Modulation of N-Methyl-D-Asparate Receptor by Transient Receptor Potential Melastatin Type-2 Regulates Neuronal Vulnerability to Ischemic Cell Death

Ishraq Alim

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Physiology University of Toronto

Ishraq Alim

Doctor of Philosophy

Department of Physiology University of Toronto

2014 Abstract

Neuronal vulnerability to ischemia is dependent on the balance between pro-survival and pro-death cellular signaling. In the latter, it is increasingly appreciated that toxic Ca2+ influx can occur not only via postsynaptic glutamate receptors, but also through other cation conductances.

One such conductance, the Transient receptor potential melastatin type-2 (TRPM2) channel, is a non-specific cation channel having similar homology to TRPM7, a conductance reported to play a key role in anoxic neuronal death. The role of TRPM2 conductances in ischemic Ca2+ influx has been difficult to study due to the lack of specific modulators. Here we used TRPM2-null mice (TRPM2(-/-)) to study how TRPM2 may modulate neuronal vulnerability to ischemia.

TRPM2(-/-) mice subjected to transient middle cerebral artery occlusion (tMCAO) exhibited smaller infarcts when compared to wild-type (WT) animals, suggesting the absence of TRPM2 to be protective. Surprisingly, field potentials (fEPSPs) recorded during oxidative stress in brain slices taken from TRPM2(-/-) mice revealed increased excitability, a phenomenon normally associated with ischemic vulnerability, whereas WT fEPSPs were unaffected. The upregulation in fEPSP in TRPM2(-/-) neurons was blocked selectively by an NR2A antagonist. This oxidative stress-induced increase in excitability of TRPM2(-/-) fEPSPs depended on the upregulation and downregulation of NR2A and NR2B-containing NMDARs, respectively, and augmented pro-

ii survival signaling via Akt and ERK pathways culminating in the inhibition of the proapoptotic factor, GSK3β. Cultured hippocampal neurons from TRPM2(-/-) animals subjected to oxygen glucose deprivation had a reduction in cell death in comparison to WT neurons, demonstrating that absence of TRPM2 is protective at the neuronal level in vitro. Our results suggest that

TRPM2 plays a role in downregulating pro-survival signals in central neurons and that TRPM2 channels may comprise a therapeutic target for preventing ischemic damage.

iii

Acknowledgments

I would first like to thank my supervisor, Dr. Michael Tymianski, for helping me through this project and giving me the opportunity to work in his lab. He has been an inspiration for my scientific career and has provided me with the skills and knowledge to be a scientist. I am forever grateful for the patience and time he invested in my work. In addition I would also like to thank my PhD Advisory Committee: Dr. James Eubanks, Dr. John MacDonald and Dr. Peter Carlen, for all their critique and insightful advice on my project. Special thanks to Dr. Eubanks, whose door has always been open to me whenever I have needed advice. I would also like to thank the entire lab for helping me and giving me advice whenever I needed it (in no particular order): Lucy Teves, Kinga Szydlowska, Xuijun Sun, Rongwen Li, Hong Cui, Hong Sun, Jane Wu, Carlo Santaguida, Andrew Bersic, Zhanxin Ji, Sandra Vetiska, Junbo Huang and Douglas Cook. Special thanks to Lucy Teves who kept me focused on my research and always questioned my work. Also to Kinga Szydlowska, Xuijun Sun, Sandra Vetiska, Junbo Huang, Rongwen Li, Hong Cui, and Zhanxin Ji who were always there to provide technical help whenever experiments weren't going well. Finally, I would like to thank my parents, my sisters and my wife for always supporting my career in science and for believing in me throughout this process.

Who did what:

Experiments were designed by Ishraq Alim and Michael Tymianski

MCAO experiments and analysis were conducted by Lucy Teves with assistance from Ishraq Alim

Immunoflouresence was done by Rongwen Li

Electophysiology, cell cultures, Western blots and most data analysis were conducted by Ishraq Alim

Table of Contents

Acknowledgments ...... iv

Table of Contents ...... v

List of Tables ...... x

List of Figures ...... xi

List of Appendices ...... xiii

Chapter 1 ...... 1

1 Introduction ...... 1

1.1 Overview of stroke ...... 1

1.1.1 Prevalence of stroke ...... 1

1.1.2 Current therapies for stroke ...... 2

1.1.3 Other neurodegenerative disorders ...... 3

1.1.4 Animal models of ischemic stroke ...... 4

1.2 Mechanism of excitotoxic cell death during ischemia ...... 4

1.2.1 Excitotoxicity ...... 5

1.2.2 Reperfusion injury following ischemia ...... 6

1.3 Ion channels involved in ischemia ...... 6

1.3.1 Glutamate receptors ...... 7

1.3.2 Non-glutamate driven cation channels ...... 12

1.4 Ca2+-induced downstream signaling and apoptosis ...... 16

1.4.1 Ca2+-mediated pro-survival pathways ...... 16

1.4.2 Ca2+-mediated pro-death pathways ...... 17

1.4.3 Caspase-mediated cell death ...... 17

1.5 Free radicals and oxidative stress ...... 21

1.6 TRPM channels and stroke ...... 23

1.6.1 Role of TRPM7 in ischemia ...... 23

1.6.2 TRPM2, a potential novel channel involved in ischemia? ...... 25

Chapter 2 ...... 30

2 Aims and Hypotheses ...... 30

2.1 Background and rationale ...... 30

2.2 Specific aims and hypotheses ...... 31

2.2.1 Hypothesis 1 ...... 31

2.2.2 Hypothesis 2 ...... 32

2.2.3 Hypothesis 3 ...... 32

2.2.4 Hypothesis 4 ...... 33

Chapter 3 ...... 34

3 Methods ...... 34

3.1 Animals and genotyping ...... 34

3.2 In vivo stroke model ...... 35

3.3 Electrophysiology ...... 36

3.4 Western blot ...... 39

3.5 Sex determination ...... 39

3.6 Cell cultures and in vitro OGD ...... 40

3.7 Immunofluorescence for quantification of NR2 subunit expression ...... 41

3.8 Statistical analysis ...... 42

Chapter 4 ...... 43

4 The role of TRPM2 in MCAO-induced ischemic damage ...... 43

4.1 Results for Aim 1 ...... 44

4.1.1 The absence of TRPM2 reduces infarct volume following transient but not permanent cerebral ischemia...... 44

4.1.2 Measuring blood flow to confirm occlusion in WT and TRPM2(-/-) mice ...... 48

4.1.3 Body temperature, and weight remain consistent between WT and TRPM2(-/-) animals following a MCAO ...... 53

4.1.4 WT and TRPM2(-/-) animals exhibit loss of motor function following stroke...... 58

4.2 Discussion ...... 61

4.3 Summary ...... 62

Chapter 5 ...... 63

5 The role of TRPM2 in synaptic activity during oxidative stress ...... 63

5.1 Results for Hypothesis 2 ...... 64

5.1.1 TRPM2(-/-) increases synaptic excitability during oxidative stress ...... 64

5.1.2 TRPM2(-/-) has no effect on LTP inhibition during oxidative stress conditions ... 66

5.1.3 Paired-pulse facilitation suggests oxidative stress-induced excitability in TRPM2(-/-)is mediated by changes in the post-synapse...... 70

5.2 Discussion ...... 72

5.3 Summary ...... 73

Chapter 6 ...... 74

6 TRPM2 modulates NMDAR-dependent cell death mechanisms ...... 74

6.1 Results for Hypothesis 3 ...... 75

6.1.1 The absence of TRPM2 alters NMDAR NR2 subunit expression ...... 75

(-/-) 6.1.2 NR2A is involved in H2O2 induced synaptic excitability in TRPM2 ...... 81

6.1.3 TRPM2(-/-) increases Akt and ERK Activity...... 86

6.1.4 TRPM2(-/-) inhibits GSK3β and PSD95, but has no effect on TRKβ ...... 86

6.1.5 Absence of TRPM2 does not alter TRPM7 expression ...... 90

6.2 Discussion ...... 91

6.3 Summary ...... 92

Chapter 7 ...... 93

7 Comparing neuroprotective properties of TRPM2(-/-) to known neuroprotective drugs ...... 93

7.1 Results for Hypothesis 4 ...... 94 vii

7.1.1 Absence of TRPM2 is neuroprotective in hippocampal neurons following OGD ...... 94

7.1.2 Neuroprotective agents do not enhance neuroprotective effects in TRPM2-null hippocampal neurons ...... 94

7.2 Discussion ...... 98

7.3 Summary ...... 99

Chapter 8 ...... 100

8 Discussion ...... 100

8.1 Proposed mechanism of TRPM2(-/-) and neuroprotection ...... 100

8.2 TRPM2(-/-) and neuroprotection ...... 102

8.2.1 TRPM2(-/-) as a model for in vitro neuroprotection ...... 102

8.2.2 Translating TRPM2 from in vitro to in vivo stroke model ...... 103

8.2.3 Reperfusion damage required for TRPM2-mediated damage ...... 105

8.3 TRPM2 in synaptic activity ...... 106

8.3.1 TRPM2 and synaptic plasticity ...... 107

8.3.2 Possible mechanisms of TRPM2(-/-) modulation of synaptic transmission ...... 108

8.4 TRPM2(-/-) modulates glutamate receptors ...... 109

8.4.1 Possible mechanisms of how TRPM2(-/-) modulates NMDAR ...... 111

8.4.2 Potential role of TRPM2 in NR2A-NR2B developmental switch ...... 112

8.4.3 NMDAR mediated survival/death pathways ...... 112

8.5 Alternate mechanisms of neuroprotection ...... 113

8.5.1 TRPM2 mediated Ca2+-overload ...... 113

8.5.2 Microglia and immune response ...... 114

8.5.3 Gender and neuroprotection ...... 115

8.5.4 Other channels modulated by TRPM2(-/-) ...... 115

8.6 TRPM2 as a therapeutic target ...... 116

8.6.1 Combined inhibition of TRPM2 and other therapeutic targets ...... 116 viii

8.6.2 TRPM2 in other neurodegenerative diseases ...... 117

8.7 Concluding remarks ...... 118

Chapter 9 ...... 119

9 Future Directions ...... 119

9.1 Resolving mechanism of how TRPM2(-/-) modulates NMDAR ...... 119

9.2 Development of acute and specific TRPM2 inhibitor ...... 120

References Cited ...... 121

Appendices ...... 139

1 Appendix 1: TRPM2(-/-) is in cortical neurons following OGD ...... 139

1.1 Introduction ...... 139

1.2 Methods ...... 139

1.2.1 Cell culture ...... 139

1.2.2 Western blot ...... 139

1.2.3 Statistical analysis ...... 140

1.3 Results ...... 140

1.3.1 Absence of TRPM2 reduces neuronal death in cortical neurons following 1h OGD ...... 140

1.3.2 Neuroprotective agents do not have an additive effect on neuroprotective properties of cortical neurons from TRPM2-null mice ...... 140

1.3.3 Cortical neurons from TRPM2(-/-) mice promote activation of pro-survival pathways via modulation of NR2 subunit expression ...... 140

1.4 Discussion ...... 141

1.5 Appendix References ...... 145

List of Tables

Table 1: Primers for RTPCR detection of TRPM2(-/-) ...... 35

Table 2: PCR primers for Sex Determination on Chromosome Y ...... 40

List of Figures

Figure 1: Synaptic and extrasynaptic mechanisms of cell survival and death ...... 20

Figure 2 Diagram of TRPM2 structure ...... 27

Figure 3 Development of TRPM2(-/-) mice ...... 28

Figure 4 Coronal sections and template for infarct volume ...... 38

Figure 5 TRPM2(-/-) mice have reduced infarct volume compared to WT following 48h transient MCAO ...... 47

Figure 6: Visual reduction of blood flow following MCAO ...... 50

Figure 7: Laser Doppler measurement of cerebral blood flow during MCAO ...... 52

Figure 8: Consistent body temperature is maintained during MCAO ...... 55

Figure 9: pMCAO and tMCAO reduces body weight in WT and TRPM2(-/-) animals ...... 57

Figure 10: Neurological score demonstrates reduction in motor function post-MCAO ...... 60

(-/-) Figure 11: 200uM of H2O2 increases evoked baseline fEPSP slope in TRPM2 with no effect on WT ...... 65

(-/-) Figure 12: 200 uM of H2O2 inhibits LTP formation in both WT and TRPM2 ...... 68

Figure 13 Removal of H2O2 following LTP induction returns fEPSP to original baseline in TRPM2(-/-) hippocampal slices ...... 69

(-/-) Figure 14: Sub lethal H2O2 has no effect on paired pulse facilitation in WT and TRPM2 hippocampus ...... 71

Figure 15: Hippocampi from TRPM2(-/-) mice have has reduced NR2B expression and increased NR2A expression...... 77

Figure 16: Absence of TRPM2 reduces NR2B-subunit expression and increases NR2A subunit expression ...... 78

Figure 17: Absence of TRPM2 has no effect on synaptic co-localization of NR2A subunit ...... 79

Figure 18: TRPM2(-/-) reduces synaptic co-localization of NR2B ...... 80

Figure 19: NR2A modulates oxidative stress induced excitability in TRPM2(-/-) hippocampal neurons ...... 83

Figure 20: NR2B has no effect on oxidative stress induced excitability in TRPM2(-/-) hippocampal neurons ...... 85

Figure 21: TRPM2(-/-) increases activity in ERK and AKT pathways ...... 88

Figure 22: Absence of TRPM2 reduces PSD95 expression but has no effect on BDNF signaling ...... 89

Figure 23: Expression of TRPM2 has no effect on TRPM7 expression ...... 90

Figure 24: Cultured hippocampal neurons from TRPM2(-/-) mice have reduced OGD-induced cell death in comparison to WT ...... 96

Figure 25: Neuroprotectants do not enhance neuroprotective effects of absence of TRPM2 for up to 1.5h OGD ...... 97

Figure 26: Proposed mechanism cell death involving TRPM2 ...... 101

Figure 27: OGD in cortical neurons from WT and TRPM2(-/-) mice ...... 142

Figure 28: Neuroprotective agents have similar level of protection in cortical neurons from WT and TRPM2(-/-) mice ...... 143

Figure 29: Absence of TRPM2 promotes pro-survival pathway in cortical neurons ...... 144

xii

List of Appendices

Appendices ...... 139

1 Appendix 1: TRPM2(-/-) is in cortical neurons following OGD ...... 139

1.1 Introduction ...... 139

1.2 Methods ...... 139

1.2.1 Cell culture ...... 139

1.2.2 Western blot ...... 139

1.2.3 Statistical analysis ...... 140

1.3 Results ...... 140

1.3.1 Absence of TRPM2 reduces neuronal death in cortical neurons following 1h OGD ...... 140

1.3.2 Neuroprotective agents do not have an additive effect on neuroprotective properties of cortical neurons from TRPM2-null mice ...... 140

1.3.3 Cortical neurons from TRPM2(-/-) mice promote activation of pro-survival pathways via modulation of NR2 subunit expression ...... 140

1.4 Discussion ...... 141

1.5 Appendix References ...... 145

xiii 1

Chapter 1

1 Introduction

During a stroke, a number of interdependent mechanisms come into play which ultimately leads to cell death. Elucidating therapeutic targets in cell death pathways have been difficult, since many targets are required for normal cellular function and due to the lack of specific antagonists. Transient receptor potential melastatin 2 (TRPM2), is a channel that has recently been suggested to be involved in stroke, since TRPM2 mRNA is increased following a stroke and it mediates oxidative stress-induced calcium influx (Olah et al., 2009; Fonfria et al., 2006; Yamamoto et al., 2008; Aarts et al., 2003). The project described in this thesis initially intended to determine if TRPM2 is involved in ischemic neuronal death and the role this protein plays in regulating cell death mechanisms. In the introduction, we will discuss the current understanding of mechanisms involved in ischemia and why TRPM2 has been suggested to be involved in neuronal death.

1.1 Overview of stroke

1.1.1 Prevalence of stroke

Some of the earliest recoded cases of stroke cases go back to the 2nd millennium BC in ancient Persia and Mesopotamia. The first person to describe the symptoms was Hippocrates (460-370 BC), who identified sudden paralysis with stroke and called it apoplexy, meaning “struck down by violence” (Paciaroni and Bogousslavsky, 2009). Our current understanding of ischemia describes it as being a rapid loss of blood supply to regions of the brain. The term "stroke" can be divided into three categories: brain ischemia (blockage), subarachnoid hemorrhage (bleeding in the subarachnoid space) and intracerebral hemorrhage (bleeding inside the brain tissue). The brain ischemia category can be further subdivided into thrombosis (development of blood clot), embolism (lodging of embolus) and hypoperfusion (sudden drop of blood pressure/flow). This loss of blood rapidly causes cell death and ultimately leads to brain dysfunction (De Freitas et al., 2009). Stroke afflicts approximately 660,000 people and causes

175,000 deaths every year in North America. Risk factors of stroke include old age, high blood pressure, previous stroke, diabetes, obesity, smoking and high cholesterol (American Stroke Association Web Site, 2000). Due to the prevalence, and high rate of mortality, studies have focused on understanding the mechanism of cell death during ischemia.

1.1.2 Current therapies for stroke

1.1.2.1 Preventing stroke

Currently, most of the therapeutic focus on stroke has revolved around preventative measures. Hypertension is the leading cause of stroke and accounts for 30-50% of stroke risk (Thom et al., 2006). Anti-hypertensive therapy is often used to prevent stroke in patients with high blood pressure (Ahmed et al., 2009). Other factors that can contribute to stroke are atrial fibrillation, blood lipids and diabetes, all of which are often treated to reduce the risk of stroke (Gorelick et al., 1999).

Beyond treatment of pre-existing conditions, stroke can be prevented by changes in lifestyle. These modifiable lifestyle factors include smoking, excessive alcohol consumption, obesity, lack of physical activity, processed red meat consumption and a high fat diet. If allowed to continue unchecked, these lifestyle choices can lead to hypertension and other health issues that lead to stroke (Gorelick et al., 1999). The effectiveness of lifestyle changes in the reduction of stroke is not well understood, however, circumstantial evidence has shown that drug therapies treating preexisting conditions (such as hypertension) in combination with healthier lifestyle results in a decreased risk of stroke (Galimanis et al., 2009).

1.1.2.2 Treatment of stroke

Since damage caused by a stroke is almost immediate, there are very few treatments that prevent cell death following onset. Treatment for acute ischemic stroke due to blockage usually involves the breakdown of clots. Thrombolysis involves the breakdown of a clot using recombinant tissue plasminogen activator (rtPA), a serine protease that breaks down plasminogen to plasmin. It is usually given within 3 hours of onset of a stroke, and increases the chance of survival by 9%, however there are some complications as rtPA can cause hemorrhage (Wardlaw et al., 2012). When clotting occurs in a larger artery, a thrombectomy is often required, where the clot is physically removed via surgery (Benmira et al., 2011). In the case of

3 hemorrhagic stroke, the patient usually needs to be evaluated to determine the cause of the bleeding and whether or not surgery will be helpful in blocking it; occasionally clotting factors are given to the patient to reduce the bleeding (Vergouwen et al., 2008).

Following the onset of stroke, there are very few treatments that prevent neuronal death, and if the patient survives, most of the treatments are focused on rehabilitation. This often requires some form of physical therapy, speech language therapy, deep brain stimulation and/or occupational therapy in order to regain some normal function (Sivenius et al., 1985). Due to the almost immediate onset stroke mediated damage and the lack of post-ischemic treatments, there has been a focus on understanding the mechanism of neuronal death following a stroke, in hopes to develop neuroprotective treatments (Besancon et al., 2008; Szydlowska and Tymianski, 2010; Tymianski, 2011).

1.1.3 Other neurodegenerative disorders

A number of chronic neurodegenerative diseases, such as Alzheimer’s Disease and Parkinson’s Disease, undergo similar cell death mechanisms as stroke. Both Alzheimer's and Parkinson's disease afflict approximately 26.6 (Brookmeyer et al., 2007) and 6.3 (Baker and Graham, 2004) million people worldwide, respectively. Currently, there is an emerging appreciation of the relationship between ischemic vascular dementia and Alzheimer’s disease, where both pathologies share common risk factors and often occur concurrently, suggesting both may have an additive effect during neurodegeneration (O'Brien et al., 2003). Also, oxidative stress due to an imbalance of reactive oxygen species (ROS) has been suggested to play a role in neuronal death in acute conditions such as stroke and chronic conditions such as Alzheimer’s (Benzi and Moretti, 1995) and Parkinson’s disease (Jenner et al., 1992). However, the extent to which oxidative damage causes neuronal death is not fully understood for chronic conditions (Halliwell and Gutteridge, 1990; Knight, 1998; Lynch et al., 2000). Understanding the mechanisms of cell death following a stroke may shed light on cell death mechanisms in chronic neurodegenerative disorders, and this, in turn, may provide neuroprotective targets for a wide range of disorders in the brain.

1.1.4 Animal models of ischemic stroke

The development of new therapies for treating ischemic stroke has always needed in vivo animal models that mimic ischemia in humans. Animal models are usually divided into two categories: global ischemia and focal ischemia. Global ischemia is intended to imitate complete loss of blood flow to the brain, often observed during cardiac arrest or some form of asphyxiation. Rodent global ischemia models include: cardiac arrest, two or three vessel occlusion (depending on whether the strain has a complete circle of Willis), aorta/vena cava cuff and hypoxia models (Ginsberg et al., 1989; Yonekura et al., 2004). Related to global ischemia, some hemorrhagic models have incomplete global ischemia where there is a loss of blood flow to a large section of the brain, but not complete loss in the entire brain.

Focal ischemia models are intended to mimic localized infarct, often caused by localized blockage of blood vessels in the brain. The two most popular focal ischemia models in rodents are endothilin-1 induced constriction and middle cerebral artery occlusion (MCAO). Endothilin- 1 is a potent vasoconstrictor that can be microinjected into a region of the brain causing blood vessels to constrict creating ischemic conditions and eventually reperfusion. The MCAO models can be divided into those that require craniotomy and those that don't. The models that don't require craniotomy are embolic MCAO, which injects particles (blood clots) into the carotid artery and endovascular filament MCAO, which inserts a surgical filament into the internal carotid artery and makes its way to the MCA (Charmichael, 2005). MCAO that requires craniotomy, usually involves a ligation or electrocautery of the MCA creating a permanent ischemia model (Charmichael, 2005).

1.2 Mechanism of excitotoxic cell death during ischemia

Within minutes of a stroke, the lack of blood reaching regions of the brain results in the modulation of a number interdependent biochemical reactions which are involved in neuronal death. The damage due to ischemia is usually secondary to the initial stroke injury and continues after post-stroke reperfusion. The initial lack of oxygen causes a disruption in normal ATP production, resulting in neurons switching from aerobic to anaerobic metabolism. The lack of ATP causes ATP-dependent channels to fail and leads to cell depolarization via the influx of cations, particularly Calcium (Ca2+) (Kristian and Siesjo, 1998). This Ca2+-influx triggers a massive release of excitatory neurotransmitters, specifically glutamate, which, in turn, stimulates

5 activation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPAR) and N-methyl-D-aspartate glutamate receptors (NMDAR), and gates more Ca2+ into the neuron (Besancon et al., 2008). This Ca2+-influx via glutamate receptors, triggers further glutamate neurotransmitter release and activates glutamate receptors in neighboring neurons (Ginsberg and Busto, 1998).

Excessive amounts of Ca2+ produce a number of chemicals involved in cell death, including free radicals and calcium dependent enzymes (ATPases, calpain, and phospholipases) (Tymianski and Tator, 1996; Siesjo et al., 1995; Kristian and Siesjo, 1998). Phospholipases trigger the breakdown of cell membranes, while ROS and free radicals damage DNA and cause mitochondrial dysfunction (Siesjo et al., 1995). Damage to the mitochondria leads to a loss of mitochondrial membrane potential, production of cytochrome-c and the production of ROS. Cytochrome-C activates caspase-9/3 dependent apoptotic pathways, which leads to DNA fragmentation and ultimately leads to apoptotic cell death (Garrido et al., 2006). Alternatively, many neurons that do not die via apoptosis die prematurely via necrosis, where the damage to the cell membrane is so severe that it forms varicosities (blebs). These blebs fuse and become larger, until the cell membrane ruptures and all the cellular content is released into the extracellular space (Bindokas and Miller, 1995). Intracellular glutamate and ROS are also released following necrosis, which causes further damage to surrounding cells (Kristian and Siesjo, 1998). The major difference between apoptosis and necrosis, is that apoptosis is natural programmed cell death, while necrosis is a direct response to severe cell damage.

1.2.1 Excitotoxicity

Glutamate is a fundamental excitatory neurotransmitter that is found in the mammalian nervous system and is responsible for synaptic communication and plasticity. Since the 1950s, studies have shown that excessive release of glutamate is associated with Ca2+- overload during a stroke. The first study to demonstrate this was by Lucas and Newhouse (1957), where injection of L-glutamate into mice retina caused retinal degeneration. John Olney (1969) later discovered that this phenomenon was not specific to the retina but also occurred in the brain. His study was the first to coin the term “excitotoxicity” to describe this phenomenon. This term aptly describes the phenomenon of excessive excitatory neurotransmitter release, which causes neurotoxicity.

1.2.2 Reperfusion injury following ischemia

Following ischemia, there is often a reperfusion of blood to the ischemic regions either via the removal of the blockage, blood supply from nearby regions and/or contralateral blood flow. This influx of oxygen increases the ROS production in already damaged cells, which results in further ROS-mediated damage (Halliwell, 2006a). Also, ROS production is used in immune response, where phagocytes produce ROS to further breakdown damaged neurons for consumption (Halliwell, 2006b). In addition to the neuronal damage caused by ischemia, reperfusion can also mediate ROS-induced breakdown in tight endothelial junctions forming the blood brain barrier (BBB) (Brown and Davis, 2002). This damage to the BBB changes the composition of the cerebral spinal fluid that normally surround neurons and allows for toxins and other harmful substances in the blood to come into contact with neurons. Disruption in BBB also allows the entry of large molecules, like albumin, into the cerebral parenchymal extracellular space, this causes water to enter the space via osmosis causing cerebral edema (Brown and Davis, 2002). Once these intravascular proteins can pass the BBB, they begin to move quickly and spread in the brain, resulting in wide spread edema (Brown and Davis, 2002).

In both ischemic and reperfusion conditions there are a number of interdependent events that lead to neuronal damage and ultimately cell death. Here, for brevity, we have given an overview of the major events that occur following a stroke, and have not gone in depth in fully describing ischemia-induced cell death. Since many of the events that occur after a stroke are interconnected, there has been a great deal of interest in finding targets and developing neuroprotective drugs that will halt initial ischemic events in cells, which in turn prevent further downstream damage (Tymianski, 2011). One of the earliest events in ischemia is a large influx of Ca2+ into the neuron. This influx is required for both apoptosis and the propagation of ischemic cell death, and thus has led many to suggest that Ca2+-overload could be a therapeutic target for neuroprotection (Besancon et al., 2008; Szydlowska and Tymianski, 2010; Tymianski, 2011).

1.3 Ion channels involved in ischemia

For almost 40 years calcium has been known to be a key signaling molecule involved in the early stages of neuronal cell death (Schlaepfer and Bunge, 1973; Schanne et al., 1979; Choi, 1985). Since Ca2+ is required for normal cellular function and cell survival (Penn and

Loewenstein, 1966), it has become complicated to try and differentiate the roles of Ca2+ in normal cellular signaling and cell death. Initially, many thought Ca2+ levels had to surpass a particular threshold to induce cell death, while concentrations below that threshold were used for other cellular functions (Manev et al., 1989; Marcoux et al., 1990). This hypothesis, assumed that neuronal death was dependent on the quantity of calcium influx and did not consider other factors such as the source of the influx. Another hypothesis, termed "source specific hypothesis" postulates that Ca2+-mediated cell survival and death was modulated by the route of Ca2+ entry, instead of concentration. This hypothesis suggests that specific channels were responsible for activation of distinct secondary messenger pathways responsible certain cellular functions. Through the use of channel blockers, it became evident that certain ion channels were involved in the Ca2+-influx involved in neurodegeneration (Tymianski et al., 1993; Sattler et al., 1998). By focusing on distinct Ca2+-influx pathways it has become easier to differentiate Ca2+-mediated cell death and survival pathways; and have provided potential therapeutic targets that would prevent ischemic death, without interfering in normal cellular function (Tymianski et al., 1993; Sattler et al., 1998).

The overload of Ca2+ during ischemic cell death is primarily mediated by two specific types of cation channels: glutamate channels and non-glutamate driven channels (Choi et al., 1987; Aarts et al., 2003; Szydlowska and Tymianski, 2010). To a lesser extent, calcium can also come from internal calcium stores, in the mitochondria or endoplasmic reticulum, through either damage to their membranes or via ion channels on the surface of these organelles (Besancon et al., 2008; Chong et al., 2005). For the purposes of this thesis, this section will discuss some of the key routes of extracellular Ca2+-influx following ischemia and how blocking these channels contribute to neuroprotection.

1.3.1 Glutamate receptors

Glutamate, a major excitatory neurotransmitter, has been suggested to play a role in Ca2+ influx during a stroke since the 1950s (Lucas and Newhouse, 1957; Olney and De Gubareff, 1978). Glutamate receptors are predominantly found in the post-synapse (although occasionally at the pre-synapse as well) and are divided into two categories: ionotropic and metabotropic. Ionotropic receptors have an ion channel pore attached to the receptor, and include AMPAR, NMDAR and Kainate Receptor (Li and Stys, 2000). Metabotropic receptors are G-coupled

8 protein receptors, which activate ion channels and other cellular functions indirectly via g- protein cascade (Bruno et al., 2001; Rothstein et al., 2005). Two ionotropic glutamate-receptors, NMDAR and AMPAR, are associated with excitotoxicity and are often the target for neuroprotection (Besancon et al., 2008; Sattler et al., 1999; Aarts et al., 2002).

1.3.1.1 AMPA receptors

AMPARs are predominantly involved in fast synaptic transmission, in response to glutamate. They are made of tetraheteromeric structures made up of a combination of GluR1- GluR2 (A-D) subunits (Keinanen et al., 1990). AMPARs are predominantly sodium (Na+) permeable, with the exception of AMPARs without GluR2 subunit which are Ca2+ permeable (Pellegrini-Giampietro et al., 1997). Initially, it was thought that following ischemia the release of glutamate activates AMPARs, leading to a large influx of sodium. This build up of sodium causes the removal of Mg2+ block at NMDARs, and thus excess extracellular glutamate can now bind to NMDARs causing Ca2+ overload (Pellegrini-Giampietro et al., 1997; Iihara et al., 2001). It has been reported that following ischemia, AMPARs permeable to calcium are sharply increased in hippocampal neurons (Pellegrini-Giampietro et al., 1997). Also, under ischemic conditions GluR2 mRNA is reduced, thus indicating that ischemia causes a change in AMPAR subunitss, specifically increasing expression of Ca2+ permeable and decreasing expression of Na+ permeable AMPARs. This switch is described as the “GluR2a hypothesis”, suggesting that increase in Ca2+ permeable AMPARs expression due to ischemia contributes to delayed Ca2+ influx causing neurodegeneration (Pellegrini-Giampietro et al., 1997; Iihara et al., 2001).

Although AMPARs have been suggested as a potential target for the treatment of ischemia, unfortunately, therapies blocking AMPARs have been unsuccessful in human trials and due to its involvement in normal synaptic communication, there is the potential for harmful side effects (Weiser, 2005). Also, since the role of AMPARs in stroke is not as well studied compared to NMDARs, most therapeutic strategies have opted to focus on NMDARs rather than AMPARs.

1.3.1.2 NMDA receptors

The most studied glutamate receptor involved in ischemia is NMDARs. It is a heterotetramer comprised of 3 subfamilies of subunits: NR1, NR2 (A-D) and NR3 (Monyer et

9 al., 1992). NR1 is expressed in all NMDARs, and most functional NMDARs contain NR2 and some cases NR3 (Moriyoshi et al., 1991; Ciabarra et al., 1995). It functions as a slow-gating channel and is highly permeable to both Ca2+ and Na+. The channel component is responsible for a number downstream Ca2+-dependent NMDA signaling pathways (Aarts and Tymianski, 2004). This includes the activation of neuronal nitric oxide synthase (nNOS), which is modulated by NMDAR-mediated Ca2+ influx and plays an important role in promoting free radical production during oxidative stress (Sattler et al., 1999).

The C-terminus of NMDARs interacts with a number of intracellular synaptic proteins and cytoskeletal proteins. The multi protein complex that is created through the interaction of the C-terminal is known as the post-synaptic density (PSD; Cho et al., 1992). The PSD complex, bound to NMDARs, is composed of primarily cytoskeletal, scaffolding proteins, and enzymes. Cytoskeletal proteins such as actin, fodrin and tubulin play a role in clustering and localizing the PSD complex to specific locations (Qualmann et al., 2004). Scaffolding proteins work as transporters, bringing signaling and other proteins into the PSD complex. Intracellular signaling from the PSD often requires phosphorylation, which is carried out by a number of kinases, including the SRC family of non-receptor protein tyrosine kinases, specifically calcium/calmodulin kinase II (CaMKII), protein kinase C (PKC), ERK2-type mitogen activated kinase and calcineurin (Zukin and Bennett, 1995; Arundine and Tymianski, 2004; Husi et al., 2000). Both the Ca2+ influx and the PSD mediates a number of mechanisms in neurons, including synaptic signaling, plasticity, cell survival and cell death (Ikonomidou and Turski, 2002; Hardingham and Bading, 2003; Papadia and Hardingham, 2007; Yao et al., 2004).

In neurons, NMDARs are the most efficient pathway for producing Ca2+ influx following ischemia (Arundine and Tymianski, 2004; Tymianski et al., 1993). However, translating neuroprotective effects of NMDAR antagonists in animal models to stroke clinical trials have thus far been unsuccessful, and many have suggested that it is because NMDAR is also involved in other mechanisms that are not involved in cell death (Davis et al., 1997; Ikonomidou and Turski, 2002; Lees et al., 2000). Since general NMDA blockers were unable to protect neurons in humans, there was a shift in focus towards specific types of NMDARs. Two candidates have emerged as being involved in cell survival and death: NR2A containing NMDARs and NR2B containing NMDARs (Liu et al., 2007b; Wyllie et al, 2013).

Activation of NR2A containing NMDARs, which only contain NR2A and NR1 subunits, has been shown to be involved in cell survival and synaptic excitation (Liu et al., 2007b; Hardingham and Bading, 2010; Terasaki et al., 2010). Low concentrations of NMDA are known to precondition neurons for cell survival following an insult. During preconditioning there is an increase in synaptic excitability, a phenomenon that is associated with cell survival (Soriano et al., 2006). Inhibition of synaptic excitability makes neurons more susceptible to insult and neuronal death. This increase in synaptic excitability has been associated with activation of NR2A containing NMDARs, where blocking NR2A containing NMDARs results in a decrease in synaptic excitability and a loss of preconditioned neuroprotection (Liu et al., 2007b; Wyllie et al, 2013). Activation of NR2A is also associated with a number of downstream pro-survival pathways, specifically the phosphorylation of extracellular-signal-regulated kinases (ERK1/2) and protein kinase B (Akt) pathways, which inhibits molecules involved in cellular death (Liu et al., 2007b; Wyllie et al, 2013; Terasaki et al., 2010).

NR2B containing NMDARs have the opposite effects of those containing NR2A, where it has been shown to be involved in cell death and inhibition of cellular excitability (Soriano et al., 2008; Sattler et al., 1999). The PSD complex of NR2B contains a number of signaling proteins including: PSD-95/SAP90, chapsyn-110/PSD93 and other members of the membrane- associated guanylate kinase (MAGUK) family (Kornau et al., 1995; Brenman et al., 1996; Muller et al., 1996). PSD-95, a scaffolding protein, is the most studied member of the MAGUK family of proteins. It is exclusively located at the PSD, contains 3 PDZ domains and is involved in signaling and anchoring NMDARs (Chung et al., 2004; Hunt et al., 1996; Sattler et al., 1999; Sheng and Sala, 2001). Our lab has shown that PSD-95’s interaction with the NR2B containing NMDA receptors is required for the lethal influx of Ca2+ during excitotoxicity (Sattler et al., 1999; Aarts et al., 2002). Anti-sense suppression of PSD-95 was shown to inhibit NMDAR- mediated cell death (Sattler et al., 1999). Since mutation and mRNA suppression is therapeutically not ideal, a PSD-95 peptide inhibitor that encodes the sequence of the NR2B C- terminus binding site was developed (Aarts et al., 2002). This peptide, known as NA-1, was shown to bind to PSD95 and thus disrupting normal NMDAR-PSD95 interaction (Aarts et al., 2002; Cui et al., 2007). In both experiments, disruption of NR2B-PSD95 interaction inhibited neurotoxic Ca2+ influx with no effect on normal NMDAR function (Aarts et al., 2002; Sattler et al., 1999). These findings were tested in rat models, where inhibiting NR2B-PSD95 reduced

11 infarct volume and cell death in following a stroke (Aarts et al., 2002). Our lab has recently translated this therapy from rodent models to non-human primates, where again NA1 was neuroprotective following ischemia (Cook et al., 2012).

Through this PSD95-NR2B interaction, a number of downstream pro-survival and pro- death mechanisms are regulated. When PSD95 and NR2B interacts, the influx of Ca2+ leads to an inactivation of ERK1/2 and Akt pathways, which prevents the inhibition of pro-death pathways and leads to cell death (Soriano et al., 2008). Also, NR2B containing NMDAR-mediated Ca2+- influx inhibits cAMP response element binding protein (CREB), which is a transcription factor that regulates the production of brain-derived neurotrophic factor (BDNF), which is required for neuronal growth and survival (Soriano et al., 2006). NR2B-mediated Ca2+-influx is also a key player in excitotoxicity and acts to suppress synaptic excitatory signals from NMDAR with other NR2 subunits (Liu et al., 2007b; Wyllie et al, 2013; Aarts et al., 2002; Sattler et al., 1999). Blocking NR2B-PSD95 interaction has been shown to restore pro-survival mechanisms, such as activation of Akt, CREB and ERK1/2, as well as increase in synaptic excitability (Aarts et al., 2002; Sattler et al., 1999; Soriano et al., 2008; Liu et al., 2007b; Wyllie et al, 2013).

In addition to the subunits contained in NMDARs, it has been suggested that the location of NMDARs on the neuron’s membrane plays a role in their involvement in survival and death mechanisms (Hardingham and Bading, 2010). Since both cell survival and death are regulated through NMDAR-mediated Ca2+ influx, it was initially proposed that the location of this influx activated different intracellular pathways; specifically synaptic NMDARs were responsible for pro-survival signaling (Hetman and Kharebava, 2006; Hardingham, 2006b), while extrasynaptic NMDARs were responsible for pro-death signaling (Hardingham et al., 2002; Hardingham and Bading, 2010). Also, this demarcation of NMDARs supported the idea that Ca2+-influx at the synapse was responsible for synaptic excitability, while influx at the extrasynapse resulted in a change in the Ca2+ gradient and synaptic inhibition (Hardingham et al., 2002; Hardingham and Bading, 2010; Massey et al., 2004). Initial localization studies of NR2A and NR2B supported this hypothesis, where NR2A was predominantly expressed at the synapse and NR2B was predominantly expressed at the extrasynapse (Li et al., 2002; Tovar and Westbrook, 1999; Hardingham et al., 2002; Thomas et al., 2006). However, recent studies have begun to question whether or not NR2A and NR2B are clearly segregated between the synapse and extrasynapse, where NR2B has been found at the synapse and vice versa (Hardingham, 2006a). The separation

12 of NR2A and NR2B at the synapse and extrasynapse is an interesting model for how NMDAR- mediated Ca2+-influx separates survival and death signaling and an area that warrants further study.

The role of NMDARs in initial excitotoxic Ca2+ influx is an area that that has been extensively studied. Its dual role in balancing pro-survival and pro-death mechanisms following an ischemic insult makes it even more difficult to find therapeutic drugs that specifically target pro-death mechanisms while having no effect on normal NMDAR function. So far, the PSD-95- NMDAR interaction inhibitor, NA1, has been the most promising drug to provide neuroprotection without any side effects (Aarts et al., 2002; Cui et al., 2007; Cook et al., 2012). Although NMDARs are the primary glutamate receptor involved in excitotoxicity, it is not the only channel that contributes to Ca2+ overload, especially delayed Ca2+ influx.

1.3.2 Non-glutamate driven cation channels

Following ischemia, glutamate receptors predominantly mediate the initial Ca2+ influx due to excess neurotransmitter release. However, glutamate-mediated Ca2+ influx is not the sole entrance for Ca2+-overload, a number non-glutamate driven channels contribute to this influx either dependent or independent of glutamate receptor activity. Non-glutamate driven channels that contribute to Ca2+ overload include: acid sensing channels, voltage-dependent sodium/calcium exchangers, hemi channels, L-type voltage dependent Ca2+ channels, volume regulated anion channels and TRP channels (Besancon et al., 2008; Szydlowska and Tymianski, 2010).

1.3.2.1 Acid sensing channels

Acid-sensing channels (ASICs) are ligand gated multimeric channels that are members of the epithelial sodium channel superfamily. These channels can be activated by: lactate, membrane stretching, arachidonic acid, low pH and decrease in extracellular Ca2+ (Allen and Attwell, 2002; Immke and McCleskey, 2001; Immke and McCleskey, 2003). All of these conditions are found following an ischemic insult, and is the primary reason why ASICs have been suggested to be involved in Ca2+ overload (Allen and Attwell, 2002; Diarra et al., 1999). It has been reported that blocking ASIC results in neuroprotection by reduction of neurotoxicity due to acidosis (Xiong et al., 2004). Also, ASIC1 KO mice were shown to have reduced infarct

13 size following MCAO and cultured cortex neurons from these mice were resistant to neurotoxic acidosis (Xiong et al., 2004). These findings show that ASICs contribute to neurotoxicity following ischemic conditions, likely due to acidosis in neurons leading to ASIC-mediated Ca2+ influx.

1.3.2.2 Voltage-dependent Na+/Ca2+ exchangers

Voltage dependent Na+/Ca2+ exchangers (NCX) is a bidirectional ion transporter and in neurons are responsible for regulating both sodium and calcium. During depolarization, where there is a large influx of sodium, NCX channels act to maintain very low intracellular Na+ concentration by transporting Na+ out of the cell and Ca2+ in (Jeffs et al., 2007). Under unchallenged conditions, inhibition of NCX prevents transport of Na+ out of the cell, increases depolarization, which in turn activates voltage-dependent calcium channels and results in calcium overload and cell death (Pignataro et al., 2004). However, during ischemic conditions blocking NCX and Na+/Ca2+ -K+ exchangers (NCKX) reduced Ca2+ influx and provided neuroprotection (Czyz and Kiedrowski, 2002). This initial paradoxical evidence suggested that different types of NCX may have different effects on neurons.

There are three distinct isoforms of NCX found in the brain, NCX1, NCX2 and NCX3, which are located in distinct regions of the brain (Secondo et al., 2007; Boscia et al., 2006). NCX3 has been suggested to be the primary isoform that is responsible for Ca2+ homeostasis under normal conditions (Secondo et al., 2007). Following a stroke, NCX1 mRNA expression is upregulated, while NCX3 mRNA is cleaved and inactivated (Boscia et al., 2006). Furthermore, early cleavage of NCX3 contributes to glutamate-driven calcium overload (Bano et al., 2005). This dual role NCX suggests that during excitotoxicity, some isoforms of NCX contribute to the Ca2+ influx, while others attempt to maintain Ca2+ homeostasis by pumping Ca2+ out of the cell. Like NMDARs, NCX may play a role in regulating both neuronal death and survival, depending on the conditions.

1.3.2.3 Hemichannels

Gap junctions connect the cytoplasm of two cells and are formed by two hemichannels. Almost every cell type in the brain contains gap junctions and allows molecules and ions to pass freely between cells (Contreras et al., 2004). A number events associated with ischemia are

14 known to activate hemichannels, including increased ROS, changes in phosphorylation pathways and redox-dependent modulation of nitrosylation of connexins (Contreras et al., 2004; Retamal et al., 2006). During ischemia, hemichannels have been suggested to play a role in the influx of cations, as well as glutamate, which activate glutamate receptors (Contreras et al., 2004). Blocking gap junctions reduces cell death following hypoxia and reducing gap junction connections reduces cell death following hypoxic insults (Frantseva et al., 2002; Pina-Benabou et al., 2005). In mammals and other vertebrates, pannexin hemmichannels are a type of channel that connects cytoplasm of two cells and has been suggested to play a role in ischemic cell death (Barbe et al., 2006; Thompson et al., 2006). Pannexin 1 (Px1) has been shown to be opened under OGD conditions and Px1 hemmichannel opening results in a loss of internal ATP stores, a phenomenon associated with neuronal death (Thompson et al., 2006). Although inhibition of hemichannels alone is not neuroprotective, there is evidence showing that they are involved in ischemic cell death.

1.3.2.4 L-type voltage-dependent Ca2+ channels

1,4-dihydropyridine sensitive L-type voltage-dependent calcium channels (CaV1.2) are predominantly found at the dendrites of cortical neurons and are activated during depolarization of the cell membrane. It is thought that during excitotoxicity, CaV1.2 channels are activated and contribute to Ca2+ influx (Gribkoff and Winquist, 2005). However, blocking these channels have shown little effect in terms of neuroprotection in patients following acute ischemia (Horn and Limburg, 2000).

In addition to voltage-dependent inward Ca2+ currents, Ca2+ pumps responsible for the removal and homeostasis of intracellular Ca2+ have also been suggested to play a role in ischemic neuronal death (Pottorf et al., 2006). The plasma membrane Ca2+ ATPase (PMCA) has been shown to reduce Ca2+ efflux from rat hippocampal neurons following excitotoxicity. During glutamate-mediated neuronal death PMCA undergoes cleavage and internalization at the plasma membrane preventing efflux of Ca2+ (Pottorf et al., 2006). This shows that stroke not only affects the influx of Ca2+, but also inhibits the efflux and maintenance of Ca2+ homeostasis to promote Ca2+ overload.

1.3.2.5 Volume-regulated anion channels

Volume-regulated anion channels (VRAC) are predominantly found in astrocytes and are involved in neurotransmitter and ATP release following astrocytic swelling (Kimelberg, 2005; Kimelberg et al., 2003; Kimelberg et al., 2006). Unregulated activation of VRAC leads to a number of conditions associated with toxicity, including, additional fluid via aquaporins; depletion of ATP; lowering of pH due to ammonia build up and dysfunction in glutamate metabolism; and activation of pH-dependent cation channels (Kimelberg, 2005). Blocking these channels following an acute stroke has been shown to be neuroprotective in rats (Kimelberg et al., 2003). Since astrocytes are used to maintain extracellular ion and neurotransmitter homeostasis, dysfunction of VRAC will prevent astrocytes from balancing extracellular environments of neurons and thus not only facilitate cell death in astrocytes but also in neurons.

1.3.2.6 TRP Channels

Transient receptor potential (TRP) channels were first described in Drosophila, where trp mutants demonstrated transient voltage response to constant light (Cosens and Manning, 1969). TRP channels often contain a sensory component and can be activated by light temperature, odor, taste, metabotropic factors and mechanical forces (Agam et al., 2000; Aarts and Tymianski, 2005). All TRP channels share 6 transmembrane segments and form tetrameric channel pores (Figure 2). The TRP channel superfamily is comprised of seven protein sub-families: TRPC (canonical), TRPV (vanilloid), TRPA (ankyrin), TRPM (melestatin), TRPP (polycystin), TRPML (mucolipin) and TRPN (no mechanoreceptor potential C). TRP channels are known to detect intracellular cations and oxidative stress (Hara et al., 2002; Wehage et al., 2002; Prawitt et al., 2003). Two members of the TRPM family, TRPM7 and TRPM2, have been suggested to play a role in neurodegeneration following a stroke (Kaneko et al., 2006; Hara et al., 2002; Aarts et al., 2003; Sun HS et al., 2009; Nicotera and Bano, 2003). We will discuss these channels in greater detail later in the introduction. Briefly, both these channels are non-specific cation channels that are activated by oxidative stress conditions, which is why many have focused on them as a potential contributors to Ca2+ overload (Hara et al., 2002; Aarts et al., 2003).

Whereas TRPM channels are suggested to be involved in ischemic neuronal death, other TRP channels have been suggested to be involved in survival mechanisms in neurons. Members of TRPC family (“C” for canonical), TRPC3 and TRPC6, have been shown to be neuroprotective

16 when overexpressed in cerebellar granual cells. Downregulation of TRPC3 and TRPC6 in cerebellum neurons induce neuron apoptosis, and can be rescued by reintroduction of TRPC3 and TRPC6 (Jia et al., 2007). The same study also demonstrated that overexpression of TRPC3 and TRPC6 increased CREB-dependent reporter gene transcription and are involved in BDNF- mediated neuroprotection (Jia et al., 2007).

1.4 Ca2+-induced downstream signaling and apoptosis

As mentioned in the prior section, Ca2+-influx via cation channels, particularly NMDARs, can initiate a number downstream pathways involved with survival and death following ischemic conditions. In addition to neuronal viability, NMDARs modulate a number of cellular functions including: synaptic plasticity, translation, transcription, cell growth, and proliferation. The number of Ca2+-dependent pathways in neurons are far too great to discuss in this introduction, however we will focus on those related to cell survival and death, particularly pathways related to the findings of this thesis.

1.4.1 Ca2+-mediated pro-survival pathways

As stated before, NR2A containing NMDAR activation is required in cell survival. Ca2+ influx from these synaptic receptors activates via phosphorylation Akt, ERK1/2 and Ca2+/calmodulin-dependent protein kinase II (CaMK II) pathways (Soriano et al., 2008; Hardingham and Bading, 2002). Akt is activated by Ca2+ influx via phosphatidylinositide 3- kinases (PI3K), and goes on to modulate a number of transcription regulatory proteins, particularly, forkhead box (FOXO) and glycogen synthase kinase 3 beta (GSK3B) (Soriano et al., 2008; Liu et al., 2007b; Wyllie et al, 2013). FOXO is a transcription factor, which is inhibited by active Akt and regulates metabolism, cell proliferation and stress tolerance. FOXO upregulates transcription of a number of genes, including: fas-l, p27, and kip1, all of which are involved in apoptosis (Lehmann et al., 2003). GSK3β is deactivated by phosphorylation from Akt and is involved in energy metabolism, neuronal cell development and proliferation (Plyte et al., 1992; Soriano et al., 2006). Inhibition of GSK3β prevents the inhibition of transcription factors such as NFAT and GATA4, which are required for cell proliferation and survival (Hardingham and Bading, 1999).

ERK1/2 is activated by Ca2+ via the Ras (meaning “rat sarcoma”) subfamily of small GTPase proteins responsible for signal transduction. Both phosphorylated ERK1/2 and CAMKII are known activators of CREB, a transcription factor. CREB is involved in the regulation transcription of a number of genes, including: c-fos, brain derived neurotrophin factor (BDNF), tyrosine hydroxelase and a number of neuropeptides (Downward, 2003;Stornetta and Zhu, 2011). BDNF is involved in neuronal growth, as well as synaptic activity, particularly facilitating long term memory. Increased production of BDNF is also associated neuroprotection following ischemia (Soriano et al., 2006).

1.4.2 Ca2+-mediated pro-death pathways

NR2B containing NMDARs are one of the primary sources of toxic Ca2+ influx (Soriano et al., 2008; Sattler et al., 1999; Aarts et al., 2002). NR2B mediated Ca2+ has been shown to inhibit the pro-survival mechanisms that are induced by NR2A-NMDARs, specifically the inhibition of ERK1/2 and CREB phosphorylation (Soriano et al., 2008; Liu et al., 2007b). NR2B mediated Ca2+ influx inhibits CREB activation either by inhibition of ERK1/2 or by inhibiting Jacob, a protein involved in NMDAR plasticity and activation of CREB. This mechanism prevents the production of BDNF, which is required for cell survival (Soriano et al., 2008). An alternate pathway that also promotes cell death is via calpains (Lee et al., 1991). Calpains are proteases that cleave proteins, such as striatal enriched protein tyrosine phosphatase (STEP), which is involved in development and synaptic function. Under normal conditions STEP inhibits P38, a mitogen-mediated protein kinase that promotes apoptosis, and during stress STEP is cleaved by calpains, which stops the inhibition of P38 (Munoz et al., 2003; Paul et al., 2003; Xu et al., 2009). Pro-death Ca2+ influx can also activate NO synthase (NOS), which is involved in the production of free radicals. These free radicals go on to damage the cell leading to death (Arundine and Tymianski, 2004). Free radicals and their role in apoptosis will be discussed later in this introduction.

1.4.3 Caspase-mediated cell death

In cell death models, caspases play a central role in apoptosis, necrosis and inflammation. Caspases are highly conserved family of proteins that are cystine dependent aspirate-specific proteases (Fesik and Shi, 2001). Caspases are divided into two groups: initiator caspases (caspase 8,9,10,2) and effector caspases (caspase 3,7,6). Initiator caspases are activated by

18 oligomeric activator protein, which then activates effector caspases via proteolytic cleavage (Thornberry and Lazebnik, 1998). Following Ca2+-overload and free radical production, damaged mitochondria releases the cytochrome complex (cytochrome-c), which under normal condition is used in the ion-transport chain for energy production (Dejean et al., 2006). Cytochrome-c then activates caspase-9, which in turn cleaves both caspase-3 and 7, both of which are involved in cell damage and cell death (Slee et al., 2001; Ashkenazi and Dixit, 1998). Although not via direct activation, Ca2+ overload is still involved in initiating the pathway that leads to caspase-mediated cell death.

In summary, Ca2+ signaling is involved in both cell survival and cell death mechanisms. This paradoxical role of Ca2+ raises important questions about how the cell differentiates pro- survival and pro-death signaling from Ca2+ influx. As previously mentioned, the source, location and quantity of intracellular Ca2+ affects which pathways are activated and which pathways are inhibited (Manev et al., 1989; Marcoux et al., 1990; Sattler et al., 1998; Tymianski et al., 1993). The downstream interaction of survival and death pathways suggests that the cell maintains a kind of homeostasis between survival and death, where disruption or modifications of this pathway can tilt the scales of cell life and death.

Figure 1: Synaptic and extrasynaptic mechanisms of cell survival and death

Synaptic NMDAR (NR2A) activity induces pro-survival mechanisms by activating Akt and ERK1/2 pathways in neurons. These pathways activate transcription factors involved with cell survival, such as CREB, and inhibit transcription factors involved in cell death, such as FOXO. Extrasynaptic NMDAR (NR2B) activity promotes apoptosis by inhibiting ERK1/2 which inhibits activation of CREB. Extrasynaptic signaling is also responsible for FOXO activation.

1.5 Free radicals and oxidative stress

ROS are a natural byproduct of normal metabolism of oxygen and play a role in cell signaling and homeostasis under normal conditions. During normal conditions cells regulate ROS and prevent oxidative stress via enzyme production (alpha-1 microglobulin, superoxide dismutase, lactoperoxidases, glutathione peroxidases and peroxiredoxins); as well as intake of small molecules antioxidants (ascorbic acid, uric acid, tocopherol, and glutathione) (Larsson et al., 2004; Atamna and Ginsburg, 1995; Kumar and Bandyopadhyay, 2005). Although ROS is produced in the core of the ischemic stroke, it is heavily produced in the penumbra, where there is contralateral blood flow. The penumbra is a very appealing target for stroke treatments as the neurons in that region can survive up to 3-6 hours after initial ischemic insult (Chavez et al., 2009). For neurons maintaining this oxidative homeostasis is critical for survival, as a loss of homeostasis will quickly damage neurons and ultimately trigger cellular death cascades.

Under oxidative stress conditions, there is an overproduction of free radicals such as ROS and reactive nitrogen species (RNS), which work together to cause cell damage. This involvement in cell death is emphasized by the fact that free radical scavengers can reduce infarct in focal cerebral ischemia models. Following ischemia, these free radicals cause damage to the cell via: damage to DNA, oxidation of fatty acids in lipids, oxidation of amino acids and inactivate enzymes and cofactors via oxidation (Radi et al., 1991; Aarts et al., 2003; O'Leary et al., 1992; Crow and Beckman, 1995). Other than damage to neurons, free radicals can also play a role in apoptosis via downstream signaling and activation of ion channels involved in cell death.

It has been established that excitotoxicity produces Nitric Oxide (NO), since it is generated by both in vitro and in vivo activation of NMDAR (Dawson et al., 1991; Huang and Gean, 1994; Yang and Iadecola, 1998). Targeted disruption of NOS, which catalyzes the production of NO, prevents NMDA excitotoxicity (Dawson et al., 1991). Both NO and Ca2+ - trigger the production of superoxide (O2 ), which when combined with NOS can produce highly - - toxic peroxynitrate (ONOO ). Another product of O2 is hydrogen peroxide (H2O2), which is produced when superoxide leaks into the mitochondria and interferes with the electron transport - chain (Chanock et al., 1994). Both ONOO and H2O2, as well as other free radicals, go on to cause neuronal damage, but there is growing evidence that they also feed back to activate ion channels and further calcium influx.

As stated earlier, members of the TRPM family, TRPM7 and TRPM2, have recently been identified as being involved in neuronal death (Kaneko et al., 2006; Aarts et al., 2003; Sun HS et al., 2009). Both are activated by free radicals and they are non-specific cation permeable (Hara et al., 2002; Aarts et al., 2003; Olah et al., 2009). Calcium overload produces NO, which in turn - produces by ONOO and H2O2, which are known activators of TRPM7 and TRPM2 (Hara et al., 2002; Aarts et al., 2003), which is believed to facilitate further Ca2+-influx to produce more ROS/RNS. Due to the requirement of initial ROS/RNS production for TRPM7/TRPM2 activation, it has been suggested that these channels are involved in delayed calcium influx following the initial Ca2+ spike via NMDAR (Aarts et al., 2003). TRPM7 and TRPM2 are prime examples of how free radicals are involved in signaling increased influx of Ca2+ following ischemia, we will further discuss the specific roles of TRPM7 and TRPM2 later in the introduction.

During oxidative stress, there is often a disruption in synaptic activity, and studies have shown that ROS, specifically H2O2, can modulate both normal synaptic function and plasticity, including long term potentiation (LTP) and is involved in the disruption of synaptic plasticity following a stroke (Colton et al., 1989; Kamsler and Segal, 2003). Studies have found that concentrations of 20-500uM of H2O2 inhibit LTP formation, returning excitatory field potentials (fEPSPs) back to baseline (Kamsler and Segal, 2003; Maalouf and Rho, 2008). What is more intriguing is how different concentrations affect synaptic plasticity, where higher concentrations of H2O2, 5 mM, not only inhibits LTP formation but promotes long term depression (LTD), while lower concentrations of 1uM of H2O2 attenuated LTP (Kamsler and Segal, 2003). It is obvious that hydrogen peroxide plays a critical role in modulating plasticity, unfortunately, the mechanism of how LTP and LTD are modulated by H2O2 is not fully understood. There is evidence that voltage dependent calcium channels (VDCC), which are activated during plasticity, are involved; because blocking these channels via nifedipine prevents H2O2 induced attenuation and inhibition of LTP at 1uM and 20uM concentrations (Kamsler and Segal, 2003).

It is becoming more evident that the role of free radicals following ischemia is more than simply cellular damage. ROS and RNS are becoming key players in feedback signaling to further Ca2+ overload, as well as inhibiting excititory synaptic plasticity following a stroke. Viewing free radicals as cell death signaling molecules for ion channels provides interesting new targets to

23 prevent immediate cell death and to prevent long term effects such as disruption of synaptic plasticity.

1.6 TRPM channels and stroke

The eight member melastatin (TRPMs) subfamily are non-selective cation channels and as mentioned earlier, TRPM7 and TRPM2 have been implicated to play a role in ischemic neuronal death (Kaneko et al., 2006; Aarts et al., 2003; Sun HS et al., 2009; Olah et al., 2009). TRPM7 and TRPM2 are unique compared to other TRPM channels because they both exhibit ion channel and enzyme function (also known as chanzymes; Scharenberg, 2005). It has been suggested that NMDAR, TRPM7 and TRPM2 all play a role in the influx of neurotoxic Ca2+, which leads to activation of nitric oxide synthase (nNOS) activity, mitochondrial failure and oxidative stress via reactive oxygen/nitrogen species (ROS/RNS) (Rempe et al., 2009). Therefore blocking one or more of these channels in neurons could provide neuroprotective properties during ischemia.

1.6.1 Role of TRPM7 in ischemia

Blocking NMDARs alone does not completely abolish ischemic Ca2+-influx (Davis et al., 1997; Lees et al., 2000) and other channels have been suggested to make up for the remaining inward Ca2+ current (Aarts et al., 2003). Oxygen glucose deprivation (OGD) insult has been shown to elicit excitotoxic cortical neuronal death (Goldberg and Choi, 1993; Goldberg et al., 1987). However, our lab has demonstrated that in these cultured neurons when excitotoxicity is blocked by a cocktail of MK-801, CNQX and nimodipine (abbreviated as“MCN”) targeting NMDA, AMPA and L-type calcium channels, respectively, a new prolonged OGD-induced mechanism is unmasked by a Ca2+-permeable nonselective cation conductance originally termed

“IOGD” (Nadler et al., 2001; Aarts et al., 2003).

TRPM7 is ubiquitously expressed in the body, with high levels of expression in the brain, including the cortex and CA1 neurons in the hippocampus and cortical neurons (Wei et al., 2007; Aarts et al., 2003). TRPM7 channels have high permeability to cations, possess an outwardly rectifying current and can be blocked by Gd3+ (Wei et al., 2007; Aarts et al., 2003; Aarts and Tymianski, 2005a). Over expression of TRPM7 in HEK293 cells has been shown to result in increased cell death (Nadler et al., 2001; Monteilh-Zoller et al., 2003; Schmitz et al., 2003). By

silencing TRPM7, the IOGD current can be suppressed and OGD-induced neuronal death can be inhibited in cultured neurons (Aarts et al., 2003). Thus demonstrating that IOGD is a “TRPM7- like” current and TRPM7 is needed for OGD-induced cell death. This suggests that excitotoxicity is not the only mechanism that causes ischemic neuronal death and that TRPM7

(and perhaps similar channels) may cause neuronal death via another mechanism.

Experiments to fully assess the role of TRPM7 channels have been limited, since TRPM7 knock-out is embryonically lethal (Jin et al., 2008), and there are no commercially available specific blockers for TRPM7 (Aarts et al., 2003). Silencing TRPM7 via siRNA is specific and provides a partial loss of function of TRPM7, and is currently one of the best methods used when assessing TRPM7 function during ischemia (Aarts et al., 2003; Sun HS et al., 2009). Our lab demonstrated that selective silencing of TRPM7 provides protection for neurons following OGD stress in vitro and global ischemia in vivo (Aarts et al., 2003; Sun HS et al., 2009). A number of labs, including our own1, have been looking into small molecule inhibitors for TRPM7 channels (Castillo et al., 2010). Recently a potential small molecule candidate, M21, has emerged in our lab that blocks “TRPM7-like” currents and provides neuroprotection to cortical neurons exposed to OGD1.

Little is known about the role of TRPM channels in synaptic plasticity under both normal and anoxic conditions. Most research involving TRP channels and synaptic function has been focused on TRPV1 which is a key mediator of plasticity in the hippocampus (Alter and Gereau, 2008) and the role of non-selective cation channels in releasing internal calcium stores (Lange et al., 2009). TRPM7 is present on the membrane of synaptic vesicles and can modulate EPSP (Krapivinsky et al., 2006). Recent studies involving TRPM7 knockdowns have shown that 30 days following ischemia, LTP formation was preserved in knockdown animals and lost in the control ischemia-only group (Sun HS et al., 2009). Although there are many reasons as to why this may have occurred, it is interesting that TRPM7 channels not only protect neurons from cell death but also preserves synaptic plasticity.

1 Unpublished data from Dr. Tymianski’s lab using small molecule inhibitor for TRPM7 screenings

Both in-vitro and in-vivo data confirms that TRPM7 is a contributor to ischemia-induced neuronal cell death (Aarts et al., 2003; Sun HS et al., 2009). Furthermore, silencing TRPM7 provides neuroprotection during ischemia and protects synaptic plasticity mechanisms following ischemia. This led us to ask if similar TRPM channels, in terms of function, may also be involved in mechanisms leading to ischemia-induced neuronal death.

1.6.2 TRPM2, a potential novel channel involved in ischemia?

TRPM2, originally referred to as TRPC7, was discovered in 1998 (Nagamine et al., 1998). Microarray data has shown that during MCAO TRPM2 is significantly upregulated (Fonfria et al., 2006). TRPM2 is expressed throughout the body including the pancreas, immune cells, the heart and the brain (Togashi et al., 2006; Lange et al., 2009; Kaneko et al., 2006; Olah et al., 2009; Yamamoto et al., 2008; Yang et al., 2006). TRPM2 is stimulated by two mechanisms: (1) by overproduction of adenosine diphosphate ribose (ADPR) and (2) ROS

(particularly H2O2) (Wehage et al., 2002; Hara et al., 2002; Perraud et al., 2005).

Oxidative stress is one of two activators of TRPM2 channels and has been shown to induce an inward Ca2+ current in cortical neurons, hippocampal neurons, and monocytes (Hara et al., 2002; Kaneko et al., 2006; Yamamoto et al., 2008; Olah ME et al., 2009). Hydrogen peroxide is a major contributor for oxidative damage and its over-production often leads to neuronal death. Silencing of TRPM2 (using siRNA) in rat cortical neurons has been shown to suppress 2+ H2O2-induced Ca influx and H2O2-induced cell death (Kaneko et al., 2006). However, whether inhibition of TRPM2 alone is neuroprotective remains controversial, as some have shown that pharmacological blockers of TRPM2 have no effect on oxidative stress-induced cell death (Bai and Lipski, 2010; Wilkinson et al., 2008).

The second activator of TRPM2 is ADPR, which binds to ADPRase NUDT9H (Nudix motif) domain of the C-terminus of TRPM2 (Figure 2; Perraud et al., 2005). ADPR when applied intracellularly, induces large inward currents in cortical neurons, HEK293 cells, pancreatic β cell lines, monocytes and hippocampal CA1 neurons (Kaneko et al., 2006; Yamamoto et al., 2008; Togashi et al., 2006; Lange et al., 2009; Olah ME et al., 2009). There are other factors that may modulate the ADPR induced activation of TRPM2 channels including temperature (Togashi et al., 2006) and in the case of hippocampal CA1 neurons ADPR induced activation is dependent on voltage-dependent Ca2+ channels and NMDARs (Olah ME et al., 2009). Intracellular and

26 extracellular application of ADPR can cause an increase in intracellular Ca2+ concentration; however intracellular application is the only way to activate TRPM2-induced Ca2+ influx via the enzymatic complex in TRPM2 (Ishii et al., 2006; Lange et al., 2009).

The results from the Jia et al. (2011) study provides a potential explanation as to why in vitro experiments were inconclusive, specifically, neuroprotection by TRPM2 inhibition is dependent on the gender of the animal. Following in vitro ischemic conditions, TRPM2 inhibitors were neuroprotective in neurons from male animals, but not those from females (Jia et al., 2011). The same group also found that neurons derived from male and female animals express functional TRPM2 channels, however under oxidative stress conditions only TRPM2 channels in neurons from male animals produce inward TRPM2-like currents (Verma et al., 2012). TRPM2 being involved neuronal death in males only, provides a potential reason as to why mixed gender cell culture experiments provided opposite results.

Thus far, research trying to functionally identify TRPM2 has primarily involved TRPM2 silencing (Olah ME et al., 2009; Hara et al., 2002) or the use of TRPM2 blockers such as clotrimazole, N-(p-amylcinnamoyl)anthranilic acid (ACA) and flufenamic acid (Olah et al., 2009; Lange et al., 2009; Harteneck et al., 2007). Although, these methods have been able to show insights into the function of TRPM2 channels, they can be problematic since silencing doesn’t completely block TRPM2, and non-specific blockers can inhibit multiple channels; which leads to the need for a knockout model. Recently, a study has generated TRPM2 knockout (TRPM2(-/-)) mice, which have been shown to block ROS-induced signaling cascade and prevent oxidative stress in monocytes (Yamamoto et al., 2008). With the development of this TRPM2 KO mouse it has become easier to explore the role of TRPM2 in neuronal death during oxidative stress and ischemia.

Figure 2 Diagram of TRPM2 structure

TRPM2 channel consists of 6 transmembrane components, where the S5-S6 linker is part of the pore forming domain. A complete channel is formed by 4 individual TRPM2 protiens. On the c-terminus there is a NUDT9H (Nudix motif) domain.

1.6.2.1 Development and recent findings using TRPM2(-/-) mouse

TRPM2(-/-) was developed by an insertion of the neomycine (neo) gene into exon 3, which is encodes for transmembrane segment 5 and S5-S6 linker of the TRPM2 channel (Figure 3; Yamamoto et al., 2008). This KO has been shown to inhibit TRPM2-like currents in monocytes, CA1 neurons in the hippocampus and macrophages (Xie et al., 2011; Yamamoto et al., 2008; Kashio et al., 2012). TRPM2(-/-) also influences a number of phenomon and mechanisims that occur during ischemia, including modulation of macrophage activity, ERK activity (Yamamoto et al., 2008) and AMPA expression (Xie et al., 2011).

Figure 3 Development of TRPM2(-/-) mice

The TRPM2(-/-) mouse was generated by replacing exon 3 (shown in red) in the WT TRPM2 gene with a neomycin gene (shown in green). Neomycin gene was inserted using a targeting vector specific for exon 3. Exon 3 encodes for the S5-S6 pore forming domain of TRPM2. Image was modified from Yamamoto et al. (2008).

Following ischemic injury, the inflammatory response can often exacerbate damage and many anti-inflammatory techniques have been used to reduce damage (Wang et al., 2007).

Yamamoto et al. (2008) has recently shown that H2O2-induced TRPM2 activity is responsible for an increase production of macrophage inflammatory protein-2 (CXCL2) in monocytes via ERK activation. The absence of TRPM2 reduces H2O2-induced ERK activation and CXCL2 production, resulting in a suppression of the inflammatory response (Yamamoto et al., 2008). In 2+ macrophages, H2O2 induces heat-dependent Ca influx, which is abolished in cells from TRPM2(-/-) animals (Kashio et al., 2012). In vivo, carrageenan-induced inflammation and sciatic nerve injury increases TRPM2 mRNA expression at the site of injury and sensitivity to formalin test was reduced in TRPM2(-/-) animals (Haraguchi et al., 2012). Also, TRPM2(-/-) mice had a reduction in infarct volume following myocardial ischemia/reperfusion injury (Hiroi et al., 2012). Although these studies were not in the brain, it does hint that following a ischemia, TRPM2(-/-) may influence immune response and microglia function.

Recently it has been shown that TRPM2(-/-) modulates synaptic plasticity, specifically inhibiting LTD in the hippocampus (Xie et al., 2011). This inhibition of LTD in TRPM2(-/-) is mediated by a chronic reduction of GluR2 AMPA receptor subunit expression, which reduces downstream mechanisms responsible for LTD formation; such as the production of PSD95 and activation of GSK3B (Xie et al., 2011). Since the absence of TRPM2 can modulate synaptic plasticity, it raises important questions as to whether it can protect LTP following ischemic conditions. Our lab has shown that the lack of related channel, TRPM7, protects synaptic plasticity and the ability to form LTP following global ischemia (Sun HS et al., 2009).

The development of TRPM2(-/-) mouse model has provided new insights into the role of TRPM2 in a variety of cellular mechanisms. What is interesting is the role TRPM2 plays on pathways involved in neuronal survival/death, specifically ERK, GSK3B and PSD95 (Yamamoto et al., 2008; Xie et al., 2011; Aarts et al., 2002; Sattler et al., 1999; Soriano and Hardingham, 2007). The recent findings from TRPM2-null mice provide evidence to support the hypothesis that TRPM2 channels play a role in neuronal death following ischemic stroke.

Chapter 2

2 Aims and Hypotheses 2.1 Background and rationale

Stroke has become a major global health crisis and currently the number of viable treatments are limited. This has led many to explore targeting molecular mechanisms specific to neuronal death during ischemia. Ischemic neuronal death can be caused by an activation of both glutamate and non-glutamate dependent channels that leads to an overload of Ca2+ in the cell (Aarts and Tymianski, 2005b; Arundine and Tymianski, 2004). The traditional basis of glutamate-receptor-driven excitotoxicity is that it is caused by the over activation of N-methyl-D- aspartate glutamate receptors (NMDARs; Besancon et al., 2008; Sattler et al., 1999; Aarts et al., 2002). NMDARs are formed by coupling NR1 with NR2 subunits, specifically NR2A and NR2B, which are known to respectively govern cell survival and death signaling. An increase in the NR2A/NR2B subunit ratio of NMDARs at the synapse is known to promote pro-survival signaling, while the opposite promotes pro-death mechanisms ( Liu et al., 2007b; Wyllie et al, 2013; Hardingham and Bading, 2010; Martel et al., 2009).

Amongst other potential mechanisms of ischemic neuronal damage, two members of the transient receptor potential melastatin family (TRPM), TRPM7 and TRPM2, have been implicated to play a role in toxic Ca2+ influx (Olah et al., 2009; Fonfria et al., 2006; Kaneko et al., 2006; Aarts et al., 2003; Sun HS et al., 2009). These non-selective cation channels are unique amongst ion channels in that they contain both ion channel and an intracellular kinase domain (Nadler et al., 2001; Runnels et al., 2001). Our lab has shown previously that TRPM7 channels are important in mediating anoxic neuronal death in vitro and in vivo, by a mechanism involving oxidative stress (Aarts et al., 2003; Sun HS et al., 2009). The analogous TRPM2 protein contains a NUDT9H domain in the c-terminus that is activated directly by ADPR binding or indirectly by

H2O2 (Perraud et al., 2005; Kraft et al., 2004). H2O2 induced activation of TRPM2 channels have been reported to cause an influx of Ca2+ in cortical and hippocampal neurons (Olah et al., 2009; Kaneko et al., 2006; Xie et al., 2011), suggesting that it, like TRPM7, may mediate its effects on neuronal death via oxidative mechanisms. Additionally, TRPM2 has been ascribed a number of

31 physiological roles, including its involvement in immune and inflammatory response, insulin production, and LTD (Xie et al., 2011; Hiroi et al., 2013; Haraguchi et al., 2012; Lange et al., 2009; Yamamoto et al., 2008).

Most research to elucidate the role of TRPM2 in ischemia has been limited to using either TRPM2 silencing or non-specific antagonists (Olah et al., 2009; Bai and Lipski, 2010; Harteneck et al., 2007; Kaneko et al., 2006). However, recently, a TRPM2-null mouse has been generated (Yamamoto et al., 2008). These TRPM2(-/-) mice exhibit a loss of TRPM2 channel function in neurons (Xie et al., 2011). This has provided an unprecedented opportunity to investigate the role of TRPM2 channels in modulating neuronal vulnerability to ischemia.

Using the recent development of TRPM2(-/-) mice we have formulated the following four interconnected hypotheses to determine the general aim of whether TRPM2 plays a role in neuronal death following ischemic stroke. Although it may not be immediately apparent, hypotheses 3 and 4 were developed based on findings of previous hypotheses.

2.2 Specific aims and hypotheses

2.2.1 Hypothesis 1

Aim 1: To evaluate the role of TRPM2 channels in neuronal survival following cerebral ischemia in vivo.

Hypothesis 1: Following both transient and permanent cerebral ischemia in vivo, TRPM2(-/-) mice had reduced infarct volume and neuronal death compared to WT.

With the development of TRPM2(-/-) mice we first tested if these animals had reduced neuronal damage following both transient and permanent ischemia. By opting to use an in vivo model of stroke, we wanted to observe if TRPM2(-/-) was neuroprotective in whole mouse models before investing time for in vitro experiments. The model used was endovascular filament MCAO, where a nylon suture is inserted into the middle cerebral artery and blocks blood flow to one side of the cerebrum. We used two variations of the MCAO model: a transient model, where the suture was removed after a certain period of time and reperfusion occurs; and a permanent model, where the occlusion remained until the time of sacrifice. During the first hour of occlusion we measured blood flow and body temperature to ensure occlusion and that the animal

32 was able to maintain normal body temperature. Before MCAO and prior to sacrifice we also observed qualitative motor functions to observe any changes in motor function due to MCAO and infarct volume was measured using TTC.

2.2.2 Hypothesis 2

Aim 2: To determine if TRPM2 channels play a role in neuronal synaptic transmission and plasticity under control and oxidative stress conditions.

Hypothesis 2: Hippocampal slices from TRPM2(-/-) animals will be resistant to inhibitory effects of oxidative stress on synaptic plasticity.

Silencing TRPM7 has been shown to protect synaptic plasticity and LTP formation in rat hippocampus (Sun HS et al., 2009). TRPM2 has also been shown to be involved with LTD formation, but had no effect on LTP under non stressed conditions (Xie et al., 2011). Sub-lethal levels of H2O2 have been shown to inhibit LTP formation (Kamsler and Segal, 2003). H2O2 is also an activator of TRPM2, so therefore we hypothesized that TRPM2 plays a role in H2O2- induced inhibition of LTP formation. Using hippocampal slices, we stimulated the Schaffer collaterals of the hippocampus and measured evoked fEPSPs in the CA1 region. Baseline fEPSP (-/-) between TRPM2 and control animals were compared under normal and bath applied H2O2 conditions. We then induced LTP through high frequency stimulation and measured LTP formation under both normal and H2O2 conditions.

2.2.3 Hypothesis 3

Aim 3: To determine if TRPM2 channels modulate post synaptic glutamate receptors, resulting in neuroprotection and modulation of synaptic activity.

Hypothesis 3: TRPM2(-/-) promotes activation of NR2A containing NMDAR and inhibits NR2B containing NMDAR, leading to NR2A mediated neuroprotection.

Based on the results of the prior hypothesis and a study showing that chronic TRPM2(-/-) results in changes in AMPA receptor expression (Xie et al., 2011), we next tried to determine if TRPM2 mediates cell death through modulation of glutamate receptors. First, using western blots we determined expression and activation levels of glutamate receptor subunits in the hippocampus from control TRPM2(-/-) mice. Then using inhibitors specific to NR2A and NR2B

we determined if these subunits are responsible for H2O2 induced changes in evoked fEPSP. Finally, using western blot we determined whether hippocampi from TRPM2(-/-) mice had activated NMDAR-mediated pro-survival and pro-death mechanisms, specifically, Akt and ERK1/2 pathway.

2.2.4 Hypothesis 4

Aim 4: To determine, if TRPM2(-/-) enhances neuroprotective properties of potential stroke treatment drugs following OGD of CA1 neurons.

Hypothesis 4: Following OGD, TRPM2-null neurons increase resistant to cell death of neuroprotective drugs, when compared to similarly treated WT neurons.

Until recently, inhibition of TRPM2 in vitro was mostly limited to siRNA silencing and non-specific blockers (Hara et al., 2002; Kaneko et al., 2006; Bai and Lipski, 2010). Those methods have provided conflicting results as to whether or not loss of TRPM2 channel function is neuroprotective following stroke. With the development of TRPM2(-/-) animals we can now completely inhibit all TRPM2 function in neurons. Using both TRPM2(-/-) and control E18 prenatal pups, we cultured hippocampal neurons and incubated the cultures to allow them to mature to adult-like neurons. The cultures were then subjected to OGD stress and cell death was measured using PI staining 24 hours after insult. We also tested the effects glutamate receptor cocktail, MCN; TRPM7 inhibitor, M21; and PSD-95 inhibitor, NA1, in combination with TRPM2 null mice to determine additive effects of blocking multiple channels involved in cell death.

Chapter 3

3 Methods 3.1 Animals and genotyping

The TRPM2(-/-) model was generated using C57 black 6 (C57B6J) mice from Jackson Laboratory (Bar Harbor, Maine). In TRPM2(-/-) mice, exon 3 was replaced by a neomycine resistance gene (neo) as described in described in Yamamoto et al. (2008). Exon 3 of the trpm2 gene encodes for the transmembrane segment 5 and part of the segment 5-6 linker. Genotypes of animals were determined by polymerase chain reaction (PCR), using primers (shown in Table 1) and HotStar Taq RTPCR kit (Qiagen, Hilden, Germany). Samples were then run on a 1% agarose gel with sybr green for 60min using a 1X TRIS-EDTA buffer. RTPCR products were visualized using sybr green fluorescence detection on Las-3000 Imaging System (Fuji, Japan).

C57B6J mice are known to be more aggressive relative to other strains of mice, and mice were separated based on gender and genotype. There was no immediate observable difference between WT and TRPM2(-/-) animals and ear punching was used to identify mice. Breeding was conducted by pairing one male with two female mice and females were housed individually prior to birth. Vaginal plugs were observed to determine time of conception. TRPM2(-/-) were able to breed and had no difference in conceiving rate, number of pups born and gestation period compared to WT animals. To maintain C57B6J strain, animals were mated with pure breeding C57B6J mice from Jackson Labs after every 7-8 generations.

WT and TRPM2(-/-) mice were used for all experiments and sacrificed during or following experiments. All experiments and animal treatments were approved by the Animal Research Centre at the Toronto Western Hospital.

Table 1: Primers for RTPCR detection of TRPM2(-/-)

RT PCR Primer Name Sequence

PTRPM2-13F CTTGGGTTGCAGTCATATGCAGGC

PTRPM2-10R GCCCTCACCATCCGCTTCACGATG

Pneo-5’a GCCACACGCGTCACCTTAATATGCG

3.2 In vivo stroke model

To determine the role of TRPM2 in vivo, focal ischemia model MCAO was used as described previously (Bederson et al., 1986; Barber et al., 2004). Two variants of this model transient and permanent ischemia were tested, where in transient MCAO the suture is removed and reperfusion occurs, while in permanent MCAO there is no reperfusion.

Briefly, male C57B6J mice between 2-4 months old were anesthetized using a 2% Isoflurane-oxygen mixture. All MCAO experiments were conducted by an individual blinded to the genotype. MCAO was achieved by inserting a 6–0 monofilament suture (Doccol, Redlands CA, USA) into the MCA via the internal carotid artery. During the first hour of occlusion, body temperature was measured using a rectal probe. Body temperature was and maintained at 37.0 ± 0.5°C using either: heated blanket and overhead lamp (in transient ischemia), or a heated lamp above their cage (in permanent ischemia). A small window in the skin was made in the head on the ipsilateral side of the stroke, and blood flow was monitored using visual loss of blood below the skull and laser doppler flowmetry during the experiments. Occlusion was identified by a >80% drop in blood flow. Blood flow was monitored in both transient and permanent models for 1hr; however in transient models, we also monitored recovery of blood flow. Cerebral infarction was assessed in a blinded manner.

Animals were sacrificed 48 h and 24 h following transient MCAO (tMCAO) and permanent MCAO (pMCAO), respectively. The 48h time point for tMCAO was used because pilot experiments have shown this time point to have limited variability in damage2, and 24h for pMCAO, since that time point has low mortality rate with significant damage (Kim et al., 2012). Prior to animal sacrifice, a 5-point neurological test was used to visually determine motor deficit due to stroke, by lifting the animal by the tail and observing movement of bilateral limbs (5, normal; 0, complete loss of motor function; (Zhang et al., 2006). Although, this is a qualitative behavioral test, it will give initial insights into the relationship between the infarct volume and basic motor skills.

Cerebral infarct volume was analyzed by first removing the brain and cutting it coronally into 8 pieces. These sections were 2% 2,3,5-triphenyltetrazolium chloride (TTC) to stain infarct area. Infarct area on each slice was measured using a microcomputing imaging device (MCID) and volume was determined by combining areas of infarct. To compensate for edema or potential loss of tissue, infarct areas from each coronal level were mapped on image templates derived from a brain atlas, which were used by the MCID (Figure 4).

3.3 Electrophysiology

Hippocampal slices from 3- to 8-week-old male C57B6 mice were prepared using the following technique. After decapitation, the whole brain was cut into a block containing hippocampus, and placed for 5 min in a slicing solution containing (in mM): 87 NaCl, 2.5 KCl,

25 NaHCO3, 0.5 CaCl2, 1.25 NaH2PO4, 7 MgCl6H2O, 75 sucrose and 10 glucose (Sigma Aldrich, St. Louis MO, USA) (Bischofberger et al., 2006). The block was cut into 500 µm transverse slices with a Vibratome 1500 (Vibratome, Bannockburn, Il, USA) and incubated for 1 hr in artificial cerebral spinal fluid (ACSF) containing (in mM): 125 NaCl, 2.5 KCl, 25 NaHCO3,

2 CaCl2, 1.25 NaH2PO4, 1 MgCl6H2O and 10 glucose (Sigma Aldrich, St. Louis MO, USA), pH o 7.4, at 32 C. The ACSF was saturated with 95% O2-5% CO2 gas mixture. After incubation, the slices were placed in a submerged chamber for recording. All recordings were done at 32±0.5oC.

2 Personal communication from Dr. Mori’s lab. Pilot MCAO experiments were performed in their lab to determine if MCAO effects microglia function.

Recordings were made with a glass pipette containing ACSF (3-4 M ) placed in the dendrites of hippocampal CA1 neurons. Stimulation was delivered through a bipolar electrode (FHC Inc., Bowdoin, ME, USA) placed in the Schaffer collaterals. Baseline population EPSP (fEPSP) was recorded by 100-200uA pulses every 10 seconds. The fEPSP was quantified by measuring the peak amplitude and the declining slope (ie: rate of rise) of the negative-going waveform. LTP was induced by high-frequency stimulation (HFS) consisting of 1s of 100 Hz pulses. Paired-pulse facilitation was induced by evoking two pulses in rapid succession with varying 20-1000ms interpulse intervals.

To determine if hydrogen peroxide has any effect on synaptic activity, H2O2 was bath applied at 200 uM at 10 minutes before LTP induction. This concentration has been shown to inhibit LTP formation without effecting baseline (Maalouf and Rho, 2008). To determine effects of NR2 NMDAR subunits, 0.4 μM of the NR2A-specific antagonist PEAQX tetrasodium hydrate (also referred to NVP-AAM007; Sigma Aldrich) or 0.5 μM of the NR2B-specific antagonist Ro

25–6981 (Sigma Aldrich) was added to the bath medium 10 min after addition of H2O2. NR2A and NR2B inhibitor concentrations inhibit specific NMDAR subunits with no effect on baseline fEPSP, as shown in prior studies (Liu et al., 2007b; Liu et al., 2004). Data acquisition and analysis was performed using pClamp 10 (Mol. Dev. Sunnyvale, Ca, USA).

Figure 4 Coronal sections and template for infarct volume

To determine infarct volume, brain was cut coronally into 8 pieces approx. 1mm thickness. Using a template, infarct (core and penumbra) areas were shaded in on to corresponding template images (in blue). Templates were scanned on to MCID and volume was calculated.

3.4 Western blot

Whole hippocampal lysates from WT and TRPM2(-/-) animals were used for gel electrophoresis and blotting described previously in Soriano et al. (2008). Whole hippocampus was dissected out of animals and placed in lysis buffer containing protease and phosphotase inhibitors. Samples were incubated for 1hr and centrifuged at 10,000 RPM for 10 min, supernatant was collected and protein concentration was measured using Bradford assay. All samples were run on 7% or 12% (depending on protein size) Tris/glycine gels except poly- (ADP-ribose) polymerase-1 (PARP-1) gels. Each lane contained 30ug of protein containing 4× sample buffer consisting of the following: 0.125 M Tris-HCl, pH 6.8, 20% glycerol, 4% SDS, 5% β-mercaptoethanol, and 0.025% bromophenol blue, which was boiled for 5 min to denature the proteins. Samples were run at 70V in the stacking region of the gel and 100V in the resolving region of the gel. Samples are then transferred to methanol activated PVDF membranes at 100V for 1-2 hour.

Following transfer, membranes were incubated at 4°C overnight with primary antibodies with 5% dried milk in PBS. Primary antibodies used were: anti-GluR2 (Milipore), anti-NR2B, anti-phospho-NR2B, anti-psd95, anti-pser473 AKT, anti-AKT, anti-pthr AKT, anti-erk, anti- perk, anti-pGSK3B, Anti-pTRKA (Cell Signaling Technology), anti-NR1 (Upstate), anti-GluR1, anti-NR2A and anti-TRPm7 (Abcam). Phospho-tyrosine kinase A (pTRKA) antibody has been shown to detect phospho-TRKB on tyrosine 515 and hippocampal neurons do not express TRKA (Friedman, 2000). HRP-based secondary antibodies were used for 2 hours and visualized using chemiluminescent detection on Las-3000 Imaging System (Fuji, Japan). Bands were confirmed by size using a protein molecular weight marker (10-250kDa). Densitometry was measured using Photoshop Elements (Adobe) and all bands were normalized to loading control B-actin (Sigma).

3.5 Sex determination

Since it has been suggested that gender plays a role in the involvement of TRPM2- mediated cell death, we used only males in our adult experiments. To determine gender of E18 pups tail clippings were used for PCR (as previously described) to detect the SRY (sex determination on chromosome Y), primers shown in Table 2 (Liu et al., 2007a). Hippocampal cultures from each pup were plated individually and following sex determination, cultures from

40 male pups were identified. We have also used mixed gender cultures to determine difference between male only and mixed gender cultures. For in vitro OGD experiments, mixed gender cultures were primarily used due to the need of large number of neurons.

Table 2: PCR primers for Sex Determination on Chromosome Y

PCR Primer name Sequence

SRY F TCATGAGACTGCCAACCACAG

SRY R CATGACCACCACCACCACCAA

3.6 Cell cultures and in vitro OGD

Mixed hippocampal cell cultures containing neurons and glia were prepared from E18 pups from WT and TRPM2(-/-) mothers, as described in Olah et al. (2009). Hippocampi from pups were micro-dissected in Hanks Balanced Salt Solution (HBSS; gibco) and then incubated at 370C for 10min in 0.05% trypsin solution for lysis, followed by titration. Trypsin was then inactivated by HBSS and centrifuged to isolate cells. Hippocampal neurons were seeded in poly D-lysine-coated glass coverslips at a density of 0.5 x 105 neurons/coverslip using Neurobasal (GIBCO) at 370C. After 3 days of incubation, growth of non-neuronal cells were halted by exposure to 10 μM FDU-solution (5 μM uridine, 5 μM (+)-5-fluor-2′-deoxyuridine) for 48 h. Prior studies using this method has shown that these cultures have >85% neurons (Sattler et al.,

1997). Neuron cultures were incubated for 13 days, with regular medium changes, allowing them to mature to an adult-like stage.

For the oxygen glucose deprivation (OGD) paradigm, neurobasal was washed using HBSS solution without glucose, followed by another wash with deoxygenated HBSS solution. Cell cultures were then housed in an OGD chamber (Forma Anaerobic System, Fisher Scientific, Ottawa, Ontario) at 370C for 1-1.5hrs. Following OGD, medium was washed and replaced with phenol-red free neurobasal with 1ug/ml propidium iodide (PI staining), which is fluorescent when in contact with dead cells. Fluorescence is measured using at time zero (F0). Cell cultures

41 are then incubated for 24hrs and fluorescence intensity is measured again (F24). Then, cells are treated with 1mM NMDA which kills all remaining neurons and PI is measured (Fmax). We also applied 0.05% Triton-x 100 after NMDA treatment to kill off the remaining neurons and observed a 5-15% increase in PI fluorescence. Cell death is calculated by: (F24-F0)/(Fmax).

In some experiments, additional neuroprotectants: MCN cocktail (with 10uM MK801, 10uM CNQX and 2uM nimodapine) and NA1 (at 0.1uM) was also added. Concentrations used were shown to be neuroprotective in prior studies (Aarts et al., 2003; Aarts et al., 2002).

3.7 Immunofluorescence for quantification of NR2 subunit expression

Immunofluorescence and quantification of synaptic clusters procedures were previously described in Li et al. (2005). Briefly, mature hippocampal cell cultures were fixed in 4% paraformaldehyde and 4% sucrose in phosphate buffered saline (PBS) for 20 min, permeabilized with 0.1% Triton X-100 for 10 min at 4°C, and blocked with 10% goat serum in PBS for 1h at room temperature. Cultures were double-labeled using anti-VGLUT1 (a synaptic marker, vesicular glutamate transporter; Abcam) as well as with antibodies against NMDA receptor subunits (same as ones used in western blots). After washing, cells were incubated in fluorophore-labeled secondary anti-bodies (anti-species specific IgG). Immunofluorescence was visualized on a Nikon TE-300 inverted microscope (Japan) using 40x oil immersion lens. Images were captured using NIS-Elements (Nikon) and analyzed using Photoshop. In each culture 2-3 individual neurons were randomly selected and for each neuron a 2x50 µm area was analyzed in 3 different dendrites, giving a total of 24-27 dendrites per group. Fluorescence was normalized to the maximum intensity of the fluorophore channel and background fluorescence of each dendrite was subtracted. For each section, puncta representing NR2A/NR2B clusters and VGLUT clusters were individually counted. Images of NR2 subunit and VGLUT clusters were overlapped and each Nr2 cluster that overlapped with VGLUT cluster was used to quantify changes NMDAR NR2 subunit synaptic colocalization.

3.8 Statistical analysis

All statistical analyses were done on Graph Pad (Prism, La Jolla, CA USA). For direct comparison between two groups we used Student’s t-test. For multiple group comparisons one- way ANOVA with Bonferroni post-hoc was used. All data shown in mean ± SEM.

Raw data was used when possible; however in some cases data was expressed as a percent of the baseline. This was carried our primarily because: it makes it easier and simpler to display the data; and to reduce variability of the results, since changes are often seen relative to baseline.

The minimum level of significance used was α=0.05 for all experiments. Experiments that had a p-value less than 0.001 was shown as p<0.001. All tests had greater n-values than the minimum requirement and had a statistical power of greater than 80%.

Chapter 4

The results from this thesis have been divided into 4 chapters. Each chapter begins with the general aim of that chapter. This is followed by a brief introduction where we go over major points from the introduction that helped develop the experiment. The results section is broken down into subcategories dealing with different experiments and findings. Each result chapter will end with a brief discussion and summary of the major findings and some interpretation of that chapter. A broad discussion of all the findings of this thesis is found in Chapter 8 General Discussion.

4 The role of TRPM2 in MCAO-induced ischemic damage

Aim 1: To evaluate the role of TRPM2 channels in neuronal survival following cerebral ischemia in vivo.

Hypothesis 1: Following both transient and permanent cerebral ischemia in vivo, TRPM2(-/-) mice had reduced infarct volume and neuronal death compared to WT.

MCAO is a focal ischemia model that is widely used in in vivo stroke models. Following MCAO, TRPM2 mRNA has been shown to be upregulated, suggesting that TRPM2 may be required in ischemic cell death (Fonfria et al., 2006). Since earlier studies were limited to using non-specific blockers and RNA silencing to inhibit TRPM2 (Kaneko et al., 2006; Bai and Lipski, 2010), there has not been any in vivo study examining the contribution of TRPM2 channels to ischemia. In cultured neurons, the TRPM2 conductance is activated by reactive oxygen species and, particularly, H2O2 (Kaneko et al., 2006; Olah et al., 2009; Hara et al., 2002). However, published attempts to prevent oxidative stress-induced neuronal death by inhibiting TRPM2 have provided inconsistent results. Specifically, TRPM2 knockdown using antisense siRNA imparted neuroprotection to cortical neurons subjected to oxidative stress with H2O2 (Kaneko et al., 2006). However, reports using non-specific TRPM2 antagonists did not protect hippocampal neurons against lethal H2O2 (Bai and Lipski, 2010; Wilkinson et al., 2008). To resolve this for the in-vivo situation, we used TRPM2(-/-) mice to investigate whether the absence of TRPM2 imparted reduced vulnerability to cerebral ischemia following MCAO.

In this section the stroke models used were permanent and transient MCAO. The major difference between these two models is that tMCAO has reperfusion by removing the blockage, while in pMCAO the blockage is never removed. We employed TTC staining to measure volume of infarct and monitored both blood flow and body temperature. The advantage of using an in vivo model is that it is consistent with physiological conditions of stroke and it will determine if TRPM2 is a viable therapeutic target in whole animal, before time is invested pursuing in vitro models. The major objective of this chapter is to determine if a lack of TRPM2 protein reduces infarct volume following permanent and transient MCAO.

4.1 Results for Aim 1

4.1.1 The absence of TRPM2 reduces infarct volume following transient but not permanent cerebral ischemia.

In our first in vivo model, TRPM2-null and WT animals were subjected to a transient MCAO (tMCAO) for a duration of 1h followed by a 48h reperfusion. MCAO was achieved by inserting a plastic coated suture into the MCA via the common carotid artery. Infarct volumes were measured from each brain section after TTC staining (see Methods; Figure 5A, infarct depicted in white). Both WT and TRPM2(-/-) animals sustained cerebral infarcts as assessed 48h after the 1h tMCAO. The TRPM2(-/-) animals exhibited a 36.43 ± 16.65% (p<0.05;n=8) reduction in infarct volumes as compared with the WT controls (Figure 5A). Thus, the absence of TRPM2 protein imparts a state of reduced vulnerability to cerebral ischemia following tMCAO

Next, we explored if absence of TRPM2 had a similar effect in permanent MCAO (pMCAO). In this model the suture remains in the middle cerebral artery for 24h without reperfusion. At the 24h time point animals were sacrificed, and the brain was coronally cut into 8 sections for TTC staining. After measuring infarct volume in each section (Figure 5B, infarct in white), we found that there was no significant difference (p>0.05; n=5) in infarct volume in damaged brains from WT and TRPM2(-/-) mice (Figure 5B). This finding rejects part of our initial hypothesis, and demonstrates that lack of TRPM2 is not neuroprotective during permanent ischemia.

Overall, our findings in vivo suggest that TRPM2 plays a role in neuronal death during tMCAO, but not during pMCAO. The major difference between these two models is that

45 tMCAO allows reperfusion, while pMCAO does not. Since influx of oxygen into ischemic tissue is known to promote ROS production and H2O2 activates TRPM2-like currents (Olah et al., 2009; Hara et al., 2002), these findings support a model that TRPM2 is involved in neuronal death primarily during reperfusion following ischemia.

Figure 5 TRPM2(-/-) mice have reduced infarct volume compared to WT following 48h transient MCAO

MCAO was induced in both WT and TRPM2(-/-) animals. The brains were processed into coronal sections that were stained with TTC. (A) Infarcts assessed at 48h after a 1h tMCAO. Left panels: Representative coronal sections from WT and TRPM2(-/-) mice. White areas represent ischemic damage due to infarct. Right: Comparison of infarct volumes. There was a reduction of 36.43 ± 16.65% in infarct volumes in the TRPM2(-/-) animals (*p<0.05; n=8) compared to WT. (B) Infarcts assessed pMCAO. There were no significant differences between infarct volumes in WT and TRPM2(-/-) animals (p>0.05; n=5).

4.1.2 Measuring blood flow to confirm occlusion in WT and TRPM2(-/-) mice

In order to confirm that during MCAO there was lack of blood flow reaching the cerebrum we used both visual confirmation and laser Doppler on the ipsilateral side of the skull. First a small window was cut into the skin revealing the skull where the MCA and other blood vessels are visible prior to a stroke. Shortly after stroke induction the ipsilateral side of the skull goes white and the MCA can no longer be clearly seen in the skull (Figure 6). This discolouration of the skull visually confirms that MCAO reduced blood flow in the brain.

Using the same ipsilateral window in the head, we were able to quantitatively measure changes in blood flow using a laser Doppler. Blood flow was measured near the MCA prior to occlusion to determine normal blood flow, which was similar between WT and TRPM2(-/-) animals. Following MCAO, blood flow was again measured at consistent time intervals to determine if occlusion had occurred, these readings were normalized to the pre-MCAO measurement (Figure 7). The threshold for occlusion was set to >80% reduction in blood flow. In the case of tMCAO, the suture was removed at the 60min time interval and blood flow was measured for another 30 min post-MCAO. During the 30min reperfusion following tMCAO there was an increase in blood flow, although during that period it didn't return to pre-MCAO levels (Figure 7A). For pMCAO, we only measured occlusion up to the 60min time point to ensure that occlusion remained consistent for the first hour and there was no unexpected reflow near the MCA region. Using both visual assessment and laser Doppler, we determined that MCAO in both WT and TRPM2(-/-) mice resulted in a reduction in blood flow without any unexpected reflow during occlusion, and confirmed that the damage observed earlier was a result of a lack of blood flow.

Figure 6: Visual reduction of blood flow following MCAO

A small window in the skin on the ipsilateral side of the stroke exposed the skull and some of the vasculature along the cortex. Prior to MCAO (above image) the MCA and cortical vein are clearly visible and the skull has a red colour indicating that there is blood flow. During MCAO (bottom image) the MCA is no longer visible and the skull becomes white, indicating that region is not receiving blood. In the case of tMCAO, we observed that colour and blood flow began to return shortly after the suture occluding the MCA was removed.

Figure 7: Laser Doppler measurement of cerebral blood flow during MCAO

By exposing the skull around the MCA blood flow was quantitatively measured using laser Doppler. All measurements were normalized to pre-MCAO measurements and occlusion was determined by a >80% reduction in blood flow. (A) In tMCAO, blood flow was measured during the 60min occlusion as well as 30min post-MCAO. Both WT and TRPM2(-/-) animals had a >80% reduction in blood flow and had similar recovery reperfusion, although within the 30min of measurement blood flow did not return to pre-MCAO levels. (B) During pMCAO blood flow was measured for up to 60 min. Both WT and TRPM2(-/-) had >80% reduction in blood flow during that time period and there was no reflow in mice from either genotype.

4.1.3 Body temperature, and weight remain consistent between WT and TRPM2(-/-) animals following a MCAO

It is known that the severity of stroke damage is affected by a number of physiological factors. Two factors that can affect ischemic damage following MCAO are changes in body temperature and weight. A number of studies, have shown that hypothermia following a stroke can be neuroprotective in both pre-clinical rodent models and clinical human trials (Yanamoto et al., 2001;Reith et al., 1996). Furthermore, body temperature can influence blood flow through vasoconstriction/vasodilation in the areas surrounding the MCA as well as affecting contralateral blood flow. Both WT and TRPM2(-/-) mice had similar initial body temperatures at 34.6±1.2 and 35.0±0.9 0C, respectively (Figure 8A). Body temperature was measured using a rectal probe and maintained at a constant temperature of 37±0.50C with an electric blanket. In the tMCAO model, there was no significant difference in body temperature between WT and TRPM2(-/-) mice during the 60min of occlusion (Figure 8A). Similarly in the pMCAO model body temperatures remained the same between WT and TRPM2(-/-) animals during the 60min monitoring time (Figure 8B). Following the first 60mins of pMCAO, animals were housed in temperature controlled cages for the remaining 23hrs to ensure that body temperature was maintained at approximately 350C.

Both the initial body weight prior to stroke and the change in weight post stroke can influence the level of ischemic damage in mice. Body weight can be influenced by a number of factors including: fat content, insulin uptake and food intake, all of which can influence ischemic damage and recovery. To ensure that body weight did not have an effect on ischemic damage of post-MCAO of WT and TRPM2-null mice we compared both initial body weights and the change in body weight. Age-matched WT and TRPM2(-/-) mice had similar body weights at approximately between 22-26g prior to MCAO (Figure 9A,C). Having similar ischemic weight (and age) ensured that animals used were of similar size and health. Following tMCAO, both WT and TRPM2(-/-) animals respectively lost 18.1±2.0% and 18.4±2.0% of their original body weight (Figure 9B). While in pMCAO, WT and TRPM2(-/-) mice lost 13.5±2.0% and 15.3±1.5% of their initial body weight, respectively Figure 9D). We show here that absence of TRPM2 had no effect on body weight and the effects stroke has on weight loss, thus indicating that mice in both groups had similar health and eating habits.

Figure 8: Consistent body temperature is maintained during MCAO

Using a rectal probe body temperature of mice were measured prior and during MCAO. Body temperature was maintained at 37±0.50C for the first 60 min of MCAO (A) In tMCAO, there was no significant difference in body temperature between WT and TRPM2(-/-) mice. Furthermore, body temperature remained constant both during stroke and up to 30min post- stroke reperfusion. (B) During pMCAO, where there is no reperfusion, both WT and TRPM2(-/-) animals were able to maintain constant body temperature for the first 60 minutes of occlusion. There was also no significant difference between the two genotypes both before and during occlusion.

Figure 9: pMCAO and tMCAO reduces body weight in WT and TRPM2(-/-) animals

Body weight of both WT and TRPM2(-/-) mice were measured before MCAO and at the time of sacrifice. (A) For tMCAO experiments, age-matched WT and TRPM2(-/-) mice had similar body weights before (WT: 24.1±2.5mg; TRPM2(-/-): 22.5±1.9mg) MCAO and after 48h reperfusion (WT: 19.7±2.7g; TRPM2(-/-): 18.3±1.7g). (B) The percentage of body weight lost following tMCAO between WT and TRPM2(-/-) animals was also consistent at 18.1±2.0% and 18.4±2.0%, respectively. (C) In pMCAO, WT and TRPM2(-/-) mice also had similar pre-MCAO (WT: 25.8±1.7g; TRPM2(-/-): 26±2.2g) and post-MCAO body weights (WT: 22.3±1.5g; TRPM2(- /-): 22±1.9g). (D) As well as similar percentage loss of total body weight (WT: 13.5±2.0%; TRPM2(-/-):15.3±1.5%).

4.1.4 WT and TRPM2(-/-) animals exhibit loss of motor function following stroke.

One of the indicators of a stroke is a loss of motor function, and often motor deficit is used as an indicator of stroke or as a measurement of stroke severity. Here we used a qualitative measurement of motor deficit to determine whether a loss of motor function was correlated with the stroke. Prior to sacrifice, the mice were held by the tail and allowed to walk on a flat surface. If the animal walked normally without difficulty they were given a score of 5, and if they were unable to walk they were given a zero indicating a complete loss of motor function (Zhang et al., 2006). All animals used for MCAO had to have scored a 5 prior to MCAO, if animals did not meet this criteria they were not used. Following tMCAO both WT and TRPM2(-/-) animals had similar motor deficits of an average score of 1 (Figure 10A). While following pMCAO WT and TRPM2-null mice had an average neurological score of 2 (Figure 10B).

Although it seems that mice that underwent pMCAO had slightly better neurological score than those that had tMCAO, it is difficult to definitively say which procedure was more severe since this paradigm only uses visual observations. Furthermore, we are unable to determine if there are minor differences in motor deficits between WT and TRPM2(-/-) animals. What can be concluded from this experiment is that both tMCAO and pMCAO does result in significant motor deficits in WT and TRPM2(-/-) animals.

Figure 10: Neurological score demonstrates reduction in motor function post-MCAO

Motor deficit post MCAO was measured using a 5-point neurological score test (described in Zhang et al., 2006). Score of 5 was given to animals that had no motor deficits, while 0 was given to animals with no motor function. (A) Following tMCAO, both WT and TRPM2(-/-) animals had an average score approximately 1. (B) While in pMCAO, both genotypes had a similar average neurological of 2. Animals were tested both prior to MCAO and before sacrifice. These findings show that the damage due to MCAO results in an obvious motor deficit in WT and TRPM2(-/-) mice.

4.2 Discussion

Earlier studies trying to determine whether TRPM2 channels are involved in ischemic cell death were limited to non-specific antagonists and RNA-silencing (Bai and Lipski, 2010;Wilkinson et al., 2008;Kaneko et al., 2006). Due to the incomplete inhibition of TRPM2 and the potential inhibition of other channels, these earlier studies provided inconclusive results; whereas gene silencing of TRPM2 provided neuroprotection following ischemic conditions (Kaneko et al., 2006), non-specific antagonists against TRPM2-like currents failed to replicate this neuroprotection. With the development of a TRPM2(-/-) mouse (Yamamoto et al., 2008), we were able to use a model that had a complete absence of TRPM2 channels to determine if TRPM2 plays a role in ischemic death. Furthermore, unlike earlier studies which were done only in vitro, this study was done in vivo, to ensure that the TRPM2(-/-) model had implications in whole animal as opposed to just cultures.

By using an in vivo model of stroke there are potential limitations, primarily, there are a number of other cell-types and physiological factors that can contribute to ischemic damage. Although it is impossible to consider all the potential physiological changes caused by whole animal absence of TRPM2 we did manage to maintain similar age, blood flow, brain and body weight, and body temperature between WT and TRPM2(-/-) animals. Body temperature is major modulator of ischemic damage, where lowering cerebral temperature via drugs or by reducing cerebral temperature can reduce ischemic damage (Yanamoto et al., 2001;Corbett et al., 1990). In thus study we were limited to maintaining/measuring core temperature, which usually influences and is an indicator of cerebral temperature (Busto et al., 1989).

A potential caveat into our findings is how the absence of TRPM2 in mice may affect non-neuronal cells and vasculature. In terms of vasculature we visually compared brains from both WT and TRPM2(-/-) animals and found most major arteries (including the MCA) were present in both WT and TRPM2(-/-) mice. What we did not measure was specific neurovascular attributes such as the length and diameter of vessels, which would require histological studies. In addition to neurons, a number of other non-neuronal brain cells have been suggested to express TRPM2, primarily microglia (Kraft et al., 2004). Following ischemia, microglia act as the first and primary immune response, whereas, TRPM2 is required for microglia function (Kraft et al., 2004), absence of TRPM2 affect post-ischemic neuronal damage via microglia instead of just

62 neurons. Later in this thesis we will discuss absence of TRPM2 in primary neuronal cultures (Chapter 7) and further discuss the potential role of TRPM2 non-neuronal cells (Chapter 8).

We show here that a chronic absence of TRPM2 reduces ischemic-reperfusion damage in vivo, but not in the ischemic only model. There are two possible explanations to resolve the different results observed in these models, either TRPM2 plays a role in cell death during reperfusion or the pMCAO model was so severe that any cytoprotective effects of the TRPM2(-/-) was masked. Overall, by using an in vivo model we have validated in vitro studies using RNA silencing and have shown that TRPM2 can be a therapeutic target for stroke in whole animal, which has potential implications in clinical treatments.

4.3 Summary

In summary, we found that the absence of TRPM2 reduced infarct volume in 1hr tMCAO followed by 48hr reperfusion. In both cases, WT and TRPM2(-/-) animals had an at least 80% reduction in cerebral blood flow during the first hour and temperature was maintained at approximately 37.0 ± 0.5°C. These findings partially support our initial hypothesis, that TRPM2(- /-) is neuroprotective following in vivo transient ischemia. However, it does not support the hypothesis that TRPM2(-/-) is protective in permanent ischemia, where absence of TRPM2 failed to provide neuroprotection in pMCAO. Since tMCAO has reperfusion, unlike pMCAO, we suggest TRPM2 plays a role in cellular death primarily during reperfusion following ischemia.

Chapter 5 5 The role of TRPM2 in synaptic activity during oxidative stress

Aim 2: To determine if TRPM2 channels play a role in neuronal synaptic transmission and plasticity under control and oxidative stress conditions.

Hypothesis 2: Hippocampal slices from TRPM2(-/-) animals will be immune to inhibitory effects of oxidative stress on synaptic plasticity.

Following infarct there is often a disruption in synaptic function mediated by an over production of ROS. One ROS, H2O2, has been shown to modulate both normal synaptic function and plasticity (Colton et al., 1989; Kamsler and Segal, 2003). Previous studies examining the effects of H2O2 on LTP have been shown that sub-lethal concentrations (20-500 uM) inhibit LTP formation with no effect on baseline fEPSP (Kamsler and Segal, 2003; Maalouf and Rho, 2008). Our lab has demonstrated that silencing of TRPM7 channels protects the LTP formation in the hippocampus after 30 days of reperfusion following global ischemia. Functional TRPM2 channels are expressed in the hippocampus and have been shown to play a role in plasticity, specifically modulating LTD (Xie et al., 2011). Thus, using whole hippocampal slice, we examined the effects of sublethal concentrations of H2O2 in hippocampal slices taken from WT and TRPM2(-/-) animals.

Here we measured evoked excitatory field potentials in CA1 hippocampal neurons from WT and TRPM2(-/-) mice. We first measured changes in baseline and LTP formation under normal and H2O2 conditions. Then we used paired-pulse stimulation to determine if changes in

H2O2-induced synaptic activity were due to a post synaptic of pre-synaptic phenomenon. This section will give insights into how TRPM2 affects synaptic activity under normal and stressed conditions.

5.1 Results for Hypothesis 2

5.1.1 TRPM2(-/-) increases synaptic excitability during oxidative stress

Before investigating the effects TRPM2(-/-) in synaptic plasticity paradigms, we looked at whether absence of TRPM2 affects baseline synaptic activity under normal and oxidative stress conditions. We used hippocampal slices from WT and TRPM2(-/-) animals, stimulated the Schaffer collaterals and measured evoked fEPSP from CA1 neurons under control and stressed conditions. First we investigated evoked excitability in WT and TRPM2(-/-) hippocampal slices under control conditions (superfusion with normal ACSF). CA1 neurons from WT and TRPM2(- /-) mice were both able to produce baseline fEPSPs and we found that there was no difference in fEPSP slope between slices from the two genotypes (p>0.05, n=8; Figure 11a,b). These findings are consistent with recent data demonstrating that absence of TRPM2 had no effect on baseline fEPSP in the CA1 dendritic field of hippocampal slices under control ACSF conditions (Xie et al., 2011).

Next, we looked in to how baseline fEPSPs in hippocampi from TRPM2-null mice respond to oxidative stress. Previous studies have found that bath application of 200uM of H2O2 has no effect on baseline fEPSP, however it does inhibit LTP formation (Maalouf and Rho,

2008; Vereker et al., 2001). We confirmed that 200uM of H2O2 had no effect on baseline fEPSP in WT CA1 neurons (p>0.05; n=8; Figure 11A,B), however when applied to slices from TRPM2(-/-) animals, there was a 49.0±3.4% increase in baseline fEPSP slope (p<0.01; n=8;

Figure 11A,B). This increase in fEPSP slope lasted through the 60min duration of H2O2 application (Figure 11D), and when H2O2 was removed from the bath solution, the baseline of (-/-) TRPM2 slices returned back to control levels (Figure 11C). This suggests that H2O2-induced changes in fEPSP in TRPM2(-/-) hippocampus is long lasting and reversible.

These findings were surprising in the context of the relative resilience of TRPM2(-/-) mice to cerebral ischemia, since we would expect neuroprotection to be associated with a reduced synaptic strength in the face of oxidative stress rather than an increase in synaptic strength.

Moreover, H2O2 is a known activator of TRPM2 currents (Kraft et al., 2004), so the absence of TRPM2 should reduce excitability. Our counterintuitive findings of increased synaptic strength suggest that the absence of TRPM2 has unexpected effects on the channels that are responsible for these synaptic responses.

(-/-) Figure 11: 200uM of H2O2 increases evoked baseline fEPSP slope in TRPM2 with no effect on WT

Field potentials (fEPSP) evoked via stimulation in the Schaffer collaterals (at 100- 200µA) were recorded in the Hippocampal CA1 region in slices from TRPM2(-/-) and WT mice under control (ACSF) and oxidative stress (ACSF+ 200uM H2O2) conditions. H2O2 was administered through bath application. (A-B) Both WT and TRPM2(-/-) had similar raw fEPSP slope during baseline recordings (black bar in graph); and during H2O2 exposure (grey bar) TRPM2(-/-) slices showed a 49.0± 3.4% increase (p<0.05) in fEPSP slope. Representative fEPSP recordings (A) using grey and black to represent with and without H2O2, respectively, showed (-/-) (- this effect in TRPM2 . (C) A longer application of H2O2 caused fEPSP to increase in TRPM2 /-) (n=7; shown in grey) for as long as it is applied (60 min), without any effect in WT (n=5; (-/-) shown in black). (D) The H2O2 induced increase in fEPSP in TRPM2 (shown in grey; n=8) was found to be reversible when application of 200uM H2O2 (white bar) was removed, with no effect on WT (shown in black; n=8).

5.1.2 TRPM2(-/-) has no effect on LTP inhibition during oxidative stress conditions

TRPM2 channels have been implicated to play a role in plasticity in hippocampus, where in the absence of TRPM2 there is a reduction in LTD when compared to WT under ACSF conditions (Xie et al., 2011). Since 200uM of H2O2 inhibits LTP formation in normal hippocampal slices (Maalouf and Rho, 2008; Vereker et al., 2001) and is a known activator of (-/-) TRPM2, we predict that TRPM2 will prevent H2O2-induced inhibition of LTP in hippocampal

CA1 neurons. To examine the impact of oxidative stress on LTP induction, 200uM of H2O2 was applied 10 min prior to LTP induction. To induce LTP in hippocampal slices, we used the high- frequencey stimulation (HFS) paradigm, where 1s of 100 Hz HFS in the Schaffer collaterals, and LTP is measured in CA1 neurons.

Under ACSF conditions, hippocampal slices from TRPM2(-/-) and WT animals were able to form LTP (Figure 12A,B). There was no significant difference in the level of potentiation between WT and TRPM2(-/-) groups and both maintained LTP for the 60min recording time. This data is consistent to earlier findings from Xie et al. (2011), that showed that TRPM2(-/-) had no effect on LTP formation under control conditions.

When 200uM of H2O2 was applied on WT slices prior to LTP induction, there was no effect on baseline fEPSP (as was shown earlier). Shortly after LTP was induced via HFS, these WT slices failed to form LTP and fEPSP returned to baseline (n=10;A). These observations are similar to what was demonstrated in previous studies (Vereker et al., 2001; Maalouf and Rho, (-/-) 2008). When H2O2 was applied to hippocampal slices from TRPM2 animals there was an increase in baseline fEPSP slope prior to HFS, which is similar to observations made in this chapter. Following LTP induction, fEPSP slope returned to baseline levels with H2O2 (n=10;

Figure 12B). This result suggests either that the H2O2-induced increase in synaptic strength in TRPM2(-/-) is not occurring through the same mechanisms that induce LTP, or that pre-treating (-/-) slices with H2O2 before LTP induction causes a ceiling effect in the excitability of TRPM2

CA1 neurons. However, the latter explanation is unlikely, because when H2O2 was removed from the bath solution following LTP induction, fEPSP slope returned to the older baseline levels seen before the H2O2 application (n=5; Figure 13).

(-/-) Figure 12: 200 uM of H2O2 inhibits LTP formation in both WT and TRPM2

Synaptic plasticity was induced by stimulating the Schaffer collaterals and measuring changes in fEPSPs in the dendrites of CA1 hippocampal neurons. (A-B) LTP was induced by

HFS and fEPSP were recorded both pre-HFS (for 20 min) and post-HFS (for 60min). H2O2 (200uM) was applied at 10 min before HFS (white bar). Slices from WT (n=10; A) and TRPM2(- /-) (n=10;B) animals were able to form LTP under control conditions (ACSF; in black), and LTP (-/-) formation was blocked by 200uM H2O2 (shown in grey) in both WT and TRPM2 . Black dotted line shows baseline fEPSP under ACSF and grey dotted line shows baseline of TRPM2(-/-) with H2O2 (only in B).

Figure 13 Removal of H2O2 following LTP induction returns fEPSP to original baseline in TRPM2(-/-) hippocampal slices

H2O2 was applied 10 minutes before LTP induction and 30 minutes after LTP induction in hippocampal slices from TRPM2(-/-) mice (shown in gray bar). Initial application of 200uM

H2O2 increased baseline fEPSP to a new baseline. Following LTP induction fEPSP returned to the baseline with H2O2 and after H2O2 was removed fEPSP returned to original baseline, (-/-) suggesting that 200uM H2O2 inhibited LTP induction in TRPM2 slices (n=5). Black dotted line shows baseline fEPSP under ACSF and grey dotted line shows baseline with H2O2

5.1.3 Paired-pulse facilitation suggests oxidative stress-induced excitability in TRPM2(-/-)is mediated by changes in the post- synapse

Our data suggest that LTP induction by HFS and LTP inhibition by H2O2 are unaffected by the absence of TRPM2. Thus, to further explore the source of the hyper-excitability we evaluated whether H2O2 is affecting the presynapse, causing increased neurotransmitter release, or whether it is affecting the postsynapse by activating more excitatory glutamate receptors. To resolve this we used paired pulse facilitation. Each paired-pulse sweep consisted of two pulses evoked at rapid succession with varying (20-1000ms) interpulse intervals. Facilitation occurs solely in the presynapse, as a result of an increased probability of neurotransmitter release induced by Ca2+ entry evoked by the first pulse.

We found no change in paired-pulse ratio at different interpulse intervals when comparing hippocampal slices from WT and TRPM2(-/-) mice under normal ACSF condition

(p>0.05; n=10; Figure 14). Previously, it has been shown that 200uM of H2O2 has no effect on paired-pulse ratios in the hippocampus (Colton et al., 1989). In this study, application of 200uM (-/-) of H2O2 to hippocampal slices from WT and TRPM2 animals resulted in no change in paired- pulse ratio of WT and TRPM2(-/-) slices when compared to control slices in ACSF (p>0.05; n=10; Figure 14). This suggests that H2O2 does not function in the presynapse to increase excitability in TRPM2(-/-) CA1 neurons, and therefore, oxidative stress is likely to be modulating evoked synaptic excitability in the post synapse.

(-/-) Figure 14: Sub lethal H2O2 has no effect on paired pulse facilitation in WT and TRPM2 hippocampus

In paired pulse facilitation two pulses with the same amplitude (100-200uA) are given at different interpulse intervals. Facilitation only occurs if there are changes in the presynapse, affecting release of neurotransmitters. WT and TRPM2(-/-) hippocampal slices were used under control and H2O2 conditions. Both groups had maximum facilitation at 50ms interpulse interval and TRPM2(-/-) had no change in paired pulse facilitation when compared to WT in both control

(ACSF) and with 200uM H2O2 conditions (p>0.05; n=10).

5.2 Discussion

Prior to this study, very little was known about the role of TRPM2 in synaptic transmission and plasticity. The recent study by Xie et al. (2011) demonstrated that under control conditions, absence of TRPM2 in hippocampal slices reduced LTD. The study also showed that TRPM2-null slices were able to form similar LTP as the WT slices (Xie et al., 2011). An activator for TRPM2, H2O2, is known to be produced following ischemia and does play a role in modulation in LTP (Kamsler and Segal, 2003; Maalouf and Rho, 2008). A related channel to TRPM2, TRPM7, is activated by ROS, ONOO-, and our lab has demonstrated that the knockdown of TRPM7 channels provides neuroprotection following global ischemia and protects against ischemia-induced disruption of LTP (Sun HS et al., 2009). We find here that (-/-) hippocampal slices from TRPM2 animals are sensitive to H2O2, but the absence of TRPM2 did not provide protection from oxidative stress-induced disruption of LTP. These findings suggest that under oxidative stress conditions, TRPM2 acts as a modulator of synaptic excitation, via mechanisms that have no effect on LTP.

Since the absence TRPM2 provides neuroprotection following ischemia-reperfusion in vivo, it is paradoxical to find that oxidative stress would increase overall synaptic excitability, a phenomenon generally associated with excitotoxicity. The only time that increased synaptic excitability is associated with neuroprotection is during a phenomenon known as preconditioning, where sublethal concentrations of NMDA can prime neurons to be resistant to ischemic damage, while at the same time increase synaptic excitability(Soriano et al., 2006). Based on the results in this chapter it is premature to make that conclusion and we do further investigate this in the next chapter.

Considering our mouse model uses a chronic absence of TRPM2 channels, it is possible that the increased sensitivity to H2O2 may be mediated via a compensatory mechanism. In this scenario, the lack of TRPM2 in neurons may be influencing the activation and sensitivity of 2+ other excitatory channels to H2O2 in order to compensate for the loss of Ca -influx via TRPM2 (-/-) channels. Since paired-pulse experiments suggest that this sensitivity to H2O2 in TRPM2 slices occurs in the post synapse, it is likely that this phenomenon may be mediated by glutamate receptors, which are main contributors to synaptic excitation. How TRPM2(-/-) neurons become sensitive to H2O2 remains unclear, however we hypothesize that an absence of TRPM2-mediated

73 calcium influx leads to a hypersensitisation of glutamate receptors and other cation channels to

H2O2, ultimately leading to increased synaptic excitability.

5.3 Summary

In summary, we found that sub-lethal H2O2 exposure increased baseline evoked fEPSP in

CA1 neurons from the hippocampus. Previous findings showing 200uM of H2O2 inhibited LTP formation in the CA1 was confirmed in this chapter. Our initial hypothesis suggesting that TRPM2 channels played a role in LTP was incorrect, since the TRPM2(-/-) had no effect on LTP and absence of TRPM2 had no effect on preserving LTP following oxidative stress. Finally, we (-/-) found that TRPM2 had no effect on paired-pulse facilitation, under normal or H2O2 conditions, suggesting that H2O2-induced increase in synaptic excitability is due to changes in the post-synapse.

Chapter 6

6 TRPM2 modulates NMDAR-dependent cell death mechanisms

Aim 3: To determine if TRPM2 channels modulate post synaptic glutamate receptors, resulting in neuroprotection and modulation of synaptic activity.

Hypothesis 3: Absence of TRPM2 promotes activation of NR2A containing NMDAR and inhibits NR2B containing NMDAR, leading to NR2A mediated neuroprotection.

In the postsynapse, ionotropic glutamate receptors are responsible for the EPSPs and have roles in a numerous cellular mechanisms. Under normal conditions glutamate receptors play a role in neuronal excitability and plasticity (Bliss and Collingridge, 1993) but, under ischemic conditions they often lead to cell death (Lau and Tymianski, 2010). Given the effects of the absence of TRPM2 on excitability and ischemic vulnerability we evaluated whether this might be explained by alterations in glutamate receptor expression. Specifically, we focused on the expression of NR2A and NR2B subunits because among NMDAR subunits, NR2A and NR2B, are known to be involved in a number of cellular functions related to cell death and survival and are found in different locations in the membrane (Liu et al., 2007b; Chen et al., 2008; Zhou and Baudry, 2006; Terasaki et al., 2010). In mature neurons, NR2B containing NMDARs are predominantly located extrasynaptically and are involved in cell death mechanisms as well as synaptic inhibition (Liu et al., 2007b; Soriano et al., 2008). By contrast, NR2A containing NMDARs are concentrated at synapses and are involved in cell survival mechanisms, as well as synaptic excitation (Liu et al., 2007b; Hardingham and Bading, 2010; Terasaki et al., 2010). Recently, TRPM2-null mutation has been suggested to reduce GLUR1 expression, with no effect on NMDAR expression in the hippocampus (Xie et al., 2011).

Here, we measured expression and functional changes of glutamate receptors due to the absence of TRPM2. Using Western blots, we measured protein expression of both AMPA and NMDA receptors from WT and TRPM2(-/-) hippocampi. Next, we evaluated changes in synaptic expression of NR2 subunits by using immunofluorescence on WT and TRPM2(-/-) CA1 neurons. Then we used selective-NMDAR subunit inhibitors to determine which NR2 subunit is involved

in H2O2 sensitivity in TRPM2-null hippocampus. Finally, using western blots, we investigated if changes in glutamate receptor expressions resulted in changes in downstream pro-survival and pro-death mechanisms. This goal of this section is to provide a possible mechanistic explanation of the paradoxical findings of how chronic absence of TRPM2 results in both neuroprotection and increased synaptic excitability following stress conditions.

6.1 Results for Hypothesis 3

6.1.1 The absence of TRPM2 alters NMDAR NR2 subunit expression

Using Western blots from whole hippocampi from WT and TRPM2(-/-) mice, we first measured glutamate receptor expression. We found that there was no change in amount of protein expression of AMPAR subunits GluR1 and GluR2 (p>0.05; n=4; Figure 15A,B). However, in TRPM2(-/-) animals, NR2B expression levels were decreased by 46.21±3.18% (n=3;p<0.01;Figure 16C). Concomitantly, NR2A subunit expression and phosphorylation was increased in the TRPM2(-/-) hippocampi by 43.30±2.33% and 52.31± 3.11%, respectively (n=3; p<0.01 Figure 16B,C), as compared to WT. The expression of phospho-NR2B was also reduced by 39.55±1.22% in TRPM2(-/-) (n=3; p<0.01; Figure 16D), suggesting that this reduction in basal expression may also translate to a reduction in NR2B function. The levels of NR1 subunit protein, which binds to NR2A and NR2B was unchanged (p>0.05;Figure 16A). These findings suggest a novel role for TRPM2 in the regulation of glutamate receptors. Specifically the deficiency of TRPM2 results in marked alterations in the ratios of expressed NR2A and NR2B subunits without affecting the total expression of NMDARs.

NR2A and NR2B, are known to be involved in a number of cellular functions related to neuronal survival and death, and have also been suggested to be located at different locations in the membrane. (Liu et al., 2007b; Chen et al., 2008; Zhou and Baudry, 2006; Terasaki et al., 2010). Specifically, NR2B containing NMDARs, are located primarily located in the extrasynapse (Liu et al., 2007b; Soriano et al., 2008), while NR2A containing NMDARs, are concentrated at the synapse (Liu et al., 2007b; Hardingham and Bading, 2010; Terasaki et al., 2010). Based on the results from the Western blots, we asked if absence of TRPM2 causes changes in NR2A and NR2B expression at the synapse in CA1 hippocampal neurons.

To further interpret our findings we examined whether the changes in NR2A and NR2B expression induced by the absence of TRPM2 can be visualized in hippocampal CA1 neurons. To this end we measured the total number of NR2A and NR2B clusters and the synaptic colocalization of these clusters in individual dendrites of low-density cultures taken from the CA1 sectors of both WT and TRPM2(-/-) animals. First, we found that CA1 neurons express both NR2A and NR2B in both WT and TRPM2(-/-) (Figure 17A,B; 18A,B). By counting number of clusters of NR2A and NR2b subunits, we found that absence of TRPM2 caused a decrease in NR2B clusters (n=24 dendritic segments; p<0.05; Figure 18B), and no change in NR2A clusters (n=27; p>0.05;Figure 17B) as compared to WT. Next, using synaptic marker Vglut to determine synaptic colocalization, we found that TRPM2(-/-) exhibited a reduction in the number of synapses containing NR2B as compared to WT (n=24; p<0.05; Figure 18D), whereas there was no significant change in the number of synapses containing NR2A (n=27; p>0.05; Figure 17D). All CA1 neurons shown here came from E18 male pups, however E18 mixed gender cultures showed similar results. These findings suggest that a chronic absence of TRPM2 reduces the numbers of NR2B clusters and NR2B localization at the synapse, and thus changes the ratio of NR2A to NR2B containing NMDARs at the synapse to promote signaling through NR2A- containing NMDARs.

Figure 15: Hippocampi from TRPM2(-/-) mice have has reduced NR2B expression and increased NR2A expression.

Western blots detecting glutamate receptor expression in hippocampal protein extracts from 3-8 week old TRPM2(-/-) and WT animals. Densitometric analysis showed that there was no significant change in AMPA subunits GluR1 and GLur2 (p>0.05; n=4; A-B).

Figure 16: Absence of TRPM2 reduces NR2B-subunit expression and increases NR2A subunit expression

Western blots detecting glutamate receptor expression in hippocampal protein extracts from 3-8 week old TRPM2(-/-) and WT animals. (A) NMDA subunit NR1, which binds to NR2 subunits did not have any significant change in expression. Expression of NR2 subunits in NMDA did however change, (B-C) where basal and phosphorylated NR2A expression was increased by 43.30 ± 2.33% and 52.31± 3.11%, respectively in the TRPM2(-/-) (grey bar) when compared to WT (black bar; n=3; *p<0.01); and (D-E) both basal and active NR2B is reduced in TRPM2(-/-) by 46.21 ± 3.18% and 39.55±1.22%, respectively, when compared to WT (n=3; *p<0.01).

Figure 17: Absence of TRPM2 has no effect on synaptic co-localization of NR2A subunit

Immunofluorescence of detecting NR2A clusters in hippocampal CA1 neurons from WT and TRPM2(-/-) animals. (A) To detect synaptic NR2A, equal sized dendritic segments were used to count labeled clusters, where green fluorescence represents synaptic marker VGLUT expression and red represents NR2A subunit expression. Overlaid images were used to count colocalized clusters. (B) WT and TRPM2(-/-) had no difference in number of NR2A clusters (n=27; p>0.05). (D-E) There was also no change in number of synapses and NR2A synaptic colocalization between neurons from WT and TRPM2(-/-) animals (n=27; p>0.05).

Figure 18: TRPM2(-/-) reduces synaptic co-localization of NR2B

Immunofluorescence of detecting NR2A NR2B clusters in hippocampal CA1 neurons from WT and TRPM2(-/-) animals. (A) To detect synaptic NR2B, equal dendritic segments were used to count labeled clusters, where red fluorescence is VGLUT and green is NR2B. Overlaid images were used to count colocalized clusters. (B) The number of NR2B clusters were reduced in hippocampal neurons from TRPM2(-/-) animals in comparison to those from WT animals (n=24;*p<0.05). (C-D) NR2B synaptic colocalization was also reduced in TRPM2-null neurons when compared to WT (n=24; *p<0.05), with no effect on total number of synapses. (H) Colocalization of NR2B and VGLUT was reduced in TRPM2(-/-), when compared to WT.

(-/-) 6.1.2 NR2A is involved in H2O2 induced synaptic excitability in TRPM2

Thus far we have shown that chronic absence of TRPM2 results in increased excitability in hippocampal neurons in response to oxidative stress and that synaptic NR2A/NR2B ratios are altered in a direction that may promote pro-survival signaling. We next explored whether the change in NMDAR subunit expression was functionally related to the observed increase in excitability. As previously noted, bath application of 200uM of H2O2 increased excitability of CA-1 neurons in slices taken from TRPM2(-/-) mice. However, an application of the NR2A- selective antagonist, PEAQX (0.4uM; >10 fold selectivity of NR2A over NR2B; (Auberson et al., 2002;Feng et al., 2005) 10 min after the H2O2 challenge significantly reduced the fEPSP slope, though fEPSPs did not return to the same baseline levels as in the absence of H2O2 (n=7 p<0.05; Figure 19a). By contrast, the NR2B-selective antagonist, RO25-6981 (0.5 uM), had no (-/-) effect on the H2O2-induced increased excitability in the TRPM2 hippocampal slices (Figure 19A).

Neither the NR2A, nor the NR2B antagonists affected baseline fEPSP slope in both the WT and TRPM2(-/-) hippocampus (Figure 19B, 20B) Moreover, when antagonists were applied prior to the application of H2O2, the NR2B antagonist did not inhibit the increase in fEPSP slope, while the NR2A antagonist significantly blunted this increase in the TRPM2(-/-) neurons (Figure 19B,D). These findings demonstrate that NMDARs containing the NR2A subunit are responsible for mediating the increase in synaptic excitability observed when TRPM2(-/-) hippocampal neurons are subjected to oxidative stress and that this phenomenon is paralleled by an increase in NR2A expression in the absence of TRPM2. It should be noted that the NR2A specific antagonist does not completely inhibit the increase in synaptic excitability due to oxidative stress, suggesting that other channels may still be involved.

Figure 19: NR2A modulates oxidative stress induced excitability in TRPM2(-/-) hippocampal neurons

By stimulating Schaffer collaterals (100-200uA), evoked field potentials (fEPSPs) were measured in the CA1 hippocampal dendritic region in slices from TRPM2(-/-) and WT animals.

(A) Following 10 min of baseline fEPSP recording, 200uM H2O2 (white bar) was applied to induce an increase in fEPSP slope in TRPM2-null slices, with no effect on WT (as described earlier). Bath application of NR2A-specific antagonist (0.4uM of PEAQX, black bar) at 20min (-/-) reduced the H2O2-induced increase in fEPSP slope in TRPM2 slices (n=7 p<0.05), with no effect on WT. (B) When NR2A-specific antagonist was applied at 10min without H2O2 there was no change in baseline fEPSP slope in both WT and TRPM2(-/-) slices. At 20min 200uM H2O2 was applied, resulting in a smaller increase (when compared to H2O2 only) in fEPSP slope in TRPM2(-/-) slices and had no effect in WT. All compounds were diluted in ACSF.

Figure 20: NR2B has no effect on oxidative stress induced excitability in TRPM2(-/-) hippocampal neurons

By stimulating Schaffer collaterals (100-200uA), evoked field potentials (fEPSPs) were measured in the CA1 hippocampal dendritic region in slices from TRPM2(-/-) and WT animals. (A) Application of NR2B-specific antagonist (0.5uM RO256981, grey bar) following 10 min of (-/-) 200uM H2O2 had no effect on the H2O2 induced increase in fEPSP slope seen in TRPM2 slices (n=7). (b) NR2B- specific antagonist, when applied prior to H2O2 application, had no effect on baseline fEPSP slope (n=7) in both TRPM2(-/-) and WT slices, nor did it have any effect on (-/-) oxidative-stress induced excitability in TRPM2 slices when H2O2 was applied. All compounds were diluted in ACSF.

6.1.3 TRPM2(-/-) increases Akt and ERK Activity.

Changes in NR2A and NR2B expression and activity in normal cells can result in the activation of pro-survival downstream pathways that allows neurons to be preconditioned to neuroprotection. An increase in NR2A containing NMDAR activity results in an increase in phosphorylation of Akt and ERK (Papadia et al., 2005; Hardingham, 2006b; Hardingham et al., 2001a), leading to the promotion of pro-survival mechanisms in the cell. By contrast, an increase in NR2B containing NMDAR activity results in an inhibition of pro-survival mechanisms and promotion of pro-death mechanisms (Liu et al., 2007b; Soriano et al., 2008). Using western blots from whole hippocampi dissected from WT and TRPM2(-/-) mice, we found that the basal expression levels of Akt and ERK proteins were similar in both (Figure 21A,D). However, upon comparing the levels of the active (phosphorylated) form of Akt and ERK, there was an increase in both phospho-Akt (by 49.39±8.26% in Ser 473; 65.11±13.78% in Thr-308;n=4;p<0.01) and phospho-ERK (by 76.44±23.12%;n=4;p<0.01) ), respectively, in the TRPM2(-/-) hippocampus (Figure 21B,C,E).

6.1.4 TRPM2(-/-) inhibits GSK3β and PSD95, but has no effect on TRKβ

Glycogen synthase kinase (GSK3β) is an important target of Akt, which is involved with mitochondrial BAX translocation during apoptosis (Linseman et al., 2004). Akt is involved in neuroprotection by inhibiting GSK3β via phosphorylation of serine 9 (Crowder and Freeman, 2000;Hetman et al., 2000). If TRPM2(-/-) is neuroprotective via pro-survival NMDAR signalling then Akt should be inhibiting GSK3β through its phosphorylation. Compatible with this hypothesis, we found using Western blots that in TRPM2(-/-) hippocampi there was a 38.96±4.97% (n=3; p<0.01) increase in phospho-GSK3β as compared to WT (Figure 21F). These findings suggest that TRPM2(-/-) is neuroprotective via the Akt-pathway, by inhibiting downstream GSK3β.

PSD-95 is known to interact with NR2B and is involved in maintaining NR2B surface clustering (Chung et al., 2004) and in signalling neuronal death (Aarts et al., 2002; Sattler et al., 1999). Since, TRPM2(-/-) results in a reduction of NR2B clustering in hippocampal neurons, we questioned whether it might also inhibit the expression of PSD-95. To determine this we used whole hippocampus for Western blots and found that absence of TRPM2 resulted in a 45.61 ±

12.76% (n=3;p<0.01) reduction in PSD-95 expression as compared with WT (Figure 22a). The reduction of PSD-95 and NR2B further suggests that TRPM2(-/-) inhibits both NR2B clustering and pro-death signaling.

Neuroprotection by preconditioning with NMDA has been shown to cause a rapid release in BDNF, which activates postsynaptic BDNF receptor TRKB (Jiang et al., 2005). We used Western blots to see whether TRPM2(-/-) is involved in regulating BDNF release leading to neuroprotection. The phosphorylation of TRKB was the same between WT and TRPM2(-/-) hippocampus (Figure 22B), suggesting that TRPM2 does not play a role in BDNF mediated neuroprotection.

Figure 21: TRPM2(-/-) increases activity in ERK and AKT pathways

Western blots of downstream cell survival mechanisms show that TRPM2(-/-) promotes pro- survival conditions. Basal levels of AKT had no change in expression between WT and KO (A), however phospho-AKT at both phosphorylation sites showed an increase (49.39±8.26% in Ser 473; 65.11±13.78% in Thr-308; n=4;*p<0.01) in TRPM2(-/-) compared to WT(B,C). There was no difference in basal ERK expression when comparing WT and TRPM2(-/-) (D); but active ERK was increased by 76.44±23.12% (n=4;*p<0.01) in TRPM2(-/-) when compared to WT (E). GSK3B is phosphorylated by AKT, in TRPM2(-/-) there is a 38.96±4.97% (n=3;*p<0.01) increase in phosphorylation of GSK3B when compared to WT (F).

Figure 22: Absence of TRPM2 reduces PSD95 expression but has no effect on BDNF signaling

Western blots of hippocampi from WT and TRPM2(-/-) mice of post synaptic density and BDNF signalling. PSD-95 expression was decreased by 45.61 ± 12.76% (n=3;*p<0.01) in TRPM2(-/-) when compared to WT (A). Also, BDNF mediated TRKB phosphorylation, which is known to be involved NMDA mediated neuroprotection, was not affected by the absence of TRPM2 (n=3;B).

6.1.5 Absence of TRPM2 does not alter TRPM7 expression

In addition to NMDA-mediated cell death, our lab has shown that TRPM7 channels also contribute to ischemic cell death (Aarts et al., 2003; Sun HS et al., 2009). Silencing TRPM7 has 2+ been shown to inhibit OGD-dependent Ca currents (IOGD) and provide neuroprotection following ischemia both in vivo and in vitro (Aarts et al., 2003; Sun HS et al., 2009). It was also reported that siRNA inhibition of TRPM7 was associated with a reduction of TRPM2 protein expression, suggesting that TRPM7 and TRPM2 expression may be interdependent (Aarts et al., 2003). However, when we compared TRPM7 levels in WT and TRPM2(-/-) hippocampi we found that absence of TRPM2 had no effect on total TRPM7 expression (n=3, p>0.05; Figure 23). Overall, these results suggest that TRPM7 expression is not dependent of TRPM2 and the neuroprotective properties of TRPM2(-/-) is not mediated by modulation of TRPM7.

Figure 23: Expression of TRPM2 has no effect on TRPM7 expression

TRPM7 channels share similar function and homology to TRPM2 channels and has been suggested to be interdependent. Western blots show that absence of TRPM2 channels has no effect on TRPM7 protein expression (n=3, p>0.05).

6.2 Discussion

Both glutamate receptors and non-glutamate cation channels contribute to Ca2+-overload during ischemia. Our lab has identified both NMDAR and TRPM7 as major contributors to this Ca2+-influx (Aarts et al., 2003; Sattler et al., 1999). The NR2A and NR2B subunits of NMDAR are known to govern cell survival and death mechanisms respectively (Soriano et al., 2008; Hardingham and Bading, 2010; Martel et al., 2009). Activation of NR2A and inhibition of NR2B mediates pro survival mechanisms via Akt and ERK1/2 pathways (Soriano et al., 2008; Hardingham et al., 2002). Prior to this study, it has been implicated that NMDAR and TRPM2 channels may have interdependent modulation the function. Where in hippocampal neurons TRPM2 activation is dependent on NMDAR activity (Olah et al., 2009), and in GABAergic neurons of the substantia nigra (SNr GABAergic) TRPM2 activation is required for NMDAR- mediated bursting pattern (Lee et al., 2013). Here we demonstrate that the TRPM2 channel is a modulator of NR2 subunit expression and function.

The modulation of NR2 subunits by the absence of TRPM2 leads to similar changes in downstream pro-survival pathways as seen NMDA-mediated preconditioning. In preconditioning, low doses of NMDA preferentially activate synaptic NMDAR, which in turn, mediates Akt and ERK dependent pro-survival pathways (Soriano et al., 2006). This leads us to ask if TRPM2 channels are involved in NMDA-mediated preconditioning and if so, does an absence of TRPM2 in vivo mimic low doses of NMDA in vitro. How low levels of NMDA treatment preconditions neurons is not fully understood, however the common element between NMDA preconditioning and TRPM2(-/-) preconditioning is an increase in synaptic activity (Papadia et al., 2005). A difference between these two models, is that NMDA-mediated preconditioning induces BDNF-mediated TRKB activation (Soriano et al., 2006), a phenomenon we did not observe in TRPM2-null hippocampi. We suggest here that the absence of TRPM2 and NMDA-mediated preconditioning uses common neuroprotective mechanisms.

A major question from this chapter is how does TRPM2 channels modulate NMDAR subunit expression? Since the lack of TRPM2 is chronic, we suspect that it may be due to a compensatory mechanism. Unfortunately, since downstream pathways of TRPM2 channels are not fully understood, it will be a difficult to determine at what stage of translation and/or transcription is TRPM2 required for the modulation of NR2 subunits. Alternatively, TRPM2

92 channels may also be part of the NMDAR complex and its absence may disrupt complex formation; unfortunately, due to the lack of accurate TRPM2 antibodies this hypothesis will be difficult to pursue. We know that PSD-95 binds to the NR2B subunit and is NR2B-clustering (Chung et al., 2004; Cui et al., 2007). Since both PSD-95 and NR2B levels were inhibited in TRPm2(-/-) hippocampi, it is possible that PSD-95 may play a role in TRPM2-mediated modulation of NR2B subunits.

The findings in this chapter provide a novel function for TRPM2, as a modulator NR2 subunits of NMDAR, and this provides a new mechanism of how TRPM2 is involved in neuronal cell death. Although we were unable to pursue all NMDA-mediated cell survival and death mechanisms; nor were we able to elucidate the exact mechanism of how TRPM2 modulates NR2 subunits, we have opened the door to potential new projects that counld understand the relationship between NMDAR and TRPM2 channels.

6.3 Summary

Overall, our findings show that chronic absence of TRPM2 modulates expression and activation of NMDARs in unstressed whole hippocampus; specifically, increased expression of NR2A and decreased activation and expression of NR2B. In the case of TRPM2(-/-), this change affects pro-survival and pro-death pathways. Where, increase in NR2A expression leads to an increase in pro-survival Akt and ERK1/2 pathways; and results in the inhibition of GSK3B, which is known to cause apoptosis. We show here a novel mechanism by which TRPM2 involved in cell death mechanisms.

Chapter 7

7 Comparing neuroprotective properties of TRPM2(-/-) to known neuroprotective drugs

Aim 4: To determine, if TRPM2(-/-) enhances neuroprotective properties of potential stroke treatment drugs following OGD of CA1 neurons.

Hypothesis 4: Following OGD, TRPM2-null neurons increase resistant to cell death of neuroprotective drugs, when compared to similarly treated WT neurons.

Thus far, we have demonstrated that TRPM2(-/-) genotype confers neuroprotection following in vivo ischemia-reperfusion injury and that the absence of TRPM2 modulates NMDAR NR2 subunits to promote pro-survival mechanisms. In vitro attempts to prevent oxidative damage by inhibiting TRPM2 have provided inconsistent results. Specifically, TRPM2 knockdown using antisense siRNA imparted neuroprotection to cortical neurons subjected to oxidative stress with H2O2 (Kaneko et al., 2006). However, reports using non-specific TRPM2 antagonists did not protect hippocampal neurons against lethal H2O2 (Bai and Lipski, 2010; Wilkinson et al., 2008). Unlike prior studies, we used cultured neurons from TRPM2(-/-) animals to determine if loss of TRPM2 function results in neuroprotection in vitro. Furthermore, we also used the TRPM2(-/-) model to observe the combined neuroprotective effects of the TRPM2(-/-) and potential neuroprotective drugs.

We first cultured prenatal hippocampal neurons from WT and TRPM2(-/-) pups and incubated the cultures until they matured to an adult-like condition. These neurons were then subjected to OGD stress and neuronal death was measured 24h after OGD. Also, we treated WT and TRPM2(-/-) cultures with known neuroprotectants, specifically, MCN (glutamate receptor antagonist), M21 (TRPM7 antagonist) and NA1 (PSD95-NR2B inhibitor) to determine if blocking conductances involved in neuronal death had an additive neuroprotective effect on TRPM2(-/-) neurons. This section we will resolve whether complete inhibition of TRPM2 is neuroprotective in vitro and will provide insights into the therapeutic effects of blocking multiple channels involved in ischemia.

7.1 Results for Hypothesis 4

7.1.1 Absence of TRPM2 is neuroprotective in hippocampal neurons following OGD

We first explored whether the neuroprotective properties of TRPM2(-/-) mice we observed in vivo could be translated to an in vitro ischemia model. Using cultured primary CA1 hippocampal neurons from both WT and TRPM2(-/-) animals we measured cell death under control (unchallenged) and 1h OGD conditions. Number of neurons per plate were consistent between WT and TRPM2(-/-) groups (~50,000 cells per plate) and level of cell death was measured using PI staining, which enters into dying/dead cells and fluoresces. Both WT and TRPM2(-/-) neurons exhibited similar morphology under the microscope prior to stress and both had similar levels of cell death after 24h control conditions (n=6 cultures; P<0.01; Figure 24). When exposed to 1h OGD followed by 24h incubation in neurobasal (which mimics reperfusion), neurons derived from TRPM2(-/-) mice had significantly reduced level of neuronal death when compared to WT neurons under the same conditions (n=6; P<0.01; Figure 24). Post- OGD neurons were observed under a microscope to visually confirm neuronal death. This study confirms our in vivo findings that absence TRPM2 in neuroprotective and confirms results from prior experiments using siRNA against TRPM2 to provide neuroprotection (Kaneko et al., 2006; Hara et al., 2002).

7.1.2 Neuroprotective agents do not enhance neuroprotective effects in TRPM2-null hippocampal neurons

Currently, there are a number of drugs in development that have been shown to provide neuroprotection following ischemia. Many of these drugs target different cation channels and glutamate receptors suggesting that there are multiple routes for Ca2+-influx following ischemia. TRPM2 has been suggested to be one of those routes (Hara et al., 2002), and furthermore, we showed earlier that an absence of TRPM2 also reduces NR2B containing NMDARs, which could be another route for toxic Ca2+-influx (Soriano et al., 2008). Here we used known neuroprotective drugs in combination with our TRPM2(-/-) model to investigate whether a combination of TRPM2 absence, and inhibition of other channels can have an additive effect on neuroprotection following 1h OGD. The known neuroprotective agents used were: MCN, a glutamate receptor inhibitor containing MK-801, nimodipine and CNQX (described in Aarts et al. 2003); NA-1, a psd-95 inhibitor (described in Aarts et al. 2002) and M21, a small molecule

95 inhibitor that has been shown to inhibit TRPM7-like currents and provide neuroprotection (unpublished results). In the 1h OGD experiment, three drugs provided neuroprotection to hippocampal neurons from both WT and TRPM2(-/-) mice with no significant difference between the level of protection of the two genotypes. In TRPM2(-/-) neurons the three drugs provided greater protection than the absence of TRPM2 alone (Figure 25). Since there was no difference in the neuroprotective effects of the 3 drugs between neurons from WT and TRPM2(-/-) animals, we next tried a 1.5h OGD time period in case we may be seeing a flooring effect. Again, in the 1.5h OGD model there was no significant difference in neuronal death between WT and TRPM2(-/-) neurons treated with NA1 and MCN (n=4; Figure 25). Also, following the 1.5h OGD experiments, we still see that neurons from TRPM2(-/-) mice had a lower neuronal death than those from WT mice (Figure 25). This data suggests that there is no therapeutic benefit against ischemia to combining cytoprotective drugs with TRPM2-null neurons when compared to WT neurons treated with the same drugs.

Figure 24: Cultured hippocampal neurons from TRPM2(-/-) mice have reduced OGD- induced cell death in comparison to WT

Primary cultures of hippocampal CA1 neurons from WT and TRPM2(-/-) mice exposed to 1h OGD followed by 24h control conditions. Level of cell death was measured using PI staining. TRPM2(-/-) neurons had a similar levels of cell death as WT neurons under control conditions (p>0.05; n=6). Following 1h OGD-reperfusion conditions, TRPM2(-/-) neurons demonstrated a resistance to cell death when compared to WT (*p<0.01; n=6). This confirms in vivo findings that absence of TRPM2 provides neuroprotection following ischemia-reperfusion.

Figure 25: Neuroprotectants do not enhance neuroprotective effects of absence of TRPM2 for up to 1.5h OGD

Primary hippocampal neurons from WT and TRPM2(-/-) mice were treated with neuroprotective drugs and exposed to 1-1.5h OGD. Neuroprotective drugs used were: MCN, NA1 and M21. Treated and untreated WT and TRPM2(-/-) neurons had similar levels of cell death following no-OGD and 24h of unchallenged control conditions. Cytoprotective drugs provided similar resistance to OGD between WT and TRPM2(-/-) neurons (p>0.05; n=6). Cytoprotective drugs also provided greater resistance to OGD-induced cell death when compared to untreated TRPM2(-/-) neurons for up to 1.5h of OGD (p>0.05; n=4).

7.2 Discussion

In Chapter 4, we demonstrated that TRPM2(-/-) mice had reduced infarct volume compared to WT following ischemia-reperfusion injury. Since that experiment was done in vivo, it was difficult to determine whether absence of TRPM2 provided neuroprotection directly or indirectly through other non-neuronal cell types. To exacerbate this problem, in vitro experiments using non-specific TRPM2 antagonists failed to provide neuroprotection following oxidative stress (Bai and Lipski, 2010;Wilkinson et al., 2008). In this chapter we showed that primary hippocampal neuronal cultures (>85% neurons) from TRPM2(-/-) mice were more resistant to OGD-induced cell death than neurons from WT mice. This finding suggests that absence of TRPM2 provides protection directly in the neuron. It also implies that earlier experiments using non-specific TRPM2 antagonists to provide neuroprotection, likely also inhibited other cellular mechanisms required for survival.

When using known neuroprotective agents, we found that TRPM2(-/-) neurons had similar neuroprotection as WT neurons following OGD, as opposed to having an additive protective effect. A possible explanation is that there may be a flooring effect in terms of neuroprotection which is reached by these drugs, and neuroprotection from the absence of TRPM2 cannot add to it. Both MK-801 and NA1 inhibit NMDAR-mediated Ca2+-influx, while we have shown that the TRPM2(-/-) promotes cell survival by a similar mechanism by reducing NR2B expression/activation. It is possible that the MK-801 and NA1 treatments have a maximum effect on NMDAR-mediated excitotoxicity, rendering no additional effect from the absence of TRPM2. It is unknown whether TRPM7 modulates NMDAR-dependent excitotoxicity, and the ceiling effects in M21 may be due to a mechanism independent of TRPM2.

A limitation to using neurons from TRPM2(-/-) mice to observe the effects of blocking multiple sources of neurotoxic Ca2+ is that chronic absence of TRPM2 already alters pro-survival and death mechanisms used by those drugs, thus potentially interfering with the efficacy of neuroprotective drugs. This is very likely true for any drug targeting NMDAR, like Mk-801 and NA1. Due to the lack of specific antagonists it is difficult to observe the effects of acute inhibition of TRPM2, with the inhibition of other cation channels/glutamate receptors. In order to fully understand the therapeutic effects of TRPM2 inhibition in stroke there needs to be a development of a TRPM2-specific antagonist.

7.3 Summary

Here we show that neurons from TRPM2(-/-) mice are resistant to OGD in vitro, thus confirming our earlier findings in vivo. When treated with neuroprotective drugs, MCN, NA1 and MK-801, there is no difference in resistance to 1h OGD between WT and TRPM2(-/-) neurons, suggesting that our initial hypothesis was incorrect because the absence of TRPM2 did not enhance the cytoprotective effects of the drugs. Furthermore, when comparing resistance to 1h OGD between untreated neurons from TRPM2(-/-) mice and other neurons (from WT and TRPM2(-/-) mice) treated drugs, we found that the drugs provided greater resistance to OGD- mediated cell death than neurons lacking TRPM2 alone. Overall, these in vitro studies confirm our in vivo findings and suggest that therapeutic strategies combining absence of TRPM2 and inhibition of NMDAR or TRPM7 may not provide additive neuroprotection.

100

Chapter 8

8 Discussion

In this study we used TRPM2(-/-) mice to describe a novel mechanism of how TRPM2 is involved in regulating cell death following an ischemic insult. First, using an in vivo model, we showed that TRPM2(-/-) is neuroprotective only when there is ischemia followed by reperfusion, but not when there is permanent ischemia. Second, using hippocampal slices, we found that (-/-) 200uM of H2O2 increased baseline synaptic strength in CA1 neurons from TRPM2 animals but had no effect on WT. Third, using western blots, we found that whole hippocampus from TRPM2(-/-) mice had a reduction in NR2B expression and an increase in NR2A expression when compared to WT. NR2A antagonists were found to block the increase in oxidative stress induced synaptic potentiation in TRPM2(-/-) neurons. These changes in NR2A and NR2B expression also led to an increase in downstream Akt and ERK1/2 activation. Finally, using cell cultures, we found that TRPM2(-/-) mice were resistant to OGD induced cell death.

8.1 Proposed mechanism of TRPM2(-/-) and neuroprotection

Using these findings we propose a novel mechanism of how TRPM2 is involved in cell death following ischemia, specifically, TRPM2 channel expression is required for the expression of NR2B and plays a role in inhibiting NR2A. Consequently, regulation of NR2A and NR2B leads to inhibition of downstream Akt and ERK1/2 pathways and ultimately leads to promotion of pro-death transcription factor, GSK3β (Figure 26). Overall, this model shows that chronic presence of TRPM2 is required for maintaining NMDAR-mediated cell death pathways, and absence of TRPM2 preconditions neurons for neuroprotection.

101

Figure 26: Proposed mechanism cell death involving TRPM2

TRPM2 channels are required to promote expression of PSD-95 and inhibit NR2A subunits in NMDA receptors. PSD-95 is responsible for transporting NR2B containing NMDAR to the cell surface and activating NR2B by binding. When PSD95 activates NR2B containing NMDAR, there is an influx of extrasynaptic Ca2+ which leads to inhibition of phosphorylation of ERK1/2 and promotes cell death. The inhibition of NR2A expression reduces synaptic Ca2+ influx and prevents downstream activation of MEK and PI3 kinases, which are required for phosphorylation of ERK1/2 and Akt, respectively. Phosphorylation of Akt leads to inactivation of pro-apoptotic factor, GSK3β. Overall, TRPM2 channel reduces expression of pro-survival NMDARs and increases expression of pro-death NMDARs.

102

8.2 TRPM2(-/-) and neuroprotection

TRPM2 was initially implicated to play a role in ischemic cell death because it is activated by products of oxidative stress, specifically H2O2 (Kaneko et al., 2006; Hara et al.,

2002) and ADPR (Perraud et al., 2005). It was hypothesized that over production of H2O2 and ADPR would lead to the activation of TRPM2 which would contribute to Ca2+ overload and ultimately cell death (Besancon et al., 2008; Tymianski, 2011; Xie et al., 2011). Until recently, inhibition of TRPM2 was limited to either mRNA silencing (Kaneko et al., 2006;Olah et al., 2009) and non-specific blockers (Bai and Lipski, 2010). In vitro attempts to block TRPM2 to provide neuroprotection have given conflicting results, specifically, TRPM2 silencing prevents oxidative stress induced neuronal cell death (Kaneko et al., 2006), while the use of nonspecific blockers does not (Bai and Lipski, 2010).

Here, we use TRPM2(-/-) mice to show that chronic absence of TRPM2 is neuroprotective in both in vivo and in vitro models. Using in vitro OGD model we showed that neuronal-enriched cultures from TRPM2(-/-) mice were resistance to ischemia. It should be noted that we used an

OGD model instead of H2O2 application, like in prior studies, because it provided physiological ischemic conditions, instead of solely oxidative stress conditions. In vivo tMCAO confirms in vitro findings, however pMCAO did not provide any neuroprotection. In this section we hope to address some questions that arise from the use of TRPM2(-/-) animals, as well as the ischemic models we used.

8.2.1 TRPM2(-/-) as a model for in vitro neuroprotection

A number of in vitro models have been used to inhibit TRPM2, these include: pharmacological blockers, mRNA silencing, dominant negative isoform of TRPM2 and TRPM2(- /-) mice (Harteneck et al., 2007; Zhang et al., 2003; Bai and Lipski, 2010; Hara et al., 2002; Yamamoto et al., 2008; Xie et al., 2011; Fonfria et al., 2005; Olah et al., 2009). Not all these models provide neuroprotection, which raises the question as to why some models work and some models don't.

Pharmacological blockers failed to provide neuroprotection in hippocampal slices following oxidative stress conditions (Wilkinson et al., 2008; Bai and Lipski, 2010). They are often the easiest and cheapest way to acutely inhibit TRPM2, however the disadvantage is that

103 they are known to be non-specific (Hill et al., 2004a; Hill et al., 2004b). Unlike channel blockers, silencing TRPM2 mRNA prevents neuronal death following oxidative stress. Silencing TRPM2 is specific to only TRPM2 channels, however there isn't a complete inhibition of TRPM2, resulting in reduced rather than abolished channel function. As we have shown, chronic TRPM2- absence in hippocampal neurons is also neuroprotective following in vitro OGD. Unlike acute models, TRPM2(-/-) prevents the production of functional TRPM2 channels. A caveat into this model is that chronic absence of TRPM2 may lead to compensatory changes in neurons, such as changes in NMDAR expression we observed in the hippocampus.

Based on prior methods used to inhibit TRPM2, it seems that specific TRPM2 inhibition is required for neuroprotection. Also, a complete loss of TRPM2 function is not required, as mRNA silencing provides neuroprotection (Kaneko et al., 2006). From our own results, TRPM2(- /-) also prevented ischemic cell death and any secondary effects from the chronic absence of TRPM2 did not prevent neuroprotection. However, the mechanisms of neuroprotection in acute and chronic absence of TRPM2 may be different, since chronic absence also affects expression of NMDAR subunits, which may not occur in acute cases. This also provides a possible explanation as to why pharmacological blockers failed to be neuroprotective, because non- specific effects on other channels may inhibit pathways required for survival or compensate for the loss of TRPM2 function.

8.2.2 Translating TRPM2 from in vitro to in vivo stroke model

Prior studies attempting to demonstrate the effects of TRPM2 inhibition on neuronal survival have been limited to in vitro models. Since we used a TRPM2(-/-) mouse model, we are the first to demonstrate that absence of TRPM2 is neuroprotective in vivo. The TRPM2(-/-) mouse had no observable difference compared to WT animals and both produced normal healthy offspring, unlike TRPM7(-/-) which are pre-natal lethal (Jin et al., 2008). Thus far the only behavioral difference between TRPM2-null and WT mice is that TRPM2(-/-) mice have a decreased sensitivity to formalin test (Haraguchi et al., 2012). Since these knockouts do not have observable survival or reproduction deficits, it makes the move from in vitro to in vivo model easier and increases the potential of TRPM2 channels being a therapeutic target.

The in vivo stroke model we used was MCAO, which has been used successfully in a number of studies with C57B6 mice (Yi et al., 2007; Wiegler et al., 2008). Using a transient

104

MCAO model of 1hr occlusion followed by 48hr reperfusion, we found that brains from TRPM2(-/-) animals had a reduced infarct volume when compared to WT. However, the neuroprotective effects of the TRPM2(-/-) genotype were not seen in permanent 24h MCAO. In both WT and TRPM2(-/-) mice we confirmed a >80% reduction of blood flow following MCAO, with a return of similar levels of reperfusion in tMCAO, thus showing that mice of both genotypes had similar blood flow. There was also no significant difference in mortality between TRPM2(-/-) and WT animals, indicating that TRPM2(-/-) did not affect survival rate following MCAO.

The tMCAO model is similar to the in vitro OGD model we used, specifically both models first deprive neurons of oxygen and glucose, followed by an immersion/reperfusion of oxygen and glucose for a certain period of time. A major difference between these two models is that our in vitro model used hippocampus neuronal cultures, while MCAO predominantly affects neurons in the cortex. The reason why we used hippocampal neuronal cultures was because it was consistent with our hippocampal slice model that was used in electrophysiology experiments. None the less, both hippocampal and cortical neurons express functional TRPM2 channels and we confirmed that TRPM2(-/-) was neuroprotective in cortical neurons following in vitro OGD (see Appendix 1). Furthermore, to assess in vivo stroke damage in just the hippocampus using global ischemia is very difficult, as this model is technically challenging and results in high mortality rate. It is even more difficult in C57B6 mice, since, unlike other rodent models, the strain has a complete circle of Willis3; so in addition to clipping the common carotid arteries, the basilar artery must also be clipped in order to achieve global ischemia (Yonekura I. et al., 2004). Since in vitro OGD experiments in TRPM2(-/-) cortical neurons showed similar neuroprotective properties as in vitro TRPM2(-/-) hippocampal neurons (Appendix 1), we predict that TRPM2(-/-) mice will have reduced hippocampal damage following global ischemia, similar to what we saw in MCAO. Overall, our findings suggest that TRPM2 is involved in neuronal death in both the hippocampus and cortex.

3 A complete circle of Willis in C57B6 is found in approximately 70% of animals, while other mice strains such as Balb/C often lack posterior communicating artery. Therefore the brain in C57B6 mice can still have blood flow from the basilar artery.

105

A question that arises when using in vivo stroke models, is whether absence of TRPM2 in neurons specifically contribute to neuroprotection. TRPM2 is found in a number of different non neuronal cell types and is not always found in neurons, like in the case of the cerebellum. Determining what cells express TRPM2 channels is difficult since there is a lack of specific TRPM2 antibodies4. TRPM2 mRNA is found in hippocampal neurons and TRPM2-like current were found in both the hippocampal and cortical neurons. Here we demonstrate that in neuronal- enriched hippocampal cultures TRPM2(-/-) reduced neuronal death following OGD when compared to control. We visually confirmed presence of neurons based on morphology both before and after OGD, however there were some non neuronal cell types in cultures. Our in vitro results demonstrate that in a predominantly neuron environment TRPM2(-/-) is neuroprotective, suggesting that during in vivo stroke the absence of TRPM2 in neurons contribute to neuroprotection, in addition to other potential neuroprotective mechanisms via non-neuronal cell types.

8.2.3 Reperfusion damage required for TRPM2-mediated damage

Although, TRPM2(-/-) provided neuroprotection following tMCAO, it failed to reduce infarct volume following 24h pMCAO. A possible explanation to these findings is that TRPM2 is involved in cell death primarily during reperfusion. Reperfusion of oxygen into ischemic areas often overproduces ROS (Simonson et al., 1993), which in turn is known to activate TRPM2. With the absence of functional TRPM2 channels, we predict that during reperfusion in tMCAO there is a loss of TRPM2-mediated Ca2+ influx; while in pMCAO, the TRPM2-null mutant had little effect on cell survival because TRPM2 channels are not activated during ischemia. Alternatively, another explanation for our findings is that pMCAO was so severe that any protection from TRPM2(-/-) was not detectable. We observed in Figure 5 that infarct volumes in pMCAO was much higher than in tMCAO, suggesting that pMCAO had greater ischemic damage. This explanation would suggest that TRPM2 does play a role in neuronal death during ischemia, but our pMCAO model was so severe that any cytoprotective effects of TRPM2 deficiency was not detectable. Overall, we found that TRPM2(-/-) reduces infarct volume due to

4 In addition to Xie et al. 2011, we have also attempted to use commercially available TRPM2 antibodies and found them to be non specific.

106 reperfusion damage, however whether it also protects against ischemic only damage is an area that warrants further research.

8.3 TRPM2 in synaptic activity

During ischemic cell death there are a number of mechanisms that come into play that influence synaptic activity, including: release of excitatory neurotransmitters, glutamate mediated Ca2+ overload and production of ROS (Arundine and Tymianski, 2004). The production of ROS primarily occurs during reperfusion (Chan, 1996) and ischemic levels of ROS have been shown to disrupt synaptic plasticity (Kamsler and Segal, 2003; Maalouf and Rho, 2008). Furthermore, silencing TRPM7 protects LTP formation following in vivo ischemia, suggesting TRPM7 may play a role in ischemia-induced disruption of synaptic plasticity. Consistent with results from Xie et al. (2011), we found that TRPM2(-/-) had no effect on baseline fEPSP in CA1 region of hippocampal slices under normal ACSF conditions. However, when

200uM of H2O2+ACSF was applied, there was an increase in baseline fEPSP slope and magnitude in TRPM2(-/-) slices with no effect on WT baseline fEPSP. This increased synaptic sensitivity to H2O2 was found to be in the post synapse, since 200uM H2O2 had no effect on paired-pulse facilitation. In the post synapse, evoked synaptic excitability is caused by activation of glutamate receptors, NMDAR and AMPAR. We found here that blocking NR2A containing (-/-) NMDAR (using 0.4uM of PEAQX) inhibited the effects of H2O2 on TRPM2 mice while NR2B antagonists did not. The concentrations of NR2A and NR2B antagonist we used is known to specifically inhibit their respective NMDAR currents without any affect on fEPSP.

Together these findings suggest that absence of TRPM2 makes hippocampal CA1 neurons more sensitive evoked excitability under oxidative stress conditions. This increase in H2O2-induced fEPSP is mediated NR2A containing NMDAR. Supporting these findings, we also found that NR2A expression was increased in TRPM2(-/-) hippocampus, which may be the cause of the increased synaptic sensitivity to H2O2. From the perspective of neuroprotection, we know that NR2A is responsible for pro-survival mechanisms and that neuroprotective preconditioning causes a NR2A-mediated increase in excitability. These findings in TRPM2(-/-) hippocampal slices are consistent with the neuroprotective preconditioning, where under stress conditions pro- survival mechanisms are associated with an increase in synaptic excitability; while pro-death mechanisms are associated with inhibition of synaptic excitability.

107

8.3.1 TRPM2 and synaptic plasticity

Following a stroke, oxidative stress disrupts synaptic plasticity mechanisms often affecting cognitive functions such as learning and memory. Hydrogen peroxide has been identified as a modulator of LTP formation; where a low concentration (<5uM) facilitate potentiation, while higher concentrations (>20 uM) have an inhibitory effect (Kamsler and Segal, 2003). Also, inhibition of TRPM7 has been shown to protect LTP formation following in vivo stroke, suggesting that neuroprotection also protects synaptic plasticity. A recent study using hippocampus from TRPM2(-/-) animals has shown that an absence of TRPM2 leads to inhibition of long term depression (LTD) via phosphorylation, suggesting that TRPM2(-/-) may play a role in synaptic plasticity. Like in the Xie et al. (2011) study we found that hippocampal slices from both WT and TRPM2(-/-) animals were able to form normal LTP lasting up to 60 minutes. We next found that 200uM of H2O2 inhibited LTP formation in WT hippocampal slices, as been shown in other studies (Kamsler and Segal, 2003; Maalouf and Rho, 2008). In slices from (-/-) TRPM2 animals, H2O2 caused an initial increase in baseline fEPSP, and following LTP induction, fEPSP slope returned to baseline levels with H2O2. There are two possible (-/-) explanations why TRPM2 did not prevent H2O2-induced inhibition of LTP formation: 1) the (-/-) TRPM2 genotype has no effect on the loss of LTP formation by 200uM H2O2 or 2) that the (-/-) increase in baseline fEPSP induced by H2O2 in hippocampi from TRPM2 animals results in a ceiling effect, thus preventing any additional potentiation from high frequency stimulation. It is unclear which of the two possiblities is the likely reason, however we showed that removal of

H2O2 following LTP induction in TRPM2-null hippocampi results in a return to baseline without

H2O2; suggesting that an absence of TRPM2 in the hippocampus did not prevent the H2O2- induced loss of LTP formation. Overall, these findings show that the absence of TRPM2 has no effect on normal LTP formation and does not protect LTP from the inhibitory effects of (-/-) oxidative stress. Since the removal of H2O2 following LTP induction in TRPM2 hippocampal slices returned fEPSP to the original baseline, we suspect that LTP and H2O2-induced excitability in TRPM2-null slices may be via independent mechanisms, however further study is required to determine if this is the case.

108

8.3.2 Possible mechanisms of TRPM2(-/-) modulation of synaptic transmission

(-/-) We have reported that 200uM of H2O2 increased baseline fEPSP in TRPM2 slices, which raises the question how does absence of TRPM2 cause synaptic sensitivity to H2O2? Since (-/-) the sensitivity to H2O2 observed in TRPM2 hippocampi can be inhibited by NR2a antagonist; we determined that that sensitivity to oxidative stress in TRPM2(-/-) is partially mediated by NR2A containing NMDARs. Furthermore, since TRPM2(-/-) had no effect on LTP formation nor does it protect LTP from oxidative stress, we have suggested that H2O2 sensitivity is independent of LTP mechanisms. How H2O2 increases excitability and sensitivity of NR2A subunit is unknown, however we speculate three possible mechanisms for the increase in sensitivity:

2+ 1) H2O2 sensitizes NMDARs to glutamate, independent of intracellular Ca concentration.

In this model H2O2 directly modulates NMDAR complex, making NMDAR more sensitive to glutamate mediated activation. In WT neurons, 200uM H2O2 would increase glutamate sensitivity in both NR2A and NR2B subunits; respectively balancing excitatory and inhibitory modulators of synaptic transmission. In the case of hippocampi from TRPM2(-/-) mice, where there was an increase in NR2A subunit and a decrease in NR2B subunit expression,

200uM H2O2 increased synaptic strength which was dependent on NR2A activity. In addition, since PSD95 coupling with NR2B is required for excitotoxic Ca2+ influx (Sattler et al., 1999), and we here observed a reduction in PSD95, it is possible that a reduction in NR2B function may not be due to just a reduction in NR2B expression, but also due to a reduction of PSD95-NR2B coupling. How presence or absence of TRPM2 is involved with this mechanism or whether TRPM2 is part of the NMDAR complex is currently unknown.

2+ 2) Sensitivity to H2O2 is dependent on intracellular Ca concentration.

The lack of TRPM2 during oxidative stress results in a lack of intracellular Ca2+ (Olah et al., 2009; Yamamoto et al., 2008), which makes it easier for Ca2+ to move along the concentration gradient. This in turn could lead to a hypersensitization of glutamate receptors 2+ (specifically containing NR2A) and other Ca channels to H2O2, which results in an increase in synaptic efficacy.

3) H2O2 promotes increased glutamate release in the presynapse.

109

Since, H2O2 is involved in presynaptic neurotransmitter regulation, it is possible that TRPM2 modulates Ca2+-mediated neurotransmitter release. Similarly, TRPM7 is involved in acetylcholine release in the peripheral nervous system (Krapivinsky et al., 2006). However, paired-pulse experiments comparing WT and TRPM2(-/-) hippocampal slices under H2O2 conditions showed no change in facilitation; suggesting that H2O2-induced excitation in TRPM2(-/-) hippocampus is not caused by changes in neurotransmitter release in the presynapse. This mechanism is the least likely to be the cause of H2O2 sensitivity in TRPM2(-/-) and thus the first two proposed mechanisms should be the primary focus of future research.

(-/-) Thus far we have determined TRPM2 modulates synaptic sensitivity to H2O2 via a mechanism that is independent of presynaptic plasticity, is independent of LTP formation and is mediated by NR2A containing NMDARs. Based on our findings we have suggested 2 mechanisms that may explain how TRPM2(-/-) sensitizes synaptic excitability of hippocampal neurons to H2O2; and further study is required to determine whether one or both mechanisms are involved.

8.4 TRPM2(-/-) modulates glutamate receptors

The relationship between TRPM2 channels and glutamate receptors have recently become an area of interest, where studies have given evidence that glutamate receptor and TRPM2 channel function and expression may be interdependent. A recent study has shown TRPM2 function is dependent on the activation of NMDAR, specifically, ADPR-mediated TRPM2 Ca2+-influx is dependent on the presence of NMDA (Olah ME et al., 2009) Furthermore, it has been shown that inhibition of TRPM2 can modulate NMDAR mediated bursting in GABAergic neurons of the substantia nigra (Lee et al., 2013) and TRPM2(-/-) reduces AMPA subunit GLUR2 expression in hippocampal neurons (Xie et al., 2011). Our study has also demonstrated a relationship between TRPM2 channels and glutemate receptors. Using western blot in whole hippocampus, we demonstrated TRPM2(-/-) modulated NMDAR subunit expression without any effect on AMPAR subunit expression. Specifically, absence of TRPM2 reduces NR2B expression and activation, and increases NR2A expression, with no effect on NR1. Since NR1 is an obligatory subunit, this suggests that TRPM2 regulates NR2 subunit expression without any effect on overall NMDAR expression.

110

Different NR2 subunits are known to play a role in cell survival/death and in synaptic excitability. Specifically, NMDARs containing NR2A are responsible for synaptic excitability and cell survival (Liu et al., 2007b; Terasaki et al., 2010; Soriano et al., 2006), and NMDARs containing NR2B regulate synaptic depression and cell death mechanisms (Liu et al., 2007b; Soriano et al., 2006). This demonstrates that data from our Western blots are consistent with our (-/-) earlier findings showing that TRPM2 is neuroprotective and causes H2O2-induced synaptic potentiation. Together this data suggests that TRPM2(-/-) chronically increases expression of pro- survival pathways by increasing NR2A activation and reduces pro-death pathways by reducing NR2B expression, making TRPM2(-/-) neurons prone to neuroprotection.

In addition to regulating cell survival/death, NR2 subunits are suggested to segregate on the cell membrane, specifically NR2A subunit is predominantly expressed at the synapse, while NR2B is at the extra-synapse (Liu et al., 2007b; Hardingham et al., 2002). Our immunofluorescence results show that TRPM2(-/-) had reduced NR2B clusters at the synapse compared to WT, with no significant change in NR2A synaptic clustering. Considering that TRPM2(-/-) reduced NR2B expression in western blots, this reduction in NR2B clustering may be a due to overall reduction in NR2B expression in the neuron. Also, since TRPM2(-/-) increased NR2A levels in whole hippocampus, it is possible that TRPM2(-/-) has no effect on number on NR2A synaptic clusters, but increases NR2A concentration at each cluster. Alternatively, the increase in NR2A expression may not be occurring at synaptic clusters, but instead at extrasynaptic locations in the neuron. Unfortunately, since immunofluorescence is not as sensitive as Western blot in determining changes in NR2 expression, it is difficult to determine when in the membrane changes in expression occurs. Collectively, immunofluorescence shows that TRPM2(-/-) changes the ratio of NR2A to NR2B containing NMDARs at the synapse, by making the post synapse predominantly NR2A containing NMDARs, which is associated with neuroprotection.

A recent study using hippocampal slices from TRPM2(-/-) animals, has shown that an absence of TRPM2 reduces LTD formation (Xie et al., 2011). This inhibition of LTD was attributed to an increase in GSK3β phosphorylation, and inhibition of PSD-95 and AMPA subunit GLUR1 expression, phenomenon which are associated with loss of LTD. The same study found no change to NMDAR subunits NR2A and NR2B expression (Xie et al., 2011). We also show here that in TRPM2(-/-) hippocampus there is a reduction in PSD-95 and an increase in

111 phosphorylation of GSK-3B. However, we didn't detect any changes in AMPA receptors and instead found that a lack of TRPM2 modulates NR2A and NR2B expression. The only difference between ours and their western blot protocol was that we used freshly isolated hippocampus, while Xie et al. (2011) used sliced hippocampus that was incubated in ACSF. The differences in our hippocampus preparation may have been the cause this discrepancy, however together these finding suggest that TRPM2(-/-) may play a role in regulating multiple glutamate receptors. Another caveat when using whole hippocampus is that neurons cannot be readily isolated from non-neuronal cells, which may contribute to NR2A and NR2B expression. However, our results showing reduction of NR2B synaptic clusters and increases NR2A-mediated synaptic sensitivity (-/-) to H2O2, demonstrates that TRPM2 has an effect on NR2 subunits in neurons.

8.4.1 Possible mechanisms of how TRPM2(-/-) modulates NMDAR

A major unanswered question that remains from this thesis is how TRPM2 modulates NMDA subunit expression? We have shown that in TRPM2(-/-) hippocampi PSD-95 expression is reduced and since PSD-95 is required for NR2B clustering (Chung et al., 2004), we hypothesize that TRPM2 may regulate NR2B subunit levels via regulation of PSD-95 expression. Now, this leads to another mystery, which is how does TRPM2 regulate PSD-95 expression? Due to the lack of TRPM2-specific antibodies it is very difficult to determine if the TRPM2 protein has direct interactions with PSD-95 or the post-synaptic density complex of NMDARs. Alternatively, TRPM2 channels may regulate PSD-95/NR2B levels indirectly via downstream signalling pathways from either TRPM2's channel function or intracellular kinase function. It remains unclear how and when TRPM2 modulates NR2A subunit, although it has been suggested that NR2B transporter kinesin superfamily motor protein 17 is also involved in ubiquitination of NR2A resulting in post-translational degredation (Yin et al., 2011). It is known that intracellular Ca2+ can modulate a number of translation and transcription factors, including CREB, which can regulate pro-survival and pro-death pathways (Hardingham et al., 2002; Bell et al., 2013). Unfortunately the downstream pathways of both enzymatic and channel function of TRPM2 proteins are unknown, and any possible translation or transcriptional modification via TRPM2 would require further investigation.

112

8.4.2 Potential role of TRPM2 in NR2A-NR2B developmental switch

It is well known that during early brain development NR2B containing NMDARs are predominantly expressed in the prenatal and early postnatal brain, while NR2A expression begins shortly before birth and eventually the NR2A subunit outnumbers NR2B (Monyer et al., 1994; Wenzel et al.,1997). This phenomenon is known as the NR2B-NR2A developmental switch. Unlike in the adult brain, in the pre-natal brain NR2B plays a role in synaptic excitability and in pro-survival signaling (Zhou and Baudry, 2006). Since the adult NR2A-NR2B ratio is established during postnatal brain development (Wenzel et al.,1997), it is possible that TRPM2 may play a role in modulating the NR2B-NR2A developmental switch. Unfortunately, the role of TRPM2 during development has yet to be studied and thus it is difficult to conclude when TRPM2 modulates NR2A-NR2B ratio. We hypothesize that during development TRPM2 may regulate NR2B and NR2A expression which in turn would affect the NR2B-NR2A ratio in adult brain.

8.4.3 NMDAR mediated survival/death pathways

Signaling from NMDARs with NR2A and NR2B subunits lead to the activation of a number of downstream mechanisms that are responsible for cell survival and cell death. Sub- lethal concentrations of NMDA, which activates NR2A containing NMDAR results in the phosphorylation of ERK and Akt, which has been associated with neuroprotective preconditioning (Papadia et al., 2005; Soriano et al., 2006; Hardingham, 2006b; Hardingham et al., 2001a). This study showed that an increase in NR2A expression and sensitivity to oxidative stress is associated with an increase in ERK and Akt activation, and is consistent with earlier findings that show this mechanism contributes to neuroprotection following stroke. This increase in ERK activation does not appear to be ubiquitous across all cell types, as a recent study using monocytes from TRPM2(-/-) mice has demonstrated that lack of TRPM2 inhibits ERK1/2 activation via protein tyrosine kinase 2 (Yamamoto et al., 2008). This suggests that our findings may be unique to neurons and that TRPM2 plays different roles in various cells in modulating pro-survival/death signaling.

In addition to NR2A-mediated neuroprotection, modulation of NR2B activity may also prevent neuronal death. Our lab has previously shown that specifically inhibiting NR2B and PSD-95 interactions results in neuroprotection (Aarts et al., 2002; Sattler et al., 1999; Soriano et

113 al., 2008). A recent study has shown that the uncoupling of NR2B and PSD95 not only reduces excitotoxic Ca2+ influx but also activates CREB-dependent pro-survival pathways (Bell et al., 2013). We observe here that a lack of TRPM2 reduces both PSD-95 expression and NR2B activation, which suggests that in TRPM2(-/-) mice there could be a reduction in PSD95-NR2B coupling, that would provide neuroprotection via similar pro-survival mechanisms as previously observed (Bell et al., 2013; Sattler et al., 1999; Aarts et al., 2002).

We also show here that in TRPM2(-/-) hippocampus there is a reduction in PSD-95 and an increase in phosphorylation of GSK-3β. However, we were unable to show any changes in BDNF-mediated TRKB activation, which has been shown to be associated with pro-survival mechanisms (Soriano et al., 2006). These findings are consistent with previous studies showing NMDAR activity modulates GSK-3B via Akt signaling pathway (Soriano et al., 2006) and suggests that TRPM2 channels plays a role in GSK-3β regulation without any affect on BDNF.

8.5 Alternate mechanisms of neuroprotection

Prior to this study, there have been a number of suggestions to the role of TRPM2 in 2+ ischemic neuronal death, primarily, that during ischemic ADPR and H2O2 induce Ca influx via TRPM2 channels (Harteneck et al., 2007; Hara et al., 2002; Kaneko et al., 2006). Here we will briefly discuss some alternative mechanisms that may contribute to the neuroprotection seen in TRPM2(-/-) mice.

8.5.1 TRPM2 mediated Ca2+-overload

TRPM2 was initially suggested to play a role in ischemic cell death because it is a non- specific cation channel, that is stimulated by: (1) by overproduction of ADPR and (2) ROS (specifically hydrogen peroxide) (Wehage et al., 2002; Hara et al., 2002; Perraud et al., 2005). It was shown that both ROS and ADPR induce inward Ca2+ current in a variety of cell types(Hara et al., 2002; Kaneko et al., 2006; Yamamoto et al., 2008; Olah ME et al., 2009). Since both ADPR and ROS are heavily produced during ischemic conditions, our initial hypothesis was that the TRPM2(-/-) was neuroprotective only by limiting Ca2+-overload. Although we did not pursue our initial hypothesis, Ca2+-influx via TRPM2 can still be a potential alternative mechanism as to why TRPM2(-/-) mice had reduced ischemic damage compared to WT. There has been a number of studies in vitro showing that TRPM2 gene silencing does protect against redox

114

damage and prevents H2O2-induced Ca2+ influx (Hara et al., 2002; Kaneko et al., 2006). Unfortunately, due to the lack of specific TRPM2 inhibitors it is difficult to observe the acute affects of inhibiting TRPM2 function without modulating the channel expression. Whether or not the loss of TRPM2 channel function in TRPM2(-/-) animals contributed to neuroprotection is not described in this study, but considering the research done using TRPM2 silencing, it is likely that absence of TRPM2 also contributed to neuroprotection in our mice.

8.5.2 Microglia and immune response

In our MCAO model, we attributed a reduction of infarct volume in TRPM2(-/-) mice to cytoprotective effects of neurons, and did not consider the effects the lack of TRPM2 may have on non-neuronal cells, particularly those involved in immune response. One potential non- neuronal cell that may influence ischemic damage are microglia, a type of glia found in the brain. These cells are the resident macrophages of the brain and are responsible for removing foreign material, and damaged/apoptotic cells. Microglia are also known to secrete large amounts of

H2O2 and NO, which are used to further break down damaged cells (Gehrmann et al., 1995). Earlier studies have suggested that TRPM2 plays a role in microglia function following ischemia. Oxidative stress has been shown to increase Ca2+ influx in microglia and after MCAO TRPM2 mRNA in microglia is elevated (Fonfria et al., 2006; Kraft et al., 2004). The faction of TRPM2 in microglia is not fully understood, which makes it difficult to determine what effect, if (-/-) any, the TRPM2 has in microglia-mediated neuronal protection. Since H2O2 is secreted by microglia and is an activator of TRPM2 channels, it is possible that TRPM2 may regulate ROS secretion in microglia, which in turn may reduce neuronal damage.

In addition to microglia, TRPM2 may also regulate the function of other glia cells. It is unknown whether TRPM2 is expressed in other glial cells, with the exception of astrocytes where studies have shown the absence of TRPM2 mRNA (Kraft et al., 2004). However, TRPM2 has been shown to play a role in the innate immune system of the body. TRPM2 is expressed in monocytes, where the channel is responsible for the production of the cytokine CXCL2. In the absence of TRPM2, monocytes have reduced CXCL2 production under oxidative stress conditions and a suppression of inflammatory response (Yamamoto et al., 2008). It has been reported that CXCL2 is expressed in glia cells (primarily astrocytes and microglia), however the mechanism of the chemokine secretion is unknown (Otto et al., 2000). This suggests that

115

TRPM2(-/-) glia, like monocytes, may have reduced CXCL2 production and thus lower inflammatory response. It should be noted that gliosis following an ischemic stroke occurs at different time points in the penumbra, where movement of microglia, cytokine response and inflammation occurs within 1-6 hours after a stroke; the remyelination and proliferation of astrocytes occurs about 3-5 days after injury resulting in a glial scar (Chavez et al., 2009). It is unknown the effects the absence of TRPM2 has on later stages of gliosis as we did not observe those time points.Here we suggest there is strong evidence that absence of TRPM2 may regulate immune signaling and microglia function in the brain, which in turn may have provided additional neuroprotective effects in our in vivo model.

8.5.3 Gender and neuroprotection

A recent study has shown that susceptibility to neuroprotection by TRPM2 inhibition is sex dependent, where as TRPM2 inhibition (using non-specific antagonists) provided neuroprotection in neurons from males, but not females, following ischemic conditions (Jia et al., 2011). These findings may explain why earlier studies in vitro using gender mixed cell cultures were inconclusive, since those studies may have used neurson from both male and female mice. This study has used only male mice both in vivo and in vitro experiments. However, in our in vitro OGD experiments we found that in the absence of TRPM2 both male-only and gender mixed cell cultures had similar levels of neuroprotection. To avoid gender bias, gender mixed cell cultures from multiple litters were used (4 litters). The difference between this study and that by Jia et al. (2011) is that we used the TRPM2(-/-) model as opposed to non-specific antagonists, which may explain our results, as these antagonists may affect gender-dependent cell survival/death via mechanisms independent of TRPM2. Overall, our experiments did take into consideration the possible influence of gender in the neuroprotection observed in TRPM2(-/-) mice, and therefore most of our findings should be considered for males.

8.5.4 Other channels modulated by TRPM2(-/-)

We have shown that absence of TRPM2 is correlated with a modulation of NR2 subunits, which leads to the obvious question as to whether this lack of TRPM2 modulates other channels, particularly those involved in cell survival/death. Xie et al. (2011) demonstrated that absence of TRPM2 reduces AMPAR subunit GluR2 levels. The GluR2 subunit is responsible for mediating cation conductances via AMPAR, specifically Ca2+ and Na+ and during cell death, GluR2 has

116 been shown to play a role in Na+-mediated neurotoxicity (Iihara et al., 2001). Although, we were unable to detect any changes to GluR2 expression, it should be taken into consideration that TRPM2 may regulate AMPAR expression and AMPAR-mediated Na+ conductance. In addition to glutamate receptors, TRPM7 is known to play a role in Ca2+-overload following ischemia and mRNA silencing of TRPM7 reduces TRPM2 levels (Aarts et al., 2003). In TRPM2(-/-) Hippocampi, we found there was no change in TRPM7 protein expression, suggesting that TRPM7 expression may not be dependent on presence of TRPM2. In addition to modulation of cation channels, the chronic absence of TRPM2 may have other compensatory mechanisms downstream in neurons which may need to be explored further.

8.6 TRPM2 as a therapeutic target

When starting this study one of the first questions we asked was whether TRPM2 channels could be a viable therapeutic target for treating stroke. Using the TRPM2(-/-) mouse we found that absence of TRPM2 was neuroprotective using in vivo and in vitro stroke paradigms. Furthermore we found that the TRPM2(-/-) did not affect the viability, reproduction or observable behavior of the mutant mice, suggesting a lack of TRPM2 has few side effects under normal conditions, making it a more appealing therapeutic target. One of the major limitations to using TRPM2 inhibition to treat stroke is the lack of TRPM2 specific antagonists. Here we used a mouse model with a genetic knockout, something that cannot be done at the human level. Furthermore, the role of TRPM2 channels in humans may be different from mice, and thus blocking TRPM2 in humans may have side effects or no neuroprotection at all, as was seen in the case of NMDAR inhibitor clinical trials (Davis et al., 1997; Ikonomidou and Turski, 2002; Lees et al., 2000). The translation from mouse models to clinical trials may require intermediary animals such as gyrencephalic non-human primates (Cook et al., 2012). The lack of specific TRPM2 antagonists in combination with the unknown effects of translating TRPM2 inhibition to clinical trials, tells us that there is a long way to go before TRPM2 can be a viable therapeutic target in humans.

8.6.1 Combined inhibition of TRPM2 and other therapeutic targets

Following the failure of NMDAR antagonists in treating clinical stroke patients, a number of neuroprotective drugs have been discovered in the lab that target specific channels including TRPM7 channels and PSD-95-NR2B interactions of NMDAR (Sattler et al., 1999;

117

Aarts et al., 2003; Aarts et al., 2002). Since a number of channels involved Ca2+-overload during ischemia, theoretically a combination of neuroprotective agents targeting these channels may have additive cytoprotective effects in neurons. Unfortunately, we found that during OGD, using known neuroprotective drugs in TRPM2(-/-) neurons did not provide enhanced the neuroprotection as compared to these drugs on WT neurons. The combined effects of these drugs and TRPM2(-/-) mutation did reduce cell death and dying cells when compared to TRPM2(-/-) neurons alone following OGD. We predict our observations can be explained by a "ceiling effect" of neurotoxic Ca2+-influx, where there is a maximal level of protection achieved by blocking certain Ca2+ entry points. Also, in TRPM2(-/-) neurons, which provide some neuroprotection, does not reach that neuroprotection "ceiling effect" observed in the drugs we tested. Although we were unable to demonstrate additive neuroprotection by blocking multiple Ca2+ entry points, we did therapeutic insights that blocking multiple channels may be redundant when it comes to treating stroke.

8.6.2 TRPM2 in other neurodegenerative diseases

Although we have described a role for TRPM2 in ischemic stroke, it remains to be seen whether the findings in this thesis have implications of the role TRPM2 plays in other pathologies. Oxidative stress is common in a number of neurodegenerative diseases and has been shown to contribute to neuronal death of a number of disorders including Alzheimer’s (Benzi and Moretti, 1995) and Parkinson’s disease (Jenner et al., 1992). It has been suggested that ROS- induced TRPM2 activation may play a key role in cell death during Alzheimer's Disease (Olah et al., 2009;Fonfria et al., 2005). In vitro studies using amyloid-beta (Aβ) peptides to mimic Alzheimer's conditions has shown that neurons treated with a TRPM2 dominant negative allele had reduced cell death compared to WT. Recently, a similar study using neurons from TRPM2(-/- ) mice demonstrated similar neuroprotection in TRPM2-null neurons following Aβ treatment5. Also, NMDARs have been implicated to play a role in Alzheimer's, Aβ is can regulate NMDAR expression and trafficking in neurons (Snyder et al., 2005). Since we have shown that TRPM2 can also modulate NMDAR subunit expression, it is possible that the relationship we describe between NMDAR and TRPM2 channels may also affect Aβ-mediated neuronal death.

5 Personal communication and poster presentation at BRAIN Platform day from JF Macdonald Lab

118

A recent study has shown that TRPM2 inhibition (using intracellular TRPM2 antibodies) inhibits NMDA-mediated bursting pattern in SNr GABAergic neurons (Lee et al., 2013). This bursting pattern is known to be associated with the depletion of dopamine in SNr GABAergic neurons, phenomenon that has been implicated in Parkinson's disease Wichmann et al., 1999; Wichmann and Dostrovsky, 2011). Furthermore the duration of the NMDA-mediated bursting pattern is increased in the presence of H2O2, a known TRPM2 activator (Lee et al., 2013). Both this study and the study by Lee et al. (2013) provides evidence suggesting that TRPM2 channels may be a modulator of NMDAR expression and function, however the exact mechanism in different neuronal types remains unknown. In addition to burst firing, like in Alzheimer's disease, ROS also plays a role in neurodegeneration of Parkinson's disease, suggesting that TRPM2 may also play a role in neuronal death (Turnbull et al., 2001;Dauer and Przedborski, 2003).

8.7 Concluding remarks

Overall this study provides a novel mechanism on how chronic TRPM2(-/-) results in neuroprotection by changes in NMDAR subunit expression. We are also the first to show that lack of TRPM2 is neuroprotective in vivo, providing a therapeutic role in whole animal model. The findings we presented raise some interesting questions about how TRPM2 can be a target for therapeutic techniques following ischemia. It is presently unknown whether acutely blocking TRPM2 will have a similar effect on NMDAR expression. Therefore, a fuller understanding of how TRPM2 regulates NMDARs and potentially other channels is an area that needs to be further studied in order to provide insights on ischemic mechanisms.

119

Chapter 9

9 Future Directions

Although this study has given some novel findings, there are still a number of questions that remain unanswered. Two major issues to come out the discussion are: how are NMDAR subunits modulated by TRPM2 and the need to develop a TRPM2 specific antagonist. Here, we will discuss some possible future projects to resolve these two issues.

9.1 Resolving mechanism of how TRPM2(-/-) modulates NMDAR

There is very little known about downstream signaling pathways of TRPM2, which makes it difficult to determine how TRPM2 modulates NR2 subunits. Furthermore, since our model is chronic, it also raises the question as to whether TRPM2 modulates NR2 subunits during development. We propose 2 experiments that could shed some light on the mechanism by which TRPM2 modulates NR2 subunits

1) NR2 levels during development

To resolve at what stage of development does TRPM2 channels modulate NR2 subunits we propose measuring both mRNA and protein levels in the hippocampus for pre-natal, shortly after post-natal and adult WT and TRPM2(-/-) mice. These time points were selected because that is when the NR2A-NR2B switch occurs and it will help us determine if TRPM2 plays a role during development. If TRPM2 is involved in modifying the NR2A-NR2B switch, then it raises questions as to the role of TRPM2 in pre-natal cell death, because during that stage of development NR2A is primarily involved in neuronal death instead of NR2B (Zhou and Baudry, 2006).

2) Measuring expression of known transcription and translation factors

Once we have determined at what stage of development TRPM2 plays a role in NR2 modulation, the next experiment is to explore whether TRPM2 modulates transcription, translation or post-translational factors that may affect NR2 expression. A way to narrow down the candidates is to see if NR2 levels are changed in mRNA or in protein when comparing

120 hippocampi from WT and TRPM2(-/-) animals. If there is a change in mRNA level then one can focus on transcription factors, if not then TRPM2 channels are likely acting on translational or post-translational factors. Both activation and expression levels of transcription/translation factors of interest should be measured.

It should be noted that it is possible that TRPM2 is part of the NMDAR complex, and TRPM2 is required in forming that complex. Since current commercially available TRPM2 antibodies are non-specific, it is difficult to conduct association studies for TRPM2 and NMDAR complex. One method that may work is to use a protein tag inserted into the TRPM2 sequence that can be easily detected by antibodies and can be used to see if TRPM2 is physically associated with NMDARs. The obvious pitfall to this experiment is that insertion of foreign sequences may disrupt function and folding of TRPM2 channels.

9.2 Development of acute and specific TRPM2 inhibitor

The development of a TRPM2-specific inhibitor would be invaluable for future research in TRPM2 channels and potential therapeutic implications. The first step towards generating a specific antagonist would be to develop an assay where drugs can be used to test their inhibition of TRPM2 channel function. The most likely assay would use TRPM2-like currents, where various small molecules could be tested using electrophysiology to see whether they inhibit TRPM2-like conductances. The obvious imitation to using this assay is asking whether the drugs directly affect TRPM2 or are indirectly affecting TRPM2 signaling (ie: via ADPR). Another limitation is whether the drug affects channel function or enzyme function (NudT9-H) in the TRPM2 protein, both of which could be independent of each other. Finally, the TRPM2-like current may consist of other channels activated by either ADPR or H2O2 and drugs may give false positive results in the assay. Overall, without fully understanding the physiology of TRPM2 channels, the development of a TRPM2-specific inhibitor will be difficult, and it will be a number of years before we see a therapeutic drug that targets TRPM2 channels.

121

References Cited

Aarts M, Iihara K, Wei WL, Xiong ZG, Arundine M, Cerwinski W, MacDonald JF, Tymianski M (2003) A key role for TRPM7 channels in anoxic neuronal death. Cell 115: 863-877.

Aarts M, Liu YT, Liu LD, Besshoh S, Arundine M, Gurd JW, Wang YT, Salter MW, Tymianski M (2002) Treatment of ischemic brain damage by perturbing NMDA receptor-PSD-95 protein interactions. Science 298: 846-850.

Aarts MM, Tymianski M (2004) Molecular mechanisms underlying specificity of excitotoxic signaling in neurons. Current Molecular Medicine 4: 137-147.

Aarts MM, Tymianski M (2005a) TRPM7 and ischemic CNS injury. Neuroscientist 11: 116-123.

Aarts MM, Tymianski M (2005b) TRPMs and neuronal cell death. Pflugers Archiv-European Journal of Physiology 451: 243-249.

Ahmed N, Wahlgren N, Brainin M, Castillo J, Ford GA, Kaste M, Lees KR, Toni D (2009) Relationship of blood pressure, antihypertensive therapy, and outcome in ischemic stroke treated with intravenous thrombolysis: retrospective analysis from Safe Implementation of Thrombolysis in Stroke-International Stroke Thrombolysis Register (SITS-ISTR). Stroke 40: 2442-2449.

Allen NJ, Attwell D (2002) Modulation of ASIC channels in rat cerebellar Purkinje neurons by ischaemia-related signals. J Physiol 543: 521-529.

Alter BJ, Gereau RW (2008) Hotheaded: TRPV1 as mediator of hippocampal synaptic plasticity. Neuron 57: 629-631.

American Stroke Association Web Site. Stroke. 200. Ref Type: Report

Arundine M, Tymianski M (2004) Molecular mechanisms of glutamate-dependent neurodegeneration in ischemia and traumatic brain injury. Cellular and Molecular Life Sciences 61: 657-668.

Ashkenazi A, Dixit VM (1998) Death receptors: signaling and modulation. Science 281: 1305- 1308.

Atamna H, Ginsburg H (1995) Heme degradation in the presence of glutathione. A proposed mechanism to account for the high levels of non-heme iron found in the membranes of hemoglobinopathic red blood cells. J Biol Chem 270: 24876-24883.

Auberson YP, Allgeier H, Bischoff S, Lingenhoehl K, Moretti R, Schmutz M (2002) 5- Phosphonomethylquinoxalinediones as competitive NMDA receptor antagonists with a preference for the human 1A/2A, rather than 1A/2B receptor composition. Bioorg Med Chem Lett 12: 1099-1102.

122

Bai JZ, Lipski J (2010) Differential expression of TRPM2 and TRPV4 channels and their potential role in oxidative stress-induced cell death in organotypic hippocampal culture. Neurotoxicology 31: 204-214.

Baker MG, Graham L (2004) The journey: Parkinson's disease. British Medical Journal 329: 611-614.

Bano D, Young KW, Guerin CJ, Lefeuvre R, Rothwell NJ, Naldini L, Rizzuto R, Carafoli E, Nicotera P (2005) Cleavage of the plasma membrane Na+/Ca2+ exchanger in excitotoxicity. Cell 120: 275-285.

Barbe MT, Monyer H, Bruzzone R (2006) Cell-cell communication beyond connexins: the pannexin channels. Physiology (Bethesda ) 21: 103-114.

Barber PA, Hoyte L, Colbourne F, Buchan AM (2004) Temperature-regulated model of focal ischemia in the mouse - A study with histopathological and behavioral outcomes. Stroke 35: 1720-1725.

Bederson JB, Pitts LH, Tsuji M, Nishimura MC, Davis RL, Bartkowski H (1986) Rat Middle Cerebral-Artery Occlusion - Evaluation of the Model and Development of A Neurologic Examination. Stroke 17: 472-476.

Bell KF, Bent RJ, Meese-Tamuri S, Ali A, Forder JP, Aarts MM (2013) Calmodulin Kinase IV- dependent CREB activation is required for neuroprotection via NMDA receptor-PSD95 disruption. J Neurochem.

Benmira S, Banda ZK, Bhattacharya V (2011) The start of a new era for stroke treatment: mechanical thrombectomy devices. Curr Neurovasc Res 8: 75-85.

Benzi G, Moretti A (1995) Are Reactive Oxygen Species Involved in Alzheimers-Disease. Neurobiology of Aging 16: 661-674.

Besancon E, Guo SZ, Lok J, Tymianski M, Lo EH (2008) Beyond NMDA and AMPA glutamate receptors: emerging mechanisms for ionic imbalance and cell death in stroke. Trends in Pharmacological Sciences 29: 268-275.

Bindokas VP, Miller RJ (1995) Excitotoxic degeneration is initiated at non-random sites in cultured rat cerebellar neurons. J Neurosci 15: 6999-7011.

Bischofberger J, Engel D, Li L, Geiger JR, Jonas P (2006) Patch-clamp recording from mossy fiber terminals in hippocampal slices. Nat Protoc 1: 2075-2081.

Bliss TVP, Collingridge GL (1993) A synaptic model of memory:long-term potentiation in the hippocampus. Nature 361: 31-39.

Boscia F, Gala R, Pignataro G, de Bartolomeis A, Cicale M, Ambesi-Impiombato A, Di Renzo G, Annunziato L (2006) Permanent focal brain ischemia induces isoform-dependent changes in the pattern of Na+/Ca2+ exchanger gene expression in the ischemic core, periinfarct area, and intact brain regions. J Cereb Blood Flow Metab 26: 502-517.

123

Brenman JE, Christopherson KS, Craven SE, McGee AW, Bredt DS (1996) Cloning and characterization of postsynaptic density 93, a nitric oxide synthase interacting protein. J Neurosci 16: 7407-7415.

Brookmeyer R, Johnson E, Ziegler-Graham K, Arrighi HM (2007) Forecasting the global burden of Alzheimer's disease. Alzheimers & Dementia 3: 186-191.

Brown RC, Davis TP (2002) Calcium modulation of adherens and tight junction function: a potential mechanism for blood-brain barrier disruption after stroke. Stroke 33: 1706-1711.

Bruno V, Battaglia G, Copani A, D'Onofrio M, Di Iorio P, De Blasi A, Melchiorri D, Flor PJ, Nicoletti F (2001) Metabotropic glutamate receptor subtypes as targets for neuroprotective drugs. J Cereb Blood Flow Metab 21: 1013-1033.

Busto R, Dietrich WD, Globus MY, Ginsberg MD (1989) The importance of brain temperature in cerebral ischemic injury. Stroke 20: 1113-1114.

Castillo B, Porzgen P, Penner R, Horgen FD, Fleig A (2010) Development and optimization of a high-throughput bioassay for TRPM7 ion channel inhibitors. J Biomol Screen 15: 498-507.

Carmichael ST (2005) Rodent models of focal ischemia: size, mechanism and purpose. NeuroRx 23(2): 273-279.

Chan PH (1996) Role of oxidants in ischemic brain damage. Stroke 27: 1124-1129.

Chanock SJ, el Benna J, Smith RM, Babior BM (1994) The respiratory burst oxidase. J Biol Chem 269: 24519-24522.

Chavez JC, Hurko O, Barone FC, Feurstein GZ (2009) Pharmacologic interventions of stroke. Stroke 40: e558-e563.

Chen M, Lu TJ, Chen XJ, Zhou Y, Chen Q, Feng XY, Xu L, Duan WH, Xiong ZQ (2008) Differential Roles of NMDA Receptor Subtypes in Ischemic Neuronal Cell Death and Ischemic Tolerance. Stroke 39: 3042-3048.

Cho KO, Hunt CA, Kennedy MB (1992) The rat brain postsynaptic density fraction contains a homolog of the Drosophila discs-large tumor suppressor protein. Neuron 9: 929-942.

Choi DW (1985) Glutamate neurotoxicity in cortical cell culture is calcium dependent. Neurosci Lett 58: 293-297.

Choi DW, Maulucci-Gedde M, Kriegstein AR (1987) Glutamate neurotoxicity in cortical cell culture. J Neurosci 7(2): 357-368.

Chong ZZ, Li F, Maiese K (2005) Oxidative stress in the brain: novel cellular targets that govern survival during neurodegenerative disease. Prog Neurobiol 75: 207-246.

124

Chung HJ, Huang YH, Lau LF, Huganir RL (2004) Regulation of the NMDA receptor complex and trafficking by activity-dependent phosphorylation of the NR2B subunit PDZ ligand. J Neurosci 24: 10248-10259.

Ciabarra AM, Sullivan JM, Gahn LG, Pecht G, Heinemann S, Sevarino KA (1995) Cloning and characterization of chi-1: a developmentally regulated member of a novel class of the ionotropic glutamate receptor family. J Neurosci 15: 6498-6508.

Colton CA, Fagni L, Gilbert D (1989) The Action of Hydrogen-Peroxide on Paired Pulse and Long-Term Potentiation in the Hippocampus. Free Radical Biology and Medicine 7: 3-8.

Contreras JE, Sanchez HA, Veliz LP, Bukauskas FF, Bennett MVL, Saez JC (2004) Role of connexin-based gap junction channels and hemichannels in ischemia-induced cell death in nervous tissue. Brain Research Reviews 47: 290-303.

Cook DJ, Teves L, Tymianski M (2012) Treatment of stroke with a PSD-95 inhibitor in the gyrencephalic primate brain. Nature 483: 213-217.

Corbett D, Evans S, Thomas C, Wang D, Jonas RA (1990) MK-801 reduced cerebral ischemic injury by inducing hypothermia. Brain Res 514: 300-304.

Cosens DJ, Manning A (1969) Abnormal electroretinogram from a Drosophila mutant. Nature 224: 285-287.

Crow JP, Beckman JS (1995) Reactions between nitric oxide, superoxide, and peroxynitrite: footprints of peroxynitrite in vivo. Adv Pharmacol 34: 17-43.

Crowder RJ, Freeman RS (2000) Glycogen synthase kinase-3 beta activity is critical for neuronal death caused by inhibiting phosphatidylinositol 3-kinase or Akt but not for death caused by nerve growth factor withdrawal. J Biol Chem 275: 34266-34271.

Cui H, Hayashi A, Sun HS, Belmares MP, Cobey C, Phan T, Schweizer J, Salter MW, Wang YT, Tasker RA, Garman D, Rabinowitz J, Lu PS, Tymianski M (2007) PDZ protein interactions underlying NMDA receptor-mediated excitotoxicity and neuroprotection by PSD-95 inhibitors. Journal of Neuroscience 27: 9901-9915.

Czyz A, Kiedrowski L (2002) In depolarized and glucose-deprived neurons, Na+ influx reverses plasmalemmal K+-dependent and K+-independent Na+/Ca2+ exchangers and contributes to NMDA excitotoxicity. J Neurochem 83: 1321-1328.

Dauer W, Przedborski S (2003) Parkinson's disease: mechanisms and models. Neuron 39: 889- 909.

Davis SM, Albers GW, Diener HC, Lees KR, Norris J (1997) Termination of acute stroke studies involving selfotel treatment. Lancet 349: 32.

Dawson VL, Dawson TM, London ED, Bredt DS, Snyder SH (1991) Nitric oxide mediates glutamate neurotoxicity in primary cortical cultures. Proc Natl Acad Sci USA 88: 6368-6371.

125

De Freitas GR, De HC, Bogousslavsky J (2009) Topographic classification of ischemic stroke. Handb Clin Neurol 93: 425-452.

Dejean LM, Martinez-Caballero S, Kinnally KW (2006) Is MAC the knife that cuts cytochrome c from mitochondria during apoptosis? Cell Death Differ 13: 1387-1395.

Diarra A, Sheldon C, Brett CL, Baimbridge KG, Church J (1999) Anoxia-evoked intracellular pH and Ca2+ concentration changes in cultured postnatal rat hippocampal neurons. Neuroscience 93: 1003-1016.

Downward J (2003) Targeting RAS signalling pathways in cancer therapy. Nat Rev Cancer 3: 11-22.

Feng B, Morley RM, Jane DE, Monaghan DT (2005) The effect of competitive antagonist chain length on NMDA receptor subunit selectivity. Neuropharmacology 48: 354-359.

Fesik SW, Shi Y (2001) Structural biology. Controlling the caspases. Science 294: 1477-1478.

Fonfria E, Marshall IC, Boyfield I, Skaper SD, Hughes JP, Owen DE, Zhang W, Miller BA, Benham CD, McNulty S (2005) Amyloid beta-peptide(1-42) and hydrogen peroxide-induced toxicity are mediated by TRPM2 in rat primary striatal cultures. J Neurochem 95: 715-723.

Fonfria E, Mattei C, Hill K, Brown JT, Randall A, Benham CD, Skaper SD, Campbell CA, Crook B, Murdock PR, Wilson JM, Maurio FP, Owen DE, Tilling PL, McNulty S (2006) TRPM2 is elevated in the tMCAO stroke model, transcriptionally regulated, and functionally expressed in C13 microglia. Journal of Receptors and Signal Transduction 26: 179-198.

Frantseva MV, Kokarovtseva L, Perez Velazquez JL (2002) Ischemia-induced brain damage depends on specific gap-junctional coupling. J Cereb Blood Flow Metab 22: 453-462.

Friedman WJ (2000) Neurotrophins induce death of hippocampal neurons via the p75 receptor. J Neurosci 20: 6340-6346.

Galimanis A, Mono ML, Arnold M, Nedeltchev K, Mattle HP (2009) Lifestyle and stroke risk: a review. Curr Opin Neurol 22: 60-68.

Garrido C, Galluzzi L, Brunet M, Puig PE, Didelot C, Kroemer G (2006) Mechanisms of cytochrome c release from mitochondria. Cell Death Differ 13: 1423-1433.

Gehrmann J, Matsumoto Y, Kreutzberg GW (1995) Microglia: intrinsic immuneffector cell of the brain. Brain Res Brain Res Rev 20: 269-287.

Ginsberg MD, Busto R (1989) Rodent models of cerebral ischemia. Stroke 20: 1627-1642.

Ginsberg MD, Busto R (1998) Combating hyperthermia in acute stroke: a significant clinical concern. Stroke 29: 529-534.

126

Goldberg MP, Choi DW (1993) Combined Oxygen and Glucose Deprivation in Cortical Cell- Culture - Calcium-Dependent and Calcium-Independent Mechanisms of Neuronal Injury. Journal of Neuroscience 13: 3510-3524.

Goldberg MP, Pham PC, Choi DW (1987) Dextrorphan and dextromethorphan attenuate hypoxic injury in neuronal culture. Neurosci Lett 80: 11-15.

Gorelick PB, Sacco RL, Smith DB, Alberts M, Mustone-Alexander L, Rader D, Ross JL, Raps E, Ozer MN, Brass LM, Malone ME, Goldberg S, Booss J, Hanley DF, Toole JF, Greengold NL, Rhew DC (1999) Prevention of a first stroke: a review of guidelines and a multidisciplinary consensus statement from the National Stroke Association. JAMA 281: 1112-1120.

Gribkoff VK, Winquist RJ (2005) Voltage-gated cation channel modulators for the treatment of stroke. Expert Opin Investig Drugs 14: 579-592.

Halliwell B (2006a) Oxidative stress and neurodegeneration: where are we now? J Neurochem 97: 1634-1658.

Halliwell B (2006b) Phagocyte-derived reactive species: salvation or suicide? Trends Biochem Sci 31: 509-515.

Halliwell B, Gutteridge JMC (1990) Role of Free-Radicals and Catalytic Metal-Ions in Human- Disease - An Overview. Methods in Enzymology 186: 1-85.

Hara Y, Wakamori M, Ishii M, Maeno E, Nishida M, Yoshida T, Yamada H, Shimizu S, Mori E, Kudoh J, Shimizu N, Kurose H, Okada Y, Imoto K, Mori Y (2002) LTRPC2 Ca2+-permeable channel activated by changes in redox status confers susceptibility to cell death. Molecular Cell 9: 163-173.

Haraguchi K, Kawamoto A, Isami K, Maeda S, Kusano A, Asakura K, Shirakawa H, Mori Y, Nakagawa T, Kaneko S (2012) TRPM2 contributes to inflammatory and neuropathic pain through the aggravation of pronociceptive inflammatory responses in mice. J Neurosci 32: 3931- 3941.

Hardingham GE (2006a) 2B synaptic or extrasynaptic determines signalling from the NMDA receptor. J Physiol 572: 614-615.

Hardingham GE (2006b) Pro-survival signalling from the NMDA receptor. Biochem Soc Trans 34: 936-938.

Hardingham GE, Arnold FJ, Bading H (2001a) A calcium microdomain near NMDA receptors: on switch for ERK-dependent synapse-to-nucleus communication. Nat Neurosci 4: 565-566.

Hardingham GE, Arnold FJ, Bading H (2001b) Nuclear calcium signaling controls CREB- mediated gene expression triggered by synaptic activity. Nat Neurosci 4: 261-267.

Hardingham GE, Bading H (1999) Calcium as a versatile second messenger in the control of gene expression. Microsc Res Tech 46: 348-355.

127

Hardingham GE, Bading H (2002) Coupling of extrasynaptic NMDA receptors to a CREB shut- off pathway is developmentally regulated. Biochim Biophys Acta 1600: 148-153.

Hardingham GE, Bading H (2003) The Yin and Yang of NMDA receptor signalling. Trends Neurosci 26: 81-89.

Hardingham GE, Bading H (2010) Synaptic versus extrasynaptic NMDA receptor signalling: implications for neurodegenerative disorders. Nat Rev Neurosci 11: 682-696.

Hardingham GE, Fukunaga Y, Bading H (2002) Extrasynaptic NMDARs oppose synaptic NMDARs by triggering CREB shut-off and cell death pathways. Nat Neurosci 5: 405-414.

Harteneck C, Frenzel H, Kraft R (2007) N-(p-amylcinnamoyl)anthranilic acid (ACA): A phospholipase A(2) inhibitor and TRP channel blocker. Cardiovascular Drug Reviews 25: 61-75.

Hetman M, Cavanaugh JE, Kimelman D, Xia ZG (2000) Role of glycogen synthase kinase-3 beta in neuronal apoptosis induced by trophic withdrawal. J Neurosci 20: 2567-2574.

Hetman M, Kharebava G (2006) Survival signaling pathways activated by NMDA receptors. Curr Top Med Chem 6: 787-799.

Hill K, Benham CD, McNulty S, Randall AD (2004a) Flufenamic acid is a pH-dependent antagonist of TRPM2 channels. Neuropharmacology 47: 450-460.

Hill K, McNulty S, Randall AD (2004b) Inhibition of TRPM2 channels by the antifungal agents clotrimazole and econazole. Naunyn Schmiedebergs Arch Pharmacol 370: 227-237.

Hiroi T, Wajima T, Negoro T, Ishii M, Nakano Y, Kiuchi Y, Mori Y, Shimizu S (2012) Neutrophil TRPM2 channels are implicated in the exacerbation of myocardial ischaemia/reperfusion injury. Cardiovasc Res.

Hiroi T, Wajima T, Negoro T, Ishii M, Nakano Y, Kiuchi Y, Mori Y, Shimizu S (2013) Neutrophil TRPM2 channels are implicated in the exacerbation of myocardial ischaemia/reperfusion injury. Cardiovasc Res 97: 271-281.

Horn J, Limburg M (2000) Calcium antagonists for acute ischemic stroke. Cochrane Database Syst Rev CD001928.

Huang CC, Gean PW (1994) Paired-pulse depression of the N-methyl-D-aspartate receptor- mediated synaptic potentials in the amygdala. Br J Pharmacol 113: 1029-1035.

Hunt CA, Schenker LJ, Kennedy MB (1996) PSD-95 is associated with the postsynaptic density and not with the presynaptic membrane at forebrain synapses. J Neurosci 16: 1380-1388.

Husi H, Ward MA, Choudhary JS, Blackstock WP, Grant SG (2000) Proteomic analysis of NMDA receptor-adhesion protein signaling complexes. Nat Neurosci 3: 661-669.

128

Iihara K, Joo DT, Henderson J, Sattler R, Taverna FA, Lourensen S, Orser BA, Roder JC, Tymianski M (2001) The influence of glutamate receptor 2 expression on excitotoxicity in GluR2 null mutant mice. Journal of Neuroscience 21: 2224-2239.

Ikonomidou C, Turski L (2002) Why did NMDA receptor antagonists fail clinical trials for stroke and traumatic brain injury? Lancet Neurol 1: 383-386.

Immke DC, McCleskey EW (2001) Lactate enhances the acid-sensing Na+ channel on ischemia- sensing neurons. Nat Neurosci 4: 869-870.

Immke DC, McCleskey EW (2003) Protons open acid-sensing ion channels by catalyzing relief of Ca2+ blockade. Neuron 37: 75-84.

Ishii M, Shimizu S, Hagiwara T, Wajima T, Miyazaki A, Mori Y, Kiuchi Y (2006) Extracellular- added ADP-ribose increases intracellular free Ca2+ concentration through Ca2+ release from stores, but not through TRPM2-mediated Ca2+ entry, in rat beta-cell line RIN-5F. Journal of Pharmacological Sciences 101: 174-178.

Jeffs GJ, Meloni BP, Bakker AJ, Knuckey NW (2007) The role of the Na(+)/Ca(2+) exchanger (NCX) in neurons following ischaemia. J Clin Neurosci 14: 507-514.

Jenner P, Dexter DT, Sian J, Schapira AHV, Marsden CD (1992) Oxidative Stress As A Cause of Nigral Cell-Death in Parkinsons-Disease and Incidental Lewy Body Disease. Annals of Neurology 32: S82-S87.

Jia J, Verma S, Nakayama S, Quillinan N, Grafe MR, Hurn PD, Herson PS (2011) Sex differences in neuroprotection provided by inhibition of TRPM2 channels following experimental stroke. J Cereb Blood Flow Metab.

Jia Y, Zhou J, Tai Y, Wang Y (2007) TRPC channels promote cerebellar granule neuron survival. Nat Neurosci 10: 559-567.

Jiang X, Tian F, Mearow K, Okagaki P, Lipsky RH, Marini AM (2005) The excitoprotective effect of N-methyl-D-aspartate receptors is mediated by a brain-derived neurotrophic factor autocrine loop in cultured hippocampal neurons. J Neurochem 94: 713-722.

Jin J, Desai BN, Navarro B, Donovan A, Andrews NC, Clapham DE (2008) Deletion of Trpm7 Disrupts Embryonic Development and Thymopoiesis Without Altering Mg2+ Homeostasis. Science 322: 756-760.

Kamsler A, Segal M (2003) Hydrogen peroxide modulation of synaptic plasticity. Journal of Neuroscience 23: 269-276.

Kaneko S, Kawakami S, Hara Y, Wakamori M, Itoh E, Minami T, Takada Y, Kume T, Katsuki H, Mori Y, Akaike A (2006) A critical role of TRPM2 in neuronal cell death by hydrogen peroxide. Journal of Pharmacological Sciences 101: 66-76.

129

Kashio M, Sokabe T, Shintaku K, Uematsu T, Fukuta N, Kobayashi N, Mori Y, Tominaga M (2012) Redox signal-mediated sensitization of transient receptor potential melastatin 2 (TRPM2) to temperature affects macrophage functions. Proc Natl Acad Sci U S A 109: 6745-6750.

Keinanen K, Wisden W, Sommer B, Werner P, Herb A, Verdoorn TA, Sakmann B, Seeburg PH (1990) A family of AMPA-selective glutamate receptors. Science 249: 556-560.

Kim HA, Brait VH, Lee S, De Silva TM, Diep H, Eisenhardt A, Drummond GR, Sobey CG (2012) Brain infarct volume after permanent focal ischemia is not dependent on Nox2 expression. Brain Res 1483: 105-111.

Kimelberg HK (2005) Astrocytic swelling in cerebral ischemia as a possible cause of injury and target for therapy. Glia 50: 389-397.

Kimelberg HK, Jin Y, Charniga C, Feustel PJ (2003) Neuroprotective activity of tamoxifen in permanent focal ischemia. J Neurosurg 99: 138-142.

Kimelberg HK, MacVicar BA, Sontheimer H (2006) Anion channels in astrocytes: biophysics, pharmacology, and function. Glia 54: 747-757.

Knight JA (1998) Free radicals: Their history and current status in aging and disease. Annals of Clinical and Laboratory Science 28: 331-346.

Kornau HC, Schenker LT, Kennedy MB, Seeburg PH (1995) Domain interaction between NMDA receptor subunits and the postsynaptic density protein PSD-95. Science 269: 1737-1740.

Kraft R, Grimm C, Grosse K, Hoffmann A, Sauerbruch S, Kettenmann H, Schultz G, Harteneck C (2004) Hydrogen peroxide and ADP-ribose induce TRPM2-mediated calcium influx and cation currents in microglia. Am J Physiol Cell Physiol 286: C129-C137.

Krapivinsky G, Mochida S, Krapivinsky L, Cibulsky SM, Clapham DE (2006) The TRPM7 ion channel functions in cholinergic synaptic vesicles and affects transmitter release. Neuron 52: 485-496.

Kristian T, Siesjo BK (1998) Calcium in ischemic cell death. Stroke 29: 705-718.

Kumar S, Bandyopadhyay U (2005) Free heme toxicity and its detoxification systems in human. Toxicol Lett 157: 175-188.

Lange I, Yamamoto S, Partida-Sanchez S, Mori Y, Fleig A, Penner R (2009) TRPM2 Functions as a Lysosomal Ca2+-Release Channel in b Cells. Science Signaling 2: 1-11.

Larsson J, Allhorn M, Kerstrom B (2004) The lipocalin alpha(1)-microglobulin binds heme in different species. Arch Biochem Biophys 432: 196-204.

Lau A, Tymianski M (2010) Glutamate receptors, neurotoxicity and neurodegeneration. Pflugers Arch 460: 525-542.

130

Lee CR, Machold RP, Witkovsky P, Rice ME (2013) TRPM2 channels are required for NMDA- induced burst firing and contribute to H(2)O(2)-dependent modulation in substantia nigra pars reticulata GABAergic neurons. J Neurosci 33: 1157-1168.

Lee KS, Frank S, Vanderklish P, Arai A, Lynch G (1991) Inhibition of proteolysis protects hippocampal neurons from ischemia. Proc Natl Acad Sci USA 88: 7233-7237.

Lees KR, Asplund K, Carolei A, Davis SM, Diener HC, Kaste M, Orgogozo JM, Whitehead J (2000) Glycine antagonist (gavestinel) in neuroprotection (GAIN International) in patients with acute stroke: a randomised controlled trial. Lancet 355: 1949-1954.

Lehmann OJ, Sowden JC, Carlsson P, Jordan T, Bhattacharya SS (2003) Fox's in development and disease. Trends Genet 19: 339-344.

Li B, Chen N, Luo T, Otsu Y, Murphy TH, Raymond LA (2002) Differential regulation of synaptic and extra-synaptic NMDA receptors. Nat Neurosci 5: 833-834.

Li S, Stys PK (2000) Mechanisms of ionotropic glutamate receptor-mediated excitotoxicity in isolated spinal cord white matter. J Neurosci 20: 1190-1198.

Linseman DA, Butts BD, Precht TA, Phelps RA, Le SS, Laessig TA, Bouchard RJ, Florez- McClure ML, Heidenreich KA (2004) Glycogen synthase kinase-3 beta phosphorylates bax and promotes its mitochondrial localization during neuronal apoptosis. J Neurosci 24: 9993-10002.

Liu L, Wong TP, Pozza MF, Lingenhoehl K, Wang Y, Sheng M, Auberson YP, Wang YT (2004) Role of NMDA receptor subtypes in governing the direction of hippocampal synaptic plasticity. Science 304: 1021-1024.

Liu M, Hurn PD, Roselli CE, Alkayed NJ (2007a) Role of P450 aromatase in sex-specific astrocytic cell death. J Cereb Blood Flow Metab 27: 135-141.

Liu YT, Wong TP, Aarts M, Rooyakkers A, Liu LD, Lai TW, Wu DC, Lu J, Tymianski M, Craig AM, Wang YT (2007b) NMDA receptor subunits have differential roles in mediating excitotoxic neuronal death both in vitro and in vivo. Journal of Neuroscience 27: 2846-2857.

Lucas DR, Newhouse JP (1957) The toxic effect of sodium L-glutamate on the inner layers of the retina. AMA Arch Ophthalmol 58: 193-201.

Lynch T, Cherny RA, Bush AI (2000) Oxidative processes in Alzheimer's disease: the role of A beta-metal interactions. Experimental Gerontology 35: 445-451.

Maalouf M, Rho JM (2008) Oxidative Impairment of Hippocampal Long-term Potentiation Involves Activation of Protein Phosphatase 2A and Is Prevented by Ketone Bodies. Journal of Neuroscience Research 86: 3322-3330.

Manev H, Favaron M, Guidotti A, Costa E (1989) Delayed increase of Ca2+ influx elicited by glutamate: role in neuronal death. Mol Pharmacol 36: 106-112.

131

Marcoux FW, Probert AW, Jr., Weber ML (1990) Hypoxic neuronal injury in tissue culture is associated with delayed calcium accumulation. Stroke 21: III71-III74.

Martel MA, Soriano FX, Baxter P, Rickman C, Duncan R, Wyllie DJ, Hardingham GE (2009) Inhibiting pro-death NMDA receptor signaling dependent on the NR2 PDZ ligand may not affect synaptic function or synaptic NMDA receptor signaling to gene expression. Channels (Austin ) 3: 12-15.

Massey PV, Johnson BE, Moult PR, Auberson YP, Brown MW, Molnar E, Collingridge GL, Bashir ZI (2004) Differential roles of NR2A and NR2B-containing NMDA receptors in cortical long-term potentiation and long-term depression. J Neurosci 24: 7821-7828.

Monteilh-Zoller MK, Hermosura MC, Nadler MJ, Scharenberg AM, Penner R, Fleig A (2003) TRPM7 provides an ion channel mechanism for cellular entry of trace metal ions. J Gen Physiol 121: 49-60.

Monyer H, Sprengel R, Schoepfer R, Herb A, Higuchi M, Lomeli H, burnashev N, Sakmann B, Seeburg PH (1992) Heteromeric NMDA receptors: Molecular and functional distinction of subtypes. Science 256: 1217-1221.

Moriyoshi K, Masu M, Ishii T, Shigemoto R, Mizuno N, Nakanishi S (1991) Molecular cloning and characterization of the rat NMDA receptor. Nature 354: 31-37.

Muller BM, Kistner U, Kindler S, Chung WJ, Kuhlendahl S, Fenster SD, Lau LF, Veh RW, Huganir RL, Gundelfinger ED, Garner CC (1996) SAP102, a novel postsynaptic protein that interacts with NMDA receptor complexes in vivo. Neuron 17: 255-265.

Munoz JJ, Tarrega C, Blanco-Aparicio C, Pulido R (2003) Differential interaction of the tyrosine phosphatases PTP-SL, STEP and HePTP with the mitogen-activated protein kinases ERK1/2 and p38alpha is determined by a kinase specificity sequence and influenced by reducing agents. Biochem J 372: 193-201.

Nadler MJS, Hermosura MC, Inabe K, Perraud AL, Zhu QQ, Stokes AJ, Kurosaki T, Kinet JP, Penner R, Scharenberg AM, Fleig A (2001) LTRPC7 is a Mg center dot ATP-regulated divalent cation channel required for cell viability (vol 411, pg 590, 2001). Nature 412: 660.

Nagamine K, Kudoh J, Minoshima S, Kawasaki K, Asakawa S, Ito F, Shimizu N (1998) Molecular cloning of a novel putative Ca2+ channel protein (TRPC7) highly expressed in brain. Genomics 54: 124-131.

Nicotera P, Bano D (2003) The enemy at the gates. Ca2+ entry through TRPM7 channels and anoxic neuronal death. Cell 115: 768-770.

O'Brien JT, Erkinjuntti T, Reisberg B, Roman G, Sawada T, Pantoni L, Bowler JV, Ballard C, DeCarli C, Gorelick PB, Rockwood K, Burns A, Gauthier S, DeKosky ST (2003) Vascular cognitive impairment. Lancet Neurology 2: 89-98.

132

O'Leary VJ, Darley-Usmar VM, Russell LJ, Stone D (1992) Pro-oxidant effects of lipoxygenase- derived peroxides on the copper-initiated oxidation of low-density lipoprotein. Biochem J 282 ( Pt 3): 631-634.

Olah ME, Jackson MF, Li H, Perez Y, Sun HS, Kiyonaka S, Mori Y, Tymianski M, MacDonald JF (2009) Ca2+-dependent induction of TRPM2 currents in hippocampal neurons. J Physiol 587: 965-979.

Olney JW, De Gubareff T (1978) Glutamate neurotoxicity and Huntington's chorea. Nature 271: 557-559.

Otto VI, Heinzel-Pleines UE, Gloor SM, Trentz O, Kossmann T, Morganti-Kossmann MC (2000) sICAM-1 and TNF-alpha induce MIP-2 with distinct kinetics in astrocytes and brain microvascular endothelial cells. J Neurosci Res 60: 733-742.

Paciaroni M, Bogousslavsky J (2009) The history of stroke and cerebrovascular disease. Handb Clin Neurol 92: 3-28.

Papadia S, Hardingham GE (2007) The dichotomy of NMDA receptor signaling. Neuroscientist 13: 572-579.

Papadia S, Stevenson P, Hardingham NR, Bading H, Hardingham GE (2005) Nuclear Ca2+ and the cAMP response element-binding protein family mediate a late phase of activity-dependent neuroprotection. J Neurosci 25: 4279-4287.

Paul S, Nairn AC, Wang P, Lombroso PJ (2003) NMDA-mediated activation of the tyrosine phosphatase STEP regulates the duration of ERK signaling. Nat Neurosci 6: 34-42.

Pellegrini-Giampietro DE, Gorter JA, Bennett MV, Zukin RS (1997) The GluR2 (GluR-B) hypothesis: Ca(2+)-permeable AMPA receptors in neurological disorders. Trends Neurosci 20: 464-470.

Penn RD, Loewenstein WR (1966) Uncoupling of a nerve cell membrane junction by calcium- ion removal. Science 151: 88-89.

Perraud AL, Takanishi CL, Shen B, Kang S, Smith MK, Schmitz C, Knowles HM, Ferraris D, Li WX, Zhang J, Stoddard BL, Scharenberg AM (2005) Accumulation of free ADP-ribose from mitochondria mediates oxidative stress-induced gating of TRPM2 cation channels. Journal of Biological Chemistry 280: 6138-6148.

Pignataro G, Gala R, Cuomo O, Tortiglione A, Giaccio L, Castaldo P, Sirabella R, Matrone C, Canitano A, Amoroso S, Di Renzo G, Annunziato L (2004) Two sodium/calcium exchanger gene products, NCX1 and NCX3, play a major role in the development of permanent focal cerebral ischemia. Stroke 35: 2566-2570.

Pina-Benabou MH, Szostak V, Kyrozis A, Rempe D, Uziel D, Urban-Maldonado M, Benabou S, Spray DC, Federoff HJ, Stanton PK, Rozental R (2005) Blockade of gap junctions in vivo provides neuroprotection after perinatal global ischemia. Stroke 36: 2232-2237.

133

Plyte SE, Hughes K, Nikolakaki E, Pulverer BJ, Woodgett JR (1992) Glycogen synthase kinase- 3: functions in oncogenesis and development. Biochim Biophys Acta 1114: 147-162.

Pottorf WJ, Johanns TM, Derrington SM, Strehler EE, Enyedi A, Thayer SA (2006) Glutamate- induced protease-mediated loss of plasma membrane Ca2+ pump activity in rat hippocampal neurons. J Neurochem 98: 1646-1656.

Prawitt D, Monteilh-Zoller MK, Brixel L, Spangenberg C, Zabel B, Fleig A, Penner R (2003) TRPM5 is a transient Ca2+-activated cation channel responding to rapid changes in [Ca2+]i. Proc Natl Acad Sci U S A 100: 15166-15171.

Qualmann B, Boeckers TM, Jeromin M, Gundelfinger ED, Kessels MM (2004) Linkage of the actin cytoskeleton to the postsynaptic density via direct interactions of Abp1 with the ProSAP/Shank family. J Neurosci 24: 2481-2495.

Radi R, Beckman JS, Bush KM, Freeman BA (1991) Peroxynitrite-induced membrane lipid peroxidation: the cytotoxic potential of superoxide and nitric oxide. Arch Biochem Biophys 288: 481-487.

Reith J, Jorgensen HS, Pedersen PM, Nakayama H, Raaschou HO, Jeppesen LL, Olsen TS (1996) Body temperature in acute stroke: relation to stroke severity, infarct size, mortality, and outcome. Lancet 347: 422-425.

Rempe DA, Takano T, Nedergaard M (2009) TR(I)Pping towards treatment for ischemia. Nature Neuroscience 12: 1215-1216.

Retamal MA, Cortes CJ, Reuss L, Bennett MV, Saez JC (2006) S-nitrosylation and permeation through connexin 43 hemichannels in astrocytes: induction by oxidant stress and reversal by reducing agents. Proc Natl Acad Sci U S A 103: 4475-4480.

Rothstein JD, Patel S, Regan MR, Haenggeli C, Huang YH, Bergles DE, Jin L, Dykes HM, Vidensky S, Chung DS, Toan SV, Bruijn LI, Su ZZ, Gupta P, Fisher PB (2005) Beta-lactam antibiotics offer neuroprotection by increasing glutamate transporter expression. Nature 433: 73- 77.

Runnels LW, Yue LX, Capham DE (2001) TRP-PLIK, a bifunctional protein with kinase and ion channel activities. Science 291: 1043-1047.

Sattler R, Charlton MP, Hafner M, Tymianski M (1997) Determination of the time course and extent of neurotoxicity at defined temperatures in cultured neurons using a modified multiwell plate fluorescence scanner. Journal of Cerebral Blood Flow and Metabolism 17: 455-463.

Sattler R, Charlton MP, Hafner M, Tymianski M (1998) Distinct influx pathways, not calcium load, determine neuronal vulnerability to calcium neurotoxicity. Journal of Neurochemistry 71: 2349-2364.

Sattler R, Xiang ZG, Lu WY, Hafner M, MacDonald JF, Tymianski M (1999) Specific coupling of NMDA receptor activation to nitric oxide neurotoxicity by PSD-95 protein. Science 284: 1845-1848.

134

Schanne FA, Kane AB, Young EE, Farber JL (1979) Calcium dependence of toxic cell death: a final common pathway. Science 206: 700-702.

Scharenberg AM (2005) TRPM2 and TRPM7: channel/enzyme fusions to generate novel intracellular sensors. Pflugers Archiv-European Journal of Physiology 451: 220-227.

Schlaepfer WW, Bunge RP (1973) Effects of calcium ion concentration on the degeneration of amputated axons in tissue culture. J Cell Biol 59: 456-470.

Schmitz C, Perraud AL, Johnson CO, Inabe K, Smith MK, Penner R, Kurosaki T, Fleig A, Scharenberg AM (2003) Regulation of vertebrate cellular Mg2+ homeostasis by TRPM7. Cell 114: 191-200.

Secondo A, Staiano RI, Scorziello A, Sirabella R, Boscia F, Adornetto A, Valsecchi V, Molinaro P, Canzoniero LM, Di Renzo G, Annunziato L (2007) BHK cells transfected with NCX3 are more resistant to hypoxia followed by reoxygenation than those transfected with NCX1 and NCX2: Possible relationship with mitochondrial membrane potential. Cell Calcium 42: 521-535.

Sheng M, Sala C (2001) PDZ domains and the organization of supramolecular complexes. Annu Rev Neurosci 24: 1-29.

Siesjo BK, Katsura K, Zhao Q, Folbergrova J, Pahlmark K, Siesjo P, Smith ML (1995) Mechanisms of secondary brain damage in global and focal ischemia: a speculative synthesis. J Neurotrauma 12: 943-956.

Simonson SG, Zhang J, Canada AT, Jr., Su YF, Benveniste H, Piantadosi CA (1993) Hydrogen peroxide production by monoamine oxidase during ischemia-reperfusion in the rat brain. J Cereb Blood Flow Metab 13: 125-134.

Sivenius J, Pyorala K, Heinonen OP, Salonen JT, Riekkinen P (1985) The significance of intensity of rehabilitation of stroke--a controlled trial. Stroke 16: 928-931.

Slee EA, Adrain C, Martin SJ (2001) Executioner caspase-3, -6, and -7 perform distinct, non- redundant roles during the demolition phase of apoptosis. J Biol Chem 276: 7320-7326.

Snyder EM, Nong Y, Almeida CG, Paul S, Moran T, Choi EY, Nairn AC, Salter MW, Lombroso PJ, Gouras GK, Greengard P (2005) Regulation of NMDA receptor trafficking by amyloid-beta. Nat Neurosci 8: 1051-1058.

Soriano FX, Hardingham GE (2007) Compartmentalized NMDA receptor signalling to survival and death. J Physiol 584: 381-387.

Soriano FX, Martel MA, Papadia S, Vaslin A, Baxter P, Rickman C, Forder J, Tymianski M, Duncan R, Aarts M, Clarke P, Wyllie DJA, Hardingham GE (2008) Specific Targeting of Pro- Death NMDA Receptor Signals with Differing Reliance on the NR2B PDZ Ligand. Journal of Neuroscience 28: 10696-10710.

135

Soriano FX, Papadia S, Hofmann F, Hardingham NR, Bading H, Hardingham GE (2006) Preconditioning doses of NMDA promote neuroprotection by enhancing neuronal excitability. J Neurosci 26: 4509-4518.

Stornetta RL, Zhu JJ (2011) Ras and Rap signaling in synaptic plasticity and mental disorders. Neuroscientist 17: 54-78.

Sun HS, Jackson MF, Martin LJ, Jensen A, Teves L, Cui H, Kiyonaka S, Mori Y, Jones M, Forder JP, Golde TE, Orser BA, Macdonald JF, Tymianski M (2009) Suppression of hippocampal TRPM7 protein prevents delayed neuronal death in brain ischemia. Nat Neurosci 12: 1300-1307.

Szydlowska K, Tymianski M (2010) Calcium, ischemia and excitotoxicity. Cell Calcium 47: 122-129.

Terasaki Y, Sasaki T, Yagita Y, Okazaki S, Sugiyama Y, Oyama N, Omura-Matsuoka E, Sakoda S, Kitagawa K (2010) Activation of NR2A receptors induces ischemic tolerance through CREB signaling. J Cereb Blood Flow Metab 30: 1441-1449.

Thom T, Haase N, Rosamond W, Howard VJ, Rumsfeld J, Manolio T, Zheng ZJ, Flegal K, O'Donnell C, Kittner S, Lloyd-Jones D, Goff DC, Jr., Hong Y, Adams R, Friday G, Furie K, Gorelick P, Kissela B, Marler J, Meigs J, Roger V, Sidney S, Sorlie P, Steinberger J, Wasserthiel-Smoller S, Wilson M, Wolf P (2006) Heart disease and stroke statistics--2006 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation 113: e85-151.

Thomas CG, Miller AJ, Westbrook GL (2006) Synaptic and extrasynaptic NMDA receptor NR2 subunits in cultured hippocampal neurons. J Neurophysiol 95: 1727-1734.

Thompson RJ, Zhou N, MacVicar BA (2006) Ischemia opens neuronal gap junction hemichannels. Science 312: 924-927.

Thornberry NA, Lazebnik Y (1998) Caspases: enemies within. Science 281: 1312-1316.

Togashi K, Hara Y, Tominaga T, Higashi T, Konishi Y, Mori Y, Tominaga M (2006c) TRPM2 activation by cyclic ADP-ribose at body temperature is involved in insulin secretion. EMBO Journal 25: 1804-1815.

Togashi K, Hara Y, Tominaga T, Higashi T, Konishi Y, Mori Y, Tominaga M (2006) TRPM2 activation by cyclic ADP-ribose at body temperature is involved in insulin secretion. EMBO Journal 25: 1804-1815.

Tovar KR, Westbrook GL (1999) The incorporation of NMDA receptors with a distinct subunit composition at nascent hippocampal synapses in vitro. J Neurosci 19: 4180-4188.

Turnbull S, Tabner BJ, El Agnaf OM, Moore S, Davies Y, Allsop D (2001) alpha-Synuclein implicated in Parkinson's disease catalyses the formation of hydrogen peroxide in vitro. Free Radic Biol Med 30: 1163-1170.

136

Tymianski M (2011) Emerging mechanisms of disrupted cellular signaling in brain ischemia. Nat Neurosci 14: 1369-1373.

Tymianski M, Charlton MP, Carlen PL, Tator CH (1993) Source Specificity of Early Calcium Neurotoxicity in Cultured Embryonic Spinal Neurons. Journal of Neuroscience 13: 2085-2104.

Tymianski M, Tator CH (1996) Normal and abnormal calcium homeostasis in neurons: A basis for the pathophysiology of traumatic and ischemic central nervous system injury. Neurosurgery 38: 1176-1195.

Vereker E, O'Donnell E, Lynch A, Kelly A, Nolan Y, Lynch MA (2001) Evidence that interleukin-1 beta and reactive oxygen species production play a pivotal role in stress-induced impairment of LTP in the rat dentate gyrus. Eur J Neurosci 14: 1809-1819.

Vergouwen MD, de Haan RJ, Vermeulen M, Roos YB (2008) Statin treatment and the occurrence of hemorrhagic stroke in patients with a history of cerebrovascular disease. Stroke 39: 497-502.

Verma S, Quillinan N, Yang YF, Nakayama S, Cheng J, Kelley MH, Herson PS (2012) TRPM2 channel activation following in vitro ischemia contributes to male hippocampal cell death. Neurosci Lett 530: 41-46.

Wang Q, Tang XN, Yenari MA (2007) The inflammatory response in stroke. J Neuroimmunol 184: 53-68.

Wardlaw JM, Murray V, Berge E, del Zoppo G, Sandercock P, Lindley RL, Cohen G (2012) Recombinant tissue plasminogen activator for acute ischaemic stroke: an updated systematic review and meta-analysis. Lancet 379: 2364-2372.

Wehage E, Eisfeld J, Heiner I, Jungling E, Zitt C, Luckhoff A (2002) Activation of the cation channel long transient receptor potential channel 2 (LTRPC2) by hydrogen peroxide - A splice variant reveals a mode of activation independent of ADP-ribose. Journal of Biological Chemistry 277: 23150-23156.

Wei WL, Sun HS, Olah ME, Sun XJ, Czerwinska E, Czerwinski W, Mori Y, Orser BA, Xiong ZG, Jackson MF, Tymianski M, MacDonald JF (2007) TRPM7 channels in hippocampal neurons detect levels of extracellular divalent cations. Proceedings of the National Academy of Sciences of the United States of America 104: 16323-16328.

Weiser T (2005) AMPA receptor antagonists for the treatment of stroke. Curr Drug Targets CNS Neurol Disord 4: 153-159.

Wichmann T, Bergman H, Starr PA, Subramanian T, Watts RL, DeLong MR (1999) Comparison of MPTP-induced changes in spontaneous neuronal discharge in the internal pallidal segment and in the substantia nigra pars reticulata in primates. Exp Brain Res 125: 397-409.

Wichmann T, Dostrovsky JO (2011) Pathological basal ganglia activity in movement disorders. Neuroscience 198: 232-244.

137

Wiegler K, Bonny C, Coquoz D, Hirt L (2008) The JNK inhibitor XG-102 protects from ischemic damage with delayed intravenous administration also in the presence of recombinant tissue plasminogen activator. Cerebrovasc Dis 26: 360-366.

Wilkinson JA, Scragg JL, Boyle JP, Nilius B, Peers C (2008) H2O 2-stimulated Ca2+ influx via TRPM2 is not the sole determinant of subsequent cell death. Pflugers Arch 455: 1141-1151.

Wyllie DJ, Livesey MR, Hardingham GE (2013) Influence of GluN2 subunit identity on NMDA receptor function. Neuropharm 74:4-17.

Xie YF, Belrose JC, Lei G, Tymianski M, Mori Y, MacDonald JF, Jackson MF (2011) Dependence of NMDA/GSK-3beta Mediated Metaplasticity on TRPM2 Channels at Hippocampal CA3-CA1 Synapses. Mol Brain 4: 44.

Xiong ZG, Zhu XM, Chu XP, Minami M, Hey J, Wei WL, MacDonald JF, Wemmie JA, Price MP, Welsh MJ, Simon RP (2004) Neuroprotection in ischemia: Blocking calcium-permeable acid-sensing ion channels. Cell 118: 687-698.

Xu J, Kurup P, Zhang Y, Goebel-Goody SM, Wu PH, Hawasli AH, Baum ML, Bibb JA, Lombroso PJ (2009) Extrasynaptic NMDA receptors couple preferentially to excitotoxicity via calpain-mediated cleavage of STEP. J Neurosci 29: 9330-9343.

Yamamoto S, Shimizu S, Kiyonaka S, Takahashi N, Wajima T, Hara Y, Negoro T, Hiroi T, Kiuchi Y, Okada T, Kaneko S, Lange I, Fleig A, Penner R, Nishi M, Takeshima H, Mori Y (2008) TRPM2-mediated Ca2+ influx induces chemokine production in monocytes that aggravates inflammatory neutrophil infiltration. Nature Medicine 14: 738-747.

Yanamoto H, Nagata I, Niitsu Y, Zhang Z, Xue JH, Sakai N, Kikuchi H (2001) Prolonged mild hypothermia therapy protects the brain against permanent focal ischemia. Stroke 32: 232-239.

Yang G, Iadecola C (1998) Activation of cerebellar climbing fibers increases cerebellar blood flow: role of glutamate receptors, nitric oxide, and cGMP. Stroke 29: 499-507.

Yang KT, Chang WL, Yang PC, Chien CL, Lai MS, Su MJ, Wu ML (2006) Activation of the transient receptor potential M2 channel and poly(ADP-ribose) polymerase is involved in oxidative stress-induced cardiomyocyte death. Cell Death Differ 13: 1815-1826.

Yao WD, Gainetdinov RR, Arbuckle MI, Sotnikova TD, Cyr M, Beaulieu JM, Torres GE, Grant SG, Caron MG (2004) Identification of PSD-95 as a regulator of dopamine-mediated synaptic and behavioral plasticity. Neuron 41: 625-638.

Yin X, Takei Y, Kido MA, Hirokawa N (2011) Molecular Motor KIF17 Is Fundamental for Memory and Learning via Differential Support of Synaptic NR2A/2B Levels. Neuron 70: 310- 325.

Yi JH, Park SW, Kapadia R, Vemuganti R (2007) Role of transcription factors in mediating post- ischemic cerebral inflammation and brain damage. Neurochem Int 50: 1014-1027.

138

Yonekura I., Kawahara N, Kirino T (2004) A Model of Global Ischemia in C57 BL/6 Mice. Journal of Cerebral Blood Flow and Metabolism 24: 151-158.

Zhang Q, Wei EQ, Zhu CY, Zhang WP, Wang ML, Zhang SH, Yu YP, Chen Z (2006) Focal cerebral ischemia alters the spatio-temporal properties, but not the amount of activity in mice. Behavioural Brain Research 169: 66-74.

Zhang W, Chu X, Tong Q, Cheung JY, Conrad K, Masker K, Miller BA (2003) A novel TRPM2 isoform inhibits calcium influx and susceptibility to cell death. J Biol Chem 278: 16222-16229.

Zhou M, Baudry M (2006) Developmental changes in NMDA neurotoxicity reflect developmental changes in subunit composition of NMDA receptors. J Neurosci 26: 2956-2963.

Zukin RS, Bennett MV (1995) Alternatively spliced isoforms of the NMDARI receptor subunit. Trends Neurosci 18: 306-313.

139

Appendices 1 Appendix 1: TRPM2(-/-) is in cortical neurons following OGD 1.1 Introduction

In this thesis we have shown that hippocampal neurons from TRPM2(-/-) mice are more resistant to in vitro OGD when compared to WT neurons. Furthermore, we have described that absence of TRPM2 modulates NR2 subunit expression in the hippocampus. However, our in vivo stroke model primarily affected the cortex. Also earlier studies in vitro using TRPM2 silencing has shown that inhibition of TRPM2 protects cortical neurons (Kaneko et al. 2006). It has also been suggested that not all neurons express TRPM2 mRNA (Olah et al. 2009; Kraft et al. 2004) and thus there may be a difference in the effect the TRPM2-null mutant has on hippocampal and cortical neurons. To resolve the potential different roles TRPM2 play in cell survival/death between cortical and hippocampal neurons, we investigated whether cortical neurons from TRPM2(-/-) mice exhibited similar properties as hippocampal neurons from the same mice. We report here that following in vitro OGD, cortical neurons that lack TRPM2 are more resistant to cell death than neurons from WT animals. We also show that that the cortex from TRPM2(-/-) animals have increased and reduced NR2A and NR2B expression, respectively, when compared the cortex from WT animals.

1.2 Methods

1.2.1 Cell culture

Cell culture methods were the same as those found in methods section 3.6. The only difference was that cortical neurons from whole cortex were plated instead of hippocampal neurons. Experiments using cortical and hippocampal cultures were conducted concurrently.

1.2.2 Western blot

Western blot protocols were similar to those described in section 3.4, with the exception of using whole cortex instead of whole hippocampus.

140

1.2.3 Statistical analysis

See section 3.8 for further information on statistical analyses used in the appendix.

1.3 Results

1.3.1 Absence of TRPM2 reduces neuronal death in cortical neurons following 1h OGD

In hippocampal neurons, the absence of TRPM2 was found to be neuroprotective following 1h OGD. Here we show that cortical neurons from TRPM2(-/-) mice also had a reduced level of cell death in comparison to those from WT animals following 1h OGD (n=6, p<0.05; Figure 27). Cortical neurons from both WT and TRPM2(-/-) mice had similar levels of cell death under unchallenged conditions. Overall, we show here that chronic absence of TRPM2 has similar neuroprotective effects in cortical neurons as compared to hippocampal neurons.

1.3.2 Neuroprotective agents do not have an additive effect on neuroprotective properties of cortical neurons from TRPM2-null mice

Like in hippocampal neurons, absence of TRPM2 in cortical neurons did not enhance the protective effects of known neuroprotective agents, NA1, M21 and MCN, following 1h OGD (n=4, p>0.05; Figure 28). These drugs provided WT and TRPM2(-/-) neurons similar levels of neuroprotection when exposed to 1h OGD. We did not investigate the 1.5h OGD time point in these neurons. Furthermore, these neuroprotective drugs had even less fraction of neuronal death in comparison to untreated neurons from TRPM2(-/-) mice. Like in hippocampal neurons, we show here that cortical neurons also hit a neuroprotective "flooring effect" when NA1, M21 and MCN are used in neurons with and without the presence of TRPM2 channels.

1.3.3 Cortical neurons from TRPM2(-/-) mice promote activation of pro- survival pathways via modulation of NR2 subunit expression

Using western blot we investigated whether absence of TRPM2 also modulated NR2 subunit expression in the cortex. Using protein from whole cortex we found that cortices from TRPM2(-/-) mice had significantly increased NR2A, and decreased active pNR2B expression, in comaprison from cortices from WT mice (n=3;p<0.05; Figure 29 A,B). These changes in NR2 expression in the cortex of TRPM2(-/-) mice were associated with an increase in the

141 phosphorylation of ERK and Akt when compared to cortical protein samples from WT mice (n=3;p<0.05; Figure 29 C,D). This change in ERK and Akt activation is also associated with NR2-mediated preconditioning which promotes neuronal survival (Soriano et al., 2006). Overall, both the hippocampus and cortex exhibit similar latent NMDAR-mediated pro-survival mechanisms in the TRPM2(-/-) mouse model.

1.4 Discussion

Here we resolve the question as to whether cortical neurons from TRPM2(-/-) mice have similar neuroprotective properties as their hippocampal counterparts. We show that absence of TRPM2 reduces fraction neuronal death in cortical neurons following 1h OGD and this neuroprotection is associated with changes in NR2 subunits that promote pro-survival mechanisms. This data also demonstrates that our in vivo results, showing that TRPM2(-/-) animals have reduced infarct in the cortex in comparison to WT animals, can also be compared to our in vitro hippocampal findings. One pitfall when interpreting our results from conrtical in vivo results to hippocampal in vitro results is that neurons in the cortex and the hippocampus have different surrounding environments and circuitry, which may also be modulated by the absence of TRPM2 and affect neuronal survival. As such, it is difficult to determine if absence of TRPM2 in the cortex and the hippocampus have similar physiological properties such as the increased synaptic sensitivity to oxidative stress. Overall, we have confirmed in vitro, that a lack of TRPM2 provides neuroprotection in both hippocampus and cortex by modulating NR2- subunits of NMDARs to promote pro-survival signaling.

142

Figure 27: OGD in cortical neurons from WT and TRPM2(-/-) mice

In vitro 1h OGD experiments in cortical neurons demonstrates that absence of TRPM2 significantly reduces cell death in cortical neurons (n=6; p<0.05). Both WT and TRPM2(-/-) cortical neurons had similar levels of cell death under unchallenged control conditions.

143

Figure 28: Neuroprotective agents have similar level of protection in cortical neurons from WT and TRPM2(-/-) mice

Primary cortical neurons from WT and TRPM2(-/-) mice were treated with neuroprotective drugs (MCN, NA1 and M21) and exposed to 1OGD followed by 24h of unchallenged control condition. Treated and untreated WT and TRPM2(-/-) neurons had similar levels of cell death following no OGD and 24h of unchallenged control conditions. Cytoprotective drugs provided similar resistance to OGD between WT and TRPM2(-/-) neurons (p>0.05; n=4) and were not enhanced by the absence of TRPM2 channels.

144

Figure 29: Absence of TRPM2 promotes pro-survival pathway in cortical neurons Western blots detecting NR2, ERK and Akt expression in cortical protein extracts from 3-8 week old TRPM2(-/-) and WT animals. (A-B) Densitometry of NR2 levels show that absence of TRPM2 reduces active pNR2B expression (n=3;p<0.05) and increases basal NR2A expression (n=3;p<0.05). (C-D) Absence of TRPM2 also increased phosphorylation of ERK (P42-ERK; n=3;p<0.05) and Akt (P-thr 308 Akt n=3;p<0.05).

145

1.5 Appendix References

Olah ME, Jackson MF, Li H, Perez Y, Sun HS, Kiyonaka S, Mori Y, Tymianski M, MacDonald JF (2009) Ca2+-dependent induction of TRPM2 currents in hippocampal neurons. J Physiol 587: 965-979.

Soriano FX, Papadia S, Hofmann F, Hardingham NR, Bading H, Hardingham GE (2006) Preconditioning doses of NMDA promote neuroprotection by enhancing neuronal excitability. J Neurosci 26: 4509-4518.