessary for activation (Renatus et al 2001). Mice null for different mammalian caspases have been useful in beginning to understand the roles of individual caspases. As mentioned in the Introduction, both caspase-3 and caspase-9 null mice have a severe neuronal phenotype, seemingly a result of reduced neuronal apoptosis in development (Kuida et al 1996, 1998, Hakem et al 1998). There is no obvious neuronal phenotype in mice lacking many of the other caspases including caspase-1 (Kuida et al 1995), caspase-2 (Bergeron et al 1998), caspase-11 (Wang et al 1998) and caspase-12 (Zheng et al 2000); however, this does not mean that they could not play a more subtle role. Caspase-1 null mice are deficient in cytokine processing (Kuida et al 1995). Caspase-8 null mice are early embryonic lethal (Varfolomeev et al 1998). Caspase-2 null mice have mild defects affecting 21 oocyte ablation (Morita et al 2001). Caspase-14 is involved in keratinocyte differentiation (Elchart et al 2000, Lippens et al 2000). Since the discovery of mammalian caspase-3 in 1993 there has been plenty of evidence for the involvement of caspase-3 in apoptosis, and some highlights follow. In 1996 it was demonstrated that in cell-free extracts of healthy non-neuronal cells, addition of caspase-3 is sufficient to trigger apoptosis (Enari et al 1996). It was subsequently shown that pro-caspase-3 is cleaved into its active form during apoptosis of many cell types, including cerebellar granule neurons (Du et al 1997). Specific peptide active-site inhibitors of caspase-3 block death of cultured neurons in many different situations (reviewed by Vil la et al 1997). However, caspase-3 activation is not involved in all types of neuronal cell death. Other caspases may be used as a function of cell type or, interestingly, as a function of the death stimulus. Thus, death of sympathetic neurons induced by NGF removal requires a caspase-2-like enzyme, whereas following superoxide dismutase depletion, apoptosis of the same neurons involves caspase-1 (Troy et al 1996). There is evidence for caspase activation in a number of neurodegenerative diseases, including Alzheimer's disease (caspases-3, -6 and -9; Chan et al 1999, LeBlanc et al 1999, Stadelmann et al 1999, Lu et al 2000), Parkinson's disease (caspases-3, -8 and -9; Anglade et al 1997, Jeon et al 1999, Viswanath et al 2001), Huntington's disease (caspases-1 and -8; Sanchez et al 1999), and amyotrophic lateral sclerosis (caspases-1 and -3; Pasinelli et al 1998). There are limitations to the interpretation of these data, reviewed in Troy and Salvesen 2002. Inhibitors of caspase-3 are able to prevent or delay neuronal apoptosis in several clinically relevant model systems. For example, specific caspase-3 inhibitors have been shown to prevent apoptosis following axotomy in rat retinal ganglion cells in vivo (Kermer et al 1998), to reduce apoptosis following traumatic brain injury in rats (Yakovlev et al 1997), to reduce neuronal loss in various ischemia models in the rat (Lam et al 1999, L i et al 2000), and specifically to delay or reduce apoptosis following ischemia in the rodent (Fink et al 1998, Chen et al 1998, M a et al 1998). In addition, the broader caspase inhibitor zVAD-fmk has been used successfully to prevent apoptosis in pneumococcal meningitis in human neurons in vitro (Braun et al 1999), and to reduce apoptosis in a cold injury-induced model of brain trauma in mice (Morita-Fujimura et al 1999). A broad caspase inhibitor can also delay caspase-3-associated apoptotic death following global ischemia in the gerbil (Himi et al 1998), and reduce overall neuronal loss in neonatal hypoxic-ischemic brain injury in the rat (Cheng et al 1998). 22 FIGURE 2.1.1 Features of apoptosis In response to a proapoptotic signal (black circles), cells undergo the definitive morphological changes of apoptosis such as shrinkage, membrane blebbing, chromatin condensation and marginization, and finally formation of apoptotic bodies and phagocytosis (top row). Concomitant with this are molecular changes such as selective proteolysis, D N A fragmentation and modification of membrane proteins by phosphatidyl serine exposure on the cell surface (bottom row), which serves as a signal for macrophage (M) infiltration (from www.nih.gov/sigs/aig/aboutapo.html). 2 3 a) bulbectomy nasal cavity b) olfactory nerve transection FIGURE 2.1.2 Experimental models to manipulate ORN death ORNs can be induced to die by one of three experimental paradigms: (a) olfactory bulbectomy, which is both a target deprivation and a deafferentation; (b) transection of the olfactory nerve at the cribriform plate; or (c) direct intranasal application of a toxic chemical such as methyl bromide, (d) ORN death can also occur by manipulated sensory deprivation (naris occlusion). 24 FIGURE 2.1.3 Pathways implicated in ORN apoptosis Of all the dozens of different signals which have been implicated in initiating apoptosis (at the cell membrane or intracellularly), propagating pro-apoptotic signals (through mitochondrial or non-mitochondrial pathways) or acting to terminate apoptosis (activating caspase-3 or its cellular homologues), the above pathways have been implicated in some way in the ORN. Specific apoptotic molecules that there is some evidence for in the O R N are underlined. 25 C-jun p53 fas P75NGFR FIGURE 2.1.4 Apoptotic molecules within Olfactory Receptor Neurons p75 is found at the ORN synaptic terminal, consistent with the possibility that it might initiate an apoptotic signal from the olfactory bulb. The fas ligand is found at the ORN dendrites, where it would be in a position to initiate apoptosis following an environmental insult. The proapoptotic transcription factors p53 and c-jun are found in the cell bodies of some ORNs, although their exact role in ORN apoptosis is not clear. The missing link is that until this thesis, no effectors of apoptosis have yet been placed in the ORN, and no apoptotic molecules have yet been demonstrated along the ORN axon, to propagate a signal between receptor and cell body. 26 CHAPTER 3: AIMS AIM1: TO TEST THE HYPOTHESIS THAT CASPASE-3 AND CASPASE-9 ARE INVOLVED IN MATURE ORN APOPTOSIS IN VIVO These two caspases are the most strongly implicated in neuronal apoptosis out of those thus far detected in the OE. Knockout mice lacking each of these proteins have been generated, and show similar gross abnormalities of the brain, attributed to lack of developmental apoptosis, as will be described further in Aim 2. Caspase-3 has also been shown to cleave a number of proteins that play a role in neurodegenerative diseases, such as huntingtin (Goldberg et al 1996), APP (Barnes et al 1998, Gervais et al 1999), presenilins 1 and 2 (Kim et al 1997, Loetscher et al 1997), and ataxin-3 (the affected gene product in spinocerebellar ataxia type 3, Wellington et al 1998). Parkin also has a putative caspase-3 cleavage site (unpublished observation). Inhibitors of caspase-3 are able to prevent or delay neuronal apoptosis in several clinically relevant model systems, such as traumatic brain injury and stroke. Some specific examples are listed in the Literature Review, section 2.2. The importance of caspase-9 in neuronal apoptosis in vivo has not been as widely studied, but based on the knockout phenotype one might expect it to be as important as caspase-3, at least during development. To address the involvement of caspase-3 and caspase-9,in mature O R N apoptosis in vivo, I examine the expression and activation of caspase-3, caspase-9 and their downstream targets, at key timepoints during apoptosis in mouse ORNs. TUNEL analysis after bulbectomy Olfactory bulbectomies are performed to deprive ORNs of target-derived trophic factors and hence induce neuronal apoptosis in vivo, which is confirmed by D N A laddering and T U N E L . OE tissue is obtained 4-72 hours following bulbectomy, and spatial and temporal analysis of patterns of T U N E L labelling will be performed. Analysis of caspase expression and activation The T U N E L analysis data will be correlated with the expression pattern and cleavage pattern of proteins of interest, which are monitored by western blotting and immunohistochemistry over the timecourse of apoptosis. Proteins of interest include caspase-3, caspase-9, and the specific downstream target of caspase-3, APLP2.1 use western blotting to determine which cleavage forms of caspases are found in the OE at which key times postbulbectomy, and immunohistochemistry (including double immunofluorescence) to confirm the cellular and subcellular localization. TUNEL on adult caspase-3 knockout mice postbulbectomy To confirm whether caspase-3 is necessary for mature ORN apoptosis after bulbectomy, I analyse bulbectomized tissue from caspase-3 null mice by T U N E L and histology, and ask whether apoptosis occurs. I expect that mature ORNs will be able to die without caspase-3, but that this death will be delayed. 27 It is known that ORNs dying after bulbectomy do so by apoptosis (Morrison and Costanzo 1989, Michel et al 1994, Holcomb et al 1995). This would be the first time that anyone has attempted to demonstrate the three-dimensional pattern of retrograde apoptosis throughout different turbinates of the OE. Such a demonstration will be necessary in order to fully analyse caspase activation in the ORN. This is the first time that caspase involvement in ORNs in vivo has been addressed. I describe in section 1.3 why the olfactory bulbectomy paradigm is a good system in which to study mechanisms of neuronal apoptosis. One of the major aims of this thesis is to determine which pathways act in ORNs in vivo as the neurons undergo apoptosis. The olfactory bulbectomy model gives us a means to study and analyse neuronal cell death in real neurons as they are dying. Despite significant progress in understanding the enzymological process of neuronal apoptosis, essentially nothing is known about how a neuron, which exists in a complex three-dimensional environment in vivo, integrates the pro and antiapoptotic signals received from different parts of the neuraxis. I take advantage of the unique structural and organizational features of the OE to examine where caspases are activated in mature ORNs after deafferentation, and how this activity is propagated in space and time. Considering the number of developmental mechanisms that are also conserved between olfactory neurons and other peripheral and central neurons (reviewed in Roskams et al 1996), I believe that studying the signal transduction mechanisms that contribute to apoptosis in the OE will provide insight not only into how olfactory neurons die, but also into how this process may be regulated in other neuronal groups. 28 AIM 2: TO INVESTIGATE THE ROLE OF CASPASE-3 AND CASPASE-9 IN OLFACTORY SYSTEM DEVELOPMENT USING NULL MICE. Both caspase-3 and caspase-9 null mice are known to have grossly abnormal brain development. I will assess whether they have abnormal olfactory system development. Almost nothing is known about the involvement of caspases in programmed cell death in the developing olfactory system. The main questions asked in this section are: • What is the expression pattern of caspase-3 and caspase-9 in the normal mouse embryo OE? • Is apoptosis abnormal in the olfactory system of caspase-3 and caspase-9 null mice? • Does the caspse-3 null mouse have gross olfactory structural abnormalities? As mentioned in the Introduction, caspase-3 and caspase-9 null mice both have dramatically abnormal brains, with no gross abnormalities in any other organ. The gross morphology of the two brain phenotypes is very similar. The major features are ectopic neuronal tissue protruding from the skull, disruption of cortical structure, expansion of the ventricular zone, and sometimes hydrocephaly, seemingly all a result of decreased neuronal apoptosis (Kuida et al 1996, al 1998, Hakem et al 1998). The caspase-9 phenotype is more severe than the caspase-3 in that caspase-9 null mice generally do not live beyond P3, whereas caspase-3 null mice are reported to live 3-6 weeks postnatally (Kuida et al 1996). The caspase-3 and caspase-9 null phenotypes have been cited as evidence that these two caspases play a crucial role in apoptosis in the developing brain. In addition, it has been shown that there is reduced T U N E L in caspase-3 and caspase-9 null mouse embryonic brains (Srinivasan et al 1998 and Hakem et al 1998 respectively). However, the question of whether caspase-3 and caspase-9 are important in developmental apoptosis specifically in the olfactory system has never been addressed. Apoptosis in the developing olfactory epithelium I will use T U N E L to define the timecourse of apoptosis in the OE at a range of embryonic and neonatal timepoints. There is reported to be an early peak of apoptosis at E l 1-13 (Pellier and Astic 1994), but it is not quite clear what happens subsequently. Caspase-3 and caspase-9 expression and activation in the normal mouse embryo OE I will examine the expression of caspase-3 and caspase-9 during embryonic development of olfactory neurons in normal mice. I will carry out immunohistochemistry with antibodies against caspase-3 and caspase-9 on sections from a range of embryonic and neonatal timepoints, and ask whether the expression pattern of these proteins puts them in the right place at the right time to be potentially involved in apoptosis. TUNEL in the olfactory system of developing caspase-3 and caspase-9 null mice The next question would be, what is the olfactory phenotype of caspase-3 and -9 nulls? Do they display a developmental profile that suggests caspases mediate olfactory neuronal cell death during embryogenesis? There is no mention in the literature of an olfactory phenotype (Kuida et al 1996, 1998, Hakem et al 1998), although caspase-9 knockout mice are said to have enlarged OE and olfactory bulb (Keisuke Kuida, personal communication). I will carry out T U N E L staining in both strains of null mutant embryos at E l 3 (the usual peak of TUNEL+ ORNs) and 29 compare TUNEL-positivity in the developing olfactory system with that seen in control littermates. I will also carry out T U N E L staining in caspase-3 null mice and controls at P4, when I can ask whether glomerular formation and bulb formation in general have been impaired, and by which time loss of caspase-3 is causing some lethality. Does the caspase-3 knockout mouse have gross olfactory structural abnormalities? As mentioned, there has been a suggestion that caspase-9 knockout mice do have enlarged OE and olfactory bulbs as a result of failure of caspase-9-dependent developmental apoptosis (Kuida), but there is no mention in the literature of any olfactory abnormality in caspase-3 knockout mice. I will address the question of whether the caspase-3 null mice have enlarged olfactory bulbs. To test this hypothesis I will prepare sets of 20 u,m sections of entire bulbs from three wildtype and three caspase-3 null mice. I will measure the cross-sectional area of every 8 th section, plot these measurements against the depth into the bulb, and from the area under the curve find the volume of the bulb. A subset of the sections measured will be immunolabelled with neural cell adhesion molecule (NCAM), OMP or T H to help determine whether connections are properly set up, which cells are affected, and whether the glomeruli have formed normally. Do the caspase-3 null mice attempt to compensate for their deficiency by up-regulating caspase-9? The similarity of the caspase-3 and caspase-9 knockout phenotypes has been used to support the argument that the two act as part of the same pathway. There is also strong in vitro evidence in support of the ability of caspase-9 to activate caspase-3, and furthermore there is evidence that caspase-3 can also activate caspase-9 in a positive feedback loop (Thornberry et al 1997, Fujita et al 2001). I therefore wanted to investigate whether the congenital lack of one of these caspases influences the expression level of the other, either negatively due to the lack of positive feedback, or positively in an attempt at compensation. Any such compensation might account in part for the variation in phenotype seen in the caspase-3 nulls in terms of longevity. 30 CHAPTER 4: METHODS Mice and Genotyping CD-I mice were used except where specified otherwise. Caspase-3 null transgenic mice were obtained from D.W. Nicholson, Merck Frosst. Caspase-9 null mice were obtained from K . Kuida, Vertex Pharmaceuticals. Obtaining a D N A sample from caspase-3 and caspase-9 mice Tissue used to obtain a D N A sample was ~5 mm of tail tip from an adult; whole tail from a neonate or embryo E l 5 and older, or tail and hindquarters from an E10 or E l 3 embryo. For general genotyping in the colony, tail tip samples were taken from anaesthetized animals at time of weaning. If an adult was subsequently used in an experiment, a further tail tip sample was taken at time of sacrifice for confirmation. In the case of embryos and neonates whose tissue was used for an experiment, a D N A sample was obtained at the time of sacrifice. The tissue sample was stored when necessary prior to processing at -20°C. A l l tissue samples were processed in the following way: add 300 pi modified STE buffer (50 m M tris-HCl, 10 m M EDTA, 100 mM NaCl, 1% SDS) and 18 ul of 20 mg/ml proteinase K ; digest at 55°C overnight; vortex; heat at 100°C for 10 minutes. 100 ul of this digest was taken for phenol/chloroform extraction. The remainder was stored at 4°C. Phenol/chloroform extraction was performed as follows: to 100 ul add 100 ul phenol, mix for 5 minutes; spin for 5 minutes at 13,000 rpm; transfer aqueous layer and add 100 ul of phenol/chloroform (1:1), mix for 5 minutes; spin for 5 minutes at 13,000 ipm; transfer aqueous layer and add 125 ul chloroform, mix for 5 minutes; spin for 5 minutes at 13,000 rpm; transfer aqueous layer and add 250 ul 95% ethanol, invert and allow to precipitate at -20°C for 30-60 minutes; spin for 5 minutes at 3,500 rpm; wash D N A pellet in ethanol, pellet again, dry D N A and resuspend in 20 ul water. 1 pi of this final D N A solution was then used in PCR. Polymerase chain reaction (PCPQ Primers used for caspase-3 PCR: Common forward primer: 5' > A A G CTG TCT TCG TCC A G T G A G < 3' Null reverse primer: 5' > GTC G A T C C A C T A GTT C T A G A G C G G C < 3' Wildtype reverse primer: 5' > C T A A G T T A A C C A A A C T G A G C A C C G A < 3' PCR mix: 50 pi containing 1 ul D N A sample, 200 uM dNTP, 2.5 ul dimethyl sulphoxide, 50 pmol primers, 750 u,M M g C l 2 , 0.25 ul taq polymerase (Gibco), PCR buffer (Gibco). PCR conditions: hold (2 minutes 94°C), 30 cycles (30 seconds 94°C denaturing, 1 minute 60°C annealing, 1 minute 72°C extension), hold (5 minutes 72°C). Primers used for caspase-9 PCR: Null forward primer: 5' > TCT CCT CTT CCT C A T CTC C G G GCC < 3' Null reverse primer: 5' > G A A C A G TTC G G C T G G C G C G A G C C C < 3' Wildtype forward primer: 5' > A G G C C A GCC A C C TCC A G T TCC < 3' Wildtype reverse primer: 5' > C A G A G A TGT G T A G A G A A G CCC A C T < 3' PCR mix: 50 pi containing l u l D N A sample, 200 p M dNTP, 40 pmol primers, 1.5 m M M g C l 2 , 0.25 pi taq polymerase (Gibco), PCR buffer (Gibco). PCR conditions: hold (2 minutes 94°C), 31 30 cycles (30 seconds 94°C denaturing, 30 seconds 61°C annealing, 1 minute 72°C extension), hold (5 minutes 72°C). Bulbectomy Olfactory bulbectomies were performed on adult mice as described previously (Roskams et al 1994, 1996, von Koch et al 1997) (figure 1.3.3). Bulbs were removed unilaterally for subsequent analysis of OE by immunohistochemistry, T U N E L or in situ hybridization (n=3-4/timepoint); or bilaterally for OE protein preparation (n=6/timepoint). Animals were sacrificed 0, 4, 8, 12, 24, 36, 48 or 72 hours following surgery. For partial bulbectomies (used in section 5.4), the centre of the olfactory bulb (including the mitral cell bodies) was removed stereotaxically, leaving the nerve fibre layer and glomerular layer relatively intact (confirmed both visually upon dissection and histologically). Bulbectomy data presented in Chapter 5 were obtained from CD-I mice unless stated otherwise in the text. Where so stated, bulbectomies were also performed on adult caspase-3 null mice and their littermate controls. Tissue Preparation To prepare tissue for immunohistochemistry, T U N E L or in situ hybridization: Adults, fixed and fresh frozen: Adult mice were anaesthetized with 120 pi of Xylaket (25% Ketamine H C L (MTC Pharmaceuticals), 2.5% Xylazine (Bayer Inc.), 15% ethanol, 0.55% NaCl) and rapidly perfused with cold phosphate-buffered saline (PBS). If the tissue was to be fresh frozen, brains, olfactory bulbs and olfactory epithelia were dissected, and I then proceeded directly to the embedding step. If the tissue was to be paraformaldehyde (PFA)-fixed, then perfusion with PBS was immediately followed by perfusion with 4% PFA in PBS. Brains, olfactory bulbs and olfactory epithelia were dissected out and post-fixed in 4% PFA for 2 hours at 4°C. Tissue was then equilibrated in 10% sucrose followed by 30% sucrose for 12-20 hours each at 4°C, before proceeding to the embedding step. Fresh frozen adult tissue was used for T U N E L , alkaline phosphatase method (section 5.1, figures 5.7.1 and 6.5.1), and for immunofluorescence (sections 5.2, 5.4, 5.5 and 6.5). Fixed adult tissue was used for investigation of the abnormalities in unlesioned caspase-3 null adults (section 6.6), as some antibodies such as anti-tyrosine hydroxylase work optimally with this fixation condition. Neonatal PFA-fixed tissue: P4 mice were anaesthetised with AErrane inhalation anaesthetic (Janssen, Toronto, ON), and cardiac perfused by hand with 2 ml of ice-cold PBS followed by 3 ml of 4% PFA in PBS. The mice were decapitated, and whole heads were post-fixed in 4% PFA for 2 hours. Heads were then equilibrated in 10% sucrose followed by 30% sucrose for 12-20 hours each at 4°C. The skin and the tips of the noses were then removed, before proceeding immediately to the embedding step. Neonatal fixed tissue was used in sections 6.2, 6.3 and 6.7. PFA-fixed embryos: In the case of embryos, mothers were anaesthetised as indicated for adults, and the embryos were dissected out and immersion-fixed in 4% PFA overnight. They were then equilibrated in 10% sucrose followed by 30% sucrose for 12-20 hours each at 4°C, before proceeding to embedding. Fixed embryo tissue was used in sections 6.2, 6.3, 6.4 and 6.5. Embedding: A l l tissues were then equilibrated (under suction except for embryos) in warm Tissue-Tek embedding medium (Sakura Finetek, Torrance, CA) for 5 minutes; and frozen in liquid nitrogen. 20 u,m carefully numbered coronal sections of adult OE and olfactory bulb, 32 sagittal sections of P4 heads, and sagittal sections of whole embryos were prepared by cryostat sectioning and stored at -20°C for subsequent analysis. For experiments involving immunohistochemistry, TUNEL, in situ hybridization, or measurements of anatomical structures, tissue was always from a minimum of three animals per age, genotype or lesion condition. Preparation of protein homogenates for immunoblotting or immunoprecipitation Adult mice were sacrificed by decapitation following anaesthesia as indicated above. Olfactory tissue and control brain areas were dissected out without perfusion (Roskams et al 1995, 1996, 1997), snap frozen in liquid nitrogen, before being homogenized in buffer A (50 m M Tris-HCl pH 8; 150 m M NaCl; 1% Triton X-100; 1 pg/ml aprotinin; 1 /*g/ml leupeptin; 100 #g/ml phenylmethylsulfonyl fluoride). Each set (timepoint, genotype or treatment group) consisted of pooled tissues of three animals. Western blot analysis was performed on at least two different sets of animals (total n=6). Following estimation of protein by the B C A method (Pierce), tissue homogenates were frozen at -80°C for subsequent analysis. Tissue prepared in this way was used in figures 5.2.1,5.5.2 and section 5.6. " „ , Immunoblotting Aliquots of tissue homogenates (25 fig protein) were subjected to SDS-polyacrylamide gel electrophoresis and transferred to Immobilon membrane (Millipore). Equal loading of lanes was confirmed by Ponceau S staining. Membranes were blocked for 1 hour at room temperature with 5% non-fat milk in tris-buffered saline (TBS); incubated for 12-20 hours at 4°C in primary antibody in 2% milk/TBS; washed for 3x 5 minutes in 0.1% Tween-20 in TBS; incubated for 1 hour at room temperature in peroxidase-coupled goat anti-rabbit IgG (BioRad) diluted 1:10,000 in 2% milk/TBS; washed 3x 5 minutes; and detected with chemiluminescent substrate (Pierce). For details of all the primary antibodies used for immunoblotting see table 4.1. Immunoprecipitation Immunoprecipitation of Delta-C cleavage products (figure 5.3.1) was performed by incubating 200 pg of tissue lysate (prepared as described above) with affinity-purified D2-IIAPLP-2-specific antibody (gift from Gopal Thinakaran) at a dilution of 1:250, in a final volume of 500 pi modified RIPA buffer (150 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholate, 0.1% SDS, 50 m M Tris pH 8.0, 1 pg/ml aprotinin, 1 p,g/ml leupeptin, 100 pg/ml PMSF). After incubation at 4°C, APLP2-containing immunocomplexes were isolated by further incubation at 4°C with protein A-conjugated Sepharose CL-2B beads (Sigma), washed and separated on gel containing a 4-20% poly aery lamide gradient. 33 Immunohistochemistry For light microscopy detection of one antigen When fresh frozen sections were used, they were first fixed in 4% PFA in PBS for 10 minutes. A l l sections were then permeabilized in 0.1% triton X-100 in PBS for 30 minutes; blocked with 4% normal serum in PBS for 20 minutes; incubated at 4°C for 12-20 hours with primary antibody in PBS; and rinsed for 2x 5 minutes with PBS. To detect primary antibody binding, sections were incubated for 30 minutes at room temperature in biotin-conjugated anti-rabbit, anti-mouse or anti-goat secondary antibodies diluted 1:200 in 1 part 4% normal horse serum to 50 parts PBS. Sections were treated to quench endogenous peroxidase with 0.5% H 2 0 2 in PBS for 10 minutes; rinsed in PBS; incubated with avidin-biotin-HRP complex kit (Vector Laboratories, Burlingame, CA) for 30 minutes, as per manufacturer's instructions; and rinsed for 2x 5 minutes in PBS. For colour detection of the peroxidase, I use either the peroxidase substrate kit Vector VIP (purple, figures 5.1.7, 5.5.3, 5.6.3, 6.2.1, 6.2.2, 6.5.1, 6.5.2c, 6.6.4, 6.7.2 and 6.8.1), or D A B (brown, figures 5.2.2 and 5.3.1) (again, both from Vector Laboratories, used as per manufacturer's instructions). The colour reactions were quenched with distilled water. Sections were either mounted with aquapolymount (Polysciences) or, for long-term storage, sections were dehydrated through a series of alcohols and xylene, and mounted with Permount (Fisher Scientific). In every run I included a negative control section which did not receive primary antibody. When trying a new tissue, or antibody of unknown expression pattern, I also included a positive control of an established antibody made in the appropriate host. For fluorescent detection of one or more antigens Immunofluorescence (used in figures 5.2.2, 5.4.1, 5.5.3, 6.3.5, 6.5.1, 6.5.2 and 6.8.1) was performed by incubating fresh frozen sections with primary antibody as described above, followed by 1 hour incubation at room temperature with a fluorescently conjugated secondary antibody in PBS. Where multiple secondaries were employed, they were used sequentially with 3 5 minutes PBS washes in between. Sections were then washed with water and coverslipped using citiflour mounting medium (Marivac) or VectaShield mounting medium (Vector Laboratories). Fluorescent secondary antibodies used were: Cy2-conjugated goat or donkey anti-rabbit IgG, Texas red-labelled rabbit anti-goat or anti-mouse IgG (1:100, all from Jackson ImmunoResearch); cascade blue-conjugated goat anti-mouse IgG (1:100, Molecular Probes); goat anti-rabbit or goat anti-mouse Alexa-488 (1:200, green, Molecular Probes). Details of all primary antibodies used for immunohistochemistry are given in table 4.2. Variations used for some specific primary antibodies • Anti-active caspase-3 (MF397), fluorescent detection (used in figure 5.2.2): to reduce background, primary antibody was diluted in 2% normal goat serum (NGS), a second block of 30 minutes in 4% NGS was introduced after the primary antibody incubation, and the secondary antibody (anti-rabbit-Cy2) was diluted in 2% normal goat serum and 0.05% triton X-100. • Anti-active caspase-8 (MF) (section 5.6): to reduce background, the washes following primary antibody incubation were in 0.2% tween-20 in PBS. • Anti-single-strand D N A (used in figure 5.1.7): treat with 1 N NaOH for 10 minutes; permeabilize in 0.1% triton X-100 in PBS for 30 minutes; incubate in 10 _g/ml proteinase K in PBS for 5 minutes at room temperature; rinse for 2x 5 minutes with distilled water; 34 incubate in 50% formamide in distilled water in a 56-60°C water bath for 30 minutes; rinse in ice-cold PBS; block in 4% normal horse serum for 20 minutes; incubate overnight at 4°C in anti-single-stranded D N A antibody 1:75 in 2% normal horse serum in PBS. The primary antibody was then detected for VIP light microscopy, as described in the main protocol. Terminal dUTP nick-end labelling (TUNEL) A . For solid colour detection with alkaline phosphatase (purple/black) This method is basically as previously described (Gavrieli et al 1992). Briefly, fresh frozen sections were fixed in 4% PFA for 15 minutes; permeabilized with 0.5% triton X-100 for 25 minutes; acetylated for 10 minutes; incubated with digoxigenin (DIG)-conjugated UTP and terminal transferase enzyme (both from Boehringer Mannheim) and allowed to hybridize for 1 hour at 37°C; blocked in 5% NGS for 1 hour; incubated with 1:5000 anti-DIG antibody coupled to alkaline phosphatase (Boehringer Mannheim) for 12-16 hours at 4°C; and visualized using NBT-BCIP substrate for the alkaline phosphatase. The reaction was stopped with water and sections were mounted using aquapolymount (Polysciences). I used this method on fresh frozen adult post-bulbectomy tissue (section 5.1, and figures 5.7.1 and 6.5.2). B. For solid colour detection with TACS blue The TACS Blue TUNEL kit (R&D Systems) was used as per the manufacturer's instructions. Briefly, slides were permeablised with proteinase K; incubated with biotin-conjugated nucleotides and terminal transferase enzyme and allowed to hybridise; the biotin signal was amplified by complexing with streptavidin-HRP; and detected by a solid blue colour substrate. Slides were then dehydrated through a series of alcohols and xylene, and mounted with Permount (Fisher Scientific). I used this method on fixed embryos and neonates (figures 6.3.1, 6.3.3, 6.3.6, 6.3.7 and 6.7.3) because the alkaline phosphatase method above (A) does not work optimally on fresh tissue. C. Solid colour TACS Blue, double-labelled with one fluorophore I used this variation for double-labelling of T U N E L with GAP-43 (figure 6.3.5). In a modified version of the TACS Blue T U N E L protocol described above (B), I used cytonin instead of proteinase K to permeabilize; completed the T U N E L run apart from the final step, the formation of the blue precipitate; then proceeded with the entire immunohistochemistry run (as described earlier for fluorescent detection), using the red fluorescent secondary antibody anti-mouse Alexa 594 (Molecular Probes); and finally completed the last detection step of the T U N E L run. In this case slides were then coverslipped with the fluorescence mounting media VectaShield (Vector), and viewed immediately. This method is only adequate if one does not need to demonstrate direct overlap in the same cells, as was the case in figure 6.3.5. Furthermore VectaShield is not suitable for long term preservation of the TACS Blue solid colour signal. I subsequently improved the protocol, below (D), with a further modification to the TACS Blue kit (B) which allows both T U N E L and another antigen to be detected by fluorescence, thereby allowing the signals to be overlapped in the conventional way on the fluorescence microscope. D. Double fluorescence TUNEL. modified from TACS Blue protocol For double-labelling with N C A M (figure 6.5.1) or GAP-43 (figure 6.5.2c), I used the TACS Blue T U N E L kit (R&D Systems, described above, B) with the following modifications: I used cytonin 35 instead of proteinase K to permeabilize; substituted dig-conjugated nucleotides in the hybridization; and incorporated an overnight primary antibody incubation with anti-dig-Rhodamine and anti-NCAM or anti-GAP-43. Ant i -NCAM or anti-GAP-43 binding was subsequently detected using a green fluorescent secondary antibody anti-rabbit or anti-mouse Alexa 488 (Molecular Probes). Slides were coverslipped with the fluorescence mounting medium Citifluor (Marivac). In Situ Hybridization In situ hybridization was performed as previously described (Blackshaw and Snyder, 1997) using dig-labelled full-length sense and antisense probes (2 Kb) to the OMP (gift of F. Margolis, U . Maryland). Briefly, sections were post-fixed in 4% PFA; permeabilized in 0.1% Triton X -100; acetylated in triethanolamine/HCl/acetic anhydride; incubated for 2 hours at room temperature with prehybridization buffer; and incubated overnight at 65°C with hybridization solution containing dig-labelled OMP R N A probe or control, protected by siliconized coverslips. The signal was then detected by overnight incubation with anti-dig antibody, followed by colour development with alkaline phosphatase as described for T U N E L (A). In situ hybridization was used in figure 5.1.5. Histological Stains Haematoxylin and Eosin Sections were hydrated in water; placed in Harris haematoxylin (Sigma) for 5 minutes; washed briefly in water; placed in alcoholic eosin (Fisher Scientific) for 5 minutes, washed briefly in water; and mounted with aquapolymount (Polysciences). This stain was used in conjunction with N C A M immunohistochemistry for measurement of adult axon bundles and neuronal layer of theOE (figure 5.7.2). Methyl Green Sections were equilibrated in PBS brought to pH 4.8 with acetic acid; stained for 30 min in methyl green working solution (0.4% aqueous methyl green, 25% glycerol, 0.5x pH 4.8 buffer); rinsed briefly in pH 4.8 buffer, and mounted with aquapolymount (Polysciences). This stain was used on PFA-fixed sections of adult and neonatal olfactory bulb, which were to be measured for size and shape (figures 6.5.1, 6.6.3 and 6.7.1). Image Capture Images of tissue sections, both brightfield and fluorescent, were captured from a Zeiss Axioskop 2 M O T SPOT digital camera (Diagnostic Instruments, Inc.) and Northern Eclispe software. They were then imported into Adobe Photoshop for final compilation. Confocal microscopy was performed using a Zeiss Axiovert S100 T V microscope fitted with Bio-Rad Radiance Plus confocal hardware and LaserSharp software. Confocal Z-series were processed using NIH Image software version 1.62 (Wayne Rasband, National Institutes of Health) and imported into Adobe Photoshop for colourization and determination of signal co-localization. 36 Methods of Sampling, Counting and Measurement Spatial evaluation of T U N E L at 24 hours after bulbectomy (figure 5.1.3) At 24 hours after bulbectomy, coronal sections from the back, middle and front of the OE (defined as 1, 2.5±0.45, and 4 mm from the cribriform plate respectively) were assessed to provide spatial patterns of the number of TUNEL-labelled ORNs. Turbinates in each section were assessed as belonging to superior (close to bulb), mediolateral, and inferior (distant from bulb). Temporal evaluation of T U N E L after bulbectomy (figure 5.1.61 This was another aspect of the evaluation of caudo-rostral progression of apoptosis, in conjunction with spatial evaluation. For temporal evaluation of olfactory neuron loss, we counted the number of TUNEL+ neurons in the OE of the mediolateral turbinate, 2.5 (2+0.45) mm from the cribriform plate at 12-72 hrs post-bulbectomy. Adjacent sections were assayed for the expression of OMP (by in situ hybridization). A l l cell counts were tabulated and analyzed statistically for mean +SD using Microsoft Excel. Determination of intensity of Delta-C immunoreactivity (figure 5.4. le) To assess the relationship between axonal caspase activation and the likelihood of an O R N to undergo apoptosis, sections of successive mediolateral turbinates were sorted into quadrants, based on axonal Delta-C-APP/NST (neuron-specific tubulin) labelling on a scale from 0 to 100%. 0% represents NST+ only (red), and 100% represents complete overlap (completely yellow axon bundle). The four categories were 0-20% (red), 20-50% (orange), 50-75% (yellow) and 75-100% (completely yellow). ORN soma within each quadrant were then assessed for activation for caspase-3 (green) and TUNEL+ (black nuclei) in correlation with their adjacent axonal caspase-3 activation (mean per quadrant ± SD; three independent sections per animal). Measurement of OE thickness in post-bulbectomy caspase-3 null adults (figure 5.7.2) Coronal sections of OE (n=3 animals per genotype) were stained with haematoxylin and eosin in order to be able to identify the neuronal layer. For each section, photographs were taken of 3 locations on each side. Files were renamed in code and the thickness of the neuronal layer was then measured blindly using Northern Eclipse software. Each lesioned side value was divided by its corresponding unlesioned side value to give a set of ratios. The sets of ratios for each genotype were compared using a one-tailed t-test. Counting the number of TUNEL-positive cells per linear mm of OE, in embryos and neonates rfigures 6.3.2. 6.3.4 and 6.7.31 At least 3 animals were used per group (genotype and age), and from each animal a section was included in the analysis from at least 3 separate T U N E L runs with positive and negative controls. Northern Eclipse software was used to measure the entire length of the OE visible on each section, and I counted the total number of apoptotic nuclei in the neuronal layer of the OE on each section. For each group, I then divided the total number of apoptotic cells by the total length of OE measured. 37 Measurement of OE thickness atE13 (figures 6.4.1, 6.4.2) As above, at least 3 animals per genotype were used, and at least 3 sections per animal from different controlled T U N E L runs. Within each section 6 positions along the OE were selected for thickness measurements to get a good representation of the thickness. I used Northern Eclipse software to blindly measure the perpendicular thickness of the OE neuronal layer at the selected points. After the results were decoded, they were either averaged for each genotype, or plotted as individual points on a scatter graph. Measurement of size and shape of adult olfactory bulbs (figures 6.6.1 and 6.6.3) and P4 bulbs (figure 6.7.1) 3 wildtype and 5 caspase-3 null animals were used for the adult analysis. I calculated the length (rostro-caudal) of each bulb by assessing how many of the numbered 20 urn coronal sections contained bulb. For bulbs to be assessed quantitatively, every 8 th section was selected (i.e. every 160 u.m) for measurements. Those sections which were not required for immunohistochemistry were stained with methyl green. Northern Eclipse software was used to measure areas of bulbs, and to analyse the shapes of olfactory bulbs. This was done by calculating a shape factor, defined as 4jta/c2, where a=area of the shape and c=circumference, such that a perfect circle would have a shape factor of 1 and a line would have a shape factor of 0. The typical ellipse of a wildtype olfactory bulb has a shape factor of about 0.7. To measure the total volume of a bulb, the area of every section was plotted against the position of that section along the length of the bulb, so that the area under the graph provided the volume. For the P4 timepoint, I used 3 animals of each genotype. In this case sagittal sections were used, so I initially calculated the width (medial - lateral) of the bulb by assessing how many of the numbered 20 urn sections contained bulb. For quantitative analysis I measured every 4 t h 20 u,m section, and then calculated area and volume in the same way as described above for adult. Measurement of size and number of axon bundles in adult OE (figure 6.6.5) Sections were from 4 caspase-3 +/- mice and 3 caspase-3 - / - mice, labelled with N C A M and with haematoxylin and eosin. Within each section I made measurements in 9 specific fields of view, comprising 3 fields within each of the 3 turbinates: superior, middle and inferior. For each field of view I: counted the number of ORNs in the perpendicular height of the OE; measured 1 mm of OE with a graticule and counted the number of axon bundles in this mm; measured with a graticule the maximum and minimum diameters of every axon bundle falling within this mm. T-tests were performed using Microsoft Excel on the following data: 1- tailed t-test: figure 5.7.2, OE thickness post-bulbectomy 2- tailed t-tests: figure 6.3.4, fewer dying OE cells at E l 3 in caspase null figure 6.4.1, thicker OE at E l 3 in caspase null figure 6.7.3, no difference in number of apoptotic ORNs at P4 38 o S H H O CQ O Q w in on W Q O PQ H 2 1-CD «-H .3 ON ON ON 03 O ca B o\ o ON < —< ea ea -*-» •*-» tu CD i -a c Vi Vi <a <D s CN v i o fi" T 3 C3 XI T3 v i v i v i v i v i v i v ^ v i v i v i v i v i v i v i v i v i o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o v i o o v i o "5b '5b <£ eg <C <C '5b '5b '5b '5b '5b 43 '5b| o o o >^ 2? x> o o o o o o o o o o o o o o I -*  I *^~. *^ I I <^  I o o o o o o o o o o o o o c a a a a a a a c i a a D J 2 3 2 S S S S S S S S S S S S S X X X O f i X X X X X X i X X X X X c a c a c a e ^ c a c a c a c a c a c a c a c a ca ca ca ca ~ y <M .a o\ CO •* tS N » r-~ oo oo c i ro c i c i -3-' b b h b s s s § M n m t * iOS C l /—\ VI OS rt _ O w O 5B- w O OO 0O ON ON ON — ' D D -O 'cl o 1) B U ca Q : IS" ca <, r- r- o o o o ^ (N CS (S (N D - O - O - O - D . O , J J 3 ^ £ B £ CN m" n i n ^ oo" • • CQ CD O O O <1> CD <D Vi c/> (/) c/> c/3 Vi Vi ca ca ca ca ca ca ca D - D . D - Q - O - D . Oh Vi vi vt vt vt Vi Vi ca ca ca ca ca ca ca o o o o o o ca ca a a a Vi Vi C3 C3 O .5 CD .3 oo Ol Ol CD o C3 s v i oo vd „ o —; CN CN v i N© v© 0 0 VI v i CN CN o" CN ~ CN CN CN C l C l VI c i c N c N C N v o v i c i v i c N ^ ^ O T r v i v o v i v i v i v i v o v d v i v i ^ o v i o o o o o o o o o o o o o o o o o o o O O O O O O V I O O O O O C N — • C N V l r t r < l C N C l v l - H V l — i '5b '5b S 6 CD CD J 3 4 3 . c c Q>. O O Ml .y .a -3 S 6 £ CD CD CD x * o o o m _>> ^ _>% a. D . a. o o o o. a, a. •1= .TS .TS .t; .t; .ts m .ts x> J O x> xi J O xi x x J O jo x x ca ca ca ca ca ca ca 1 13 I o g o ° o x> S D . ^ C •S o 2 ca I I O o x> o ° 8 Ol i t s _ m < & n u o o * < ^  2 U ca ca C B O U I Z O J3 -—- c i c i oo CD D -O ' & CD O CD a u M Q CN Pi OH OH < m m m oo ON TO 03 03 03 CU Q- &- &-t/i tn e/D P3 P3 C3 03 03 39 CHAPTER 5: APOPTOSIS AND CASPASES IN MATURE OLFACTORY NEURONS OF THE ADULT MOUSE The majority of the data presented in this chapter has been published as: Cowan C M , Thai J, Krajewski S, Reed JC, Nicholson DW, Kaufmann SH and Roskams AJ (2001). Caspases-3 and -9 send a proapoptotic signal from synapse to cell body in olfactory receptor neurons. J. Neurosci. 21(18):7099-7109. There has been some reorganization to better fit the themes of this thesis. 5.1 Apoptosis of mature ORNs after olfactory bulbectomy A gradient of apoptosis away from the bulb site To define a spatiotemporal pattern for ORN apoptosis following bulbectomy, we employed several techniques to follow the loss of the mature neuron population. Agarose gel electrophoresis was performed on D N A derived from OE taken at a range of timepoints after bulbectomy. This revealed D N A laddering characteristic of apoptosis which was first seen at 12 hours post-bulbectomy and peaked at 24-36 hours post-bulbectomy (figure 5.1.1), a profile slightly earlier than previously reported (Michel et al 1994). This profile was then used as a basis for subsequent immunocytochemical and immunoblotting experiments, in which the use of unilateral bulbectomy (indicated in figure 1.3.3) allowed me to directly compare expression patterns of lesioned and unlesioned neurons within each animal. We used T U N E L staining to identify that the cells responsible for the D N A laddering pattern lie within the O R N cell layer (figures 5.1.2 and 5.1.4). For example, figure 5.1.2a shows intense T U N E L on the lesioned side and minimal T U N E L on the unlesioned side in a coronal section of OE 24 hours after unilateral bulbectomy. These data were analysed to show that apoptosis progressed in a caudo-rostral gradient both in space and time, where ORNs closest to the olfactory bulb removal site underwent apoptosis earlier than more distant ORNs. Figure 5.1.2b-f shows that in successive coronal sections of OE 24 hours after bulbectomy, T U N E L generally became less intense in more rostral sections, further from the bulb. However, merely assessing successive coronal sections gives an overly simplistic view. The olfactory epithelium is comprised of turbinates, which are cartilagenous extensions or folds projecting from the lateral wall of the nasal cavity toward the nasal septum. As can be seen in figure 5.1.2a, in any one coronal section, superior turbinates are closer to the bulb in the dorso-ventral plane than are inferior turbinates, and that should be controlled for. Therefore to get an accurate picture of the progression of apoptosis in space one must consider both the caudo-rostral and dorso-ventral plane. To better investigate the spatial dynamics of apoptosis in mature wildtype ORNs, we evaluated the number of TUNEL+ cells in the superior, mediolateral and inferior turbinates at 24 hours post-bulbectomy from 3 sites at 1 (± 0.2) mm, 2.5 (± 0.45) mm and 4 (± 0.56) mm from the cribriform plate (figure 5.1.3). Now the picture is a little more complex. Overall, the most rostral sections were still found to have the least labelling. However, looking for example at the mediolateral turbinate, the peak of apoptosis here at 24 hours had progressed to the middle sections. Meanwhile the inferior turbinate, far from the bulb, had yet to show high levels of apoptosis anywhere. So overall, the number of TUNEL+ cells was greater closest to the bulb (superior turbinate of sections from 1-2.5 mm rostral to the bulb) and the least at the furthest distance from the bulb (inferior turbinate at each position rostrally, figure 5.1.3). 40 Analysis of T U N E L was also used to illustrate the progression of apoptosis away from the bulb site over time. Looking at the same turbinate in sagittal sections at successive times after bulbectomy, one can see apoptotic ORNs close to the bulb removal site at 24 hours, progressing further away at 36 hours and further still at 48 hours (figure 5.1.4). This figure also illustrates that the earliest TUNEL-positive cells were detected at 12 hours post-bulbectomy. However, these were not ORNs, but were located in the lamina propria. The lamina propria is a connective tissue, the other major layer of the olfactory mucous membrane in addition to the olfactory epithelium, and is located beneath the OE separated by a thin basement membrane. The lamina propria consists of bundles of ORN axons, olfactory ensheathing glia, blood vessels, mucous-secreting Bowman's glands, fibroblasts, and collagen and elastin fibres. These cells may possibly be olfactory ensheathing glia whose processes were directly damaged by the bulbectomy, which has never been shown before. We used in situ hybridization to confirm that cells being lost due to apoptosis were expressing OMP, an exclusive marker for mature olfactory neurons (figure 5.1.5). The contralateral OE retained OMP expression and did not show an increase in TUNEL-positivity. As ORNs underwent apoptosis (between 24 and 48 hours), the thickness of the OE decreased as the number of ORNs expressing OMP mRNA decreased. This loss of OMP expressing cells was quantified in the graph in figure 5.1.6. The proportion of OMP expressing cells in the OE declined from >60% to <5% over a period of 72 hours after bulbectomy. Meanwhile, adjacent sections were assayed by T U N E L to further analyse the progression of apoptosis over time. The number of TUNEL-positive cells increased rapidly after bulbectomy, reaching a maximum at 36 hours, and returning to very low levels by 72 hours. These data demonstrate that the peak of loss of the mature O R N population by apoptosis is from 24-36 hours post-bulbectomy, and that almost complete loss of mature ORNs has occurred by 72 hours. I also used an alternative technique to confirm the results seen with TUNEL. A monoclonal antibody against single-stranded D N A is marketed as being very specific for apoptotic nuclei, and not giving the occasional false positive for necrotic cells, in the way that it has been alleged that T U N E L can (Rink et al 1995). Using this reagent, exactly the same labelling pattern was obtained at 24 and 48 hours postbulbectomy on lesioned and unlesioned sides, as that observed with T U N E L (figure 5.1.7). 5.2 Axonal activation of caspase-3 during ORN apoptosis TUNEL-positivity reflects, in part, the action of C A D (Caspase Activated DNAse), a unique endonuclease that is liberated in its active form after caspase-3-mediated cleavage of its inhibitor ICAD (Enari et al 1998). Caspase-3 is thought to be one of the more important caspases for neuronal apoptosis in development (Kuida et al 1996), and it has been shown to be activated in dying neurons in some injury models, as mentioned in the Literature Review. Therefore I wished to examine the expression and activation of caspase-3 in ORNs after bulbectomy, to see whether it plays a role in this apoptosis. The caspase-3 proenzyme is made up of an N-terminal prodomain and a large and small subunit. Upon proteolytic activation of the zymogen dimer, this is cleaved into its component parts, and 41 two large and two small subunits complex to form an active heterotetramer (Walker et al 1994) (figure 5.2.1a). This is also the case for caspase-1, and is widely assumed to be the case for most of the other caspases except the apical caspases. The main variation on this theme is that effector caspases have small prodomains, whereas initiator caspases have larger prodomains that can bind other proteins and play a role in the recruitment and aggregation of the procaspases in order to facilitate proteolytic processing (reviewed in Earnshaw et al 1999). To assess caspase-3 proenzyme expression and activation during O R N apoptosis, I examined protein extracts from unlesioned and bulbectomized mouse OE by immunoblotting. Using two different antisera to caspase-3,1 found that the-full-length form of caspase-3 (32 kDa) was present in normal OE (figure 5.2.lb,d). Over the first 24 hours following bulbectomy, endogenous levels of procaspase-3 increased significantly (figure 5.2.1b). The large subunit produced following cleavage of caspase-3 (17 kDa) was detected at 48 hours - shown with two different antibodies (figure 5.2.1c,d). These results clearly demonstrated an accumulation in procaspase-3 immediately prior to maximal cleavage and activation but did not provide any information about the site of this process in ORNs. To determine the site of caspase-3 activation, immunohistochemistry was performed using two antisera that specifically recognize active caspase-3 (figure 5.2.2a-c). Both reagents yielded identical results. The active form of caspase-3 was detected in axon bundles and cell bodies of ORNs post-bulbectomy. Active caspase-3 could be detected with this more sensitive technique as early as 24 hours post-bulbectomy (figure 5.2.2a), but was not detectable in the contralateral (unlesioned) ORNs (figure 5.2.2b). The signal persisted on the lesioned side until 48 hours (figure 5.2.2c). A brightfield photograph of the unlesioned side immunolabelled for OMP is included for orientation of the fluorescence images, and to illustrate that axons and O R N cell bodies on the unlesioned side still retain expression of OMP (figure 5.2.2d). The immunocytochemical activation profile of caspase-3 for both antisera mirrored the pattern of activation observed on western blot - active caspase-3 in some axons and cell bodies at 24 hours post-bulbectomy, with more widespread (maximal) immunoreactivity throughout every axon bundle, and maximal ORN soma activation at 48 hours post-bulbectomy (figure 5.2.2c). 5.3 Caspase-3-dependent cleavage of axonal APLP2 Because the detection of active caspase-3 in axonal bundles was unanticipated, we wanted to examine whether active caspase-3 was being transported retrogradely without cleaving targets, or whether it was actively cleaving downstream targets throughout the neuraxis during O R N apoptosis. Few specific caspase-3 targets have been localized to ORNs and, in particular, their axons. APP is cleaved by caspase-3 to reveal an epitope (VEVD) recognized by the Delta-C-APP antiserum (Gervais et al 1999). I detected the epitope recognized by the Delta-C antiserum by immunocytochemistry in subsets of axons and cell bodies of apoptotic ORNs but not their unlesioned counterparts, at 24 hours post-bulbectomy (figure 5.3.1a-c, low and high magnification). This pattern resembles the cellular distribution of activated caspase-3 (shown in figure 5.2.2), but is actually more useful, as it allows us to visualize where caspase-3 is being activated and where it has been activated (the cleavage product persists during degeneration, whereas the caspase-3 activation is more transient). At low power, the Delta-C epitope first appeared most predominantly in subsets of axon bundles and dying ORNs at 24 hours following 42 bulbectomy (figure 5.3.1a). The epitope was distributed in distinct regions of axon bundles of apoptotic ORNs and, in ORN cell bodies, accumulated in a c-shape deposit surrounding the nucleus (figure 5.3.1c). Because APP is only found in embryonic olfactory neurons, it is not likely to be responsible for the Delta-C-APP epitope revealed in apoptotic ORNs. Its homologue, APLP2, which is predominantly expressed throughout the axons of ORNs in development and in adulthood also contains the conserved cleavage site (VEVD) for caspase-3-mediated cleavage (Thinakaran et al 1995). To determine whether APLP2 might be responsible for the observed staining, antisera to the N-terminal (D2-II) and C-terminal (CT12 APLP2) fragments that could be produced by caspase-mediated cleavage of APLP2 were utilized to immunoprecipitate APLP2. Blotting with the Delta-C-APP antisera confirmed that this reagent did recognize an APLP2 cleavage product in apoptotic ORNs (figure 5.3.Id). In addition, the C-terminus (CT-12) of APLP2 was also liberated in OE extracts from wildtype mice, but not from mice with a null mutation for caspase-3 (figure 5.3.Id). Collectively, these results demonstrate that caspase-3-dependent cleavage of APLP2 occurs in the same spatial pattern as appearance of activated caspase-3, i.e. predominantly in the axons and later in cell bodies. 5.4 Evidence that proteolysis starts at the presynaptic complex The preceding results suggest that caspase-3 activation may begin in O R N axons and is propagated in a retrograde fashion to ORN cell bodies from the site of lesion. To confirm this hypothesis, we altered our lesion paradigm to that of a partial bulbectomy (removing the cell bodies of the target neurons, but leaving many of their synaptic complexes intact) in an attempt to encompass the pattern of caspase activation in a complete single slice. Following partial bulbectomy, sections of olfactory bulb and epithelium were initially examined by T U N E L , followed by co-immunolocalization of the Delta-C-APP cleavage product and the axonal protein Type III (neuron-specific) tubulin (figure 5.4.1a,c) (Roskams et al 1998). In the olfactory bulb, 24 hours following partial lesion (figure 5.4.1a), the caspase-3-liberated Delta-C-APP neoepitope first appeared as bright, punctate dots of immunoreactivity in glomerular complexes, each of which contain the synapses of up to 1,000 different ORNs. The neoepitope was also detected in a gradient back from the glomeruli into some NST-positive axons of the nerve fibre layer projecting back to the OE (figure 5.4.1a). The Delta-C neoepitope was only seen in axons of lesioned olfactory bulb and OE, where unlesioned axons appeared NST-positive (red) only. The synaptic complex localization of the Delta-C neoepitope at 24 hours following partial lesion was confirmed by co-localization with synaptophysin in the glomeruli (figure 5.4.1b). In the OE, 36 hours after bulbectomy, axons displaying the highest degree of caspase-3-mediated cleavage (Delta-C+) were most prominent close to where the olfactory nerve (NST+) exits the mediolateral turbinate (figure 5.4.1c). Delta-C+/NST- (green) and TUNEL+ ORN cell bodies were clustered closest to the axons of highest caspase-3 activation (yellow, figure 5.4.1c). Axons adjacent to neurons in the part of the turbinate farthest from the bulb, however, were not yet labelled for active caspase-3. To further assess the direction of caspase-3 activation, we examined sagittal slices of OE for the presence of active caspase-3 and co-ordinate production of the Delta-C-APP neoepitope. Figure 5.4. Id shows caspase-3 activation (red) carried through the axon bundle and into ORN cell bodies distant 43 from the olfactory bulb. Caspase-3 activation, which is transient, occurred in advance of the detection of Delta-C-APP, whose neoepitope was seen left behind in the axon (green), in the wake of the active caspase-3 signal. To assess the relationship between axonal caspase activation and the likelihood of an O R N to undergo apoptosis, we divided successive turbinates into quadrants where axon bundles demonstrated Delta-C/NST (yellow) overlap of <20% (red), 20-50% (orange), 50-75% (yellow) and 75-100% (completely yellow) (figure 5.4.le). By comparing the number of TUNEL+ neurons and green Delta-C+ neuronal soma with adjacent axonal caspase activation, we demonstrated that those axons displaying maximal axonal caspase-3 activation (yellow, 75-100%) were adjacent to regions of OE with maximal cell body caspase activation (green) and maximal TUNEL-labelling (black). Conversely, the areas of OE most distant from the olfactory bulb displayed minimal soma caspase-3 activation, minimal TUNEL-positivity and were adjacent to the axons of lowest caspase-3 activation. Collectively, these results suggest that caspase-mediated cleavage begins in the synaptic complexes, is propagated through the axons, and reaches the O R N cell bodies, which then become TUNEL+, last. 5.5 Activation of caspase-9 during ORN apoptosis Given that caspase-9 is thought to be the preferential, although not sole, activator of neuronal caspase-3 in vitro and in vivo, I examined caspase-9 expression and activation in ORNs by using multiple different caspase-9 antisera (figure 5.5.1) for immunoblotting and immunolocalization. Two different antisera that recognize full-length procaspase-9 demonstrated that levels of this zymogen increased significantly in the OE from 4 to 24 hours post-bulbectomy and then dropped to barely detectable levels by 72 hours (figure 5.5.2a,b). An antiserum that recognizes recombinant human caspase-9 (Bur49, figure 5.5.2b) and another that recognizes a neoepitope that becomes detectable at the C-terminus of the large subunit of caspase-9 after proteolytic liberation of the small subunit (figure 5.5.2c) indicate that multiple species of cleaved caspase-9 were detectable at early (4 hours) and late stages (24 hours) following bulbectomy. These species included a 35 kDa fragment (figure 5.5.2b,c), which probably represents the prodomain still joined to the large subunit of caspase-9, having been cleaved from the linker region and small subunit. This fragment has been reported as being 35 kDa (Saleh et al 1999), and is actually one of the catalytic units (figure 5.5.1). The other fragment seen is likely to be the p l9 large subunit alone (figure 5.5.2c). Caspase-9 proenzyme expression increased prior to maximal caspase-9 activation (24 hours), following which the expression of full-length caspase-9 was significantly reduced (figure 5.5.2a,b). Loading control blots probed with beta-actin and Apaf-1 are shown (figure 5.5.2d,e). Note that the levels of Apaf-1 (a known activator of caspase-9 which complexes with it to form an apoptosome) did not change following bulbectomy (figure 5.5.2d). The appearance of maximal caspase-9 activation immediately prior to maximal caspase-3 activation and TUNEL-positivity of ORNs suggests that caspase-9 activation is a significant part of the pro-apoptotic program of deafferented ORNs. For this to be the case, caspase-9 would have to be distributed throughout the neuraxis in order to activate caspase-3 in response to appropriate extracellular stimuli. Immunohistochemical detection with anti-caspase-9 antiserum alone (figure 5.5.3a) and in combination with OMP (figure 5.5.3b) showed that 44 caspase-9 was not found in the ORN cell body, but was highly expressed in ORNs and clearly distributed throughout olfactory axons all the way into olfactory glomeruli. The overlap of caspase-9 with OMP and synaptophysin, a protein enriched in presynaptic membranes, confirmed that caspase-9 had a synaptic localization (figure 5.5.3c). One point that I cannot address using reagents currently available, is whether the retrograde wave of caspase-3 activation is preceded by a wave of retrograde caspase-9 activation. Although this was suggested temporally (caspase-9 was only found in ORN axons and was maximally activated immediately prior to caspase-3 axonal activation) the commercially available anti-active caspase-9 antibody (New England Biolabs) recognizes only rat, not mouse active caspase-9. In addition, our active caspase-9 neoepitope antibody works only on western blot (not immunocytochemistry) and also appears to recognize a differentially "processed" form of caspase-9 which may represent a cytoplasmic regulation of procaspase-9. 5.6 Expression of other caspases in the mature OE during ORN apoptosis We obtained a set of anti-caspase antibodies from Merck Frosst raised against the human caspases from -1 to -10. A l l were tested on western blots of OE tissue taken at a range of timepoints after bulbectomy, to determine which caspases other than -3 and -9 are expressed in the OE, and whether any of them are activated over the course of ORN apoptosis. In some cases a suitable positive control was available, such as mouse tissue known to express the caspase in question; a purified peptide caspase; and/or an alternative commercial antibody known to recognize the mouse caspase. Using the antibodies against caspase-4 and caspase-5, a single clean band of the expected size was detected in the OE (figure 5.6.1a, and not shown). In both cases, the intensity of the signal was very weak, and neither the intensity nor the size of the band changed over time after bulbectomy. Therefore I can conclude that caspase-4 and caspase-5 do not play a role in bulbectomy-induced ORN apoptosis. I also suggest that expression of caspase-4 and caspase-5 might be in a different cell type of the OE, rather than in ORNs. That would account for the very low levels seen, since ORNs account for the major cell type in the OE. In the case of caspase-4, a mouse liver sample was included in a positive control lane (figure 5.6.1a). A band corresponding to the 43 kDa proenzyme was seen, as in the OE, as well as two cleavage products, the larger of which was an appropriate size to potentially represent the large and small subunits (32 kDa) with their prodomain cleaved off. Caspase-4 is reported to be expressed in most tissues with the exception of brain, and especially in the liver, lung, ovaries and placenta (Munday et al 1995, Kamens et al 1995). Caspase-5 is also found in many tissues but at very low levels (Munday et al 1995, Faucheu et al 1996). Caspase-2 and caspase-7 were found to be absent from the OE (figure 5.6.1b,c). In the case of caspase-2,1 used in addition to the Merck Frosst antibody another primary antibody from Santa Cruz (figure 5.6.1b) which is reported to react with mouse. Caspase-2 is reported to be expressed in development in the brain, liver, kidney and lung (Kumar et al 1994). Later in development it is down-regulated in the brain, indeed this is how it was discovered (Kumar et al 1992, 1994) but it does remain at low levels in the adult in post-mitotic neurons and several other tissues (Kumar et al 1994). I detected the 48 kDa proenzyme caspase-2 in the mouse 45 cortex, and cleaved forms in the liver. No bands were detected in the OE, or the kidney and spleen. Confusingly, in the literature, caspase-7 is variously reported as being expressed in many foetal and adult tissues with lowest expression in the brain (Fernandes-Alnemri et al 1995), not expressed in the brain (Juan et al 1997), and constitutively expressed in a number of brain regions including the hippocampus (Bonislawski and Simon 2002) and the entorhinal cortex (Pompl et al 2003). With the Merck Frosst antibody, I detected a band corresponding to full-length caspase-7 strongly in the liver (figure 5.6.1c) and lung (not shown), slightly less strongly in the cortex, and not detectable in the OE (figure 5.6.1c). Using the Merck Frosst anti-caspase-1 antibody (MF370), no bands were detected in the mouse OE. I believe the most likely explanation for this is that caspase-1 is not expressed in the mouse OE. However, in this case the positive controls were not as clean as for caspases-2 and -7. In the liver and lung, the caspase-1 antibody gave several bands, including one the size of caspase-1 (45 kDa, not shown). Therefore this data is also open to the interpretation that the antibody raised against human caspase-1 does not adequately recognize mouse caspase-1.1 have not probed for caspase-1 with an alternative antibody, and so these data are inconclusive and more evidence would be needed to confirm that caspase-1 is absent from the OE. In the literature, it is reported that caspase-1 is highly expressed in the lung and liver, and in most tissues with the exception of brain (Cohen 1997). It is generally believed that caspase-1 is not of primary importance in apoptosis in many cells (Kuida et al 1995, L i et al 1995). It certainly has another function in inflammation, and it is known not to be involved in all forms of apoptosis, but there is some evidence for its involvement in neuronal death in vertebrates (Gagliardini et al 1994). The situation with caspases-8 and -10 is intriguing, but no satisfactory conclusions have been reached. Western blots with anti-caspase-8 (MF438) and anti-caspase-10 (MF459) both repeatably and cleanly gave an interesting pattern over the course of bulbectomy (figure 5.6.2). A very weak band the size of the proenzyme (55 kDa) was seen at all times. Three discrete cleaved forms between 20 and 32 kDa were seen very strongly, in a distinct pattern at 4, 8 and 12 hours after bulbectomy, decreasing dramatically at 16 and 24 hours, and all cleaved forms reaching a very strong maximum at 36^-8 hours. This pattern would theoretically fit in with my data for caspase-3 and -9 cleavage (figures 5.2.1 and 5.5.2). However, the similarity of the caspase-8 and caspase-10 blots is suspicious. I am concerned by the possibility that both antibodies are recognizing the same thing: either caspase-8 or caspase-10, or indeed both caspases together. This is especially possible since the two caspases have a similar structure, and neither antibody has previously been proven to detect the mouse caspase specifically. Unfortunately, multiple blots with caspase-8 and caspase-10 polypeptide positive controls did little to clarify the situation, since the peptides are not the same size as the bands seen with endogenous proteins, but it seems possible that anti-caspase-8 (MF438) might recognize both mouse caspase-8 and mouse caspase-10. Blots with an alternative caspase-8 antibody (gift from S.H. Kaufmann) gave a very similar result (three cleavage products of the same size with a distinct early pattern before a decline, figure 5.6.2) with one crucial difference: the maximal expression was seen at 24 hours, not 36 hours. This is crucial because active caspase-9 peaks at 24 hours, and within the window of the next 12 hours the bulk of O R N death occurs. One reasonable interpretation of my blots is that anti-caspase-8 (SK) recognizes caspase-8, and active caspase-8 peaks at 24 hours; anti caspase-8 (MF438) recognizes both mouse caspase-8 and mouse caspase-10; anti-caspase-10 (MF459) recognizes caspase-10, and active caspase-10 peaks at 48 hours. However, for the reasons mentioned above, interpretation of the role of 46 caspases-8 and -10 in a model caspase cascade in the ORN would be very different in these different scenarios of peaking at 24, 36 or 48 hours, and therefore it would be prudent to wait for more concrete evidence of the expression and timing of activation of these caspases before speculating on their role in events. Immunohistochemistry provides further proof of caspase-8 expression and activation, although this has not yet been used to demonstrate the timing of maximal activation. We recently obtained an alternative caspase-8 antibody from Merck Frosst specific for the active form. ORN axon bundles, and cell bodies in the neuronal layer, became immunoreactive for active caspase-8 on the lesioned side only after bulbectomy (figure 5.6.3). The potential role for caspase-8 is now being examined by other members of the Roskams' laboratory (see Future Directions, section 7.3). In the literature, it is reported that there is low expression of caspase-8 in the brain (reviewed in Cohen 1997), and no expression in the embryonic brain (Muzio et al 1996). Caspase-10 meanwhile is expressed in most tissues, but the brain is one of the organs showing lower expression (Fernandes-Alnemri et al 1996). 5.7 Mature ORNs do not undergo acute bulbectomy-induced apoptosis in caspase-3 null mice Despite published reports of embryonic and early postnatal lethality in caspase-3 null mutants, we have been able (by backcrossing within our caspase-3 null colony) to generate a caspase-3 - / - mouse that survives to at least 6 months of age (see section 6.1). If caspase-3 is required for O R N apoptosis in the bulbectomy paradigm, as suggested by the close correlation between caspase-3 activation and TUNEL, then one might expect that ORNs of caspase-3 null mice would not undergo apoptosis after bulbectomy. At 24 hours post-bulbectomy (the beginning of the peak of caspase-3-mediated ORN apoptosis), at 48 hours postbulbectomy (when 90% of olfactory neurons should have completed apoptosis), and at 72 hours (when almost all the mature ORNs should have undergone apoptosis), there was never any increase in TUNEL+ ORNs above unlesioned side background within the OE in any of the knockout mice tested (figure 5.7.1). TUNEL+ non-neuronal cells in bony tissue of the nasal cavity are, however, observed in these mice (not shown). As seen in figure 5.1.5, the lesioned OE in wildtype mice decreased in thickness as ORNs were lost after bulbectomy. Although apoptotic nuclei were not detected in ORNs of caspase-3 nulls after bulbectomy, it is possible that ORNs might nevertheless have been lost by some other mechanism. However, this is unlikely to be the case, because when the thicknesses of the OE neuronal layer at 72 hours after bulbectomy were measured in wildtypes and nulls, the lesioned wildtype OE had been reduced to 70% the thickness of the unlesioned side, while in contrast in the caspase-3 null the lesioned OE had retained its unlesioned thickness of 100% (figure 5.7.2). Therefore I conclude that ORNs of caspase-3 null mice do not undergo apoptosis and are not lost by another mechanism in the acute 72 hour time period after bulbectomy, the period in which almost all mature ORNs are lost in the wildtype. 47 DISCUSSION ORNs undergo apoptosis following deafferentation in a spatiotemporal manner, where a chief predictive factor determining the timing of terminal neuronal apoptosis is the length of the ORN axon from the site of lesion (section 5.1). Because of the three-dimensional structure of the OE, this profile has enabled me to track pro-apoptotic signals all the way from the synapse to the cell body, and even across a single turbinate, to provide data pinpointing where, and at what apoptotic stage, terminal apoptotic pathways become activated. This analysis led to several novel observations. First, endogenous levels of procaspase-3 and -9 increase beginning as early as 4 hours after bulbectomy. Secondly, caspase-3 activation is initially recorded at the synapse, then axon bundles and only later in the cell bodies. Thirdly, activated caspase-3 is capable of cleaving substrates like APLP2 throughout the olfactory neuraxis. These data have important implications for current models of a role for retrograde signalling in neuronal apoptosis. In delineating the pathways that drive retrograde ORN apoptosis, caspases-9 and -3 are both expressed by mature olfactory neurons throughout their axons, all the way into the presynaptic compartment (figures 5.4.1 and 5.5.3). When the olfactory bulb is lesioned in such a way to remove trophic support from the postsynaptic neuron and deafferent the olfactory nerve, two responses are observed. The endogenous levels of caspase-3 and -9 proenzymes increase, beginning as early as 4 hours post-bulbectomy. The peak of expression of caspase-9 (at 24 hours) coincides with its maximal activation, after which proenzyme levels decrease significantly. This peak of caspase-9 activation occurs immediately before the peak of caspase-3 proenzyme expression and cleavage (at 48 hours post-bulbectomy). I have not addressed whether the elevated proenzyme levels may be a result of transcriptional up-regulation, although this could occur marginally within a timeframe allowed by fast retrograde transport. Given the distribution of caspase-3 and -9 throughout the neuraxis prior to lesion, I believe that it is more likely that the increased levels result from a local up-regulation of proenzyme production (at the level of translation) or a change in the constitutive maintenance of caspase-3 and -9. Although further experiments would be required to distinguish between these possibilities, these data are the first to show that the commitment to death that occurs early at the initiation of neuronal apoptosis also includes procaspase-9 and -3 accumulation (Putcha et al 2000). It is possible that this phenomenon could only have been revealed effectively in a compartmentalized in vivo system such as the one utilized by our laboratory. This phenomenon also fits well with the "competence to die" hypothesis being proposed by other laboratories working in the neuronal apoptosis field. It is conceivable that inhibition of accumulation of proenzyme might prove to be an alternative way to inhibit caspase-mediated pro-apoptotic signalling. At the time of maximal caspase proenzyme accumulation, caspase activation also becomes evident in these neurons. Strikingly, caspase-3 activation, detected by conformation-sensitive antibodies that recognize only the processed, active form of caspase-3, shows the active enzyme as proximal as the synapse. Active caspase-3 then becomes spatially detectable in axons and only later in O R N cell bodies. Further support for the synapse to axon to soma propagation of the apoptotic signal comes from use of the Delta-C-APP neo-epitope antibody which provided me with a sensitive way to demonstrate active cleavage of caspase-3 target proteins in ORN synapses and axons in vivo. Although it is conceivable that an alternative caspase (e.g. caspase-6) could also cleave at the same target site, this cleavage event can still be defined as caspase-3-dependent. Cleaved 48 APLP2 cannot be immunoprecipitated or detected immunohistochemically from the OE of lesioned caspase-3 knockout mice (figure 5.3.Id). Also, because the APLP2 cleavage product persists in cells and axons during degeneration, whereas active caspase-3 does not, examination of cleaved APLP2 provides a useful map of where caspase-3 is currently active and also has been active. The first site where APLP2 cleavage is detected is within the synaptic complexes of olfactory bulb glomeruli, as evidenced by overlap with synaptophysin, a synaptic protein that within the glomeruli is primarily found in ORN presynaptic terminals (Kasowski et al 1999). The active caspase-3 signal is also seen spreading out from the synaptic glomeruli into the nerve fibre layer of the olfactory bulb (figure 5.4.1a,b) and then into the axon bundles of the lamina propria (figure 5.4c,d). The persistence of the Delta—C-APP neo-epitope in axon bundles following cleavage of APLP2 reveals that there is a level of organization within O R N axon bundles at which synchronously apoptotic ORNs are subtly divided into groups of mesaxons (figure 5.3.1). Olfactory axons thus appear grouped together on the basis of functional or developmental state, which may underlie their differential vulnerability to apoptosis. I show that caspase-4 and caspase-5 proenzymes are weakly expressed in the OE, and they are not activated and their expression levels do not change over the course of bulbectomy-induced O R N apoptosis (figure 5.6.1). Therefore I conclude that they are not involved in O R N apoptosis, and they may possibly not even be expressed in the ORNs. Cellular localization of their mRNA or protein would have to be confirmed in situ by either in situ hybridization or immunohistochemistry. The same figure also shows that caspase-2 and caspase-7 are absent from the OE. It is potentially important to establish the absence of caspase-7 in this system, since caspase-7 is from the caspase-3 subfamily of caspases, has some overlap of function with caspase-3, and is structurally similar to caspase-3 such that many of the same inhibitors act on both. For these reasons, in some systems it has proved difficult to distinguish between the functions of caspase-3 and caspase-7.1 obtained tantalizing preliminary evidence that caspase-6 might be present in the OE, but unfortunately, due to technical difficulties with the antibodies, I never obtained a convincing or repeatable blot, either with anti-caspase-6 (MF424) or with anti-caspase-6 (StressGen). This is unfortunate because caspase-6, like caspase-7, has high identity with caspase-3 and can apparently subserve some of the same functions. Caspase-6 is known to be constitutively expressed in the brain (LeBlanc et al 1999). A sample of mouse liver was also included in the blots because there has been a report of caspase-6 in hamster liver. The critical role of caspase-3 in ORN apoptosis is illustrated by the lack of TUNEL+ ORNs and maintenance of OE thickness in caspase-3 null mice after bulbectomy (section 5.7). I conclude that ORNs of caspase-3 null mice do not undergo classical apoptosis and are not lost by another mechanism in the acute 72 hour time period after bulbectomy, the period in which almost all mature ORNs are lost in the wildtype. However, I have not addressed the possibility that the lesioned ORNs might eventually die by some much-delayed mechanism. The ORNs on the lesioned side after bulbectomy have no target, no synaptic contact, and indeed in most cases have undergone a deafferentation and have lost the ends of their axons. Therefore it seems somewhat unlikely that they would survive indefinitely, and almost impossible that they would remain completely unchanged in all their cellular characteristics. Any eventual death that might take place, may or may not be by apoptosis. There have been reports that under some circumstances, cells in vitro exposed to an insult which normally causes apoptosis, may instead undergo a delayed death lacking some of the features of classical apoptosis when their caspases are inhibited (McCarthy et al 1997, Brunet et al 1998). The delayed death is thought to be due to 49 the depletion of cytochrome c from the mitochondria and ensuing collapse of the electron transport chain. Alternatively, it has been shown in vivo that some neurons do not die after axotomy, but become atrophic, and can be rescued if appropriately treated up to a year later (Kwon et al 2002). Time (hrs) Af t e r Bulbec tomy 0 12 16 24 36 48 72 96 FIGURE 5.1.1 Apoptosis after bulbectomy: DNA laddering pattern D N A laddering, a characteristic feature of apoptosis, is shown in olfactory epithelial tissue different times after bulbectomy. Laddering is seen as early as 12 hours post-bulbectomy, reaching a maximum at 24-36 hours. unlesioned lesioned FIGURE 5.1.2 Apoptosis after bulbectomy: TUNEL labelling of ORNs proceeds away from the bulb a: low power coronal section of the olfactory epithelium 24 hours after bulbectomy, labelled with T U N E L to detect nicked D N A (dark nuclei). ORN nuclei show a huge increase in TUNEL-positivity on the lesioned versus unlesioned side. The midline septum is indicated, b: diagram of mouse head indicating site of bulbectomy and planes of coronal section shown in c-f. c-f: successive coronal sections of OE 24 hours post-bulbectomy. A wave of neuronal apoptosis is seen proceeding in a caudo-rostral direction from rear turbinates (f, where T U N E L density is highest), to the front (c, where the density is lowest). (Sep indicates septum). 52 200 back middle front distance from bulb — FIGURE 5.1.3 Spatial evaluation of TUNEL at 24 hours after bulbectomy At 24 hours after bulbectomy, coronal sections from the back, middle and front of the OE (defined as 1, 2.5 ± 0.45, and 4 mm from the cribriform plate respectively, corresponding approximately to f, e and c in figure 5.1.2b) were assessed to provide spatial patterns of the number of TUNEL-labelled ORNs. Turbinates in each section were assessed as belonging to superior (close to bulb), mediolateral, and inferior (distant from bulb). The number of T U N E L -positive ORNs was then plotted against distance from the olfactory bulb. 53 FIGURE 5.1.4 Apoptosis after bulbectomy: TUNEL labelling of ORNs indicates progression of apoptosis over time The caudo-rostral wave of apoptosis is also demonstrated in sagittal sections of OE at different times postbulbectomy. a-d: TUNEL-labelled sagittal sections of OE at 12, 24, 36, and 48 hours after bulbectomy respectively. ORNs closest to the olfactory bulb removal site (OB) are TUNEL-positive at 24 hours (b, black arrows indicate TUNEL-positive nuclei), while those further away die later (c, 36 hours; d, 48 hours). A red asterisk indicates the same turbinate in all sections to illustrate this. The earliest TUNEL-positivity (a, 12 hours) is actually detected in the lamina propria beneath the OE, not in the ORNs. Arrows indicate GFAP-positive OEG. 54 FIGURE 5.1.5 OMP decrease after bulbectomy confirms loss of ORNs Coronal sections of OE at 24 hours (a,c) and 48 hours (b,d) after bulbectomy, labelled for T U N E L (a,b) or OMP in situ hybridization (c,d). From 24 to 48 hours after bulbectomy, the density of OMP-expressing neurons detected by in situ hybridization decreases significantly on the lesioned side of the nasal septum, confirming that the cells lost to apoptosis are ORNs. 55 0 20 40 60 80 time after bulbectomy (hrs) FIGURE 5.1.6 Temporal analysis of ORN apoptosis by TUNEL and in situ hybridization For temporal evaluation of apoptosis, coronal sections of OE were analysed at 12, 24, 36,48 and 72 hours following bulbectomy by TUNEL. Three independent sites in a single mediolateral turbinate were assessed for TUNEL-positive neurons per linear mm at a single distance (2.5 ± 0.45mm) from the cribriform plate (n=3 animals/group). Numbers shown on y-axes are overall mean ± standard deviation for each group. Adjacent sections were analysed for number of cells found by in situ hybridization to express OMP mRNA, which is displayed as a percentage of the total cells above the basal lamina per linear mm of OE. 56 lesioned unlesioned FIGURE 5.1.7 Single-stranded DNA immunodetection shows the same pattern of apoptotic nuclei post-bulbectomy as TUNEL Lesioned (a) and unlesioned (b) sides of a coronal section of OE 24 hours after bulbectomy, labelled with a monoclonal antibody raised against single-stranded D N A . Positive ORNs (arrows) are seen on the lesioned side, in the same pattern as the labelling seen with the T U N E L technique. 57 proenzyme dimer 12 12 active tetramer 12 12 hours after bulbectomy 4 8 12 24 36 48 caspase-3 (MF 393) B-actin 24 48 caspase-3 (MF 393) B-actin 0 48 caspase-3 (SK) B-actin FIGURE 5.2.1 FIGURE 5.2.1 Caspase-3 proenzyme expression and activation in the OE during ORN apoptosis a: schematic diagram of the processing and activation of caspase-3. The 32 kDa proenzyme has its 3 kDa prodomain cleaved off, and is then further cleaved into 17 and 12 kDa large and small subunits. Two of each of these subunits then complex to form the active enzyme in the form of a heterotetramer. b-d: western blot analysis of pooled olfactory epithelium from mice recovering from bilateral bulbectomies (n=3 per group), b: the endogenous level of caspase-3 proenzyme (detected with MF393) increases during the first 24 hours after bulbectomy and then falls (because of activation). Using MF393 in a blot which has been resolved further (c) and previously described serum raised against the large subunit of human caspase-3 (d), the 32 kDa proenzyme is clearly present in unlesioned olfactory epithelium (0) and at 24 and 48 hours after bulbectomy. In addition, maximal activation of caspase-3 is demonstrated by the 17 kDa large subunit (arrow), detected 48 hours after bulbectomy. Loading control blots were probed with B-actin. MF393 = polyclonal caspase-3 antibody from D.W. Nicholson, Merck Frosst. SK = polyclonal large subunit caspase-3 antibody from S.H. Kaufmann. FIGURE 5.2.2 Caspase-3 activation in ORN axons and cytoplasm during apoptosis Coronal sections of OE from the lesioned (a) and unlesioned (b) sides of a 24 hour post-bulbectomy section, the lesioned side of a 48 hour section (c) and the unlesioned side of a different 48 hour section (d), labelled for active caspase-3 with MF397 (a,b), active caspase-3 with commercial antibody (PharMingen 67341A) (c), all with fluorescence secondary detection, or with OMP detected by the peroxidase method for light microscopy (d). a: with the more sensitive technique of immunocytochemistry, submaximal levels of active caspase-3 were found in a subpopulation of axons (Ax, partially stained axon bundles indicated with *) and some ORN cell bodies (ORN) at 24 hours after unilateral bulbectomy (MF397 antisera). b: axons and cell bodies of the side contralateral to the lesion at 24 hours (unlesioned) do not contain detectable levels of active caspase-3 (MF397). c: at 48 hours, there is an increase in axonal caspase-3 activation, with all axon bundles now containing active caspase-3 (PharMingen 67341A), and in partially labelled axon bundles (*) some breakdown in the integrity of the axon bundle can be seen, d: the unlesioned side still has a stable population of ORNs, expressing OMP throughout their cell bodies and axon bundles. 59 lesioned unlesioned a O R N b O R N A x 24h active caspase-3 (MF 397) active caspase-3 (Pharmingen) O M P F I G U R E 5.2.2 • < 9 WT • P i t t . csionecl AC-APP OE OE48 OB OB OE OE +/+_-/-IP: D2-II + + IB: Anti-AC-APP CT-12 C3 cut + + C3 cut APP APLP2 22 11 20 1 — , . 16 6  t i 22 19 20 ' 1 1 14 6 1 ( 1 — 1—| --- •— A .A 4 1 D2-II AC CT-12 F I G U R E 5.3.1 61 FIGURE 5.3.1 APLP2 is cleaved in the axons and cell bodies of apoptotic ORNs a: coronal section of OE 24 hours postbulbectomy labelled with the Delta-C-APP neoepitope antiserum, which recognizes a caspase-3-dependent epitope found in axons (arrowheads) and cell bodies (small arrows) of ORNs, only on the lesioned side of the septum, b: at higher power, there is a distinct mesaxon subcompartmentalization of the product in each axon bundle (arrowhead), and in the box indicated at higher magnification (c) there is a clear perinuclear distribution of the cleavage product (arrow) in the ORN soma. The basal cell layer is free of apoptotic cells containing this product, d: cell lysates from unlesioned OE, OE 48 hours after bulbectomy (OE48), and olfactory bulb (OB) were immunoprecipitated (IP) with an APLP2 affinity-purified N-terminal antiserum (D2-II) and probed (IB) with Delta-C-APP antiserum to confirm that the V E V D epitope is revealed within APLP2 during ORN apoptosis. To confirm that caspase-3 is required for APLP2 cleavage, OE extracts that were similarly prepared from wildtype (+/+) and caspase-3 null (-/-) mice were blotted with an APLP2 C-terminal antiserum, CT-12, which detects the cleavage product in wildtype (+/+) mice but not their caspase-3 null (-/-) littermates. FIGURE 5.4.1 Caspase-3-mediated APLP2 cleavage occurs first in the synaptic complexes, then axons and cell bodies of ORNs 24 hours after partial unilateral removal of the internal layers of the OB, transverse and sagittal sections (site and direction of section indicated on f) of OB (a,b) and OE (c,d) were analysed for the compartmentalization of the APLP2 Delta-C epitope (green, using the Delta-C-APP antibody) in comparison with TUNEL+ cells (black nuclei in a,c) and neuron-specific tubulin (NST, red in a,c), synaptophysin (red in b), and active caspase-3 (red in d) expression. In the lesioned OB, the Delta-C-APP epitope is concentrated in the synaptic glomerular complexes (Gl) and axons (Ax) of the nerve fibre layer (NFL), in which the signal becomes yellow where colocalization with axonal NST (a) or glomerular synaptophysin (b) occurs, a: NST+ (red) axons and glomeruli of the unlesioned bulb (the midline between bulbs indicated by white dotted line) do not show any evidence of caspase-3 activation or APLP2 cleavage, c: by 36 hours after bulb lesion, a low-power image of a mediolateral turbinate shows that the Delta-C-APP epitope (green) production is highest in NST+ (red) axon bundles (overlap in yellow) closest to the olfactory bulb (marked «- to OB). Neurons (ORN) most distant from the midline were less immunoreactive for Delta-C-APP (green) and T U N E L (*) and lie adjacent to axons with minimal Delta-C-APP/NST overlap (red), d: a retrograde pattern of caspase-3 activation is represented on a sagittal section of turbinate immediately adjacent to the septum, in which active caspase-3 (red) is transiently activated in axons and cell bodies (ORN), leaving cleaved APLP2 (Delta-C-APP, green) in its axonal wake. OB marks direction of olfactory bulb location, and arrow aligns the plane of section with that shown in f. e: sections of successive mediolateral turbinates were sorted into quadrants, based on axonal Delta-C-APP/NST labelling on a scale from 0 to 100%. 0% represents NST+ only (red), and 100% represents an entirely yellow axon bundle (complete overlap). ORN soma within each quadrant were then assessed for activation for caspase-3 (green) and TUNEL+ (black nuclei) in correlation with their adjacent axonal caspase-3 activation (mean per quadrant ± SD; three independent sections per animal). Scale bars: a-d, 50 pm; f, 500 um. 62 % DeltaC-APP / NST overlap FIGURE 5.4.1 63 a epitopes recognized by the caspase-9 antibodies CPEPD C9(SK) neoepitope 130 \ / . \ / I prodomain 1 p!7 1 pi 2 I C9(MF443) 330 C9(Bur49) b activation scheme for caspase-9 proenzyme 15 17 12 active form 15 17 • 12 15 17 1 12 12 15 17 | FIGURE 5.5.1 Schematic diagram of caspase-9 protein a: regions of the caspase-9 proenzyme to which the various antisera used in this study have been raised, b: the activation scheme of caspase-9. This is slightly different from the scheme shown for caspase-3 in figure 5.2.1a in that the large subunit attached to the prodomain comprises the active form, rather than the large subunit alone. 64 hrs post-bulbectomy 0 4 8 12 16 24 36 48 72 caspase-9 (MF 443) caspase-9 (Bur 49) 32 KDa - m 18KDa caspase-9 (SK Neoepitope) 132 KDa-apaf-1 B-actin F I G U R E 5.5.2 Caspase-9 proenzyme expression and activation in the O E during O R N apoptosis a-c: examination of caspase-9 levels in OE after bulbectomy by western blotting, using three of the different antisera depicted in figure 5.5.1. Caspase-9 proenzyme levels rise significantly in OE at 4-24 hours following bulbectomy as assessed using two different sera raised against recombinant human caspase-9 (MF443, a; and anti-C9/Bur49, b). In addition, a cleavage product (-*b,c) corresponding to the 35 kDa activated form of caspase-9 is detected initially at 4 and maximally at 24 hours following bulbectomy. c: a neoepitope antiserum recognizing the cleavage junction between the p 19 and pl2 subunits of caspase-9 detects the 35 kDa product representing the large subunit and prodomain (as well as a 19 kDa product which could represent the individual large subunit) at 4 and 24 hours (maximal) following bulbectomy. Control blots utilized antibodies to (d) Apaf-1 (Stressgen) and (e) mouse anti-beta-actin (Sigma). 65 ORN F I G U R E 5.5.3 Caspase-9 expression in O R N axons and synapses a: a coronal section of normal OE labelled with anti-caspase-9 (Bur49) shows that caspase-9 is present in the axon bundles (Ax) but not the cell bodies (ORN) of the ORN. b: caspase-9 (green, 200x) is expressed in axons of the normal adult OE, emanating from OMP-positive (red) cell bodies (ORN) and co-localizing with axonal OMP as olfactory axons cross the cribriform plate (CP), course through the nerve fibre layer (NFL) and enter the glomeruli (G) of the olfactory bulb, c: caspase-9 (red, 400x) colocalizes at the presynaptic compartment of OMP+ (green) olfactory neurons with synaptophysin (blue) to produce a purple overlap highly localized to the inner layers of glomeruli. White areas indicate regions of triple overlap of caspase-9, synaptophysin and OMP. Caspase-9 OMP synaptophysin 66 OE, hours after bulbectomy 0 72 48 36 24 16 12 8 a) caspase-4 (MF374) liver 45 — 32 — 18 — 6 — u > u c -a M X 8 o u w O - 4 5 - 3 2 u V > *-> n> t> a © e c) caspase-7 (MF413) o - 4 2 - 3 2 -18 b) caspase-2 (sc) F I G U R E 5.6.1 Expression of other caspases in the mouse O E Western blots to show that (a) the proenzyme form of caspase-4 (43 kDa) is expressed at very low levels in the OE and does not become upregulated or activated after bulbectomy. In the liver the proenzyme and cleaved forms are present. Caspase-2 (b, 48 kDa) and caspase-7 (c, 35 kDa) are not expressed in the mouse OE, but are detected in positive control tissues. Antibodies shown are anti-caspase-4 (MF374), anti-caspase-2 (Santa Cruz), and anti-caspase-7 (MF413). 67 hours after bulbectomy 0 4 8 12 16 24 36 48 72 32 kD — 18 kD — a) anti-caspase-10 (MF459) 33 kD 18kD — b) anti-caspase-8 (MF438) 32 kD-18 kD — c) anti-caspase-8 (SK) d) caspase-81 P25 I P^u I piu i = 5 5 k D caspase-101 P23/27 ~ | P2U I p!2 1 = 55/59 kD F I G U R E 5.6.2 Caspases-8 and -10 in the olfactory epithelium after bulbectomy Western blots of olfactory epithelium tissue taken 0-72 hours after bulbectomy, probed with anti-caspase-10 (MF459) (a), anti-caspase-8 (MF438) (b) or anti-caspase-8 (SK). Equal loading was confirmed by Ponceau S staining. In each blot, three discrete cleaved forms are detected between 20 and 32 kDa, which could possibly represent large subunit alone, prodomain alone, and large plus small subunit respectively. The expected sizes of these species are indicated in (d), a schematic diagram of caspase-8 and caspase-10 proteins. 68 unlesioned lesioned / III a F I G U R E 5.6.3 Caspase-8 is activated on the lesioned side of the O E after bulbectomy Transverse section of the unlesioned side (a) and lesioned side (b) of the OE 16 hours after bulbectomy, labelled with an antibody specific for the activated form of caspase-8. Active caspase-8-positive cells are seen in the neuronal layer of the OE on the lesioned side only. wildtype caspase-3 -/-F I G U R E 5.7.1 Caspase-3 is required for O R N apoptosis after bulbectomy in the adult mouse T U N E L was used to detect nicked D N A (dark nuclei), indicative of apoptosis. TUNEL-labelled coronal sections of OE are shown from wildtype (a,c,e) and caspase-3 null (b,d,f) mice, sacrificed either 24 (a,b), 48 (c,d) or 72 hours (e,f) after removal of the ipsilateral olfactory bulb. At 24 hours post-bulbectomy there is intense TUNEL labelling on the lesioned side of the OE in wildtype animals (a). The number of positive cells decreases at 48 hours (c) and 72 hours (e) post-bulbectomy (by which time many ORNs have completed apoptosis and been removed, see figure 5.1.6), but is still significantly increased above the very low levels seen on the unlesioned side. In the caspase-3-deficient mice, however, no such increase in TUNEL+ ORNs was seen at 24 (b), 48 (d) or 72 (f) hours in any of the null mice tested (n=3 animals per genotype and timepoint; minimum of three sections in three different T U N E L runs tested per animal). 70 OE of caspase-3 null mice retains normal thickness 72 hours after bulbectomy caspase-3 -/- wildtype F I G U R E 5.7.2 O E of caspase-3 null mice retains normal thickness 72 hours after bulbectomy Graph to show that 72 hours after unilateral bulbectomy the lesioned side of the OE in wildtype mice is only 0.70 the thickness of the unlesioned side. However, in caspase-3 knockout mice the ratio of lesioned:unlesioned side is 1.09. Coronal sections of OE (n=3 animals of each genotype) were stained with haematoxylin and eosin in order to be able to identify the neuronal layer. For each section, photographs were taken of three locations on each side. The thickness of the neuronal layer was then measured blindly. Each lesioned side value was divided by its corresponding unlesioned side value to give a set of ratios. The sets of ratios for each genotype were compared using a one-tailed t-test, giving p=0.009. 71 CHAPTER 6: CASPASE-3 AND CASPASE-9 IN DEVELOPMENT OF THE MOUSE OLFACTORY SYSTEM The majority of this chapter is in a manuscript (title as above, authors Cowan and Roskams) to be submitted to J. Comp. Neurol. Data in figures 6.6.5 and 6.8.1 have been published (Cowan et al 2001). 6.1 The caspase-3 and caspase-9 null mice It has been reported that caspase-3 null mice and caspase-9 null mice are smaller than their littermates, and often have hydrocephaly. I was able to confirm that this is also the case in mice from the Roskams' laboratory colony. Figure 6.1.1 shows a young adult caspase-3 - / - mouse shorter than his wildtype littermate (6.1.1a-c), and an example of another young adult caspase-3 - / - mouse which was found upon dissection to have a domed and fragile skull with prominent blood vessels visible underneath (6.1. Id). Approximately one-third of caspase-3 - / - mice born alive were small with hydrocephaly. The remaining majority of viable caspase-3 nulls had an apparently normal external appearance (i.e. normal size and head shape), but upon sectioning the brain and viewing under the microscope, they also did have very obvious structural abnormalities (see section 6.6). The caspase-9 null embryos generated in our colony also displayed abnormally shaped heads, which were often more prominent than normal in the region of the neocortex. Regarding frequency and lifespan of the caspase null mice, it has been reported that caspase-3 null mice generally live to 1-3 weeks of age. Brain abnormalities begin to be discernible by E12, at which stage caspase-3 - / - mice occur with Mendelian frequency. However, at birth only 7% of offspring of heterozygous parents are - / - , so there has been some embryonic lethality (Kuida et al 1996). In our mice the onset of visible abnormalities was also E12. The large ventricle of the E l 3 caspase-3 - / - mouse photographed in figure 6. L i e provides an example. However, the frequency and lifespan of the caspase-3 mice in our colony were quite different from those reported in the original publication of these mice (table 6.1). Out of 268 mice born alive to caspase-3 heterozygous parents in the Roskams' laboratory colony over the past 4.5 years, 43 (16%) were caspase-3 - / - , twice as many as reported (Kuida et al 1996). More striking, however, was the lifespan of those caspase-3 - / - mice that did reach birth. Out of 74 caspase-3 - / - mice born alive in our colony (some were offspring of a caspase-3 - / - parent and were therefore not included in the previous statistic), approximately 10% died within the first postnatal day. However, 65% survived beyond 6 weeks of age, and 18% (13 mice) survived beyond 6 months. The oldest surviving caspase-3 - / - mouse that I documented reached 10 months of age. It should be noted that only a minority of these mice died of natural causes, but many were sacrificed for experiments and therefore the true lifespan will be longer. Note also that mice sacrificed for experiments at neonatal timepoints were excluded from this statistic, for this reason. Caspase-9 null mice in our colony showed a frequency and lifespan a little closer to those reported in the literature, but in this case our mice seemed to experience slightly earlier lethality. Onset of abnormalities is reported to be E12.5 (Kuida et al 1998, Hakem et al 1998). One group reports 17% of caspase-9 - / - mice surviving to E13, and 2.6% surviving to P5-P20 (Kuida et al 72 1998). The other group reports that 7% survive to PI , but die by P3 (Hakem et al 1998). In our colony only 2 of 213 mice born alive to heterozygous parents were caspase-9 - / - , and both died within 24 hours. A further four caspase-9 - / - mice died around the time of birth. These six mice together accounted for 2.5% of births. This is very close to the percentage reported to survive postnatally by Kuida's group, except of course that none of mine survived to P20. Out of 35 mice sacrificed between the ages of E14 and E19, none was caspase-9 null. At E13, four of 50 embryos (8%) were caspase-9 - / - , less than previously reported. 6.2 Caspase-3 and caspase-9 expression in the developing olfactory system Caspase-9 was strongly expressed in ORN axons during development (figure 6.2.1). As early as E l l , caspase-9 expression was detected in the presumptive OE of the olfactory pit (6.2.1a). Expression persisted through E13 (6.2.1b) and E15 (6.2.1c). By the time evagination of the olfactory bulb is complete at E17, caspase-9 was seen throughout the ORN axons from the cell bodies in the neuronal layer of the OE, through the cribriform plate and into the olfactory bulb, where caspase-9-filled axons formed the nerve fibre layer (6.2.Id). Caspase-9 was still expressed throughout the ORN axon by P5, when caspase-9-expressing axonal projections were seen forming new glomeruli (6.2.le). We have also detected caspase-9 expression in axonal projections throughout the neocortex from E13-15, and in the axons of many different sensory neurons (DRG, trigeminal nerve, optic nerve) throughout embryonic development (data not shown). However, this axonal expression was lost postnatally in all sensory nerves apart from the olfactory nerve, where it remained strongly expressed. Caspase-3 was also strongly expressed in ORN axons during development, as well as in the cell bodies of ORNs, and in many cell types of the olfactory bulb (figure 6.2.2). Caspase-3 protein expression was not seen in the presumptive olfactory system at E10 (6.2.2a). The earliest time caspase-3 could be detected by immunohistochemistry in cells of the presumptive OE was at E l 3 (6.2.2b), and by E l 5 caspase-3 immunoreactivity could be clearly seen in the cytoplasm of ORN cell bodies and axons (6.2.2c). This ORN expression persisted through PI (6.2.2d) and P5 (6.2.2e) to adulthood (see chapter 5). By E17, when ORN axons form the nerve fibre layer of the olfactory bulb, caspase-3 immunoreactivity was detected here (not shown at E17, although the nerve fibre layer is visible in 6.2.2g,f: P5 and P7), and diffusely in the bulb itself. By neonatal timepoints, strong caspase-3 expression was seen in many areas of the bulb including the glomeruli, mitral cells, granule cells, and the external plexiform layer which contains local interneurons (e.g. at P5, 6.2.2g), as well as in the accessory olfactory bulb (not shown). Outside of the olfactory system, caspase-3 was widely expressed in the neocortex and spinal cord throughout development. I also observed caspase-3 immunoreactivity in many sensory nerves from E l 2 to E l 7 , notably dorsal root ganglia, optic nerve (and other cells of the developing eye), and the hypoglossal nerve (data not shown). Expression in the retina persisted through PI and P5. 6.3 Decreased apoptosis in the olfactory system of caspase-3 and caspase-9 null E13 embryos In order to obtain a baseline for apoptosis in wildtype mice, I first labelled sections of OE from 73 wildtype mice at a variety of stages of development using TUNEL, and counted the number of positive cells in the OE neuronal layer. The peak time of apoptosis in the OE was at E l 3 , after which time apoptosis decreased to very low levels. Figure 6.3.1 shows some representative examples of TUNEL-labelled sagittal sections. In this example, the E l 3 section has 33 apoptotic ORNs (a), E15 has 0 (b), E17 has 4 (c), and P7 has 0 (d). This was quantified by counting the number of TUNEL-positive nuclei in the OE and measuring the length of the OE, for each sagittal section. Figure 6.3.2 shows that at E13, on average 16 cells per mm of OE were apoptotic. Thereafter only 2 cells per mm or less were apoptotic. In contrast to wildtype mice, both caspase-3 and caspase-9 null mice displayed significantly lower levels of apoptosis in the OE at E13. Figure 6.3.3 shows some representative examples of TUNEL-labelled E l 3 sagittal sections. In this example, the wildtype control section contains 33 apoptotic ORNs (a), caspase-3 - / - has 8 (b) and caspase-9 - / - has 11 (c). Figure 6.3.4 shows that while both wildtype and caspase-3 heterozygous embryos contained 10-15 T U N E L -positive ORNs on average per mm of OE, caspase-3 and caspase-9 null embryos contained only about 2.5. This is consistent with the situation reported in the brain, where both caspase null mice have dramatically fewer apoptotic neurons, but some neuronal apoptosis is still able to occur in the absence of caspase-3 or caspase-9 (Kuida et al 1996,1998, Hakem et al 1998). To confirm the OE identity of the region studied in this section, I double-labelled an E l 3 wildtype section with an antibody for GAP-43, in conjunction with TUNEL. GAP-43 is transiently expressed in maturing ORNs. GAP-43 expression is found in the same epithelial population as the TUNEL-positive cells shown and counted in figures 6.3.1-6.3.4 (figure 6.3.5). One structure that contained noticeably more TUNEL-positive cells in the wildtype than in the caspase-3 null at E l 3 , was the developing eye (optic recess of the diencephalon, figure 6.3.7b). In other parts of the brain, a difference was not obvious without performing a cell count, although such a difference is reported (Kuida et al 1996). TUNEL labelling appeared sparser in the caspase-9 null E l 3 in general, when compared to both the caspase-3 null and wildtype. This is illustrated in figure 6.3.6, which shows a low power view of TUNEL-labelled caspase-3 - / -and caspase-9 - / - embryos. One exception to this is the wall of the oesophagus, which is intensely T U N E L labelled in the caspase-9 null (figure 6.3.7c,d). An interesting point that I noticed during these experiments was that apoptosis in the wildtype E l 3 OE was often zonal, that is, TUNEL-positive cells were often clustered close together in one area (figure 6.3.7a). 6.4 Caspase-3 and caspase-9 null mice have a thicker OE neuronal layer at E13 It was apparent from figure 6.3.4, showing fewer apoptotic neurons in caspase-3 - / - and caspase-9 - / - embiyos than in wildtype, that the OE also appears noticeably thicker in caspase-3 - / - and caspase-9 - / - than in wildtype. I have highlighted the basement membrane with a dotted line to facilitate seeing the extent of the OE neuronal layer. Parts of the OE in caspase-3 - / -and in caspase-9 - / - were of normal thickness, and parts had pocketed out to 5 times that thickness. To quantify this, I measured the perpendicular thickness of the OE neuronal layer in each genotype. Figure 6.4.1 shows that caspase-3 - / - and caspase-9 - / - did indeed have a thicker OE than wildtype and both heterozygotes. It can be seen from this graph what was already apparent from viewing the sections: that the sample variance was considerably bigger in both null mice. 74 This is better illustrated in a scatter graph (figure 6.4.2). Almost all the measurements taken for wildtype, caspase-3 heterozygote and caspase-9 heterozygote were between 12 and 75 pm. Only one wildtype measurement was above this (107 pm), and four caspase-3 heterozygote measurements (out of 78) were above this range. In contrast, caspase-3 - / - measurements ranged from 13 to 294 p,m and caspase-9 - / - measurements ranged from 10 to 386 p,m. 6.5 Identity of OE cells which undergo caspase-dependent or -independent apoptosis As demonstrated in sections 6.3 and 6.4, many extra cells are seen in the OE of caspase-3 null and caspase-9 null E13 embryos, due to decreased apoptosis, However, a small proportion of cells in the OE retain the ability to undergo apoptosis in both null embryos (as seen in figure 6.3.4). This section addresses whether there is an identifiable difference in identity between those cells which are resistant to apoptosis in the absence of caspase-3 or caspase-9, versus those that are still susceptible. In figure 6.3.5, GAP-43 immunolabelling was used in conjunction with T U N E L to confirm that the cells in question were indeed localized in the neuronal layer of the OE (as opposed to the respiratory epithelium). However, since GAP-43 is expressed transiently and, at any one moment, only a proportion of ORNs are GAP-43-positive, this does not confirm whether individual cells definitely possess O R N identity. To achieve this I immunolabelled sections for N C A M . ORNs express N C A M throughout their life, from commitment to a neuronal fate onwards. Figure 6.5.1a-c shows that the expanded population of cells seen in the OE in the caspase-9 null E l3 embryo were indeed NCAM-positive, and therefore developing ORNs. The same result was seen in the caspase-3 null embryo. Double fluorescence labelling for N C A M and TUNEL in sections of OE from a caspase-3 - / - E13 embryo (figure 6.5.1d,e) revealed that TUNEL-positive nuclei were not surrounded by N C A M -positive cytoplasm. The same result was seen in the caspase-9 null embryo. Therefore, cells which can die in the E l 3 OE in the absence of caspase-3 are probably not ORNs, whereas cells of the E l 3 OE which survive abnormally in the caspase-3 null are ORNs. The situation in the adult contrasts with that in the E l 3 embryo. I have shown that mature ORNs dying acutely after bulbectomy in the adult require caspase-3 for this apoptosis, as it does not occur in the caspase-3 null (section 5.8). However, the occasional TUNEL-positive apoptotic O R N was seen in adult OE, whether on the unlesioned side of bulbectomized animals, or in unoperated animals. This may possibly represent either immature O R N death, or normal O R N turnover, which occurs every 4-6 weeks. Interestingly, these occasional apoptotic cells were also seen in caspase-3 - / - animals, in lesioned, unlesioned and unoperated OE (figure 6.5.2). These sporadic dying cells were sometimes seen close to the base of the epithelium. Positive overlap in double immunofluorescence with T U N E L and GAP-43 proved that at least some of these cells which underwent apoptosis in the absence of caspase-3 were immature ORNs (figure 6.5.2c). This is in contrast to the situation in the E l 3 embryo in which TUNEL-positive cells in the caspase-3 or caspase-9 null were not NCAM-positive. The rationale for the choice of N C A M or GAP-43 as a marker of immature ORNs in these experiments is based on the fact that GAP-43 is expressed transiently in immature ORNs, while N C A M is expressed throughout the life of immature and mature ORNs. Therefore N C A M is an appropriate marker of immature ORNs at E l 3 , since at this time there are not yet any mature ORNs. Furthermore, as mentioned, the entire population of immature ORNs expresses N C A M while only a subpopulation is GAP-43-positive. However, in the adult, N C A M staining does not distinguish between immature and 75 mature ORNs. GAP-43 is more appropriate since it only detects immature ORNs (although not all of them). 6.6 Abnormal shape and size of olfactory bulbs in caspase-3 null adults Wildtype adult olfactory bulbs were of a fairly uniform size (figure 6.6.1a). In contrast, caspase-3 - / - adult olfactory bulbs could be much smaller or larger than normal (figure 6.6.1b,c). This applies to overall volume (6.6.1c), cross-sectional area, and rostro-caudal depth (6.6.1b). For each section measured, I also calculated the shape factor using Northern Eclipse imaging software. The formula incorporates the area and circumference of the shape to generate a number between 0 and 1, where 1 is a perfect circle and 0 is a line. Figure 6.6.3b shows the average shape factor for each genotype. Caspase-3 - / - adults with a normal external physical appearance had normal shaped bulbs. In contrast, those caspase-3 - / - adults with hydrocephalus had abnormally shaped bulbs. Most notably, in the hydrocephalic animals the minimum shape factor seen in the most irregularly shaped part of each bulb was much lower. Figure 6.6.2b shows an example of an olfactory bulb cross section from a hydrocephalic caspase-3 - / -individual with a shape factor approximately half of normal. Figure 6.6.2a shows an example of a normal shaped wildtype bulb. Caspase-3 - / - bulbs may take on unusual shapes more often when hydrocephalus is present, because the bulb is under pressure from its caudal surface, and so is often seen to break through the cribriform plate so that it can extend rostrally into the nasal cavity. Interestingly, these pieces of ectopic bulb might be functional, as they showed normal expression of tyrosine hydroxylase (figure 6.6.2c). In addition, several of the adult caspase-3 null bulbs examined had unusually thick nerve fibre layers, and two animals had noticeably enlarged glomeruli (figure 6.6.4). To examine whether caspase-3 - / - mice have more ORNs due to reduced developmental apoptosis, I tested whether caspase-3 - / - mice have more ORNs per linear mm of OE, more ORN axon bundles per mm of OE, and bigger ORN axon bundles. I performed immunohistochemistry on coronal sections of OE from caspase-3 - / - and caspase-3 +/- young adult animals with anti-NCAM to label all ORNs, and counterstained with haematoxylin and eosin. The results were somewhat surprising. There was no significant difference in the ORN cell body number/linear mm of OE between knockouts and littermates (data not shown). This could be because the population density within the OE in the basal-apical dimension is tightly controlled by local factors. A population expansion could be accommodated by additional folding of OE turbinates (instead of increasing the thickness of epithelium), and this appeared to be the case. In caspase-3 - / - mice the average axon bundle was significantly larger (2-6x) than in littermate controls (figure 6.6.5b, and see examples in figure 6.6.4). Contrary to what may have been predicted, however, caspase-3 nulls actually had fewer axon bundles than their littermates (figure 6.6.5c). But overall, combining these two sets of data, there was indeed an increased total cross-sectional area of axon bundle in the caspase-3 -/— (figure 6.6.5a). 76 6.7 Normal size and shape of olfactory bulbs in caspase-3 null neonates The size and shape of the olfactory bulb were abnormal in caspase-3 - / - adults. But since there are too many ORNs already at E l 3 , which might be expected to have a profound effect on the development of the bulb, is the bulb already an abnormal size after embryonic development is complete? Or, is the extensive olfactory-activity-dependent development which takes place in the first few postnatal weeks more important? To answer this question I measured the size of olfactory bulbs from mice at P4. Figure 6.7.1 shows that at P4, olfactory bulbs of caspase-3 null mice had a normal size and shape, indistinguishable from wildtype. There were no gross differences in the bulb between caspase-3 and control bulbs in the neonate, but differences were detected by immunochemistry. Such a difference which might be relevant here is that caspase-3 nulls seemed to have more intense immunoreactivity for tyrosine hydroxylase, which is a marker in the bulb for periglomerular interneurons (figure 6.7.2). I also counted the number of TUNEL-positive nuclei in the OE at P4 in caspase-3 null and control, using the same strategy as that employed to analyse the E l 3 data set. I found that at P4 there was now no difference in levels of OE apoptosis between caspase-3 null and control (figure 6.7.3). 6.8 Increased caspase-9 expression in the caspase-3 null mouse To examine whether lack of caspase-3 has any effect on regulation of other molecules in the pathway, I compared OE from 1-week-old wildtype and caspase-3 - / - mice by immunohistochemical and protein expression analysis of sibling groups. A measurable increase in the already high levels of caspase-9 was evident throughout the neuraxis of the caspase-3 null mice at P7 when equivalent sagittal sections were compared with their wildtype littermates (figure 6.8.1a). Fluorescence confocal microscopy (Z-series) was utilized to concurrently visualize and quantify OMP and procaspase-9 expression in caspase-3 - / - and caspase-3 +/+ littermates (see figure 6.8.1b,c for pseudocoloured image). OMP expression was almost 2-fold greater in each caspase-3 null mouse than in its littermate, and caspase-9 expression almost 5-fold higher. The increased OMP expression reflects the expansion of the mature neuron population in this mutant (see section 6.6), but the additional specific increase in caspase-9 represents increased caspase-9 expression in each neuron. Note that the 1.8x increase in OMP expression found by confocal analysis is consistent with the 1.5- to 2-fold expansion in total axon area seen in the caspase-3 null (figure 6.6.5). DISCUSSION Caspase-3 and caspase-9 are expressed in the presumptive OE by E l 3 (figures 6.2.1 and 6.2.2), and this is the peak period of apoptosis in this tissue {figures 6.3.1 and 6.3.2), therefore the caspases are in the right place at the right time to potentially mediate apoptosis. I show that wildtype levels of apoptosis in the OE are 8-fold higher at E l 3 than at subsequent developmental stages. Establishing the timecourse of normal embryonic apoptosis in the OE is especially important since it has remained ill-defined for many years, and the two existing 77 publications on the subject appear potentially contradictory. The central point on which the two papers and my data are consistent, is in describing a peak of apoptotic nuclei in the OE at E l 2 to E13 (Pellier and Astic 1994a,b, Voyron et al 1999). As mentioned in the Introduction, this corresponds to the time just after the involution of the olfactory pit when the olfactory turbinates are being formed. There is also apoptosis in the area prior to this time (Voyron et al 1999 see many apoptotic nuclei at E l 1), but until E l 2 or 13 the OE is not sufficiently delineated for quantitative analysis to be practical (Voyron et al 1999), and this is the reason my analysis began at E l 3 . My work agrees with Pellier and Astic in finding very few apoptotic nuclei in the neuronal layer of the OE after E l 4 . However, a subsequent very thorough quantitative study (Voyron et al 1999) describes a second wave of apoptosis peaking at E l 6 , which the authors suggest represents selective apoptosis as ORNs are making synapses with the bulb. I believe that differences in sampling strategy reconcile this difference, and actually add to our understanding of OE development. The most important difference is that Voyron et al 1999 performed cell counts only at the septum (at the midline of the OE), whereas I assessed fewer sections but counted all the nuclei in the entire sagittal plane. Given that ORNs make synapses with the bulb over a protracted period from E14 to birth, and that the OE is a complex three-dimensional structure with each area of each turbinate a different distance form the bulb (see section 5.1), it seems probable that ORNs from different local areas would undergo connection-related apoptosis in a temporally staggered manner. Therefore if one concentrated on a specific area of OE, a second peak of apoptosis might be seen at some specific time between E14 and birth (which would vary from region to region), whereas if one assessed the OE as a whole, numbers of apoptotic nuclei would remain low and unchanging over this period. Having confirmed that the peak of apoptosis in wildtype OE is at E13,1 went on to demonstrate that caspase-3 and caspase-9 null E l 3 embryos have 4- to 5-fold fewer apoptotic ORNs than wildtype littermates (figures 6.3.3 and 6.3.4). This indicates that a majority, but not all, of the programmed cell death that occurs in the OE at E l 3 is caspase-dependent. The caspase-3 - / - and caspase-9 - / - E13 embryos are comparable in that they both have a dramatically decreased number of apoptotic neurons in the OE, when compared to wildtype (figures 6.3.3 and 6.3.4). However, the situation in the rest of the E13 brain is quite different between caspase-3 - / - and caspase-9 - / - . In the caspase-9 - / - embryo, apoptosis appears to be dramatically decreased everywhere in the brain. I could only detect very, very sparse labelling in the cortex and DRG. In contrast, in the caspase-3 - / - embryo, there are many apoptotic cells in the cortex, spinal cord, and striatum. I could not say without doing a detailed cell count whether there are less apoptotic cells in the brain in general in caspase-3 - / - than in wildtype, as reported (Kuida et al 1996). However, there are clearly fewer apoptotic cells in the brain in caspase-9 - / - compared to either caspase-3 - / - or wildtype. Interestingly, there is one structure of the brain where I consistently noticed strong T U N E L labelling in wildtype El3s, but never in caspase-3 - / - El3s, and that is in the developing eye (figure 6.3.7b). As mentioned earlier, strong caspase-3 immunoreactivity was seen in the developing retina in wildtype embryos (section 6.2). It has recently been shown that developing mice which are null for the proapoptotic Bcl-2 family member bax or bak, or both, display a dramatic reduction in developmental apoptosis of the retina (Hahn et al 2003). It would be interesting to perform double-labelling with T U N E L and immunochemistry for markers of specific retinal cell types in the caspase-3 null and caspase-9 null mice, and conversely to double label with T U N E L and 78 immunochemistry for ORN markers in the bax and bak null mice, to investigate whether all these proteins are involved in the same pathways in these two developing sensory neuron populations. There also appear to be fewer than normal apoptotic cells present in other parts of the body in the caspase-9 null embryo (figure 6.3.6). In caspase-3 null and wildtype E13 embryos, I observed many TUNEL-positive nuclei in the tongue, the wall of the oesophagus, the lower jaw, non-neuronal cells of the nose, and very strong labelling in the liver. The only place that appeared to have equivalent labelling in the caspase-9 null was the wall of the oesophagus (figure 6.3.7c,d). The liver had moderate labelling, which seemed weaker than control. Elsewhere labelling was very sparse. This was my observation from careful examination of 9 sections from 3 animals for each genotype, but cell counts were not performed. If true, this would be in contrast to the literature, which states that the caspase-9 null phenotype is restricted to the brain. I find that apoptosis in the wildtype E l 3 OE is often zonal (figure 6.3.7a). Pellier and Astic (1994a) documented a particular region in the posterior part of the olfactory pit where apoptosis is particularly prevalent at E l 3, but zonal apoptosis in the sense of small clusters of cells undergoing apoptosis together has never before been reported in the developing OE. Zonal apoptosis is known to occur in the adult OE, where the phenomenon is attributed to many ORNs expressing the same receptor and projecting to the same glomerulus undergoing turnover within the same period of time. Perhaps in development it could simply be a mechanism for shaping the turbinates of the OE, as is seen in so many other tissues. I show that caspase-3 and caspase-9 null E l 3 embryos have a thicker OE neuronal layer, with more variance in thickness, than wildtype littermates (figures 6.4.1 and 6.4.2). Perhaps the fact that this thickness increase is not uniform, but rather occurs as pockets of vastly increased thickness, reflects the fact that the apoptosis which should be occurring is zonal. Given that the caspase-3 null and caspase-9 null E l 3 embryos have a thicker OE (in the basal-apical direction), and my measure of the number of apoptotic ORNs was per linear mm of OE, it is possible that my data showing 75-80% fewer apoptotic ORNs in both transgenic mice (figure 6.3.4) could be an underestimate. Had it been feasible to express the death as the proportion of all ORNs that are apoptotic, the decrease in apoptosis would surely have been even more dramatic. The thicker OE seen at E13 in the caspase null mice is hardly surprising: if fewer neurons are dying, there might well be more of them. However, in the adult caspase-3 null (sections 6.6 and 6.8), there are more ORNs without any change in thickness of the OE. Therefore the increased thickness of the OE seen at E13 appears to return to normal as the mouse develops. One possible explanation is that local controls are operating in the more mature OE that are able to tightly regulate epithelial thickness. As a result, perhaps the OE is forced to accommodate the expanded O R N population by increasing folding during turbinate formation, but there is currently no evidence available to support this suggestion. There is a difference in requirement for caspase-3 between mature and immature olfactory neurons in the adult animal: in the caspase-3 null adult, immature ORNs can undergo apoptosis 79 (figure 6.5.2), but mature ORNs cannot undergo acute bulbectomy-induced apoptosis (section 5.7). There is also a difference in requirement for caspase-3 in ORN apoptosis between mature and immature olfactory systems: in the caspase-3 null E l 3 embryo ORNs (by definition immature) fail to die (figure 6.5.1), whereas as mentioned, immature ORNs can die in caspase-3 null adults. This cannot be fully addressed in the same way for caspase-9, since null mice do not reach adulthood, however all analyses at E l 3 have yielded the same result for the two caspase null mice. It would be interesting to know the identity of the cells that are able to die in the E13 OE in the absence of caspase-3 or caspase-9, as well as the mechanism that they are using to die. Others in the Roskams' laboratory are currently investigating this, and testing the hypothesis that Apoptosis Inducing Factor might play a role. Apoptotic nuclei at E l 3 (whether wildtype or caspase null) were observed in all regions of the OE neuronal layer, from basal to apical, as can be seen in figures 6.5.1, 6.3.1 and 6.3.3. Voyron et al 1999 also describe this in wildtype rat: at later embryonic and postnatal stages in the rat, when maturity of an ORN can be distinguished by immunoreactivity for OMP or GAP-43, apoptotic cells are found at all stages of the lineage. But even at E l 3 when no ORNs are mature, both Voyron et al (1999) and this study find apoptotic nuclei situated at different levels within the thickness of the OE. In sagittal sections of the E13 olfactory system (for example figures 6.5.1 and 6.3.3), apoptotic nuclei can be seen just outside the OE (in the nasal mesenchyme that separates the OE neuronal layer from the telencephalon; in 6.3.3 this is outside of the area marked by a dotted line, in 6.5.1 this can be identified as the NCAM-negative area). This phenomenon has been previously reported. Astic's group have published several papers studying the neurons that migrate away from the OE and into the nasal mesenchyme at this time along with accompanying glial precursors, forming the presumptive olfactory nerve. They report extensive apoptosis of non-neuronal cells of the nasal mesenchyme at E13 and E14 (Pellier et al 1996). This apoptosis is a mechanism to regulate rapid mesenchyme cell proliferation, and also facilitates outgrowth of newly formed olfactory axon fascicles by creating a space through the mesenchyme. In the figures mentioned, some TUNEL-positive examples of these cells can clearly be seen in both the caspase-3 and caspase-9 null embryo, and therefore the non-neuronal cells of the nasal mesenchyme also do not appear to require caspase-3 or -9 for apoptosis. In addition to the abnormalities of the OE discussed in the preceding paragraphs, caspase-3 null mice have severe anatomical abnormalities of the olfactory bulb. The nature and severity of these abnormalities is dependent upon whether or not individuals are hydrocephalic. As mentioned in section 6.1 hydrocephaly is a source of great variability in phenotype within the colony. However, it can arise as a result of minor variations during development. It has been suggested that slight variation in the temporal or spatial appearance of extra cells at the cerebral aqueduct can determine whether or not the cerebral aqueduct becomes obstructed, and thus lead to profoundly different developmental consequences (Kuida et al 1996). Olfactory bulbs of adult caspase-3 null mice which have been put under pressure by hydrocephalus break through the cribriform plate and form ectopic pieces in the nasal cavity. The periglomerular cells around the perimeter of these pieces of bulb show normal expression of tyrosine hydroxylase (figure 6.6.2c), often taken as evidence for functionality. Expression of tyrosine hydroxylase in 80 periglomerular cells in neonatal rodents requires innervation of the bulb by functional ORNs, and sensory input in the form of odorant-stimulated neurotransmitter release by ORNs (Baker 1990, Puche and Shipley 1999). It is remarkable that ectopic pieces of bulb might possibly be functional, as these are CNS neurons directly exposed to the air. ORN axons are sometimes seen to leave the OE through the cribriform plate as normal, before exiting back into the nasal cavity to contact the ectopic bulb. From this I surmise that the bulb probably does not develop in the nasal cavity but breaks though the cribriform plate after ORN axons have made contact with the bulb. Given the abnormal shapes and sizes of caspase-3 null olfactory bulbs (section 6.6), and the decrease in apoptosis reported in the caspase-3 null brain (Kuida et al 1996), a reasonable question would be: are there more neurons in the caspase-3 null bulb? I do not have evidence to support this, even though I have seen a few individual caspase-3 null animals with enormous bulbs. It may or may not be the case that there is less apoptosis in caspase-3 null bulbs. However, whether or not this can translate into larger bulbs seems to be more dependent upon the physical constraints under which the individual bulb finds itself due to the anatomy of the rest of the brain, which seems to vary with severity of phenotype (see figure 6.6.1). Some caspase-3 null bulbs find themselves squashed by an expanded brain into a small space, and are thus smaller than normal (animal 39A4); some are forced to break through the cribriform plate, in which case they may have room to expand to normal size (39A8, 40C31) or not (39A5), and some bulbs have room to expand dorsally, and are thus much larger than normal (animal 40A7). It would seem logical that since there are more mature ORNs in the caspase-3 null (due to their reduced ability to die) that this could cause the olfactory bulb to be bigger, regardless of any hypothetical apoptosis defect in the bulb itself. Neurogenesis in the olfactory bulb in development is dependent on incoming ORNs, and therefore if more axons are coming into the bulb, more bulb cells will be born. Several of the adult caspase-3 null bulbs examined did seem to have unusually thick nerve fibre layers, and 2 animals had noticeably enlarged glomeruli (figure 6.6.4). This is probably a direct result of the expanded ORN population found in the caspase-3 null OE. In investigating the ORN population in the caspase-3 null, as an alternative to counting cells in serial sections I was able to take advantage of the unique anatomy of the OE and demonstrate an increase in O R N axon bundle area (figure 6.6.5) as a somewhat indirect measure of an expanded O R N population. OMP western blotting and immunohistochemistry is also consistent with an expanded O R N population, although it does not prove it. OMP levels appear the same in OE from adult caspase-3 nulls compared to wildtype controls on western blots (data not shown), even though OMP immunoreactivity is increased above wildtype in caspase-3 nulls on confocal immunochemistry (figure 6.8.1). The western blot has equal protein loading in each lane, and ORNs comprise the vast majority of cells in the OE. Therefore the western blot data probably represents amount of OMP per neuron, which does not change. This statement depends upon the assumption that additional ORNs in the caspase-3 null are of a normal size, which has not been proven. If one accepts the western data as evidence that the amount of OMP per neuron does not change, this provides an explanation to reconcile the western and immunohistochemistry data, since the increased OMP intensity observed in the caspase-3 null with confocal microscopy must then represent more ORNs. (The assumption that the amount of OMP per neuron does not change was used in the selection of OMP as a control for caspase-9 in the confocal immunochemistry experiment. Since caspase-9 immunoreactivity is increasing over and above 81 the OMP increase, I concluded that the amount of caspase-9 is increased per neuron.) The postnatal onset of olfactory bulb abnormalities in the caspase-3 null is an interesting phenomenon. Even though adult caspase-3 null olfactory bulbs have abnormal shapes and sizes (section 6.6), there is no difference in shape and size between bulbs of caspase-3 - / - neonates and their littermates (figure 6.7.1). This might mean that up until birth, when the bulb has little functional input from the OE (except at the modified glomerular complex), the expanded ORN population has little effect on the bulb. Rather, it is in the first three weeks after birth, when ORN axons are sending a host of new input signals to the bulb, forming glomeruli, and refining connections, that the enormous input from extra ORNs wreaks havoc with olfactory bulb dimensions. Alternatively, the reason for postnatal onset of bulb phenotype might be that this is when the bulb itself would normally turn on strong caspase-3 expression in many of its cells (figure 6.2.2g). Developmental death in the wildtype olfactory bulb peaks at P5 (Fiske and Brunjes 2001). It is easy to speculate how, without this caspase-3, bulb cells that might be supposed to undergo apoptosis might not do so, leading to an enlarged bulb. However, it is harder to imagine how this reason alone could account for the smaller bulbs that are seen in a significant proportion of caspase-3 - / - adults (figure 6.6.1). Complicating the issue somewhat are the small but important periglomerular neurons, which may be the key here. They show high caspase-3 expression in the wildtype, yet they exert both a positive and negative influence on the connections between other bulb neurons. Their birth apparently coincides with the onset of caspase-3 - / - bulb abnormalities: there is massive periglomerular neurogenesis at P1-P21. Interestingly, the periglomerular neurons are abnormal in neonatal caspase-3 nulls, in that increased tyrosine hydroxylase immunoreactivity is seen in the periglomerular region, although this has not been quantified (figure 6.7.2). There are several possible speculative hypotheses that might explain this. 1) The increased tyrosine hydroxylase intensity in the caspase-3 null neonate indicates that there are more than normal periglomerular cells. There are more cells because a) they would have needed caspase-3 to undergo programmed cell death, and did not express it, or b) the extra input from incoming ORNs and enlarged glomeruli signals a need for more periglomerular cells and so more are generated or migrate. The extra cells misregulate glomerular function either positively or negatively, thereby altering bulb size. The postnatal onset of bulb abnormalities occurs because that is when periglomerular cells are born and begin to express caspase-3. 2) There are a normal number of periglomerular cells. The above normal levels of tyrosine hydroxylase are because the extra incoming ORNs create more activity in the mitral cells and signal for the slightly earlier functional maturation of periglomerular cells. Earlier maturation similarly misregulates glomerular function and bulb size. Complicating the issue of the potential role of caspase-3 in bulb development even further, is the fact that during the first two postnatal weeks, caspase-3 activity is high in the neuronal precursors entering the bulb in the rostral migratory stream, and in the proliferating cells in the subependymal zone in the centre of the bulb. This activity does not seem to be associated with death of these neuronal precursors, but rather it has been suggested that the caspase-3 might be involved in the processes of migration, neuronal differentiation, and plasticity (Yan et al 2001). It is not known what happens to this population of neuronal precursors, and what the ramifications are for the rest of the bulb, in the caspase-3 null. Mice lacking a different apoptosis-related protein, p73 (an anti-apoptotic p53 family member), also exhibit abnormalities of the olfactory bulb which are not evident until after birth (Pozniak 82 et al 2002). In this case since p73 is anti-apoptotic the nulls show enhanced apoptosis, the opposite of caspase-3. Bulbs from p73 nulls display a phenotype close to the opposite of caspase-3 nulls: the bulb is normal at birth but by P14 is smaller than wildtype littermates, with a thinner glomerular layer, and appears to be rounder (i.e. in my analysis it would be assigned a larger shape factor). Pozniak et al 2002 suggest that this phenotype could be the result of postnatal input from fewer ORNs, the converse of what I suggest here. Another intriguing difference that we saw between caspase-3 null and wildtype is that caspase-3 nulls have higher levels of caspase-9 protein in their ORNs. I speculated briefly in Chapter 3 about the possibility of an attempt at compensation for the loss of caspase-3 in these mice. There is a precedent for this type of mechanism in the literature. When mice are challenged with a fas ligand, this causes apoptosis of hepatocytes which (in the case of so-called "type II" cells like hepatocytes) requires involvement of the mitochondrial pathway and caspase-9 and caspase-3. However, when this experiment is performed on caspase-3 null or caspase-9 null mice, the hepatocytes still die, this time using caspases-2, -6 and -7 (Zheng et al 2000). This is not merely a case of redundancy of caspase function, because wildtype hepatocytes do not express caspase-2, -6 or -7. In the context of ORN apoptosis, it is not yet clear what the role of compensatory mechanisms may be, or whether they have a function in the wildtype setting. I have not even been able to assess whether the activation patterns of caspase-9 are the same as normal in the caspase-3 null after bulbectomy. However, considerations of possible compensation might be important in interpreting in vivo and in vitro data. Unfortunately, parallel analyses to those discussed in the preceding paragraphs could not be carried out on caspase-9 null adults and neonates because of their reduced lifespan. However, as mentioned previously, in the E13 analyses there was no difference between caspase-3 null and caspase-9 null. Bulbs from the two caspase-9 - / - animals which died at PI were analysed for size and shape, and did fit into the normal range of sizes seen in caspase-3 null and wildtype (data not shown). However, no conclusion can be drawn from two animals, and the use of post-mortem tissue is a confounding factor. The fact that caspase-3 is highly expressed in many cells of the olfactory bulb postnatally, whereas caspase-9 is not, is a potentially important difference which might affect later olfactory system development in mice null for each of these proteins. These data are the first to address the role of caspases in the developing olfactory system. I conclude that there is an early requirement for both caspase-3 and caspase-9 to properly achieve a specific early wave of programmed cell death at E l 3 . This wave of death is probably a part of turbinate and palate formation, and creates a path for olfactory axons to grow out along with their accompanying migrating cells (Pellier and Astic 1994a). If this death does not occur, turbinates do form normally, but there is a lifelong expansion in the ORN population, which could conceivably be the cause of the gross abnormality of the olfactory bulb. This expanded ORN population persists despite the fact that the requirement for caspases seems fairly specific for the E l 3 peak of death, and ORNs are able to achieve normal basal levels of apoptosis postnatally. 83 TABLE 6.1 GENOTYPE FREQUENCIES AND LIFESPAN OF CASPASE NULL MICE A: Genotype frequencies of live births to heterozygous parents caspase-3 caspase -9 genotype total +/+ +/- -/- total +/+ +/- -/-number 268 71 154 43 213 83 128 2 percentage 100 26 57 16 100 39 60 0.9 B: Lifespan of caspase-3 null mice born alive age birth PI 6 wks 3 mo 4mo 5mo 6mo 7mo 8mo 9mo lOmo number 74 66 48 46 41 28 13 10 4 3 1 percentage 100 89 65 62 55 38 18 14 5 4 1.4 84 Tfmi;imfrmjTmrnwjmi|iiuji^^ & 6 7 B <i l O 1 1 \2 V3 \ B FIGURE 6.1.1 Caspase-3 null mice are often small with hydrocephaly Young adult mice anaesthetized prior to sacrifice were photographed next to a ruler: a shows a wildtype mouse measuring approximately 15.5 cm from nose to end of tail, b shows its caspase-3 null littermate with a short domed head, measuring approximately 12 cm. c: the same two mice next to each other while awake, d: Photograph of the head of a young adult caspase-3 - / -mouse during dissection, to illustrate the domed skull and prominent blood vessels of hydrocephaly, e: sagittal section of a caspase-3 - / - E l 3 embryo to illustrate a large ventricle. 85 FIGURE 6.2.1 Caspase-9 is strongly expressed in ORN axons throughout development Sagittal sections of wildtype mice embryos and neonates from (a) E l 1 to (e) P5, immunolabelled with caspase-9. pOE, presumptive olfactory epithelium; OB, olfactory bulb; N F L , nerve fibre layer; CP, cribriform plate; OE, olfactory epithelium; G, glomerulus. Arrows indicate caspase-9 positive ORN axon buldles. FIGURE 6.2.2 Caspase-3 is strongly expressed in ORN axons and cell bodies throughout development from E13 And also in many cell types of the developing olfactory bulb postnatally. a-g: sagittal sections of wildtype embryos and neonates are immunolabelled for caspase-3. The olfactory epithelium is shown from E10-P7 (a-f), and the olfactory bulb at P5 (g). Negative controls without primary antibody are shown for P5 (h) and E l 3 (i). pOE, presumptive olfactory epithelium; ORN, olfactory receptor neuron cell body; Ax, ORN axon; OB, olfactory bulb; G, glomerulus; GC, granule cell; M C , mitral cell; EPL, external plexiform layer. 86 87 FIGURE 6.3.1 Apoptotic neurons in the olfactory epithelium of wildtype developing mice Sagittal sections of OE at E13 (a), E15 (b), E17 (c) and P7 (d), labelled with T U N E L to detect apoptotic nuclei. 88 Timecourse of apoptosis in the developing mouse OE 25 FIGURE 6.3.2 E13 is the peak of apoptosis in the olfactory epithelium Graph of the number of TUNEL-positive cells per linear mm of OE, at different stages of development of the wildtype embryo and neonate. Sections were taken from a minimum of three different animals per timepoint, and the total number of mm measured was E l 3 = 20 mm, E15 = 56 mm, E17 = 133 mm, P4 = 96 mm. 89 C3 +/- C3 -/- C9-/-FIGURE 6.3.3 Apoptotic neurons in the OE of caspase-3 and caspase-9 null E13 embryos Sagittal sections of E l 3 OE of caspase-3 heterozygote (a), caspase-3 null (b), and caspase-9 null(c), labelled with T U N E L to detect apoptotic nuclei (arrows point out examples). Dotted line indicates basement membrane. 90 Caspase-3 and caspase-9 null E13 embryos have fewer apoptotic cells in the OE than littermates FIGURE 6.3.4 Caspase-3 and caspase-9 null E13 embryos have fewer apoptotic cells in the OE than littermates Graph showing number of TUNEL-positive cells per linear mm of E l 3 OE, in caspase-3 null, caspase-9 null, and control littermates. C 9 - / - , caspase-9 null; C 3 - / - , caspase-3 null; C3het, caspase-3 heterozygote; w/t, wildtype. Sections from a minimum of three animals per genotype were used. Since there is no statistical difference between wildtype and C3het, I pooled these values to compare them in t-tests against C 3 - / - (which gave p=0.009) and against C 9 - / - (which gave p=0.001). Probability of <0.01 is indicated by **. Number of mm measured was between 20 and 26 mm for each genotype. 91 -ve controls FIGURE 6.3.5 GAP-43 expression in the same population as TUNEL provides proof of OE identity Sagittal sections of E l 3 wildtype OE labelled for growth-associated protein-43 (a, and negative control b), for T U N E L (c, and negative control d), and overlap (e, negative control f). a,b were labelled with fluorescence; c,d were labelled with solid colour TACS Blue, then in Adobe Photoshop the image was inverted and placed in the red channel. 92 F I G U R E 6.3.6 Apoptosis in caspase-3 and caspase-9 null E13 embryos TUNEL-labelled sagittal sections of caspase-3 - / - (a,b) and caspase-9 - / - (c,d) E l 3 embryos, in low power plan (a,c) and higher power shots of the OE (b,d), show that while apoptosis in the OE is low in both null mice, apoptosis in the rest of the body seems less in C 9 - / - than in C3-I-. 93 FIGURE 6.3.7 Other notable features of apoptosis in E13 embryos TUNEL-labelled sections of E l 3 embryo in (a) a C3het, showing that apoptosis in the OE is sometimes zonal; (b) a wildtype, showing strong labelling in the optic recess of the diencephalon, not seen in C 3 - / - or C 9 - / - embryos; (c,d) a C9- / - , at lOx and 20x respectively, showing apoptotic nuclei in the wall of the oesophagus, one of the few structures where there appeared to be levels of apoptosis indistinguishable from wildtype. 94 Caspase-3 and caspase-9 null El3 embryos have a thicker OE neuronal layer than littermates FIGURE 6.4.1 Caspase-3 and caspase-9 null E13 mice have a thicker OE than littermates Graph showing the thickness of the neuronal layer of the OE for each genotype. 11-78 measurements were made per genotype, comprising at least three animals. Error bars are standard error of the mean, t-tests: C 9 - / - vs w/t, p=0.01; C 3 - / - vs w/t p<0.001. 95 450 400 350 300 250 200 150 100 50 0 Caspase-3 and caspase-9 null El3 embryos have a thicker OE neuronal layer than littermates • • t * < • • • • < * • • • • • • T • t r C9-/- C3-/- C9het C3het w/t FIGURE 6.4.2 Caspase-3 and caspase-9 null E13 mice have an OE of more varied thickness Graph showing the thickness of the neuronal layer of the OE for each genotype, with every data point included. Bars indicate the mean for each genotype. 96 wildtype C9-/- wt -ve control FIGURE 6.5.1 NCAM expression provides proof of ORN identity of caspase-dependent cells at E13 Sections of wildtype (a,c) and C 9 - / - (b) E l 3 OE labelled for N C A M (a,b) and negative control (c). Dotted line indicates basement membrane, solid line indicates cribriform plate, d and (higher power) e show double fluorescence labelling for N C A M (green) and T U N E L (red) in sections of OE from a C 3 - / - E13 embryo. CP, cribriform plate; NC, nasal cavity; Ax, O R N axon; ORN, olfactory receptor neuron layer of OE. 97 98 FIGURE 6.6.1 Olfactory bulbs of caspase-3 null adult mice are of more varied sizes Coronal sections of olfactory bulbs were counterstained and the area of the bulb measured every 160 u,m. These areas were then plotted on a graph against the depth of each bulb, to get a profile of size and shape, a shows individual wildtype bulbs, b shows one line of the average of the wildtype bulbs presented in a (with standard error bar), in comparison with individual bulbs from caspase-3 nulls. The area under these curves gives the total volume of each olfactory bulb, and these are presented in a bar graph, c. Note that the caspase-3 null mice have been categorized as either hydrocephalic (h-/-) or not (-/-). 99 1 2 3 rcistro-catdal depth of bulb (mm) v l av -/- 39A4 •!- 40A7 L -/- 40A7 K h- • 39ASL a-/- 39AS R a-/- 39AIL hrf- 39 AIR h-/> 40C31 L li-f 40C31 R 1 2 3 imtnvcaudaL depth of bulh (mm) F I G U R E 6.6.2 Abnormally shaped olfactory bulbs with ectopic bulb tissue in caspase-3 null adults Examples of coronal sections of olfactory bulb from a, a wildtype and b, a hydrocephalic caspase-3 null. Axis indicates dorsal (D), ventral (V), medial (M) and lateral (L) orientation. Line in b indicates the position of the cribriform plate, such that tissue ventral to this is ectopic bulb tissue in the nasal cavity. A shape factor was determined for each. Shape factor is defined as 4:ta/c2, where a=area of the shape and c=circumference, such that a perfect circle would have a shape factor of 1 and a line would have a shape factor of 0. The shape factor of the wildtype (a) is 0.672, and the C3 null (b) is 0.365. c: tyrosine hydroxylase immunoreactivity in a coronal section of the ectopic bulb shown in b. 101 Olfactory bulbs of caspase-3 null mice with hydrocephaly have an irregular shape a w/t39A2 w/t39A3 w/t40A4 -/-39A4 -/-40A7 h-/-39A5 h-/-39A8 h-/-40C31 animal identity b 0.8 r -w/t C3-/- hydro C3-/-F I G U R E 6.6.3 Caspase-3 null mice with hydrocephaly have abnormally shaped olfactory bulbs Graphs showing the average (solid bar) and minimum (dashed bar) shape factor of coronal sections of olfactory bulb, a shows the data for individual mice, while b is grouped into wildtype mice, and caspase-3 null mice with and without the hydrocephalic phenotype. As in figure 6.5.2, shape factor is defined as 4jta/c2, where a=area of the shape and c=circumference, such that a perfect circle would have a shape factor of 1 and a line would have a shape factor of 0. The usual ellipse of a wildtype bulb in cross section has a shape factor of around 0.7. 102 FIGURE 6.6.4 Enlarged glomeruli and ORN axon bundles in caspase-3 null adults Coronal sections from (a,b) wildtype; (c,d) caspase-3 null with hydrocephaly; (e,f) caspase-3 null without hydrocephaly, immunolabelled for either N C A M (a,d,e) or O M P (b,c,f) to highlight ORNs. Left-hand panels (a,c,e) show epithelium only. Right-hand panels show bulb (b) or bulb with epithelium (d,f). Wildtype sections (a,b) are included to illustrate the typical size of axon bundles (Ax) and glomeruli (G). In the four panels of caspase-3 null (c-f), examples of axon bundles above normal size are indicated *, and glomeruli above normal size are indicated G*. 103 a. Greater total ORN axon area in C3-/-• C3 -/-• C3 net mediolateral inferior b. Bigger ORN axon bundles in C3-/- c. Fewer O R N axon bundles in C3-/-superior mediolat. inferior superior mediolat. inferior F I G U R E 6.6.5 Caspase-3 null adults have increased O R N axon bundle area Graphs showing (a) greater total axon area in the OE of caspase-3 - / - mice as a result of (b) much bigger but (c) fewer O R N axon bundles. I divided the analysis into superior, middle and inferior turbinates of the OE, since the difference in axon bundle size is most dramatic in the middle turbinate. Sections were from four C 3 - / - mice and three C3+/- mice, and within each section I made measurements in nine specific fields of view. 104 Olfactory bulbs of caspase-3 null mice are the same size and shape as littermates at postnatal day 4 - • C3het -• C3het - • C3-/-C3-/-C3-/-1 2 3 lateral distance through bulb (mm) F I G U R E 6.7.1 Olfactory bulbs of P4 caspase-3 null mice are the same size and shape as littermates Sagittal sections of P4 heads were counterstained and the area of the olfactory bulb measured every 80 pm. These areas were then plotted against the lateral distance through the bulbs. Note that this represents a different orientation than that shown for the adult mice in figure 6.5.1. As in figure 6.5.1, however, the line gives an indication of bulb shape, while the area under the curve represents total bulb volume. 105 F I G U R E 6.7.2 Tyrosine hydroxylase immunolabelling of periglomerular cells at P4 in caspase-3 nulls Sagittal sections of olfactory bulb from caspase-3 null (a) and heterozygote (b) immunolabelled for tyrosine hydroxylase, a marker for a subpopulation of periglomerular neurons. 106 FIGURE 6.7.3 No difference in number of apoptotic ORNs at P4 between caspase-3 nulls and littermates Graph showing number of TUNEL-positive cells per linear mm of P4 OE, in caspase-3 null (C3-/-) and caspase-3 heterozygote (C3het). Nine sections from three animals per genotype were used. Total number of mm of OE measured: C 3 - / - = 73 mm, control = 96 mm. Two tailed t-test: p=0.84. FIGURE 6.8.1 Caspase-9 expression is increased in caspase-3 null mice a,b: sagittal sections of OB and OE from P7 caspase-3 null mouse (b) and wildtype littermate (a), showing more intense caspase-9 immunoreactivity in the O R N axons of the caspase-3 null (arrows), c-f: double immunofluorescence for OMP (c,d) and caspase-9 (e,f) on coronal sections of OE from wildtype (c,e) and null (d,f) caspase-3 mice were visualized by a confocal Z-series and pseudocoloured for relative intensity using NIH Image (see bar for scale). Using this scale, OMP expression is greater in the - / - (d), than in the +/+ (c), but the increase in intensity of C9/OMP in the enlarged population of axon bundles is 5-fold higher in the - / - (f) than in the +/+ (e) (Cowan et al 2001). 107 C3+/+ C3-/-CHAPTER 7: CONCLUSIONS 7.1 SUMMARY OF RESULTS AND CONCLUSIONS CHAPTER 5: Apoptosis and caspases in mature olfactory neurons in the adult mouse SUMMARY OF RESULTS • After unilateral bulbectomy, ORNs on the lesioned side of the OE die by apoptosis, as revealed by D N A laddering, T U N E L labelling, and single-stranded D N A immunohistochemistry. • This acute wave of post-bulbectomy apoptosis is complete within 72 hours, the peak of both D N A laddering and T U N E L labelling falling between 24 and 36 hours. • There is a gradient of apoptosis in both space and time, whereby ORNs closest to the lesion site die first. • Caspase-3 and caspase-9 are present in ORNs in vivo. • Levels of caspase-3 and caspase-9 proenzymes in lesioned OE increase between 4 and 24 hours post-bulbectomy. • The active form of caspase-3 is seen in lesioned ORN cytoplasm at 24 hours, and is maximal by 48 hours. • Maximum levels of caspase-9 are detected earlier at 24 hours. • Caspase-3 and caspase-9 are found throughout axons, and in the glomerular synaptic complexes in the olfactory bulb. • The temporal and spatial distribution of active caspase-3 is mirrored by the distribution of a caspase-3 cleaved form of APLP2. • Caspase-dependent cleavage of APLP2 occurs first in the axon and then at the cell body. • In caspase-3 null mice, mature ORNs do not become TUNEL-positive during the first 72 hours after bulbectomy. • In caspase-3 null mice, mature ORNs are not lost from the OE during the first 72 hours after bulbectomy. CONCLUSIONS • The spatial and temporal parameters of bulbectomy-induced O R N apoptosis were determined, as a framework in which to examine the role of caspases. • Caspase-3 and caspase-9 proenzyme expression in ORNs changes in an apoptosis stage-specific manner. • Cleavage of caspase-3 and caspase-9 changes in an apoptosis stage-specific manner. • Caspase-3 neuronal targets are cleaved in an apoptosis stage-specific manner. • Caspase-3 and caspase-9 are situated to initiate the propagation of the apoptotic signal from the synapse back to the cell body following target deprivation. • Caspase-3 is necessary for ORNs to undergo acute bulbectomy-induced apoptosis. 109 CHAPTER 6: Caspase-3 and caspase-9 in development of the mouse olfactory system SUMMARY OF RESULTS • Caspase-3 and caspase-9 null mice show onset of abnormalities consistent with the literature. • Caspase-3 null mice are longer lived than reported in the literature. • Caspase-9 is strongly expressed in ORN axons throughout development. • Caspase-3 is strongly expressed in ORN axons and cell bodies throughout development. • Caspase-3 is strongly expressed in the olfactory bulb during postnatal development. • The peak of apoptosis in the OE is at E l 3 . • There are approximately 4-fold fewer apoptotic neurons in the OE at E l 3 in caspase-3 and caspase-9 null embryos. • Apoptosis in the E l 3 OE is zonal. • The neuronal layer of the OE of caspase-3 and caspase-9 null E l 3 embryos has zones of dramatically increased thickness. • The expanded population of cells in the E l 3 OE of caspase-3 and caspase-9 null embryos express N C A M . • Those cells which are able to undergo apoptosis in the neuronal layer of the E l 3 OE of caspase-3 and caspase-9 null embryos, do not express N C A M . • GAP-43 expressing immature ORNs in the adult unlesioned caspase-3 null are occasionally TUNEL-positive, and can therefore undergo apoptosis. • The olfactory bulbs of caspase-3 null adult mice have an abnormal shape and size. The nature of this abnormality is affected by whether the individual has hydrocephaly as part of its phenotype. • Caspase-3 null adults have a greater ORN axon area than wildtype, comprised of fewer but much larger axon bundles. • Caspase-3 null neonates have olfactory bulbs of normal size and shape. • Caspase-3 null neonates have normal numbers of apoptotic ORNs. • Caspase-9 expression is upregulated in the ORNs of neonatal caspase-3 null mice. CONCLUSIONS • Caspases-3 and -9 are in an appropriate time and place to play a role in olfactory system development. • Caspases-3 and -9 are required for embryonic programmed cell death of developing ORNs. • Caspase-3 is required for the eventual production of the normal size and organization of O R N axon bundles in adult mice. • The requirement for caspase-3 in ORN apoptosis varies according to maturity of the olfactory neuron: in the adult, mature ORNs require caspase-3 but immature ORNs do not. • The requirement for caspase-3 in ORN apoptosis varies according to maturity of the olfactory system: immature ORNs in the E l 3 embryo require caspase-3 (and caspase-9) for apoptosis, but immature ORNs in the adult do not require caspase-3 for apoptosis. • The onset of abnormalities in the olfactory bulb of caspase-3 null mice is postnatal. • Caspase-3 - / - mice may try to compensate for their deficiency in development by increasing expression of caspase-9, which is in the same pathway. 110 7.2 GENERAL DISCUSSION The work presented in Chapter 5, collectively demonstrating synaptic caspase-3 activation, lends strong in vivo support to earlier work targeting the synapse as a focal point for initiation of pro-apoptotic signalling, either in a developing or in a lesion state (Mattson and Duan 1999, Mattson et al 1998). The synaptic junction has historically been argued as a site where positive trophic signals are utilized to reinforce and stimulate survival of the presynaptic neuron during development and maintenance of the CNS. My data suggest that, alternatively, the postsynaptic neuron (in this case, the mitral/periglomerular neuron) could control turnover of the presynaptic olfactory neuron population by manipulating the release of pro-survival stimuli, thus shifting the balance of synaptic signalling pathways from pro-survival to pro-apoptotic. The continued expression of caspases-3 and -9 at the highly plastic (NMDA receptor-mediated) olfactory synapse also puts them in an ideal position to participate in the remodelling or local dismantling of inactive synapses via the cleavage of target proteins crucial for synaptic integrity, such as alpha- and.beta-spectrin and beta-actin (Chan and Mattson 1999, Wang et al 1998) In Chapter 5 I also provide evidence for the co-localization of axonal caspase-9 with active caspase-3. Given that ORN mitochondria are also located in axons, and we know that O R N apoptosis depends on Bcl-2, this suggests a significant role for the mitochondrial pathway in mature O R N apoptosis (Krajewski et al 1999). Procaspases-3 and -9 are also ideally situated to balance pro-apoptotic pathways with those that drive neuronal plasticity, as axonal caspase-9 is detected in many other neuronal axons embryonically, but is only retained at readily detectable levels postnatally in the olfactory neuraxis (section 6.2). Now that caspases-3 and -9 have been placed at the synapse, it will be important to determine which upstream signalling mechanisms converge to balance synaptic survival/apoptotic pathways in an intact olfactory neuraxis. A number of known pro and antiapoptotic signalling proteins (e.g. the tyrosine kinase Brain Derived Neurotrophic Factor receptor, Trk B, and the low affinity nerve growth factor receptor, P75) are present at the olfactory synapse (Roskams et al 1996). It is conceivable that differential stimulation of p75 (feeding into mitochondrial apoptotic pathways) and kinase active/inactive Trk B , by bulb-derived neurotrophins, could synergistically serve to control caspase-9 activation at the synapse. In addition, nitric oxide, released from periglomerular neurons into a glomerular synapse, could directly regulate the activation of presynaptic procaspase-3 or -9 (Li et al 1999, Tamatani et al 1998, Tenneti et al 1997). The upstream signals that may stimulate axonal caspase pathways following physical deafferentation are less obvious and require further investigation. The compartmentalization of caspase-9 in the axon, and caspase-3 in axon and cell body, introduces the possibility that different pathways could be utilized for local and distant control of ORN apoptosis, where caspase-3 not only drives apoptosis of the soma but also retrogradely dismantles proteins that maintain axonal integrity. This is corroborated by some preliminary evidence supporting a possible role of caspases in axon dismantling in vitro (see Future Directions). Several structural proteins such as actin do have sites which indicate they are caspase substrates (Kayalar et al 1996, Brown et al 1997, reviewed in Chan and Mattson 1999). In contrast, in the mature optic nerve, the process of axonal dismantling and Wallerian degeneration is caspase-3-independent, but neuronal survival is caspase-3-dependent (Finn et al 2000). This work highlights the importance of how the location of a caspase within a neuron dictates not only which local pathways may be capable of activating it, but also which substrates are 111 accessible - a key issue when examining candidate neuronal caspase target proteins that could be highly compartmentalized. These data also allow us to place caspases-9 and -3 in a more dynamic role in neuronal apoptosis than previously demonstrated - simultaneously carrying pro-apoptotic signals from a synapse to their final site of action at the neuronal nucleus and actively dismantling axonal proteins. The identification of neuronal caspase activation at the living synapse, coupled with the identification of other apoptotic mediators in the synaptosome (Mattson et al 1998), now allows us to place caspase-mediated signalling at the epicentre for target-derived trophic support, plasticity and survival of the ORN. This thesis addresses the relative importance of caspases in maturity versus development, both at the level of the neuron, and the olfactory system as a whole. The expression pattern of caspases-3 and -9 provides an opportunity to speculate further on this topic. Caspase-9 expression is detected in axonal projections throughout the neocortex from E13-15, and in the axons of many different sensory neurons including dorsal root ganglia, trigeminal nerve and optic nerve throughout embryonic development. However, this axonal expression is lost postnatally in all sensory nerves apart from the olfactory nerve, where it remains strongly expressed. Therefore in some brain regions, caspase-9 may be more important in development than it is later in life. The olfactory system, in contrast, is an exceptional system that remains in a "developing" state postnatally, undergoing regeneration as it does throughout life. This is consistent with the idea that caspase-9 is more important in developmental neuronal apoptosis, and mature systems may use a different mechanism. However, caspase-3 null bulbectomy data provide evidence that caspase-9 is not sufficient to kill mature ORNs in the adult, even when it is more highly expressed. In the olfactory system I describe an identical defect in the two null lines at E l 3 . However, given that the caspase-9 null mice have a more severe phenotype then the caspase-3 knockout mice, as evidenced by earlier lethality, it is possible that caspase-9 may be more important than caspase-3 in neuronal apoptosis in general during embryonic development. With regard to the situation in maturity, data from several inhibitor studies discussed in the Literature Review suggest that caspase-3 is the main caspase necessary in a number of paradigms of neuronal apoptosis in adult injury or disease. It is possible for the reasons just discussed that caspase-9, in contrast, is only important in maturity in certain brain areas, for example the olfactory system. A n interesting set of experiments has shown that in different cell types in different apoptotic circumstances, the following four pathways can each occur: 1) caspase-3 and caspase-9-dependent apoptotic pathway; 2) caspase-3 and caspase-9-independent apoptotic pathway; 3) caspase-3-dependent and caspase-9-independent apoptotic pathway; and 4) caspase-3-independent and caspase-9-dependent apoptotic pathway (Hakem et al 1998). I focus on caspases-3 and -9 because of their severe neuronal knockout phenotype, and their apparent importance in the olfactory system. However, it should be remembered that while caspase-9 can and does directly activate caspase-3, caspase-9 can also activate other caspases and serve some effector functions itself, while caspase-3 can of course be activated by caspases other than caspase-9 (reviewed in Earnshaw et al 1999). A differential use of caspase-3 in an immature versus a mature system in vivo has previously been shown in motoneurons (Vanderluit et al 2000). In this model, motoneurons in both neonatal and adult mice express caspase-3, and upregulate caspase-3 mRNA after axotomy. However, only in the neonate do the axotomized motoneurons actually upregulate 112 protein levels acutely, activate the caspase-3, and undergo apoptosis. I previously discussed loss of caspase-9 expression entirely during maturation, as a mechanism whereby mature and immature neurons could use different death pathways. In contrast, this study suggests a potential mechanism whereby the mature and immature neurons express the same caspase, but are able to regulate its use differently. It may be that developing neurons need to have their caspases more readily activated (e.g. perhaps less tightly controlled by endogenous inhibitors), not only because they need to undergo programmed cell death, but perhaps also because they need to retain more synaptic plasticity (Mattson and Duan 1999). Mattson's studies and mine have stressed the importance of caspases in synapses and axons, as a means of propagating an apoptotic signal to the cell body. The above-mentioned study demonstrated an instance in which regulation of caspase-3 activation after axotomy affects death (Vanderluit et al 2000). However, it is clear from Martin Raffs work that it is frequently the case in many neuronal settings that caspases are not activated in degenerating axons (even though caspases may cause eventual apoptosis of the cell soma). Indeed, in many circumstances it is of limited relevance whether caspases are involved in the eventual neuronal apoptosis at all, because the neurons have previously ceased to function, following (caspase-independent) Wallerian degeneration of axons (Finn et al 2000, Raff et al 2003). These discussions have focussed on the role of caspase-3 and caspase-9 in developmental apoptosis, but there is some recent evidence that caspases have developmental functions other than cell death. For example, there is evidence to suggest the involvement of caspases in differentiation (Sordet et al 2002, Fernando et al 2002), and a recent study demonstrated that caspase-3 is involved in axon guidance of retinal growth cones (Campbell and Holt 2003). Given that I describe strong expression of caspase-3 and caspase-9 in olfactory axons throughout development, even at times when programmed cell death is minimal, and given the interesting observation that caspase-9 is highly expressed in many embryonic sensory axons but chiefly in olfactory axons postnatally, the possibility of alternative caspase functions merits further study in the olfactory system. Returning to the topic of maturity versus development, immature ORNs in the E l 3 OE require caspase-3 and caspase-9 for normal apoptotic death, whereas apoptosis at P4 and probably in the unlesioned adult is unaffected by the absence of caspase-3. This latter apoptosis is "normal turnover", and probably involves ORNs in all layers of the OE (Voyron et al 1999). It makes sense that constitutive turnover might not require caspase-3, given that the caspase-3 null adult seems able to regulate the thickness of its OE, even though it still maintains an expanded ORN population. It may be an oversimplification to conclude, as I summarized earlier, that immature systems require caspase-3 and mature systems do not; and that within the mature system, mature neurons require caspase-3 and immature neurons do not. The massive wave of apoptosis at E l 3 appears to be an exceptional case, in which caspases are used to cause a large amount of death for a specific purpose at a specific time. Later in development of the system, those same immature ORNs do not seem to need caspase-3 for constitutive death. One explanation might be that an alternative, non-caspase-3-dependent, apoptotic mechanism is available to immature ORNs, which at E l 3 simply cannot keep up with the volume of apoptosis required. This raises the question of whether immature ORNs at E l 3 are the same as immature ORNs later in life. There is reason to believe that they are not. Immature ORNs at E l 3 are undergoing 113 significant proliferation of the population; there are few other cell types in the epithelium; and the epithelium is not well stratified. From around the time when synapses are made with the bulb (approximately E15 onwards), the immature ORNs are probably a more tightly regulated population, with specific population numbers required; confined to a specific layer of the OE; sandwiched between their immediate precursors basally and mature ORNs apically; from whom they may be receiving signals telling them to differentiate or to die. Therefore there will be different local control regulating the death of an immature O R N depending on the stage of development of the system. The fact that embryonic immature ORNs at a critical stage of development need the proapoptotic proteins caspase-3 and caspase-9 to undergo proper death, might be complementary to the fact that some embryonic sensory neurons need the anti-apoptotic protein Bcl-2 for proper maturation (Middleton et al 1998), and at critical developmental stages require Bcl-2 for survival (Pinon et al 1997). Immature ORNs, by definition, are ones that have not yet made synapses with the bulb. In the developing system I have focussed on the wave of immature ORN death at E l 3 , when no ORNs yet contact the bulb. However, as mentioned earlier, there has also been a report of a local wave of ORN apoptosis later in embryonic development, purportedly related to competition for bulb connections (Voyfon et al 1999). A recent review (Roth and D'Sa 2001) contrasts the better understood target-dependent neuronal death, with presynaptic neuronal death (such as occurs in the E l 3 OE), the importance and mechanisms of which were poorly understood before the study of knockout mice for apoptotic proteins. Roth and D'Sa note that "presynaptic" developmental death of neural precursor cells and immature neurons may serve a variety of biological functions including matching of the neuronal precursor cell population to that necessary for proper brain morphogenesis, selection of regionally appropriate neuronal phenotypes, and elimination of cells with genetic abnormalities. Given that caspases are needed for a specific wave of programmed cell death of immature ORNs, but are not needed later for constitutive turnover (which probably involves ORNs at all stages of the lineage), is seems that the apoptosis occurring after bulbectomy is mechanistically distinct from turnover. The bulbectomy lesion provides a specific instance of mature O R N apoptosis which does not occur in the caspase-3 null. I suggest that the death signal might be different in the bulbectomy model and in normal turnover. Even though parallels have frequently been drawn between the two in terms of trophic factor withdrawal (as the bulb is hypothesized to provide the ORNs with survival signals), the bulbectomy is also a partial deafferentation, as the nerve fibre layer is damaged. One could suggest that turnover (of mature ORNs at least) could be receptor-mediated (via a molecular signal from the bulb), and therefore might use a different caspase cascade that might bypass caspase-3. Bulbectomy, on the other hand, by directly damaging the ends of the axons, might directly cause mitochondrial activation of caspase-9, as well as receptor-mediated activation of caspase-8, followed by activation of caspase-3. This would be consistent with the molecular sequence of events seen in a well established axotomy model, transection of the optic nerve causing apoptosis of retinal ganglion cells. In this model, changes in Bcl2 and Bax cause mitochondrial dysfunction leading to release of cytochrome c; leading to activation of caspase-9 followed by caspase-3; with a contribution from caspase-8, which is known in this case to be downstream of mitochondrial dysfunction and 114 independent of death receptor activation (Weishaupt and Bahr 2001). Although caspase-8 and caspase-9 are generally thought of as two distinct yet converging and amplifying pathways, there is also recent precedent for caspase-8 activation in some circumstances leading to cytochrome c release and caspase-9 activation (Jimbo et al 2003). Another possibility is that normal turnover might involve a different mechanism entirely from caspases, such as apoptosis inducing factor, or the calcium-dependent cysteine proteases calpains. During the course of these studies we also generated caspase-3 null mice that survived beyond the perinatal period, in contrast to the original report of caspase-3 null mice (Kuida et al 1996). As discussed above, long-lived caspase-3 null mice have a developmentally expanded O R N population and provide evidence that caspase-3 is ordinarily involved in regulating O R N number. Other cortical neuronal populations are also expanded in long-lived caspase-3 null mice generated by crossing into a black 6 background. Two other studies using long-lived caspase-3 nulls report an increased number of cells in the cortex of adult caspase-3 nulls (Le et al 2002), and an expanded population of supporting cells in the cochlea epithelium, leading interestingly to degeneration of the sensory hair cells (Takahashi et al 2001). The longevity of caspase-3 null mice, and the severity of their neurodevelopmental abnormality in general, is dependent upon the strain of mouse studied. A recent study showed that mice with the caspase-3 mutation on a C57BL/6J background reached adulthood and had only mild brain pathology; while mice with the caspase-3 mutation on a 129Xl/SvJ background were very severely affected, dying around the time of birth with a brain phenotype similar to the original report of Kuida et al 1996 (Leonard et al 2002). It was concluded that there is a strain-dependent genetic modifier which can alter the outcome of caspase-3 deficiency. The variation in phenotype seen in caspase-3 null mutants even within one strain, may be partially due to the incidence of hydrocephalus, as discussed earlier. However, variations in protein expression (including the up-regulation of caspase-9 proenzyme levels) are also detected, suggesting that olfactory neurons may utilize pathways that attempt to compensate for the loss of caspase-3. The identification of alternative compensatory mechanisms in the olfactory system and other areas of the CNS requires further study. In conclusion, I present an updated version of figures 2.1.3 and 2.1.4 from the Literature Review, illustrating how caspases-3 and -9 now fit into a central position in the pathways implicated in ORN apoptosis (figures 7.1.1 and 7.1.2). This raises once again the issue of normal turnover in the olfactory neuroepithelium, which has not been directly addressed in this thesis. I have speculated that turnover might be able to occur in the absence of caspase-3 (even though embryonic immature ORN death, and mature ORN death after bulbectomy, do not), based on normal T U N E L levels in the P4 null, and retention of sporadic apoptosis in the adult null. Since wildtype ORNs express caspase-3 and caspase-9 throughout life, one might wonder whether these differences could be attributable to different upstream signals. In fact, in none of these apoptotic situations (embryonic immature ORN death, mature O R N death after bulbectomy, or normal adult ORN turnover) is the death signal upstream of caspases known for certain. 115 F I G U R E 7.2.1 Pathways implicated in O R N apoptosis Of all the dozens of different signals which have been implicated in initiating apoptosis (at the cell membrane or intracellularly), propagating pro-apoptotic signals (through mitochondrial or non-mitochondrial pathways) or acting to terminate apoptosis (activating caspase-3 or its cellular homologues), the above pathways have been implicated in some way in the ORN. Specific apoptotic molecules that there is some evidence for in the ORN are underlined. Caspase-3 and caspase-9 are highlighted as being implicated in ORN apoptosis for the first time in this thesis (from Cowan and Roskams 2002). 116 P75NGFR Caspase-3 Caspase-9 Caspase-3 Caspase-9 C-jun p53 / fas Caspase-3 FIGURE 7.2.2 Apoptotic molecules within Olfactory Receptor Neurons P75 and caspases-3 and -9 are found at the ORN synaptic terminal, consistent with the possibility that they initiate an apoptotic signal from the olfactory bulb. This signal may be propagated down the axon by caspases-3 and -9 to the cell body where p53, c-jun and the executioner caspase-3 are located. The fas ligand is found at the ORN dendrites, where it would be in a position to initiate apoptosis following an environmental insult (from Cowan and Roskams 2002). 117 7.3 FUTURE DIRECTIONS In vivo lesion work Based on my previous work, the Roskams' laboratory is currently working on new models of apoptosis in the olfactory system, in which different specific neuronal populations of the olfactory bulb are targeted by a chemical lesion. These lesion models could be used to address many questions about neuronal development, apoptosis, migration, neurogenesis, and repopulation. Of particular interest to me, and following on directly from my work, is the N M D A lesion of the mitral cells (the projection neurons of the bulb with which ORNs synapse) mentioned in Chapter 5. Since the ORNs are not directly damaged, this lesion could be used to distinguish the contribution made to the mechanisms involved in bulbectomy by true target-deprivation versus physical injury (deafferentation). One of the advantages of the olfactory system as an experimental system is that inhibitors and other substances can be introduced relatively non-invasively through the nose. I was instrumental in a pilot experiment of this technique, and this route of administration might be used in the Roskams' laboratory to assess the ability of specific inhibitors to delay or prevent cell death in vivo in the various lesion paradigms. A variety of caspase inhibitors such as zDEVD-fmk, zVAD-fmk, and zLEHD-fmk are now available in a biotin-conjugated form (Enzyme Systems), and these could be used to confirm by immunohistochemistry that the inhibitor is entering the ORNs. Also following on from my in vivo work, the Roskams' laboratory is now working in collaboration with Maya Saleh (Merck Frosst) to determine the spatial and temporal pattern of caspase-8 activation in the ORN after bulbectomy, and also in the chemical lesion model in which N M D A is used to selectively kill the mitral cells of the bulb. Further study is also required to investigate the long term outcome in mature ORNs in the caspase-3 null which fail to die in the acute 72 hour period after bulbectomy. We intend to investigate the status (in terms of TUNEL-positivity, OMP expression, and morphology) of lesioned ORNs of caspase-3 nulls 6 days and 14 days post-bulbectomy. The 6 day timepoint is normally the peak of neurogenesis after bulbectomy in the wildtype. By 2 weeks after bulbectomy, small patches of apoptotic cells are once again seen in the wildtype, which may possibly represent ORNs newly born in response to the deficit which tried to reach the missing bulb and failed to find a target. The above analysis of caspase-3 nulls would address whether any change has now occurred in the original mature ORNs which survive acutely in the caspase-3 null, and also whether the new ORNs can die without caspase-3 as they do in the wildtype. Furthermore, we must also establish whether the birth of new ORNs occurs as normal in the caspase-3 null after bulbectomy, in the expected numbers and at the expected time, when the mature ORNs which are supposed to signal their replacement as they undergo apoptosis have not died as normal. I hope that this work which follows directly from mine will help to shed more light on the mechanisms of ORN turnover in the OE, and give clues as to the nature of the neurogenesis signal which basal cells receive from dying ORNs. 118 Behavioural work on the caspase-3 null mice A n interesting question which I have not had the opportunity to address is whether the caspase-3 null mice have a normal sense of smell, and there is some interest in addressing this in the laboratory. One of the simplest behavioural tests of olfactory function in mice is one in which pups are placed in the corners of a clean cage, their lightly anaesthetized mother is placed in the centre of the cage, and the latency of time taken to suckle (when motor ability is controlled for) is used as a measure of olfactory function. This test is good for detecting an olfactory deficit, but it might be less good at uncovering any increase in olfactory ability, or a subtle change in olfaction such as changes in threshold or discrimination. There are a number of alternative tests which could be used in this context, mostly used by researchers investigating olfactory learning (e.g. Kucharski et al 1995). In the case of the caspase-3 nulls, the mice can survive to adulthood, and the histology of their primary connections (in terms of glomerulus structure) appears normal. Therefore I would not anticipate that they would have a profound olfactory deficit. Rather, since they have more ORNs, one might hypothesize that they might have more acute sense of smell with perhaps a lower threshold of detection; or alternatively more ORNs might cause problems with proper segregation and regulation of the overly large glomeruli, and therefore they might be hypothesized to have less discrimination, for example. In vitro analysis of apoptosis in OP6 cells We have collaboratively generated a series of olfactory placode (OP) cell lines, which were produced from olfactory progenitor cells from the neuroepithelium of E10.5 C3H mice. 48 clonal OP cell lines were generated, and conditionally immortalized (Boolay 1998). At 33°C the cells divide and can be maintained in culture; when raised to 39°C they stop proliferating and terminally differentiate, due to the inactivation of the immortalizing effect of the SV40 tag. One of these cell lines, OP6, is particularly useful because at 33°C the cells provide a model of immature ORNs, and when treated with retinoic acid and raised to 39°C they terminally differentiate into cells that, in their morphology as well as their expression profile, most consistently resemble mature ORNs (Illing et al 2002). The phenotype of OP6 cells, in terms of their morphology, expression profile, and electrophysiology, has been published (Illing et al 2002). The OP6 cells at 33°C have a tear-shaped morphology with short processes. Their transcriptional profile indicates that they represent an intermediate-late developmental stage in the ORN lineage. They express many markers of immature ORNs, including OE-1 (an ORN transcription factor), Brain Factor-1 (Boolay 1998), TrkB, TrkC, PLCy2, and p75 (Illing et al 2002). OP6 cells differentiated at 39°C in retinoic acid take on a bipolar neuronal morphology and become analogous to mature ORNs, as evidenced by their transcription of key ORN chemosensory signalling components, and their electrophysiology (Cheng 2001, Illing et al 2002). They down-regulate neuron-specific transcription factors required for early stages of neuronal differentiation such as GAP-43 (Cheng 2001), and they express many markers of mature ORNs including |3-actin, olf-1, and GnRH (Boolay 1998), NST and OMP (Cheng 2001), Golf, adenylyl cyclase III and OCNC1 (a subunit of the olfactory cyclic nucleotide-gated channel) (Illing et al 2002), and two closely related odorant receptors, OP 6-13 and OR 6-8 (Illing et al 2002). Their electrophysiology reflects that of a mature ORN, in that whole-cell voltage-clamp recordings detected appropriate 119 voltage-gated sodium and potassium channels (Illing et al 2002). The OP cell lines also have distinct advantages over other olfactory cell lines in that they are conditionally immortalized, clonal, adherent, and express endogenous odorant receptors (Illing et al 2002). The study of OP6 cells can be used as an in vitro complement to in vivo studies in the mouse olfactory system. OP6 cells can be used to further compare apoptosis in development and maturity, and allow direct comparison of the role of caspases or other death mechanisms in immature and mature neurons from the same lineage, with fewer variables than in a maturing animal. In particular, a proportion of these cells die by apoptosis during differentiation into mature ORNs, and then they can also be induced to undergo apoptosis once they are mature. This provides a model of ORN apoptosis at different developmental stages, in which the response to caspase inhibitors can be tested. 90% of immature OP6 cells die over a 3 day period during differentiation into mature bipolar OP6 cells, and 90% of mature OP6 cells die over a 2 day period when serum is withdrawn. Cells dying due to serum-withdrawal quickly lost their bipolar morphology, becoming rounded, brighter, and apparently less adherent. When cell numbers fall during differentiation or after serum withdrawal, a number of parameters of apoptosis can be investigated, including TUNEL, annexin V , MitoCapture, and fluorogenic caspase-3 substrate. OP6 cells can also be transfected with a GFP-tagged cytochrome c, so that cytochrome c can be directly viewed by fluorescence microscopy as it is exiting the mitochondria during apoptosis. A preliminary western blot confirmed that the cells express caspase-3 and caspase-9. In situ labelling of OP6 cells, dying both during differentiation and after serum withdrawal, confirmed that they became TUNEL-positive, but this has not been assessed quantitatively. Immunocytochemistry with the active-caspase-3-specific antibody confirmed that dying OP6 cells activated caspase-3, but again this has not been quantitatively assessed. In preliminary experiments, I tested the effectiveness of caspase-3, caspase-9 and pancaspase peptide inhibitors in preventing or delaying the cell death which occurs during OP6 differentiation and serum-withdrawal. Numbers of viable cells were assessed using the trypan blue exclusion method. Cell morphology was monitored at every timepoint and every condition, to confirm that differentiation itself was not affected. At 2 and 3 days after differentiation there were twice as many cells with 100 p,M pancaspase inhibitor than with control, but this protective effect was lost by 4 days after differentiation. Therefore, although the caspase inhibitors significantly delayed death, they did not prevent it. In contrast to the immature OP6 cells, in mature OP6 cells there was no significant effect of caspase inhibitors in delaying or preventing death. Even though mature OP6 cells treated with caspase inhibitors were not protected from serum-withdrawal-induced death, and died at the same rate as control cells, caspase inhibitors appeared to delay dismantling of the neuronal processes. The OP6 cells could also be used to attempt to address the question of whether the active caspase signal is being transported from the synapse back to the cell body in order to cleave substrates there. Another possibility (which I suspect is more likely, since caspase proenzyme is found throughout the ORN axon) is that molecules of active caspase could be activating other molecules near them, and the activity could reach the soma as a result of a domino affect of activation without any significant transport taking place. This type of retrograde signalling, involving regenerative waves of second messenger, is epitomized by 120 the Ca + + wave during fertilization, and can take place at a speed of 690-8640 mm per day. In contrast, fast retrograde axonal transport exemplified by the transport of NGF, involving the microtubule-associate ATP-ase dynein, can take place at about 200-275 mm per day (reviewed in Fitzsimonds and Poo 1998). We could test the hypothesis that active transport is not the mechanism by which the active caspase signal is propagated from the synapse to the cell body, by inhibiting transport in vitro using a chemical such as vincristine to disrupt microtubule integrity (as Hayes 1992) following serum withdrawal, and then at later timepoints test by immunocytochemistry for active caspase-3 in the cell body. Maya Saleh is utilizing a similar approach to investigate whether caspase-8 directly interacts with dynactin in the propagation of an apoptotic signal. We are working on a protocol for growing OP6 cells and ORNs in compartmented cultures (Campenot 1992), in order to address this and other mechanistic questions. One of the broader aims of testing caspase inhibitors in vitro, is that the most effective ones could then be tested in vivo following bulbectomy, or in one of our other lesion models. However, the ORN's environment is very different in vivo and in vitro. In the dish, all parts of an O R N are simultaneously exposed to the serum-free environment in the media. In contrast in vivo, only the very end of the ORN is exposed to the insult, and the death signal must then travel down the axon, so there is potentially more time and opportunity for rescue. This is supported by the fact that the timecourse of death is much more rapid in the mature OP6 cells in vitro than in the mature ORN in vivo after bulbectomy. Compartmented cultures would therefore also be of interest in modelling more realistic death signals at the axon terminals only. We may also be interested in inducing apoptosis in OP6 cells by a variety of stimuli other than serum-withdrawal. We could use these cells to establish whether different pathways can be involved as a result of different stimuli and to apply this to studying how olfactory cells can die in vivo during normal turnover, as a result of noxious chemicals, or in development. 121 REFERENCES Adams, J. M . and S. Cory (1998). "The Bcl-2 protein family: arbiters of cell survival." Science 281(5381): 1322-6. Allsopp, T. E., S. Wyatt, et al. (1993). "The proto-oncogene bcl-2 can selectively rescue neurotrophic factor-dependent neurons from apoptosis." Cell 73(2): 295-307. Alnemri, E. S., D. J. Livingston, et al. (1996). "Human ICE/CED-3 protease nomenclature [letter]." Cell 87(2): 171. Aloyz, R. S., S. X . Bamji, et al. (1998). "p53 is essential for developmental neuron death as regulated by the TrkA and p75 neurotrophin receptors." J Cell Biol 143(6): 1691-703. Anglade, P., S. Vyas, et al. (1997). "Apoptosis and autophagy in nigral neurons of patients with Parkinson's disease." Histol Histopathol 12(1): 25-31. Ansari, K . A . and A . Johnson (1975). "Olfactory function in patients with Parkinson's disease." J Chronic Pis 28(9): 493-7. Aravind, L . , V . M . Dixit, et al. (2001). "Apoptotic molecular machinery: vastly increased complexity in vertebrates revealed by genome comparisons." Science 291: 1279-84. Armstrong, D. M . , R. Brady, et al. (1991). "Expression of choline acetyltransferase and nerve growth factor receptor within hypoglossal motoneurons following nerve injury." J Comp Neurol 304(4): 596-607. Baba, K. , M . Ikeda, et al. (1997). "Odor exposure reveals non-uniform expression profiles of c-Jun protein in rat olfactory bulb neurons." Brain Res 774(1-2): 142-8. Baglioni, C. (1992). Mechanisms of cytotoxicity, cytolysis, and growth stimulation by TNF. Tumor Necrosis Factors: The Molecules and Their Emerging Role in Medicine. B. Beutler. New York, Raven Press: 425-38. Baker, H . (1990). "Unilateral, neonatal olfactory deprivation alters tyrosine hydroxylase expression but not aromatic amino acid decarboxylase or G A B A immunoreactivity." Neuroscience 36(3): 761-71. Bamji, S. X . , M . Majdan, et al. (1998). "The p75 neurotrophin receptor mediates neuronal apoptosis and is essential for naturally occurring sympathetic neuron death." J Cell Biol 140(4): 911-23. Barber, R. D., D. E. Jaworsky, et al. (2000). "Isolation and in vitro differentiation of conditionally immortalized murine olfactory receptor neurons." J Neurosci 20(10): 3695-704. Barde, Y . A . (1989). "Trophic factors and neuronal survival." Neuron 2(6): 1525-34. Barker, P. A . (1998). "p75NTR: A study in contrasts." Cell Death Differ 5(5): 346-56. Barnes, N . Y . , L . L i , et al. (1998). "Increased production of amyloid precursor protein provides a substrate for caspase-3 in dying motoneurons." J Neurosci 18(15): 5869-80. Barrett, G. L . and P. F. Bartlett (1994). "The p75 nerve growth factor receptor mediates survival or death depending on the stage of sensory neuron development." Proc Natl Acad Sci U S A 91(14): 6501-5. Bergeron, L . , G. I. Perez, et al. (1998). "Defects in regulation of apoptosis in caspase-2-deficient mice." Genes Dev 12(9): 1304-14. Bertin, J., S. M . Mendrysa, et al. (1996). "Apoptotic suppression by baculovirus P35 involves cleavage by and inhibition of a virus-induced CED-3/ICE-like protease." J Virol 70(9): 6251-9. 122 Black, S. C , J. Q. Huang, et al. (1998). "Co-localization of the cysteine protease caspase-3 with apoptotic myocytes after in vivo myocardial ischemia and reperfusion in the rat." J M o l Cell Cardiol 30(4): 733-42. Boolay, S. (1998). Immortalisation, Characterisation and Differentiation of Temperature Sensitive Cell Lines from the Olfactory Neuroepithelium. Department of Biochemistry. Cape Town, University of Cape Town: 212. Braun, J. S., R. Novak, et al. (1999). "Neuroprotection by a caspase inhibitor in acute bacterial meningitis." Nat Med 5(3): 298-302. Bredesen, D. E. (1995). "Neural apoptosis." Ann Neurol 38(6): 839-51. Breipohl, W., A . Mackay-Sim, et al. (1986). Neurogenesis in the vertebrate main olfactory epithelium. Ontogeny of Olfaction. W. Breipohl. Berlin, Springer-Verlag: 21-3. Brown, S. B., K. Bailey, et al. (1997). "Actin is cleaved during constitutive apoptosis." Biochem J 323 (Pt l ) : 233-7. Brunet, C. L . , R. H. Gunby, et al. (1998). "Commitment to cell death measured by loss of clonogenicity is separable from the appearance of apoptotic markers." Cell Death Differ 5(1): 107-15. Burek, M . J. and R. W. Oppenheim (1999). Cellular interactions that regulate programmed cell death in the developing vertebrate nervous system. Cell Death and Diseases of the Nervous System. V . E. Koliatsos and R. Ratan, eds. Totowa, N.J. Humana Press: 145-79. Buttyan, R., C. A . Olsson, et al. (1989). "Induction of the TRPM-2 gene in cells undergoing programmed death." Mol Cell Biol 9(8): 3473-81. Bylsma, F. W., P. J. Moberg, et al. (1997). "Odor identification in Huntington's disease patients and asymptomatic gene carriers." J Neuropsychiatry Clin Neurosci 9(4): 598-600. Caggiano, M . , J. S. Kauer, et al. (1994). "Globose basal cells are neuronal progenitors in the olfactory epithelium: a lineage analysis using a replication-incompetent retrovirus." Neuron 13(2): 339-52. Calof, A . L. , P. C. Rim, et al. (1998). "Factors regulating neurogenesis and programmed cell death in mouse olfactory epithelium." Ann N Y Acad Sci 855: 226-9. Calof, A . L . , N . Hagiwara, et al. (1996). "Neurogenesis and cell death in olfactory epithelium." J Neurobiol 30(11:67-81. Calof, A . L. , J. D. Holcomb, et al. (1996). "Factors affecting neuronal birth and death in the mammalian olfactory epithelium." Ciba Found Symp 196: 188-205; discussion; 205-10. Campbell, D. S. and C. E. Holt (2003). "Apoptotic Pathway and M A P K s Differentially Regulate Chemotropic Responses of Retinal Growth Cones." Neuron 37(6): 939-52. Campenot, R. B. (1992). "Compartmented culture analysis of nerve growth." In Cell-Cell Interactions: A Practical Approach. B. Stevenson, D. Paul, and W. Gallin, editors. Oxford, IRL Press 275-98. Carr, V . M . and A . I. Farbman (1992). "Ablation of the olfactory bulb up-regulates the rate of neurogenesis and induces precocious cell death in olfactory epithelium." Exp Neurol 115(1): 55-9. Carr, V . M . and A . I. Farbman (1993). "The dynamics of cell death in the olfactory epithelium." Exp Neurol 124(2): 308-14. Carter, L . A . and A . J. Roskams (2002). "Neurotrophins and their receptors in the primary olfactory neuraxis." Microsc Res Tech 58(3): 189-196". Castren, E., Y . Ohga, et al. (1994). "bcl-2 messenger R N A is localized in neurons of the developing and adult rat brain." Neuroscience 61(1): 165-77. 123 Chan, S. L . , W. S. Griffin, et al. (1999). "Evidence for caspase-mediated cleavage of A M P A receptor subunits in neuronal apoptosis and Alzheimer's disease." J Neurosci Res 57(3): 315-23. Chan, S. L . and M . P. Mattson (1999). "Caspase and calpain substrates: roles in synaptic plasticity and cell death." J Neurosci Res 58(1): 167-90. Chen, J., T. Nagayama, et al. (1998). "Induction of caspase-3-like protease may mediate delayed neuronal death." J Neurosci 18(13): 4914-28. Cheng, J. S. (2001). A Role for Methyl-CpG Binding Domain Protein 2 in Olfactory Neuronal Development. Psychiatry. Neuroscience Program. Vancouver, University of British Columbia. Cheng, Y . , M . Deshmukh, et al. (1998). "Caspase inhibitor affords neuroprotection with delayed administration in a rat model of neonatal hypoxic-ischemic brain injury [see comments]." J Clin Invest 101(9): 1992-9. Choi, D. W. (1996). "Ischemia-induced neuronal apoptosis." Curr Opin Neurobiol 6(5): 667-72. Clark, R. S., P. M . Kochanek, et al. (1999). "Increases in Bcl-2 and cleavage of caspase-1 and caspase-3 in human." Faseb J 13(8): 813-21. Cohen, G. M . (1997). "Caspases: the executioners of apoptosis." Biochem J 326(Pt 1): 1-16. Cohen, J. A . , A . T. Yachnis, et al. (1992). "Expression of the neu proto-oncogene by Schwann cells during peripheral nerve development and Wallerian degeneration." J Neurosci Res 31(4): 622-34. Constantinescu, C. S., E. C. Raps, et al. (1994). "Olfactory disturbances as the initial or most prominent symptom of multiple sclerosis." J Neurol Neurosurg Psychiatry 57(8): 1011-2. Costanzo, R. M . and P. P. Graziadei (1983). "A quantitative analysis of changes in the olfactory epithelium following bulbectomy in hamster." J Comp Neurol 215(4): 370-81. Cowan, C. M . and A. J. Roskams (2002). "Apoptosis in the mature and developing olfactory neuroepithelium." Microsc Res Tech 58(3): 204—15. Cowan, C. M . , J. Thai, et al. (2001). "Caspases 3 and 9 send a pro-apoptotic signal from synapse to cell body in olfactory receptor neurons." J Neurosci 21(18): 7099-109. Cryns, V . and J. Yuan (1998). "Proteases to die for [published erratum appears in Genes Dev 1999 Feb 1:13(31:3711." Genes Dev 12(11): 1551-70. Cunningham, A . M . , P. B. Manis, et al. (1999). "Olfactory receptor neurons exist as distinct subclasses of immature and mature cells in primary culture." Neuroscience 93(4): 1301-12. Cuschieri, A . and L. H . Bannister (1975). "The development of the olfactory mucosa in the mouse: electron microscopy." J Anat 119(3): 471-98. Davies, A . M . (1997). "Neurotrophin switching: where does it stand?" Curr Opin Neurobiol 7(1): 110-8. de Bilbao, F., E. Guarin, et al. (1999). "Postnatal distribution of cpp32/caspase 3 mRNA in the mouse central nervous system: an in situ hybridization study." J Comp Neurol 409(3): 339-57. Deckner, M . L . , M . Risling, et al. (1997). "Apoptotic death of olfactory sensory neurons in the adult rat." Exp Neurol 143(1): 132-40. Doty, R. L. , D. A . Deems, et al. (1988). "Olfactory dysfunction in parkinsonism: a general deficit unrelated to neurologic signs, disease stage, or disease duration." Neurology 38(8): 1237^4. 124 Doty, R. L. , C. L i , et al. (1999). "Olfactory dysfunction in multiple sclerosis: relation to longitudinal changes in plaque numbers in central olfactory structures." Neurology 53(4): 880-2. Doty, R. L. , D. P. Perl, et al. (1991). "Olfactory dysfunction in three neurodegenerative diseases." Geriatrics 46 Suppl 1(8): 47-51. Doty, R. L. , P. F. Reyes, et al. (1987). "Presence of both odor identification and detection deficits in Alzheimer's disease." Brain Res Bull 18(5): 597-600. Doty, R. L . , P. Shaman, et al. (1984). "Smell identification ability: changes with age." Science 226(4681): 1441-3. Doucette, J. R., J. A . Kiernan, et al. (1983). "Two different patterns of retrograde degeneration in the olfactory epithelium following transection of primary olfactory axons." J Anat 136(Pt 4): 673-89. Du, Y . , K . R. Bales, et al. (1997). "Activation of a caspase 3-related cysteine protease is required for." Proc Natl Acad Sci U S A 94(21): 11657-62. Earnshaw, W. C., L . M . Martins, et al. (1999). "Mammalian Caspases: Structure, Activation, Substrates and Functions During Apoptosis." Annual Review of Biochemistry 68: 383^124. Eckhart, L . , W. Declercq, et al. (2000). "Terminal differentiation of human keratinocytes and stratum corneum formation is associated with caspase-14 activation." J Invest Dermatol 115(6): 1148-51. Eichenbaum, H. , T. H. Morton, et al. (1983). "Selective olfactory deficits in case H . M . " Brain 106(Pt 2): 459-72. Ellerby, H . M . , S. J. Martin, et al. (1997). "Establishment of a cell-free system of neuronal apoptosis: comparison of premitochondrial, mitochondrial, and postmitochondrial phases." J Neurosci 17(16): 6165-78. ElShamy, W. M . and P. Ernfors (1996). "A local action of neurotrophin-3 prevents the death of proliferating sensory neuron precursor cells." Neuron 16(5): 963-72. Enari, M . , H . Sakahira, et al. (1998). "A caspase-activated DNase that degrades D N A during apoptosis, and its inhibitor ICAD [see comments] [published erratum appears in Nature 1998 May 28;393(6683):396].n Nature 391(6662): 43-50. Enari, M . , R. V . Talanian, et al. (1996). "Sequential activation of ICE-like and CPP32-like proteases during Fas." Nature 380(6576): 723-6. Endres, M . , S. Namura, et al. (1998). "Attenuation of delayed neuronal death after mild focal ischemia in mice by inhibition of the caspase family." J Cereb Blood Flow Metab 18(3): 238^17. Ernfors, P., A . Henschen, et al. (1989). "Expression of nerve growth factor receptor mRNA is developmentally regulated and increased after axotomy in rat spinal cord motoneurons." Neuron 2(6): 1605-13. Estus, S., W. J. Zaks, et al. (1994). "Altered gene expression in neurons during programmed cell death: identification of c-jun as necessary for neuronal apoptosis." J Cell Biol 127(6 Pt 1): 1717-27. Evan, G. and T. Littlewood (1998). "A matter of life and cell death." Science 281(5381): 1317-22. Faleiro, L . , R. Kobayashi, et al. (1997). "Multiple species of CPP32 and Mch2 are the major active caspases present in apoptotic cells." Embo J 16(9): 2271-81. Farbman, A . I. (1990). "Olfactory neurogenesis: genetic or environmental controls?" Trends Neurosci 13(9): 362-5. 125 Farbman, A . I., P. C. Brunjes, et al. (1988). "The effect of unilateral naris occlusion on cell dynamics in the developing rat olfactory epithelium." J Neurosci 8(9): 3290-5. Farbman, A . I., J. A . Buchholz, et al. (1999). "A molecular basis of cell death in olfactory epithelium." J Comp Neurol 414(31: 306-14. Farinas, I., C. K . Yoshida, et al. (1996). "Lack of neurotrophin-3 results in death of spinal sensory neurons and premature differentiation of their precursors." Neuron 17(6): 1065-78. Faucheu, C , A . M . Blanchet, et al. (1996). "Identification of a cysteine protease closely related to interleukin-1 beta-converting enzyme." Eur J Biochem 236(1): 207-13. Fernandes-Alnemri, T., G. Litwack, et al. (1995). "Mch2, a new member of the apoptotic Ced-3/Ice cysteine protease gene family." Cancer Res 55(13): 2737-42. Fernandes-Alnemri, T., A . Takahashi, et al. (1995). "Mch3, a novel human apoptotic cysteine protease highly related to CPP32." Cancer Res 55(24): 6045-52. Fernando, P., J. F. Kelly, et al. (2002). "Caspase 3 activity is required for skeletal muscle differentiation." Proc Natl Acad Sci U S A 99(17): 11025-30. Ferrer, I., M . Olive, et al. (1996). "Naturally occurring (programmed) and radiation-induced apoptosis are associated with selective c-Jun expression in the developing rat brain." Eur J Neurosci 8(6): 1286-98. Fink, K. , J. Zhu, et al. (1998). "Prolonged therapeutic window for ischemic brain damage caused by delayed caspase activation." J Cereb Blood Flow Metab 18(10): 1071-6. Finn, J. T., M . Weil, et al. (2000). "Evidence that Wallerian degeneration and localized axon degeneration induced by local neurotrophin deprivation do not involve caspases." J Neurosci 20(4): 1333-41. Fiske, B . K . and P. C. Brunjes (2001). "Cell death in the developing and sensory-deprived rat olfactory bulb." J Comp Neurol 431(3): 311-9. Fitzsimonds, R. M . and M . M . Poo (1998). "Retrograde signaling in the development and modification of synapses." Physiol Rev 78(1): 143-70. Frade, J. M . and Y - A . Barde (1999). "Geneetic evidence for cell death mediated by nerve growth factor and the neurotrophin receptor p75 in the developing mouse retina and spinal cord." Development 126(4): 683-90. Fujita, E., J. Egashira, et al. (2001). "Caspase-9 processing by caspase-3 via a feedback amplification loop in vivo." Cell Death Differ 8(4): 335^14. Gagliardini, V . , P. A. Fernandez, et al. (1994). "Prevention of vertebrate neuronal death by the crmA gene." Science 263(5148): 826-8. Garcia, I., I. Martinou, et al. (1992). "Prevention of programmed cell death of sympathetic neurons by the bcl-2 proto-oncogene." Science 258(5080): 302-4. Gavrieli, Y . , Y . Sherman, et al. (1992). "Identification of programmed cell death in situ via specific labeling." J Cell Biol 119(3): 493-501. Gervais, F. G., D. Xu, et al. (1999). "Involvement of caspases in proteolytic cleavage of Alzheimer's amyloid-beta precursor protein and amyloidogenic A beta peptide formation." Cell 97(3): 395-406. Goldberg, Y . P., D. W. Nicholson, et al. (1996). "Cleavage of huntingtin by apopain, a proapoptotic cysteine protease, is modulated by the polyglutamine tract [see comments]." Nat Genet 13(4): 442-9. Gong, Q., M . S. Bailey, et al. (1994). "Localization and regulation of low affinity nerve growth factor receptor expression in the rat olfactory system during development and regeneration." J Comp Neurol 344(3): 336—48. 126 Graziadei, P. P. (1973). "Cell dynamics in the olfactory mucosa." Tissue Cell 5(1): 113-31. Graziadei, P. P., R. R. Levine, et al. (1978). "Regeneration of olfactory axons and synapse formation in the forebrain after bulbectomy in neonatal mice." Proc Natl Acad Sci U S A 75(10): 5230^. Graziadei, P. P., R. R. Levine, et al. (1979). "Plasticity of connections of the olfactory sensory neuron: regeneration into the forebrain following bulbectomy in the neonatal mouse." Neuroscience 4(6): 713-27. Graziadei, P. P. and G. A . Monti Graziadei (1980). "Neurogenesis and neuron regeneration in the olfactory system of mammals. III. Deafferentation and reinnervation of the olfactory bulb following section of the fila olfactoria in rat." J Neurocytol 9(2): 145-62. Graziadei, P. P. C. and G. A . Monti Graziadei (1978). The olfactory system: a model for the study of neurogenesis and axon regeneration in mammals. Neuronal Plasticity. C. W. Cotman. New York, Raven Press 131-53. Graziadei, P. P. and G. A . Graziadei (1979). "Neurogenesis and neuron regeneration in the olfactory system of mammals. I. Morphological aspects of differentiation and structural organization of the olfactory sensory neurons." J Neurocytol 8(1): 1-18. Gross, A. , J. M . McDonnell, et al. (1999). "BCL-2 family members and the mitochondria in apoptosis." Genes Dev 13(15): 1899-911. Haanen, C. and I. Vermes (1996). "Apoptosis: programmed cell death in fetal development." Eur J Obstet Gvnecol Reprod Biol 64(1): 129-33. Hahn, P., T. Lindsten, et al. (2003). Bax/Bak-deficienct mice exhibit persistence of hyaloid vessels and suppression of developmentally programed cell death in the inner retina including ectopic photoreceptors. "Molecular Mechanisms of Apoptosis" Keystone Symposiuni, Banff, A B . Hakem, R., A . Hakem, et al. (1998). "Differential requirement for caspase 9 in apoptotic pathways in vivo." Cell 94(3): 339-52. Halasz, N . (1990). The Vertebrate Olfactory System: Chemical neuroanatomy, function and development. Budapest, Academiai Kiado. Ham, J., C. Babij, et al. (1995). "A c-Jun dominant negative mutant protects sympathetic neurons against programmed cell death." Neuron 14(5): 927-39. Hara, A . , N . Yoshimi, et al. (1995). "DNA fragmentation in granular cells of human cerebellum following global ischemia." Brain Res 697(1-2): 247-50. Hayes, R. C , R. G. Wiley, et al. (1992). "Induction of nerve growth factor receptor (p75NGFr) mRNA within hypoglossal motoneurons following axonal injury." Brain Res Mol Brain Res 15(3-4): 291-7. Hengartner, M . O. and H . R. Horvitz (1994). "Programmed cell death in Caenorhabditis elegans." Curr Opin Genet Dev 4(4): 581-6. Henshall, D. C , S. L . Skradski, et al. (2002). "Expression and differential processing of caspases 6 and 7 in relation to specific epileptiform E E G patterns following limbic seizures." Neurobiol Pis 10(2): 71-87. Himi, T., Y . Ishizaki, et al. (1998). "A caspase inhibitor blocks ischaemia-induced delayed neuronal death in the gerbil." Eur J Neurosci 10(2): 777-81. Hinds, J. W., P. L . Hinds, et al. (1984). "An autoradiographic study of the mouse olfactory epithelium: evidence for long-lived receptors." Anat Rec 210(2): 375-83. Holcomb, J. D., J. S. Mumm, et al. (1995). "Apoptosis in the neuronal lineage of the mouse olfactory epithelium: regulation in vivo and in vitro." Dev Biol 172(1): 307-23. 127 Huang, C. C , K . Chen, et al. (1995). "Immunohistochemical studies of sensory neurons in rat olfactory epithelium." Eur Arch Otorhinolaryngol 252(2): 86-91. Huard, J. M . , S. L . Youngentob, et al. (1998). "Adult olfactory epithelium contains multipotent progenitors that give rise to neurons and non-neural cells." J Comp Neurol 400(4): 469-86. Illing, N . , S. Boolay, et al. (2002). "Conditionally Immortalized Lines from the Olfactory Placode Differentiate into Olfactory Neurons." Mol Cell Neurosci 20(2): 225-43. Itoh, N . , S. Yonehara, et al. (1991). "The polypeptide encoded by the cDNA for human cell surface antigen Fas can mediate apoptosis." Cell 66(2): 233-43. Jarvis, W. D., R. N . Kolesnick, et al. (1994). "Induction of apoptotic D N A damage and cell death by activation of the sphingomyelin pathway." Proc Natl Acad Sci U S A 91(1): 73-7. Jeon, B. S., N . G. Kholodilov, et al. (1999). "Activation of caspase-3 in developmental models of programmed cell death in neurons of the substantia nigra." J Neurochem 73(1): 322-33. Jimbo, A. , E. Fujita, et al. (2003). "ER stress induces caspase-8 activation, stimulating cytochrome c release and caspase-9 activation." Exp Cell Res 283(2): 156-66. Jourdan, F., E. Moyse, et al. (1998). "Olfactory neurons are protected from apoptosis in adult transgenic mice over-expressing the bcl-2 gene." Neuroreport 9(5): 921-6. Juan, T. S., I. K . McNiece, et al. (1997). "Identification and mapping of Casp7, a cysteine protease resembling CPP32 beta, interleukin-1 beta converting enzyme, and CED-3." Genomics 40(1): 86-93. Kamens, J., M . Paskind, et al. (1995). "Identification and characterization of ICH-2, a novel member of the interleukin-1 beta-converting enzyme family of cysteine proteases." J Biol Chem 270(25): 15250-6. Kaplan, D. R. and F. D. Miller (2000). "Neurotrophin signal transduction in the nervous system." Curr Opin Neurobiol 10(3): 381-91. Kasowski, H . J., H . Kim, et al. (1999). "Compartmental organization of the olfactory bulb glomerulus." J Comp Neurol 407(2): 261-74. Kastan, M . B. , Q. Zhan, et al. (1992). "A mammalian cell cycle checkpoint pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia." Cell 71(4): 587-97. Kayalar, C , T. Ord, et al. (1996). "Cleavage of actin by interleukin 1 beta-converting enzyme to reverse DNase I inhibition." Proc Natl Acad Sci U S A 93(5): 2234-8. Kermer, P., N . Klocker, et al. (1998). "Inhibition of CPP32-like proteases rescues axotomized retinal ganglion cells from secondary cell death in vivo." J Neurosci 18(12): 4656-62. Kerr, J. F., A . H . Wyllie, et al. (1972). "Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics." Br J Cancer 26(4): 239-57. Kim, T. W., W. H . Pettingell, et al. (1997). "Alternative cleavage of Alzheimer-associated presenilins during apoptosis by a caspase-3 family protease." Science 277(5324): 373-6. Kokaia, Z., G. Andsberg, et al. (1998). "Focal cerebral ischemia in rats induces expression of P75 neurotrophin receptor in resistant striatal cholinergic neurons." Neuroscience 84(4): 1113-25. Krajewski, S., M . Krajewska, et al. (1999). "Release of caspase-9 from mitochondria during neuronal apoptosis and cerebral ischemia." Proc Natl Acad Sci U S A 96(10): 5752-7. Kucharski, D., H . M . Arnold, et al. (1995). "Unilateral conditioning of an odor aversion in 6-day-old rat pups." Behav Neurosci 109(3): 563-6. 128 Kuida, K. , T. F. Haydar, et al. (1998). "Reduced apoptosis and cytochrome c-mediated caspase activation in mice lacking caspase 9." Cell 94(3): 325-37. Kuida, K. , J. A . Lippke, et al. (1995). "Altered cytokine export and apoptosis in mice deficient in interleukin-1 beta converting enzyme." Science 267(5206): 2000-3. Kuida, K. , T. S. Zheng, et al. (1996). "Decreased apoptosis in the brain and premature lethality in CPP32." Nature 384(6607): 368-72. Kumar, S., M . Kinoshita, et al. (1994). "Induction of apoptosis by the mouse Nedd2 gene, which encodes a protein similar to the product of the Caenorhabditis elegans cell death gene ced-3 and the mammalian IL-1 beta-converting enzyme." Genes Dev 8(14): 1613-26. Kumar, S., Y . Tomooka, et al. (1992). "Identification of a set of genes with developmentally down-regulated expression in the mouse brain." Biochem Biophys Res Commun 185(3): 1155-61. Kwon, B . K. , J. Liu, et al. (2002). "Survival and regeneration of rubrospinal neurons 1 year after spinal cord injury." Proc Natl Acad Sci U S A 99(5): 3246-51. Lam, T. T., A . S. Abler, et al. (1999). "Apoptosis and caspases after ischemia-reperfusion injury in rat retina." Invest Ophthalmol Vis Sci 40(5): 967-75. Lassmann, H. , C. Bancher, et al. (1995). "Cell death in Alzheimer's disease evaluated by D N A fragmentation in situ." Acta Neuropathol (Berl) 89(1): 35^11. Le, D. A. , Y . Wu, et al. (2002). "Caspase activation and neuroprotection in caspase-3-deficient mice after in vivo cerebral ischemia and in vitro oxygen glucose deprivation." Proc Natl Acad Sci U S A 99(23): 15188-93. Le Gros Clark, W. E. (1951). "The projection of the olfactory epithelium on the olfactory bulb in the rabbit." J Neurol Psychiatry 14: 1-10. LeBlanc, A . , H . Liu , et al. (1999). "Caspase-6 role in apoptosis of human neurons, amyloidogenesis, and Alzheimer's disease." J Biol Chem 274(33): 23426-36. Leonard, J. R., B . J. Klocke, C. D'Sa, R. A . Flavell and K. A . Roth (2002). "Strain-dependent neurdevelopmental abnormalities in caspase-3-deficient mice." J Neuropathol Exp Neurol 61(8): 673-7. Levkau, B. , H . Koyama, et al. (1998). "Cleavage of p21Cipl/Wafl and p27Kipl mediates apoptosis in endothelial cells through activation of Cdk2: role of a caspase cascade." Mol Cell 1(4): 553-63. L i , H . , F. Colbourne, et al. (2000). "Caspase inhibitors reduce neuronal injury after focal but not global cerebral ischemia in rats." Stroke 31(1): 176-82. L i , J., C. A . Bombeck, et al. (1999). "Nitric oxide suppresses apoptosis via interrupting caspase activation and mitochondrial dysfunction in cultured hepatocytes." J Biol Chem 274(24): 17325-33. L i , P., H. Allen, et al. (1995). "Mice deficient in IL-1 beta-converting enzyme are defective in production of mature IL-1 beta and resistant to endotoxic shock." Cell 80(3): 401-11. Lippens, S., M . Kockx, et al. (2000). "Epidermal differentiation does not involve the pro-apoptotic executioner caspases, but is associated with caspase-14 induction and processing." Cell Death Differ 7(12): 1218-24. Loetscher, H. , U . Deuschle, et al. (1997). "Presenilins are processed by caspase-type proteases." J Biol Chem 272(33): 20655-9. Loo, D. T., A . Copani, et al. (1993). "Apoptosis is induced by beta-amyloid in cultured central nervous system neurons." Proc Natl Acad Sci U S A 90(17): 7951-5. Lowe, S. W., H . E. Ruley, et al. (1993). "p53-dependent apoptosis modulates the cytotoxicity of anticancer agents." Cell 74(6): 957-67. 129 Lu, D. C , S. Rabizadeh, et al. (2000). "A second cytotoxic proteolytic peptide derived from amyloid beta-protein precursor." Nat Med 6(4): 397-404. Ma, J., M . Endres, et al. (1998). "Synergistic effects of caspase inhibitors and MK-801 in brain injury after transient focal cerebral ischaemia in mice." Br J Pharmacol 124(4): 756-62. MacEwan, D. J. (1996). "Elevated cPLA2 levels as a mechanism by which the p70 TNF and p75 N G F receptors enhance apoptosis." FEBS Lett 379(1): 77-81. Mackay-Sim, A . and P. Kittel (1991). "Cell dynamics in the adult mouse olfactory epithelium: a quantitative autoradiographic study." J Neurosci 11(4): 979-84. Magrassi, L . and P. P. Graziadei (1995). "Cell death in the olfactory epithelium." Anat Embryol (Berll 192(1): 77-87. Martin, D. A. , R. M . Siegel, et al. (1998). "Membrane oligomerization and cleavage activates the caspase-8 (FLICE/MACHalpha 1) death signal." J Biol Chem 273(8): 4345-9. Martin, D. P., R. E. Schmidt, et al. (1988). "Inhibitors of protein synthesis and R N A synthesis prevent neuronal death caused by nerve growth factor deprivation." J Cell Biol 106(3): 829^14. Martin, L . J. (1999). "Neuronal death in amyotrophic lateral sclerosis is apoptosis: possible contribution of a programmed cell death mechanism." J Neuropathol Exp Neurol 58(5): 459-71. Mattson, M . P. and W. Duan (1999). ""Apoptotic" biochemical cascades in synaptic compartments: roles in adaptive plasticity and neurodegenerative disorders." J Neurosci Res 58(1): 152-66. Mattson, M . P., Q. Guo, et al. (1998). "Presenilins, the endoplasmic reticulum, and neuronal apoptosis in Alzheimer's disease." J Neurochem 70(1): 1-14. Mattson, M . P., J. N . Keller, et al. (1998). "Evidence for synaptic apoptosis." Exp Neurol 153(1): 35^18. McCall , K . and H. Steller (1997). "Facing death in the fly: genetic analysis of apoptosis in Drosophila." Trends Genet 13(6): 222-6. McCarthy, N . J., M . K . B. Whyte, et al. (1997). "Inhibition of ced-3/ICE-related proteases does not prevent cell death induced by oncogenes, D N A damage, or the bcl-2 homologue bak." J Cell Biol 136(1): 215-27. Mehmet, H . and A . D. Edwards (1996). "Hypoxia, ischaemia, and apoptosis." Arch Pis Child Fetal Neonatal Ed 75(2): F73-5. Meisami, E. (1976). "Effects of olfactory deprivation on postnatal growth of the rat olfactory bulb utilizing a new method for production of neonatal unilateral anosmia." Brain Res 107(2): 437-44. Mesner, P. W., Jr, K . C. Bible, et al. (1999). "Characterization of caspase processing and activation in HL-60 cell cytosol under cell-free conditions. Nucleotide requirement and inhibitor profile." J Biol Chem 274(32): 22635^15. Michaelidis, T. M . , M . Sendtner, et al. (1996). "Inactivation of bcl-2 results in progressive degeneration of motoneurons, sympathetic and sensory neurons during early postnatal development." Neuron 17(1): 75-89. Michalovitz, P . , O. Halevy, et al. (1990). "Conditional inhibition of transformation and of cell proliferation by a temperature-sensitive mutant of p53." Cell 62(4): 671-80. Michel, P. , E. Moyse, et al. (1994). "Induction of apoptosis in mouse [correction of rat] olfactory neuroepithelium by synaptic target ablation." Neuroreport 5(11): 1329-32. Michel, P . , E. Moyse, et al. (1997)."Clusterin/ApoJ expression is associated with neuronal apoptosis in the olfactory mucosa of the adult mouse." J Cell Sci 110(Pt 14): 1635-45. 130 Middleton, G., L . G. Pinon, et al. (1998). "Bcl-2 accelerates the maturation of early sensory neurons." J Neurosci 18(9): 3344-50. Milligan, C. E. and L. M . Schwartz (1997). "Programmed cell death during animal development." Br Med Bull 53(3): 570-90. Miyashita, T. and J. C. Reed (1995). "Tumor suppressor p53 is a direct transcriptional activator of the human bax gene." Cell 80(2): 293-9. Moberg, P. J. and R. L . Doty (1997). "Olfactory function in Huntington's disease patients and at-risk offspring." Int J Neurosci 89(1-2): 133-9. Moberg, P. J., G. D. Pearlson, et al. (1987). "Olfactory recognition: differential impairments in early and late Huntington's and Alzheimer's diseases." J Clin Exp Neuropsychol 9(6): 650-64. Morita, Y . , D. V . Maravei, et al. (2001). "Caspase-2 deficiency prevents programmed germ cell death resulting from cytokine insufficiency but not meiotic defects caused by loss of ataxia telangiectasia-mutated (Atm) gene function." Cell Death Differ 8(6): 614-2Q. Morita-Fujimura, Y . , M . Fujimura, et al. (1999). "Inhibition of interleukin-1 beta converting enzyme family proteases (caspases) reduces cold injury-induced brain trauma and D N A fragmentation in mice." J Cereb Blood Flow Metab 19(6): 634-42. Morrison, E. E. and R. M . Costanzo (1989). "Scanning electron microscopic study of degeneration and regeneration in the olfactory epithelium after axotomy." J Neurocytol 18(3): 393^105. Moulton, D. G. (1974). "Dynamics of cell populations in the olfactory epithelium." Ann N Y Acad Sci 237(0): 52-61. Munday, N . A. , J. P. Vaillancourt, et al. (1995). "Molecular cloning and pro-apoptotic activity of ICErelll and ICErelHI, members of the ICE/CED-3 family of cysteine proteases." J Biol Chem 270(26): 15870-6. Muzio, M . , A . M . Chinnaiyan, et al. (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(6): 817-27. Muzio, M . , B. R. Stockwell, et al. (1998). "An induced proximity model for caspase-8 activation." J Biol Chem 273(5): 2926-30. Nicholson, D. W. and N . A . Thornberry (1997). "Caspases: killer proteases." Trends Biochem Sci 22(8): 299-306. Olson, L . (1997). "The coming of age of the GDNF family and its receptors: gene delivery in a rat Parkinson model may have clinical implications." Trends Neurosci 20(7): 277-9. Oppenheim, R. W. (1991). "Cell death during development of the nervous system." Annu Rev Neurosci 14: 453-501. Pallera, A . M . , J. B. Schweitzer, et al. (1994). "192 IgG-saporin causes a major loss of synaptic content in rat olfactory bulb." Exp Neurol 127(2): 265-77. Pasinelli, P., D. R. Borchelt, et al. (1998). "Caspase-1 is activated in neural cells and tissue with amyotrophic lateral sclerosis-associated mutations in copper-zinc superoxide dismutase." Proc Natl Acad Sci U S A 95(26): 15763-8. Pellegrini, M . and A . Strasser (1999). "A portrait of the Bcl-2 protein family: life, death, and the whole picture." J Clin Immunol 19(6): 365-77. Pellier, V . and L. Astic (1994a). "Cell death in the developing olfactory epithelium of rat embryos." Brain Res Dev Brain Res 79(2): 307-15. Pellier, V . and L. Astic (1994b). "Detection of apoptosis by electron microscopy and in situ labelling in the rat olfactory pit." Neuroreport 5(12): 1429-32. 131 Pellier, V . and L . Astic (1994c). "Histochemical and immunocytochemical study of the migration of neurons from the rat olfactory placode." Cell Tissue Res 275(3): 587-98. Pellier, V. , D. Saucier, et al. (1996). "Ultrastructural and cytochemical identification of apoptotic cell death accompanying development of the fetal rat olfactory nerve layer." Anat Embrvol (Bed) 194(1): 99-109. Pettmann, B . and C. E. Henderson (1998). "Neuronal cell death." Neuron 20(4): 633-47. Pinching, A . J. (1977). "Clinical testing of olfaction reassessed." Brain 100(2): 377-88. Pinon, L . G., G. Middleton, et al. (1997). "Bcl-2 is required for cranial sensory neuron survival at defined stages of embryonic development." Development 124(20): 4173-8. Pompl, P. N . , S. Yemul, et al. (2003): "Caspase gene expression in the brain as a function of the clinical progression of Alzheimer disease." Arch Neurol 60(3): 369-76. Portera-Cailliau, C , J. C. Hedreen, et al. (1995). "Evidence for apoptotic cell death in Huntington disease and excitotoxic animal models." J Neurosci 15(5 Pt 2): 3775-87. Pozniak, C. D., F. Barnabe-Heider, et al. (2002). "p73 is required for survival and maintenance of CNS neurons." J Neurosci 22(221: 9800-9. Puche, A . C. and M . T. Shipley (1999). "Odor-induced, activity-dependent transneuronal gene induction in vitro: mediation by N M D A receptors." J Neurosci 19(4): 1359-70. Putcha, G. V. , M . Deshmukh, et al. (2000). "Inhibition of apoptotic signaling cascades causes loss of trophic factor dependence during neuronal maturation." J Cell Biol 149(5): 1011-8. Rabizadeh, S., J. Oh, et al. (1993). "Induction of apoptosis by the low-affinity N G F receptor." Science 261(51191: 345-8. Raff, M . , J. Finn, et al. (2003). Self-Destruct Programs in Neurons. "Molecular Mechanisms of Apoptosis" Keystone Symposium, Banff, A B . Ramon-Cueto, A . and J. Avila (1998). "Olfactory ensheathing glia: properties and function." Brain Res Bull 46(3): 175-87. Renatus, M . , H . R. Stennicke, et al. (2001). "Dimer formation drives the activation of the cell death protease caspase 9." Proc Natl Acad Sci U S A 98(25): 14250-5. Rink, A . , K . M . Fung, et al. (1995). "Evidence of apoptotic cell death after experimental traumatic brain injury." A m J Pathol 147(6): 1575-83. Roskams, A . J., M . A . Bethel, et al. (1996). "Sequential expression of Trks A , B, and C in the regenerating olfactory neuroepithelium." J Neurosci 16(4): 1294-307. Roskams, A . J., D. S. Bredt, et al. (1994). "Nitric oxide mediates the formation of synaptic connections in developing and regenerating olfactory receptor neurons." Neuron 13(2): 289-99. Roskams, A . J., X . Cai, et al. (1998). "Expression of neuron-specific beta-Ill tubulin during olfactory neurogenesis in the embryonic and adult rat." Neuroscience 83(1): 191-200. Roth, K . A . and C. D'Sa (2001). "Apoptosis and brain development." Ment Retard Dev Disabil Res Rev 7(4): 261-6. Roux, P. P., M . A . Colicos, et al. (1999). "p75 neurotrophin receptor expression is induced in apoptotic neurons after seizure." J Neurosci 19(16): 6887-96. Saini, K . S. and N . I. Walker (1998). "Biochemical and molecular mechanisms regulating apoptosis." Mol Cell Biochem 178(1-2): 9-25. Saleh, A. , S. M . Srinivasula, et al. (1999). "Cytochrome c and dATP-mediated oligomerization of Apaf-1 is a prerequisite for procaspase-9 activation." J Biol Chem 274(25): 17941-5. Salvesen, G. S. and V . M . Dixit (1997). "Caspases: intracellular signaling by proteolysis." Cell 91(4): 443-6. 132 Samali, A . , B. Zhivotovsky, et al. (1998). "Detection of pro-caspase-3 in cytosol and mitochondria of various tissues." FEBS Lett 431(2): 167-9. Sanchez, I., C. J. Xu , et al. (1999). "Caspase-8 is required for cell death induced by expanded polyglutamine repeats." Neuron 22(3): 623-33. Schiffman, S. S. (1993). "Perception of taste and smell in elderly persons." Crit Rev Food Sci Nutr33(l): 17-26. Schiffman, S. S. and C. A . Gatlin (1993). "Clinical physiology of taste and smell." Annu Rev Nutr 13: 405-36. Schwartz Levey, M . , D. M . Chikaraishi, et al. (1991). "Characterization of potential precursor populations in the mouse olfactory epithelium using immunocytochemistry and autoradiography." J Neurosci 11(11): 3556-64. Schwob, J. E., K. E. Szumowski, et al. (1992). "Olfactory sensory neurons are trophically dependent on the olfactory bulb for their prolonged survival." J Neurosci 12(10): 3896-919. Seol, D. W. and T. R. Billiar (1999). "A caspase-9 variant missing the catalytic site is an endogenous inhibitor of apoptosis." J Biol Chem 274(4): 2072-6. Serby, M . , J. Corwin, et al. (1985a). "Olfactory dysfunction in Alzheimer's disease and Parkinson's disease." A m J Psychiatry 142(6): 781-2. Serby, M . , J. Corwin, et al. (1985b). "Olfaction in dementia." J Neurol Neurosurg Psychiatry 48(8): 848-9. Simmons, P. A . and T. V . Getchell (1981). "Neurogenesis in olfactory epithelium: loss and recovery of transepithelial voltage transients following olfactory nerve section." J Neurophysiol 45(3): 516-28. Simmons, P. A. , J. A . Rafols, et al. (1981). "Ultrastructural changes in olfactory receptor neurons following olfactory nerve section." J Comp Neurol 197(2): 237-57. Slee, E. A. , M . T. Harte, et al. (1999). "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 Cell Biol 144(2): 281-92. Sordet, O., C. Rebe, et al. (2002). "Specific involvement of caspases in the differentiation of monocytes into macrophages." Blood 100(13): 4446-53. Springer, J. E., R. D. Azbil l , et al. (1999). "Activation of the caspase-3 apoptotic cascade in traumatic spinal cord." Nat Med 5(8): 943-6. Srinivasan, A. , K . A . Roth, et al. (1998). "In situ immunodetection of activated caspase-3 in apoptotic neurons in the developing nervous system." Cell Death Differ 5(12): 1004-16. Srinivasula, S. M . , M . Ahmad, et al. (1998). "Autoactivation of procaspase-9 by Apaf-1-mediated oligomerization." Mol Cell 1(7): 949-57. Stadelmann, C , T. L . Deckwerth, et al. (1999). "Activation of caspase-3 in single neurons and autophagic granules of granulovacuolar degeneration in Alzheimer's disease. Evidence for apoptotic cell death." A m J Pathol 155(5): 1459-66. Stefanis, L. , R. E. Burke, et al. (1997). "Apoptosis in neurodegenerative disorders." Curr Opin Neurol 10(4): 299-305. Steiner, H. , A . Capell, et al. (1999). "Proteolytic processing and degradation of Alzheimer's disease relevant proteins." Biochem Soc Trans 27(2): 234-42. Steller, H . (1995). "Mechanisms and genes of cellular suicide." Science 267(5203): 1445-9. Steullet, P., H . S. Cate, et al. (2000). "A spatiotemporal wave of turnover and functional maturation of olfactory receptor neurons in the spiny lobster Panulirus argus." J Neurosci 20(9): 3282-94. 133 Su, J. H. , A . J. Anderson, et al. (1994). "Immunohistochemical evidence for apoptosis in Alzheimer's disease." Neuroreport 5(18): 2529-33. Suzuki, Y . and A . L Farbman (2000). "Tumor necrosis factor-alpha-induced apoptosis in olfactory epithelium in vitro: possible roles of caspase 1 (ICE), caspase 2 (ICH-1), and caspase 3 (CPP32)." Exp Neurol 165(1): 35^+5. Suzuki, Y . , J. Schafer, et al. (1995). "Phagocytic cells in the rat olfactory epithelium after bulbectomy." Exp Neurol 136(2): 225-33. Takahashi, K. , K. Kamiya, et al. (2001). "Caspase-3-deficiency induces hyperplasia of supporting cells and degeneration of sensory cells resulting in the hearing loss." Brain Res 894(2): 359-67. Talamo, B. R., R. Rudel, et al. (1989). "Pathological changes in olfactory neurons in patients with Alzheimer's disease." Nature 337(6209): 736-9. Talanian, R. V . , C. Quinlan, et al. (1997). "Substrate specificities of caspase family proteases." J Biol Chem 272(15): 9677-82. Tamatani, M . , S. Ogawa, et al. (1998). "Growth factors prevent changes in Bcl-2 and Bax expression and neuronal apoptosis induced by nitric oxide." Cell Death Differ 5(10): 911-9. Telser, A . , A . I. Farbman, et al. (1982). "A low-molecular-weight soluble protein from bovine lingual epithelium. II. Purification and characterization." J Invest Dermatol 79(5): 286-92. Tenneti, L . , D. M . D'Emilia, et al. (1997). "Suppression of neuronal apoptosis by S-nitrosylation of caspases." Neurosci Lett 236(3): 139-42. Thinakaran, G., C. A . Kitt, et al. (1995). "Distribution of an APP homolog, APLP2, in the mouse olfactory system: a potential role for APLP2 in axogenesis." J Neurosci 15(10): 6314-26. Thomas, L . B., D. J. Gates, et al. (1995). "DNA end labeling (TUNEL) in Huntington's disease and other neuropathological conditions." Exp Neurol 133(2): 265-72. Thornberry, N . A . , T. A . Rano, et al. (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(29): 17907-11. Troy, C. M . and G. S. Salvesen (2002). "Caspases on the brain." J Neurosci Res 69(2): 145-50. Troy, C. M . , J. E. Friedman, et al. (2002). "Mechanisms of p75-mediated death of hippocampal neurons. Role of caspases." J Biol Chem 277(37): 34295-302. Troy, C. M . , L . Stefanis, et al. (1996). "The contrasting roles of ICE family proteases and interleukin-lbeta in apoptosis induced by trophic factor withdrawal and by copper/zinc superoxide dismutase down-regulation." Proc Natl Acad Sci U S A 93(11): 5635-40. Turner, C. P. and J. R. Perez-Polo (1992). "Regulation of the low affinity receptor for nerve growth factor, p75NGFR, in the olfactory system of neonatal and adult rat." Int J Dev Neurosci 10(5): 343-59. Turner, C. P. and J. R. Perez-Polo (1993). "Expression of p75NGFR in the olfactory system following peripheral deafferentation." Neuroreport 4(8): 1023-6. Turner, C. P. and J. R. Perez-Polo (1994). "Changes in expression of the low affinity receptor for neurotrophins, p75NGFR, in the regenerating olfactory system." Int J Dev Neurosci 12(8): 767-73. Vanderluit, J. L. , L . T. McPhail, et al. (2000). "Caspase-3 is activated following axotomy of neonatal facial motoneurons and caspase-3 gene deletion delays axotomy-induced cell death in rodents." Eur J Neurosci 12(10): 3469-80. 134 r Varfolomeev, E. E., M . Schuchmann, et al. (1998). "Targeted disruption of the mouse Caspase 8 gene ablates cell death induction by the TNF receptors, Fas/Apol, and DR3 and is lethal prenatally." Immunity 9(2): 267-76. Vaux, D. L. , G. Haecker, et al. (1994). "An evolutionary perspective on apoptosis." Cell 76(5): 777-9. Villa, P., S. H . Kaufmann, et al. (1997). "Caspases and caspase inhibitors." Trends Biochem Sci 22(10): 388-93. Viswanath, V . , Y . Wu, et al. (2001). "Caspase-9 activation results in downstream caspase-8 activation and bid cleavage in l-methyl-4-phenyl-l,2,3,6-tetrahydropyridine-induced Parkinson's disease." J Neurosci 21(24): 9519-28. Vogt, P. K . and T. J. Bos (1990). "jun: oncogene and transcription factor." Adv Cancer Res 55: 1-35. von Koch, C. S., H . Zheng, et al. (1997). "Generation of APLP2 K O mice and early postnatal lethality in APLP2/APP double K O mice." Neurobiol Aging 18(6): 661-9. Voyron, S., P. Giacobini, et al. (1999). "Apoptosis in the development of the mouse olfactory epithelium." Brain Res Dev Brain Res 115(1): 49-55. Walker, N . P., R. V . Talanian, et al. (1994). "Crystal structure of the cysteine protease interleukin-1 beta-converting enzyme: a (p20/pl0)2 homodimer." Cell 78(2): 343-52. Wang, K . K. , R. Posmantur, et al. (1998). "Simultaneous degradation of alphall- andbetall-spectrin by caspase 3 (CPP32) in apoptotic cells." J Biol Chem 273(35): 22490-7. Wang, S., M . Miura, et al. (1998). "Murine caspase-11, an ICE-interacting protease, is essential for the activation of ICE." Cell 92(4): 501-9. Wang, H . and M . Tessier-Lavigne (1999). "En passant neurotrophic action of an intermediate axonal target in the developing mammalian CNS [see comments]." Nature 401(6755): 765-9. Ward, C. D., W. A . Hess, et al. (1983). "Olfactory impairment in Parkinson's disease." Neurology 33(7): 943-6. Weishaupt, J. H . and M . Bahr (2001). "Degeneration of axotomied retinal ganglion cells as a model for neuronal apoptosis in the central nervous system - molecular death and survival pathways." Restor Neurol Neurosci 19(1-2): 19-27 Wellington, C. L . , L . M . Ellerby, et al. (1998). "Caspase cleavage of gene products associated with triplet expansion." J Biol Chem 273(15): 9158-67. Wong, C. C , W. H. Poon, et al. (2000). "Gene expressions during the development and sexual differentiation of the olfactory bulb in rats." Brain Res Dev Brain Res 119(2): 187-94. Wyllie, A . H. , J. F. Kerr, et al. (1980). "Cell.death: the significance of apoptosis." Int Rev Cvtol 68(1): 251-306. Wyllie, A . H. , R. G. Morris, et al. (1984). "Chromatin cleavage in apoptosis: association with condensed chromatin morphology and dependence on macromolecular synthesis." J Pathol 142(1): 67-77. Xue, D., S. Shaham, et al. (1996). "The Caenorhabditis elegans cell-death protein CED-3 is a cysteine protease with substrate specificities similar to those of the human CPP32 protease." Genes Dev 10(9): 1073-83. Yakovlev, A . G., S. M . Knoblach, et al. (1997). "Activation of CPP32-like caspases contributes to neuronal apoptosis and neurological dysfunction after traumatic brain injury." J Neurosci 17(19): 7415-24. Yan, X . X . , J. Najbauer, et al. (2001). "Expression of active caspase-3 in mitotic and postmitotic cells of the rat forebrain." J Comp Neurol 433(1): 4-22. 135 Yang, X . , H . Y . Chang, et al. (1998). "Autoproteolytic activation of pro-caspases by oligomerization." Mol Cell 1(2): 319-25. Yuan, J., S. Shaham, et al. (1993). "The C. elegans cell death gene ced-3 encodes a protein similar to mammalian interleukin-1 beta-converting enzyme." Cell 75(4): 641-52. Zheng, T. S., S. Hunot, et al. (2000). "Deficiency in caspase-9 or caspase-3 induces compensatory caspase activation." Nat Med 6(11): 1241-7. 136 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0091148/manifest

Comment

Related Items