Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mitochondrial potassium sequestration and neuroprotection is mediated by connexin43 Kozoriz, Michael Gregory 2010

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2012_spring_kozoriz_michael.pdf [ 4.12MB ]
Metadata
JSON: 24-1.0069988.json
JSON-LD: 24-1.0069988-ld.json
RDF/XML (Pretty): 24-1.0069988-rdf.xml
RDF/JSON: 24-1.0069988-rdf.json
Turtle: 24-1.0069988-turtle.txt
N-Triples: 24-1.0069988-rdf-ntriples.txt
Original Record: 24-1.0069988-source.json
Full Text
24-1.0069988-fulltext.txt
Citation
24-1.0069988.ris

Full Text

MITOCHONDRIAL POTASSIUM SEQUESTRATION AND NEUROPROTECTION IS MEDIATED BY CONNEXIN43  by  Michael Gregory Kozoriz B.Sc.H, Queen’s University, 2001 M.Sc., University of Calgary, 2003  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF MEDICINE AND DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Anatomy)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) May 2010  © Michael Gregory Kozoriz, 2010  Abstract Reduction in the expression of the astrocyte gap junction protein connexin43 (Cx43) increases infarct volume following middle cerebral artery occlusion (MCAO). A potential mechanism for this effect is disruption of ion buffering. During ischemia, extracellular potassium concentration ([K+]o) rises, leading to a variety of potentially detrimental effects on neuronal function. It is known that astrocytes contribute to the clearance of K+o to neighbouring cells through Cx43 based gap junctions, but mitochondria also contain Cx43 which could play a role in K+ uptake. Mitochondrial K+ sequestration was examined by loading astrocytes with the fluorescent K+ indicator PBFI. Release of K+ from mitochondria into the cytoplasm was examined after uncoupling the mitochondrial membrane potential with carbonyl cyanide mchlorophenylhydrazone (CCCP). Transient applications of elevated [K+]o led to increases in K+ within mitochondria, as assessed by increases in the magnitudes of cytoplasmic [K+] ([K+]i) transients evoked by brief exposures to CCCP. When mitochondrial K+ sequestration was impaired by prolonged application of CCCP, there was a robust increase in [K+]i upon exposure to elevated [K+]o. Blockade of plasmalemmal K+ uptake routes by ouabain, Ba2+, or a cocktail of voltage-activated K+ channel inhibitors reduced K+ uptake into mitochondria. Reductions in mitochondrial K+ uptake occurred in the presence of the mitoKATP channel inhibitor 5hydroxydecanoic acid. Rises in [K+]i evoked by brief applications of CCCP following exposure to high-[K+]o were also reduced by gap junction blockers and in astrocytes isolated from Cx43null mice, suggesting that connexins also play a role in K+ uptake into astrocyte mitochondria. In a second study, the carboxy-terminal (CT) region of Cx43, a region important for channel activity, was explored in mutant mice expressing a truncated form of Cx43 (Cx43 CT mice). These mice exhibit enhanced cerebral injury following MCAO. In the peri-infarct region, astrogliosis was reduced and inflammatory cell invasion was increased. Cultured astrocytes ii  from Cx43 CT mice were less coupled, and displayed alterations in channel gating, hemichannel activity, Ca2+ wave properties and showed impairment of mitochondrial K+ sequestration. These results suggest that astrocytic Cx43 contributes to the mitochondrial sequestration of K+ and that the CT region plays an important role in protection following stroke.  iii  Table of contents Abstract ........................................................................................................................................... ii Table of contents ............................................................................................................................ iv List of tables ................................................................................................................................... vi List of figures ................................................................................................................................ vii List of abbreviations ...................................................................................................................... ix Acknowledgments.......................................................................................................................... xi Co-authorship statement ............................................................................................................... xii Chapter 1: Introduction .................................................................................................................. 1 1.1 Introduction to stroke ...................................................................................................... 2 1.1.1 The history of stroke ................................................................................................... 2 1.1.2 Classification and clinical aspects of stroke ............................................................... 3 1.1.3 Economic impact, epidemiology, and risk factors of stroke ....................................... 4 1.1.4 The physiology of focal ischemic stroke .................................................................... 6 1.1.5 The treatment of stroke and experimental research strategies .................................. 10 1.2 Astrocytes ..................................................................................................................... 11 1.2.1 The role of astrocytes in the nervous system ........................................................ 11 1.3 Gap junctions and Cx43 ................................................................................................ 16 1.3.1 Gap junctions ............................................................................................................ 16 1.3.2 Astrocyte connexins .................................................................................................. 18 1.3.3 Cx43 expression and functional properties ............................................................... 19 1.3.4 C-terminal region of Cx43 ........................................................................................ 21 1.4 Gap junctions and cellular injury .................................................................................. 23 1.4.1 Gap junction proteins and cellular injury.................................................................. 23 1.4.2 Gap junction proteins and cellular destruction ......................................................... 25 1.4.3 Gap junction proteins and cellular protection ........................................................... 28 1.4.4 Resolving controversies ............................................................................................ 31 1.5 K+ handling by astrocytes ............................................................................................. 32 1.5.1 K+ changes during stroke .......................................................................................... 32 1.5.2 Astrocyte K+ uptake and buffering ........................................................................... 33 1.5.3 Intracellular ion sequestration ................................................................................... 35 1.5.4 Mitochondria and K+ uptake ..................................................................................... 36 1.5.5 Mitochondrial connexins .......................................................................................... 37 1.6 Thesis hypothesis and manuscript chapter introduction ............................................... 40 1.7 References ..................................................................................................................... 47 Chapter 2: Temporary sequestration of potassium by mitochondria in astrocytes ...................... 66 2.1 2.2 2.3 2.4 2.5  Introduction ................................................................................................................... 67 Materials and methods .................................................................................................. 68 Results ........................................................................................................................... 74 Discussion ..................................................................................................................... 88 References ................................................................................................................... 104  iv  Chapter 3: The connexin43 C-terminal region mediates neuroprotection during stroke. .......... 110 3.1 3.2 3.3 3.4 3.5  Introduction ................................................................................................................. 111 Materials and methods ................................................................................................ 112 Results ......................................................................................................................... 119 Discussion ................................................................................................................... 123 References ................................................................................................................... 136  Chapter 4: Truncation of the C-terminal region of connexin43 and mitochondrial potassium sequestration. .............................................................................................................................. 140 4.1 4.2 4.3 4.4 4.5  Introduction ................................................................................................................. 141 Materials and methods ................................................................................................ 142 Results ......................................................................................................................... 146 Discussion ................................................................................................................... 147 References ................................................................................................................... 157  Chapter 5: Conclusion................................................................................................................ 161 5.1 5.2 5.3 5.4 5.5 5.5  Contribution to novel knowledge................................................................................ 162 Mitochondrial Cx43 based K+ sequestration and the Cx43 C-terminal region’s role in neuroprotection....................................................................................................... 162 Strengths and weaknesses ........................................................................................... 168 Future directions ......................................................................................................... 169 Overall significance and potential applications of findings ........................................ 171 References ................................................................................................................... 173  Appendices .................................................................................................................................. 178 Appendix A: Journal club – passing potassium with and without gap junctions .................. 179 Appendix B: Supplemental figures for chapter 2 .................................................................. 181 Appendix C: Supplemental figure for chapter 3 .................................................................... 183 Appendix D: Mitochondria protein interactions with Cx43 as identified by mass spectrometry............................................................................................................................ 184 Appendix E: Co-immunoprecipitation of VDAC with Cx43 ................................................ 185 Appendix F: UBC animal care certificate .............................................................................. 186  v  List of tables Table D.1  Potential mitochondrial protein interactions with Cx43 ......................................184  vi  List of figures Figure 1.1  Pathways of ischemic cell death ............................................................................43  Figure 1.2  Events in brain ischemia ........................................................................................43  Figure 1.3  Focal cerebral ischemia induces a complex series of mechanisms that damage brain cells ...............................................................................................................44  Figure 1.4  A schematic representation of gap junction formation between two cells ............45  Figure 1.5  Human Cx43 amino acid sequence and consensus interacting protein modules ...46  Figure 2.1  In situ calibration of PBFI......................................................................................94  Figure 2.2  CCCP evokes an increase in [K+]i in cultured astrocytes.......................................95  Figure 2.3  Effects of changes in [K+]o on CCCP-induced [K+]i rises .....................................96  Figure 2.4  K+ uptake mechanisms in cultured astrocytes .......................................................98  Figure 2.5  Cx43 association with astrocyte mitochondria ......................................................99  Figure 2.6  K+ uptake in mitochondria from cultured astrocytes ...........................................100  Figure 2.7  Mitochondrial uptake of K+ is reduced in the presence of gap junction blockers ..............................................................................................................................101  Figure 2.8  Mitochondrial uptake of K+ is reduced in Cx43-/- astrocytes ...............................102  Figure 2.9  Mitochondrial membrane potential during elevated [K+]o in Cx43+/+ and Cx43-/astrocytes..............................................................................................................103  Figure 3.1  The Cx43 CT mutation results in increased infarct size after MCAO ...............129  Figure 3.2  Reactive astrogliosis is decreased in Cx43 CT mice ..........................................130  Figure 3.3  The inflammatory reaction is increased in Cx43 CT mice .................................131  Figure 3.4  Immunohistochemistry for Cx43 in Cx43 CT mutants ......................................132  Figure 3.5  Western blot of adult cortex and astrocyte culture tissue ....................................133  Figure 3.6  Gap junction coupling, electrophysiology, hemichannel activity and Ca2+ waves in culture ...................................................................................................134  Figure 4.1  K+ uptake in Cx43 +, Cx43  Figure 4.2  Mitochondrial K+ sequestration is reduced in Cx43  Figure 4.3  Resting  m  /-  and Cx43  CT/-  mitochondria .................................151 /-  does not differ between Cx43 +, Cx43  and Cx43 /-  CT/-  and Cx43  astrocytes .152  CT/-  astrocytes..............................................................................................................153 Figure 4.4  Schematic representation of the potential roles of Cx43 during stroke ...............155  vii  Figure B.1  Comparison of the [Na+]i and [K+]i transients evoked by CCCP in cultured astrocytes..............................................................................................................181  Figure B.2  Applications of 12.5 mM K+o under HEPES and bicarbonate recording conditions .............................................................................................................182  Figure C.1  Brain vasculature in Cx43+/+, Cx43+/-, Cx43  CT/+  Figure E.1  Cx43 co-immunoprecipitates with VDAC in wild-type astrocytes .....................185  and Cx43  CT/-  mice ................183  viii  List of abbreviations m  ADP AGA ATP cAMP CCCP CBX COX CT Cx43 Cx43+/+ Cx43+/Cx43-/Cx43 CT Cx43 CT/ CT Cx43 CT/+ Cx43 CT/Cx43fl/fl/hGFAP-cre DAPI DMEM ER FBS FCCP GABA GAPDH GFAP gj GZA 5-HD HBSS HEPES IBA-1 ICD IL INF i.p. IP3 K+ [K+]i [K+]o Ka Kdr Kir K2P MAPK MCAO  inner mitochondrial transmembrane potential adenosine diphosphate 18 -glycyrrhetinic acid adenosine triphosphate cyclic adenosine monophosphate carbonyl cyanide m-chlorophenylhydrazone carbenoxolone cyclooxygenase C-terminal connexin43 connexin43 wild-type connexin43 heterozygous connexin43-null connexin43 truncated at codon 258 mice/cells with both Cx43 alleles truncated mice/cells with one truncated Cx43 allele and one full length Cx43 allele mice/cells with one truncated Cx43 allele and one Cx43-null allele GFAP promoter cre-lox recombinase Cx43 knock-out 4',6-diamidino-2-phenylindole Dulbecco's modified Eagle medium endoplasmic reticulum fetal bovine serum carbonyl cyanide p-trifluoromethoxyphenylhydrazone gamma-aminobutyric acid glyceraldehyde 3-phosphate dehydrogenase glial fibrillary acidic protein junctional conductance glycyrrhizic acid 5-hydroxydecanoic acid Hanks buffered salt solution 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid ionized calcium-binding adaptor molecule 1 International Classification of Disease interleukin interferon intraperitoneally inositol trisphosphate potassium cytoplasmic K+ concentration extracellular K+ concentration A-type potassium channel delayed rectifier potassium channel inward rectifier potassium channel tandem pore domain potassium channels mitogen-activated protein kinase middle cerebral artery occlusion ix  MMP NGS PBFI PBS PKA PKC R123 ROS SBFI SDS TBS TBS-T TOM/TIM TNF VDAC ZO-1  matrix metalloproteinase normal goat serum potassium-binding benzofuran isophthalate phosphate-buffered saline protein kinase A protein kinase C rhodamine-123 reactive oxygen species sodium-binding benzofuran isophthalate sodium dodecyl sulfate Tris-buffered saline Tris-buffered saline with Tween translocase of the outer membrane/translocase of inner membrane tumor necrosis factor voltage-dependent anion channel zonula occludens-1  x  Acknowledgments I would like to thank my supervisor Dr. Christian Naus for his support, guidance, and for allowing latitude in my graduate studies. I would also like to thank my supervisory committee members, Drs. Claudia Krebs, Brian MacVicar and John Church for their advice. I would particularly like to acknowledge Dr. Church’s contribution to my studies; his dedication to students and scientific rigor has been inspirational. I would also like to thank current and former members of the Naus lab including Geralyn Bechberger, John Bechberger, Vincent Chen, Wun Chey Sin, Mark Ozog, Charles Lai, Cima Cina, Steve Bond, Dave Bates, Annie Aftab, Maxence Le Vasseur, Helen Chen, Noo-Rie Ha, Geeta Modha, Janice Duong, Irina Poustovoit, Rayne Reilly and Michael Suen for their help, constructive criticism and friendship. I was also fortunate, and am thankful, to have received funding for my research from the Canadian Institutes of Health Research - Vancouver Coastal Health Research Institute, the Michael Smith Foundation for Health Research and the Natural Sciences and Engineering Research Council of Canada. Lastly I would like to thank my parents Gregory and Evelyn Kozoriz for their support and my wife Kelsey Kozoriz and son William for their love and support during my graduate studies.  xi  Co-authorship statement This thesis contains multi-author manuscripts that have been prepared, submitted or accepted for publication. Dr. Naus served as a supervisor for chapters 2 - 4; Dr. Krebs also served as a supervisor for chapter 2. In chapter 2, Dr. Church was involved in extensive manuscript revisions and helped carry out experiments involving SBFI. In this chapter Dr. Ozog also provided a portion of the Cx43 knock-out astrocytes used in this study and was involved in manuscript revisions. In chapter 3, Dr. Maass and Dr. Willecke kindly provided the Cx43 C-terminal truncated mutant mouse and Dr. Moreno performed the electrophysiological characterization of these mutants. Geralyn Bechberger and I performed the middle cerebral artery occlusion experiments and she was instrumental in teaching me this technique. Noo-Rie Ha carried out the data analysis for the middle cerebral artery stroke experiments. Michael Suen carried out the calcium wave experiments as part of his undergraduate thesis project with my guidance. John Bechberger provided technical support and was influential in the experimental success of this project. The journal club article in Appendix A was a collaborative effort from all of the authors. The abridged list of mass spectrometry data in Appendix D is derived from Dr. Vincent Chen’s work on protein-protein interactions with Cx43 in glioma cells.  xii  Chapter 1: Introduction  1  1.1  Introduction to stroke  1.1.1 The history of stroke The earliest written account of stroke was by Hippocrates approximately 2400 years ago. In his Hippocratic writings he described a condition called apoplexy (a Greek word meaning struck with violence) that affected older adults: When persons in good health are suddenly seized with pains in the head, and straightway are laid down speechless, and breathe with stertor, they die in 7 days, unless fever comes on1. Hippocrates also described apoplectic patients with hemiplegia and an inability to move the hand. These signs are still used in the clinical diagnosis of potential stroke patients. One of the first theories of how apoplexy occurs was proposed by Galen (131 – 201 A.D.). He attributed apoplexy to the obstruction of the arteries to the brain which pass animal spirits from the ventricles. Galen also described emotion, plethora, sloth, drunkenness and gluttony as predisposing causes of apoplexy. The understanding of brain blood vessel anatomy progressed over the centuries, but it was not until the 1600s that a critical physiological explanation of apoplexy was proposed. Johann Jakob Wepfer (1620 – 1695) postulated that obstruction of blood influx to the brain could cause apoplexy after he observed several cases of cerebral hemorrhage and carotid thrombus at autopsy in people with apoplexy. Wepfer also found that obesity, irregular pulse and lividity were factors that increased the risk of apoplexy (Fields and Lemak 1989; McHenry 1981; Thompson 1996). Since Hippocrates’ time, numerous advances have been made in understanding the etiology, pathophysiology and diagnosis of stroke. It is now understood that stroke results from disruption of the blood supply to the brain, which impairs the delivery of oxygen and nutrients to  1  Adapted from Fields WS, Lemak, N.A. 1989. A history of stroke: Its recognition and treatment. New York: Oxford University Press.  2  brain tissue. Despite advances in our understanding over 2400 years, no effective cure for stroke exists. In this introductory chapter, I will briefly discuss our current understanding of stroke and its treatment. I will then follow with an introduction to a supportive class of brain cells called astrocytes and their role in stroke. This will lay the groundwork to introduce a protein that has been found to play a role in both cellular death and protection, connexin43 (Cx43), and will lead into a discussion of the role of Cx43 in sequestering potassium (K+), an ion which reaches toxic levels during stroke, into the mitochondria of astrocytes. This chapter will culminate with a section on the objectives and hypothesis of this thesis.  1.1.2 Classification and clinical aspects of stroke The World Health Organization’s International Classification of Disease (ICD) is the international standard for classifying human diseases. Any pathological abnormality relating to the brain blood vessels falls under the ICD category of cerebrovascular disease. The most common type of cerebrovascular disease is a cerebrovascular accident, commonly referred to as a stroke. Stroke can result from two broad classes of injury: hemorrhage or cerebral ischemia. Epidemiological studies report that 20% of strokes are hemorrhagic, while 80% of strokes are ischemic (Feigin et al. 2003). These classes are discussed below. Hemorrhagic stroke refers to bleeding in the brain; this is commonly due to either an intracerebral hemorrhage or a subarachnoid hemorrhage. In intracerebral hemorrhage, bleeding is directly into the brain parenchyma. This forms a hematoma which spreads through the brain white matter pathways. Common causes of intracerebral hemorrhage are amphetamine use, bleeding disorders, hypertension, trauma and vascular malformations (Caplan 1992). In a subarachnoid hemorrhage, bleeding is in the subarachnoid space surrounding the brain. This can 3  increase intracranial pressure, which may lead to rapid death. Common causes of subarachnoid hemorrhage are ruptured aneurysms and bleeding from vascular malformations (Caplan 1992). The more common ischemic stroke, the type of stroke studied in this thesis, is further classified into three general subtypes: ischemia due to embolism, thrombosis or systemic hypoperfusion (Caplan 2009). Embolism refers to particles originating outside of the brain which block arteries that serve brain tissue. Thrombosis is due to local obstruction of an artery which can occur, for example, with arteriosclerosis or vessel dissection. Lastly, systemic hypoperfusion that can occur, for example, during cardiac arrest, results in reduction of blood flow to the brain. The clinical differences between these subtypes of stroke will not be discussed because it is beyond the scope of this thesis. But it is important to note that even though there may be differences with each subtype with respect to signs, symptoms and progression of stroke, ultimately diagnostic imaging is necessary to confirm the specific diagnosis. Regardless of the mode of injury, the clinical effects of stroke on an individual depend on the region of the brain involved and extent of damage. Stroke can result in death or a host of other problems including loss of movement, sensation, vision, speech, thinking, reasoning, memory, sexual function, regulation of emotions and language (Caplan 2009; Fauci 2008; Heart and Stroke Foundation of Canada 2009).  1.1.3 Economic impact, epidemiology, and risk factors of stroke Stroke is the second leading cause of death worldwide after ischemic heart disease (Murray and Lopez 1997) and the third leading cause of death in North America (Heart and Stroke Foundation of Canada 2009). It is also the leading cause of disability in North America and the most expensive disease to treat in hospital (Chan 1998; Lindsay et al. 2008; Taylor et al. 4  1996). In Canada, 87% of stroke survivors report activities of daily living restrictions, which includes restrictions in activities such as personal hygiene, meal preparation, eating and household chores (Hodgson 1998). Also, anywhere from 11-85% of patients return to some form of work following stroke (Gabriele and Renate 2009; reviewed in Wozniak and Kittner 2002). Families often carry the additional burden of lost income and caring for the affected individual (Heart and Stroke Foundation of Canada 2009; Rigby et al. 2009). At the community level, stroke drains the economy and puts a burden on the health care system. In British Columbia, stroke is the most common neurological problem requiring hospital admission. In 2003 there were 8960 hospital admissions, with an average stay of 17 days (ranging from 1 to 420 days), and patients are frequently re-admitted (Heart & Stroke Foundation of British Columbia & Yukon 2005). In Canada, the cost is estimated at $3 billion annually (Moore et al. 1997). Annual nursing home costs of stroke are estimated at $600 million in Canada and $78 million in British Columbia, and as British Columbia’s population is aging, this number is likely to rise (Heart & Stroke Foundation of British Columbia & Yukon 2005). There are several known risk factors (shown in italics) for stroke which are briefly summarized from several sources below (Fauci 2008; Goldstein et al. 2001; Heart and Stroke Foundation of Canada 2009; Struijs et al. 2005; Wolf et al. 1991). Hypertension can promote scarring of blood vessels leading to atherosclerosis and blood vessel narrowing. Hypertension doubles the risk of stroke and raises the risk of hemorrhagic stroke 3.5 times. High cholesterol can also lead to atherosclerosis. Smoking also contributes to atherosclerosis, hypertension and increases the risk of blood clots. Smokers are 3.2 times more likely to have a stroke than nonsmokers. A type of heart arrhythmia, atrial fibrillation, can lead to emboli formation which increases the risk of stroke 3 to 5 times and approximately 15% of strokes are due to this condition. 5  There are also several contributory factors for stroke; for example, the risk of stroke is increased 3 times in diabetics likely due to hypertension and atherosclerosis often displayed in these patients. Likewise, obesity and physical inactivity increase the risk of having a stroke. Excessive alcohol consumption, stress and hormone replacement therapy may also increase the risk of stroke. Lastly, social factors can influence stroke. For example, decreased socioeconomic status correlates with increased risk and severity of stroke (Cox et al. 2006). The above risk factors can be modified either with lifestyle changes, drugs or other interventions. However there are some risk factors that cannot be modified. For example, the risk of stroke increases with age, doubling every decade after age 55. Males have a slightly higher incidence of stroke (52% versus 48% in women); however, 60% of deaths due to stroke are in females. Also, a family history plays a role. The risk of a stroke increases if a related first degree family member has had a stroke before age 55. Ethnicity also plays a role as African, First Nations and South Asian people are at higher risk. Lastly, a personal history of stroke or transient ischemic attacks (mini strokes) also increases the risk of a subsequent stroke.  1.1.4 The physiology of focal ischemic stroke Ischemic injury causes a complex cascade of events which often leads to cell death. Neurons are frequently used to study the molecular pathways of ischemic injury, but other brain constituents such as blood vessels, astrocytes and microglia also play an important role in ischemic damage (Lalancette-Hebert et al. 2007; Silver et al. 1997; Tagaya et al. 2001; reviewed by del Zoppo 2009; Ekdahl et al. 2009; Rossi et al. 2007; Takano et al. 2009). Microglia are the immune cells of the brain, and in their absence ischemic injury is larger (Lalancette-Hebert et al. 2007; Nimmerjahn et al. 2005). These cells are further discussed in chapter 3. Astrocytes, the main cell type of interest in this thesis, will be discussed in a separate section below. The 6  progression from initial ischemic insult to cell death is outlined in Fig. 1.1 (adapted from Lipton 1999). Lipton describes ischemic injury as starting with various initiators and activators, leading to activation of key pathways which perpetrate cell injury. This leads to critical functional and structural changes ultimately culminating in cell death. Connections within and between various stages leading to cell death is incompletely understood. The initial event in ischemic injury is depletion of oxygen and glucose supply to a region of the brain. Oxygen and glucose are important in the production of adenosine triphosphate (ATP), a critical molecule that supplies energy for metabolic processes. When oxygen is absent ATP can also be produced through glycolysis, although this may promote damage due to lactic acid production (Giffard et al. 1990; Goldman et al. 1989; Rehncrona and Kagstrom 1983; Swanson et al. 1997). Depletion of blood supply results in a decline of ATP production. A key consequence of this is inhibition of the Na+,K+-ATPase ion transporter, which contributes to membrane depolarization and loss of ion gradients (Hansen and Nedergaard 1988; Silver et al. 1997; reviewed by Lipton 1999). This leads to release of the excitatory neurotransmitter glutamate, which contributes to entry of Ca2+. Excessive Ca2+ accumulation is a known activator of reactive oxygen species (ROS) production, catabolic proteases and mitochondrial Ca2+ overload (reviewed by Hossmann 2006; Lipton 1999; Siesjo et al. 1999). Disruption of oxidative phosphorylation leads to the generation of ROS, which has consequent damaging effects (reviewed by Chan 1996). These processes damage cellular DNA, membranes, proteins and lead to mitochondrial permeability transition and release of apoptotic factors culminating in cell death (reviewed by Hossmann 2006; Lipton 1999; Lipton and Nicotera 1998; Rossi et al. 2007). The disruption in ATP, K+, Na+, Ca2+, H+, glutamate, and membrane voltage during ischemia are outlined Fig. 1.2A (figure adapted from Rossi et al. 2007). The rise in extracellular K+  7  concentration ([K+]o) and subsequent damage control mechanisms for curbing excess K+ accumulation are discussed in detail in section 1.5. The brain is supplied by two internal carotid arteries and two vertebral arteries that anastomose and form the circle of Willis. The main arteries of the brain arise from the circle of Willis: left and right anterior, middle and posterior cerebral arteries. Ischemia is often due to focal disruption of one of these arteries (Caplan 2009). Following occlusion, an ischemic core develops where the processes described above largely occur. The ischemic core is surrounded by a penumbra which is surrounded by normal brain tissue. The penumbra is defined as a region that is electrically silent but functionally intact (Astrup et al. 1981). This area is of particular interest because it is believed that cells in this region can be saved from further damage following stroke with therapeutic intervention. These regions occur because blood flow differences exist radially from the center of an ischemic core (0% blood flow) to the outer edge of the penumbra (approximately 70% of normal blood flow) (Symon et al. 1977; reviewed by Hossmann 2006). Formation of this gradient is interesting because cellular functions are disrupted at different flow rates. For example, at blood flows between 70% - 50%, protein synthesis is disrupted. When flow rates are reduced to approximately 40% - 20% glucose use is disrupted and lactate accumulation and acidosis occurs. In the infarct core, flow rates are less than 20%. At this level of flow ATP is depleted, the plasma membrane is terminally depolarized, ion gradients are disrupted and cell death is irreversible (Paschen et al. 1992; Schuier and Hossmann 1980; Shimada et al. 1993; reviewed by Heiss 1992; Hossmann 2006). Fig. 1.2B depicts some of the changes that occur in the penumbral region. The time course of events that occur during stroke are approximately known (Heiss 1992; Hossmann 1994; Hossmann 2006; Lipton 1999; Symon et al. 1977). When flow is significantly 8  reduced, ATP depletion, depolarization of the plasma membrane and disruption of ion gradients occur in minutes. The penumbral region that is developed at flow rates between approximately 70% – 20% is only a transient phenomenon. Following several penumbral transient ischemic depolarizations2, a terminal depolarization occurs within 4 – 6 hours and the infarct core expands into the penumbra leading to irreversible damage similar to the ischemic core (see Fig. 1.2). Around this time point other events occur such as tissue necrosis and blood brain barrier disruption. In the subsequent days to weeks, edema, inflammation and further cell death occurs. Fig. 1.3 (adapted from Dirnagl et al. 2003) outlines the approximate time course of events that occur during ischemia. Important mediators that play a role within the initial minutes of ischemia, such as glutamate release, Ca2+ accumulation and ROS generation, are shown. Also, over hours to days, inflammation and apoptosis mediators such as interleukin (IL)-1, cyclooxygenase (COX)-2, matrix metalloproteinase (MMP) and caspases play an important role in the progression of cell damage. Interestingly, the brain has numerous processes that can act as endogenous protective pathways, recruited in an attempt to attenuate ischemic damage. As depicted in Fig. 1.3, at early time points inhibitory transmitters gamma-aminobutyric acid (GABA) and adenosine are released, and KATP channels are also activated. Within hours following stroke, anti-inflammatory and anti-apoptosis agents such as cytokines, BCL proteins and erythropoietin are also released, which can limit damage. Over days to weeks there is also regeneration and repair in the form of scar formation, vasculogenesis, neurogenesis and neuronal sprouting (Dirnagl et al. 2003). Furthermore, as discussed in this thesis, other potential 2  In normal brain tissue, spreading depression is a slow wave of electrical activity across the tissue which is followed by a wave of electrical depression (Martins-Ferreira et al. 2000). Transient ischemic depolarizations that occur in the penumbral region (Nedergaard and Astrup 1986; Nedergaard and Hansen 1993) are also referred to as spreading depression-like depolarizations (see Hossmann 1996; Hossmann 2006; Rossi et al. 2007). As reviewed by Hossmann (1996, 2006), spreading depression-like depolarization waves are triggered by the release of K + and glutamate. These waves migrate across the brain tissue at a rate of 3 mm/min and increase metabolic demand, which may contribute to cell death. Infarct volume size is known to correlate with the number of spreading depression-like depolarizations (Mies et al. 1993). The mechanism of wave propagation is not entirely understood but the potential role of gap junctions in spreading depression or spreading depression-like depolarizations is discussed further in section 1.4.1.  9  endogenous protective pathways, such as the astrocyte gap junction protein Cx43 and K+ sequestration are possible mechanisms to limit damage of the penumbra.  1.1.5 The treatment of stroke and experimental research strategies In Canada, practice guidelines have been created that incorporate the best evidence for the treatment of stroke (Bayley et al. 2008; Lindsay et al. 2008). Although much emphasis is put on prevention and rehabilitation, it is obvious that stroke is a disease with limited treatment options. For acute hemorrhagic stroke the treatment is focused on decreasing intracranial pressure which often includes surgery (Bayley et al. 2008; Caplan 1992). For acute ischemic stroke, the most effective treatment is thrombolysis with an agent such as tissue plasminogen activator (National Institute of Neurological Disorders 1995; van der Worp and van Gijn 2007). Although thrombolysis can restore blood flow in an occluded artery, and offers improved outcomes, not all patients are eligible to receive this treatment. In addition, prior to treatment, a computed tomography scan is required to differentiate ischemic versus hemorrhagic stroke, which is not available in many locations. Furthermore, treatment should be administered within the first 3 hours of the onset of symptoms, which is often not practical. Also, in randomized control trials with tissue plasminogen activator given within the first three hours, 31 - 50% of patients had a favorable outcome 3 months later in comparison to 20 - 38% in the placebo group (National Institute of Neurological Disorders 1995). With these modest outcomes, there is much room for improving the treatment of stroke. As mentioned, much is known about the physiology and molecular biology of stroke, and this information has been used to design therapeutic strategies. In the literature from 1957 to 2003, O’Collins has identified 1026 candidate stroke therapies that have been tested on animals and/or clinically (O'Collins et al. 2006). Strategies to reduce damage due to stroke include 10  molecules that are designed to inhibit excitotoxicity, reduce inflammation, scavenge free radicals, inhibit apoptosis, promote regeneration, reduce coagulation, regulate fluid, deliver oxygen or reduce metabolism. To date only thrombolytic agents and acetylsalicylic acid are used clinically in the treatment of stroke (Bayley et al. 2008; Lindsay et al. 2008), and only a small fraction of drugs that show promise in animals are ever tested clinically (O'Collins et al. 2006). There may be several reasons why a gap exists between experimental treatments and practice, including unintended side effects of drugs. Furthermore there may be problems with the experimental models: for example pretreatment with a drug before the onset of stroke is not a luxury in clinical practice. Furthermore, as most strokes are occlusive, it may be difficult to deliver a drug upstream of an occluded vessel where it is needed (discussed in Cheng et al. 2004).  1.2  Astrocytes  1.2.1 The role of astrocytes in the nervous system The study of neurons has dominated the field of stroke research leading some researchers to address this “neurocentric” point of view (reviewed by Barres 2008; Nedergaard and Dirnagl 2005). Although neurons are important for carrying out activity normally attributed to the brain (such as movement, sensation and memory), these cells make up less than 5% - 10% of the cells in the brain, while the most predominant cell type in the brain are glia, which comprise 90% of the cells in the human brain (Allen and Barres 2009). Glia are divided into several types, including microglia, oligodendrocytes, polydendrocytes and the most prominent glial subtype that accounts for half of the cells in the brain, the astrocyte (reviewed by Barres 2008; Kettenmann and Verkhratsky 2008; Nishiyama et al. 2009).  11  The term astrocyte was proposed by Von Lenhossek to describe the star-like cells apparent in histological sections (Kettenmann and Verkhratsky 2008; Von Lenhossek 1893), and several classes of astrocytes were categorized based on their morphology (Andriezen 1893; Kettenmann and Verkhratsky 2008; Kolliker 1889). Today it is appreciated that there are several types of astrocytes, which are classified according to their morphology, location, electrophysiological properties or molecular markers. Protoplasmic astrocytes are bushy with numerous short branches and are found in the gray matter, where they ensheath synapses and blood vessels. Fibrous astrocytes have a stellate morphology with long processes and are found in the white matter. They make contact with blood vessels and the nodes of Ranvier. Radial astrocytes have long un-branched processes and they perpendicularly surround the ventricles. There are also specialized astrocytes that are similar to protoplasmic astrocytes called Muller glia cells in the retina and Bergmann glia in the cerebellum (Zigmond 1999). Astrocytes in the hippocampus have also been classified using electrophysiological properties (discussed further in section 1.3.2), where protoplasmic astrocytes can be further classified as “GluR” positive-type astrocytes or “GluR” negative-type astrocytes (Wallraff et al. 2004). Lastly, several molecular markers, such as glutamine synthetase, S100 and glial fibrillary acidic protein (GFAP) have been used to detect astrocytes (reviewed by Sofroniew and Vinters 2010). However, labeling with glutamine synthetase or S100 antibodies is not specific for astrocytes, and not all astrocytes are labeled by GFAP staining (discussed in Sofroniew and Vinters 2010). However, reactive astrocytes (which are examined in chapter 3 and discussed below) express GFAP that is readily detected (Sofroniew 2009). Almost 40 years ago a method was developed to study purified astrocyte populations (Booher and Sensenbrenner 1972). Using this technique, astrocytes can be isolated from neonatal cortices and grown in culture. The use of cultured tissue is a useful tool for the study a 12  cell population, because the complexities of other interacting cells are removed. The classification of cultured astrocytes differs from the histological classifications described in the preceding paragraph. In culture, astrocytes are classified as either type I or II. Type I astrocytes are flat cells that express GFAP and not the A2B5 antigen, while type II astrocytes are stellate cells that express GFAP and the A2B5 antigen (reviewed by Nishiyama et al. 2009). Type I astrocytes are believed to be the equivalent of in vivo protoplasmic cortical astrocytes, while type II astrocytes are more closely related to oligodendrocytes (see Hertz et al. 1998). Astrocytes were originally thought to be passive players in the nervous system, whose role was merely to serve as supportive “glue” for neurons. Today astrocytes are known to be active players in the nervous system with a diverse range of functions. Astrocytes are important in forming the blood brain barrier, regulating blood vessel diameter, providing nutrients such as lactate to neurons, regulating ion balance in the extracellular space, repairing brain tissue, scar formation and synapse formation (Allen and Barres 2009; Bains and Oliet 2007; Barres 2008; Gordon et al. 2008). Astrocytes also play a role in synaptic transmission by forming a tripartite synapse which includes pre- and postsynaptic neuron terminals surrounded by an astrocyte (Perea et al. 2009). Astrocytes remove excess neurotransmitter and are involved in recycling neurotransmitter precursors back to neurons (Westergaard et al. 1995). Astrocytes also have receptors for neurotransmitters and are capable of releasing “gliotransmitters” such as ATP, Dserine and glutamate (Haydon and Carmignoto 2006). An important facet of astrocytes, which is relevant for this thesis, is that they form a gap junction coupled syncytium. Astrocyte gap junctions are mainly composed of Cx43 (Dermietzel et al. 1991; Yamamoto et al. 1990a) and the astrocyte syncytium allows for direct communication between astrocytes. This may be important for the passage of nutrients, signaling molecules, and communication with neighbouring cells via Ca2+ waves (Haydon and 13  Carmignoto 2006; Rouach et al. 2008; Scemes and Giaume 2006). Another important domain of astrocytes explored in this thesis is their role in regulating extracellular ion concentrations, in particular K+. Astrocytes are extensively connected by gap junctions, which provide a potential avenue to buffer rises in extracellular K+. In astrocytes, mitochondria can exist in close proximity to the plasma membrane (Golovina 2005; Grosche et al. 1999; Kolikova et al. 2006), and may be important in buffering rises in extracellular ion concentrations. This is discussed in more detail in section 1.5.  1.2.2 Astrocytes and stroke Astrocytes have recently stepped into the spotlight as important players in stroke. Several reviews have explored their role in cellular injury, and their putative role in stroke is summarized below (Chen and Swanson 2003; Giffard and Swanson 2005; Nedergaard and Dirnagl 2005; Rossi et al. 2007; Swanson et al. 2004; Takano et al. 2009). It is thought that during ischemia astrocytes largely undergo similar pathophysiological changes as neurons, just on a longer time scale (Rossi et al. 2007; Silver et al. 1997). There may be several reasons why astrocytes survive longer than neurons. For example, astrocytes maintain ATP levels longer during ischemia, which slows ionic disruption (Silver et al. 1997). One explanation for the slower decline in ATP is that astrocytes have glycogen stores, which can mobilize glucose for ATP production (reviewed by Brown and Ransom 2007). However, production of ATP in this manner leads to lactic acid accumulation that may exacerbate ischemic damage (Giffard et al. 1990; Goldman et al. 1989; Rehncrona and Kagstrom 1983; Swanson et al. 1997). Astrocytes may also be resistant to excitotoxic damage because they have the ability to scavenge ROS with the antioxidant glutathione (Swanson et al. 2004).  14  In addition to protecting themselves, astrocytes may also be able to protect neurons during ischemia. Astrocytes can take up glutamate and buffer ions (Swanson et al. 2004), which may be beneficial to slow neuronal depolarization and attenuate rises in glutamate levels. Astrocytes can also release ATP that can be converted to adenosine, which can act on presynaptic adenosine receptors to inhibit neuronal excitatory transmitter release (Lin et al. 2008). Lastly, astrocytes scavenge neuronal ROS by recycling oxidized ascorbate from neurons (reviewed by Swanson et al. 2004). In contrast, astrocytes may also promote cellular damage during stroke. During ischemia, astrocyte endfeet that contribute to the blood-brain-barrier may swell, and contribute to brain edema (Ribeiro Mde et al. 2006). Astrocytes also play a role in the inflammatory response as they can secrete a wide array of cytokines including tumor necrosis factor (TNF)- , TNF- , IL1, IL-6, IL-10, interferon (INF)- and INF- . Whether these cytokines serve a protective or destructive role during stroke in vivo is not fully understood and is complicated by the fact that cytokines have multiple roles and may be involved in both protection and destruction (reviewed by Hossmann 1994; Swanson et al. 2004). In the hours to days following stroke astrocytes play an important role in the recovery processes. Within hours astrocytes start to undergo hypertrophy and proliferation in a process termed reactive astrogliosis. These astrocytes have increased production of GFAP, vimentin and nestin, as well as changes in the production of other molecules (e.g. cytokines, glutathione) (reviewed by Pekny and Nilsson 2005; Sofroniew 2009). In the following days, reactive astrocytes form a scar to wall off the infarct area. Although this has the advantage of isolating the region of injury, gliosis may serve as a barrier to future axonal regeneration (reviewed by Ridet et al. 1997; Sofroniew 2009). That being said, by secreting NGF, basic fibroblast growth factor, brain-derived neurotrophic factor, transforming growth factor-  platelet-derived growth 15  factor, neuregulins, ciliary neurotrophic factor, neuropilin-1 and vascular endothelial growth factor, astrocytes may have a trophic role for neurite outgrowth, neurogenesis and angiogenesis that could provide a potential avenue for neuronal growth and vascularization following stroke (Chen and Swanson 2003; Giffard and Swanson 2005; Hossmann 2006; Nedergaard and Dirnagl 2005; Swanson et al. 2004). The previous sections have highlighted the fact that stroke is a complex event with many systems acting to promote destruction and repair. In the following sections I will highlight the system of neuroprotection that is the focus of this thesis, namely K+ sequestration via the gap junction protein Cx43.  1.3  Gap junctions and Cx43  1.3.1 Gap junctions Dewey and Barr were the first to describe intercellular connections between adjacent cells in intestinal smooth muscle cells, and named this contact the “nexus” (Dewey and Barr 1962). Later, it would be found that these intercellular connections, now termed gap junctions, are composed of 6 connexin protein moieties which join to form a connexon or hemichannel. These hemichannels generally adjoin with hemichannels from a neighbouring cell to form a gap junction (Fig. 1.4). Gap junctions are water-filled pores that allow for the intercellular propagation of ions and molecules. Connexins are named for their approximate molecular weight and the first connexin to be cloned was Cx32 (Paul 1986). To date 21 mouse and 20 human connexin genes have been identified (Sohl and Willecke 2003). Subtypes of connexins differ in several ways, including their pattern of tissue expression, electrical properties, molecules which can permeate and sensitivity to modulating molecules. In some instances heterotypic gap junctions may form, where a hemichannel of one subtype of connexin can 16  interact with another hemichannel subtype (Desplantez et al. 2004; Palacios-Prado and Bukauskas 2009; Rubin et al. 1992). Heteromeric gap junctions can also form where a hemichannel contains multiple types of connexins (Bouvier et al. 2009; He et al. 1999). This arrangement has the potential to vastly increase the diversity of gap junctions. There are several human diseases which are attributable to mutations of connexins. For example, deafness can be caused by mutations of Cx26, Cx31, Cx30, Cx32 or Cx43 (reviewed by Martinez et al. 2009). Also, X-linked Charcot Marie Tooth disease, a disease that affects motor and sensory neurons, is caused by mutation of Cx32 (Bergoffen et al. 1993). Oculo-dentodigital dysplasia, a disease that causes developmental abnormalities of the face, the eyes, the limbs and the teeth, is caused by mutations of Cx43 (Paznekas et al. 2003). The understanding of the role of connexins has evolved over the years. Although traditionally these proteins have been thought to be a conduit for intercellular communication, connexins are known to have other functions. Unopposed connexons (hemichannels) can communicate directly with the extracellular space allowing release of molecules such as ATP and amino acids (Kang et al. 2008; Stridh et al. 2008; Ye et al. 2003). Connexins have also been shown to play a role in neural migration and tumor suppression, which is independent of gap junction communication (Cina et al. 2009; Fu et al. 2004; reviewed by Cronier et al. 2009; Naus 2002). Furthermore, protein-protein interactions at the C-terminal region of connexins may contribute to signaling pathways that are independent of gap junction coupling (Laird 2010). Connexins have also been found to localize to the mitochondria, which serve a role in K+ uptake (Miro-Casas et al. 2009) and cardioprotection (Boengler et al. 2005; Boengler et al. 2006; Goubaeva et al. 2007; Rodriguez-Sinovas et al. 2006). Invertebrates do not express connexins, but do form gap junctions with proteins called innexins. Using degenerate primers based on innexins, another class of gap junction proteins has 17  been identified in vertebrates, termed pannexins (Panchin et al. 2000). Three types of pannexins have been identified: pannexin1, pannexin2 and pannexin3. As discussed by Thompson, pannexin1 and pannexin2 have been found in neurons, and to a lesser extent in astrocytes, and functionally pannexin1 channels appear to act more as a hemichannel than a gap junction (Thompson and MacVicar 2008).  1.3.2 Astrocyte connexins In culture, type II astrocytes do not form gap junctions or express gap junction proteins (Belliveau and Naus 1994) while type I astrocytes form gap junctions and express Cx43 (Dermietzel et al. 1991). In intact brain tissue, all classes of astrocytes, including fibrous, protoplasmic, radial, Muller and Bergmann cell types, are known to form gap junctions (Liu et al. 2006; Muller et al. 1996; Yamamoto et al. 1990b; Zahs et al. 2003). When hippocampal astrocytes are assessed based on different electrophysiological properties in intact brain sections, “GluR” positive-type astrocytes do not gap junction couple while “GluR” negative-type astrocytes extensively gap junction couple (Wallraff et al. 2004). These astrocytes may be similar to type I-like astrocytes, which express the primary gap junction protein Cx43 (Dermietzel et al. 1989; Yamamoto et al. 1990a). Cx30 and Cx26 are also expressed in astrocytes to a lesser extent (Nagy et al. 2004; Nagy et al. 2001), and in the presence of neurons, astrocyte Cx30 and Cx43 expression increases (Koulakoff et al. 2008). There is also evidence that in Cx43 knock-out astrocytes Cx40, Cx45 and Cx46 may be expressed (Dermietzel et al. 2000).  18  1.3.3 Cx43 expression and functional properties Cx43 was named for its approximate molecular weight of 43 kDa. Mouse Cx43 is composed of 382 amino acids. This protein is widely expressed in the body, and found in at least 46 different cell types (reviewed by Laird 2006). Most tissues express Cx43, and commonly studied regions expressing Cx43 include the heart, brain, skin, bone, uterus and ovaries. In the brain Cx43 is expressed in astrocytes, leptomeningeal cells and ependymal cells. Cx43 is typically thought of as a plasma membrane protein. However, evidence exists that this protein is also in other cellular compartments such as mitochondria and potentially the nucleus (Boengler et al. 2005; Boengler et al. 2009; Dang et al. 2003; Goubaeva et al. 2007; Mennecier et al. 2008; Miro-Casas et al. 2009; Rodriguez-Sinovas et al. 2006). Mitochondrial localization of Cx43 is discussed in more detail in a separate section below. Cx43 is synthesized in the endoplasmic reticulum (ER), assembled into a hexameric connexon in the Golgi apparatus and transported to the plasma membrane where they dock with connexons from adjacent cells to form gap junctions (Laird 2006). The Cx43 peptide crosses the membrane 4 times such that it has intracellular amino and carboxy terminal domains with 2 extracellular loops and 1 intracellular loop (Yancey et al. 1989). Gap junction proteins only have half-lives of a few hours, and gap junction plaques are internalized from the central region of the plaque with both halves entering one cell (Gaietta et al. 2002). Cx43 may then be degraded by lysosomes, proteasomes and potentially endosomes (Laird 2006). As mentioned above, gap junctions allow for direct cytoplasmic communication between coupled cells. Cx43 allows passage of molecules less than 1 kDa, but not indiscriminately. Gap junctions are selective for the molecules that pass through, and some parameters of importance are molecular weight, shape and charge (Goldberg et al. 2004). Our knowledge of all endogenous molecules that pass through gap junctions is likely incomplete. However, several 19  molecules and ions are known to pass through Cx43 based gap junctions including: K+, Na+, Ca2+, inositol trisphosphate (IP3), adenosine, cyclic adenosine monophosphate (cAMP), adenosine diphosphate (ADP), ATP, glucose, glutamate and glutathione (Goldberg et al. 2004; Goldberg et al. 1998; Wang and Veenstra 1997) . The existence of Cx43 hemichannel openings is controversial (Spray et al. 2006). However, some studies have suggested that hemichannels can release substances such as amino acids, ATP, glutamate and glutathione (Kang et al. 2008; Stout et al. 2002; Stridh et al. 2008; Ye et al. 2003). The electrophysiological properties of hemichannels have been studied, and it is known that the single channel conductance is 220 pS with the presence of a 75 pS substate (the conductance of a Cx43 gap junction is half or 110 pS) (Contreras et al. 2003b). One reason to doubt that hemichannels open in vivo is that these channels open at membrane potentials greater than 60 mV (Contreras et al. 2003b), whereas the resting membrane potential in astrocytes is less than -80 mV. However, externally applied ethidium bromide, a compound permeable through hemichannels, has been shown to enter cells at resting membrane potentials in Cx43 transfected cells (Contreras et al. 2003b). It has been suggested that hemichannels have a low open probability under normal conditions, and various experimental perturbations such as oxidative stress or hypoxia may further open these channels (Lin et al. 2008; Retamal et al. 2007). In addition to voltage gating, hemichannel activity is also regulated by Ca2+, pH, channel phosphorylation and channel nitrosylation (Contreras et al. 2003b; Retamal et al. 2007; Saez et al. 2005). Regulation of gap junctions has been extensively studied and an exhaustive list of endogenous and exogenous modulators has been compiled (Salameh and Dhein 2005). As with hemichannels, pH and phosphorylation are important modulators of Cx43 channel activity. A decline in pH or dephosphorylation typically closes gap junctions (Delmar et al. 2004; Lampe and Lau 2000; Lampe and Lau 2004). 20  1.3.4 C-terminal region of Cx43 It is recognized that connexins interact with other proteins. The study of the connexin proteome is in its infancy, but to date, as discussed in a recent review by Laird, over 40 proteins have been found to interact directly or indirectly with connexins (Laird 2010). These interacting proteins include: kinases, phosphatases, proteins for proteasome-based degradation, as well as proteins involved in scaffolding, trafficking, cytoskeletal interactions and growth regulation. Many of these interactions are known to occur at the C-terminal domain of connexins. The cytoplasmic C-terminal of connexins is a region of diversity amongst the connexin family. Cx43 has several known interaction sites of importance. Fig. 1.5 (adapted from Giepmans 2004) depicts many of the known interactions with the C-terminal region of Cx43. Cx43 is known to bind proteins such as -tubulin, -tubulin, CCN3, c-src, zonula occludens-1 (ZO-1) and has phosphorylation sites for protein kinase A (PKA), protein kinase C (PKC) and mitogen-activated protein kinase (MAPK) (Fu et al. 2004; Giepmans et al. 2001a; Giepmans and Moolenaar 1998; Giepmans et al. 2001b; Herve et al. 2007; Sin et al. 2009). These interactions are believed to play a role in gap junction assembly, activity and gating (Duffy et al. 2002; Liu et al. 1993; Solan and Lampe 2009; Herve et al. 2007; Kim et al. 1999; Lampe and Lau 2000; Lampe et al. 2000; Martinez et al. 2003; Musil et al. 1990). Although the significance of many of these interactions is not fully understood, one area of particular interest is the role of C-terminal region interactions and channel gating. As indicated in Fig. 1.5, the C-terminal region has many sites for phosphorylation. Dephosphorylating agents are known to decrease gap junction coupling (Duthe et al. 2000), while stimulation of kinases (PKC, PKA or MAPK) increases gap junction coupling (Kwak et al. 1995; Kwak and Jongsma 1996; Zhang et al. 1999). However, it is important to note that Cx43 phosphorylation is 21  complicated and in some instances agents that phosphorylate or dephosphorylate may have the opposite effect than that described above (Lampe and Lau 2000). The reason for this is likely due to the specific amino acid residues that are being acted upon in the context of numerous other interacting molecules. In addition to gating by phosphorylation, Cx43 is known to be gated by a decline in pH. The C-terminal region of Cx43 is known to be a pH sensor and is important in channel gating (Liu et al. 1993; Stergiopoulos et al. 1999). Because so many molecules interact at the C-terminal domain, several studies have investigated changes that occur when this region is truncated. Truncation of the C-terminal region prolongs the mean open time of the Cx43 channel and increases channel conductance (Fishman et al. 1991; Moreno et al. 2002). A mouse with a truncated Cx43 C-terminal region has been generated by mutating the adenine base at codon 258 to thymine, resulting in a translational stop codon (Maass et al. 2004). The truncated form of Cx43 is referred to as Cx43 CT in this thesis. Mice with both alleles truncated (Cx43  CT/ CT  ) die shortly after birth  due to an epidermal barrier deficit (Maass et al. 2004), but mice with one truncated allele (either Cx43  CT/+  or Cx43  CT/-  mice) survive (Maass et al. 2007). It has been shown that  cardiomyocytes in mice with one truncated allele and one knock-out allele (Cx43  CT/-  ) couple  (Maass et al. 2009; Maass et al. 2007) and have channel properties similar to studies where truncated Cx43 has been transfected into cells (Fishman et al. 1991; Moreno et al. 2002). One notable change with the Cx43 CT mutation is a reduction in the total number of gap junction plaques, yet these plaques are larger in size (Maass et al. 2007). This could be explained by the possibility that the C-terminal region may be involved in channel degradation (Li et al. 2008) and a lack of this region increases the channel half-life (Maass et al. 2004). The hearts of these mutant mice have been subjected to ischemic conditions by occluding the left descending  22  coronary artery (Maass et al. 2009). In this study it was found that Cx43  CT/-  mice had increased  infarct size and a reduced sensitivity to acid-induced uncoupling. The work in this thesis is based, in part, by the finding that reduced Cx43 enhances damage due to stroke (Lin et al. 2008; Nakase et al. 2003a; Nakase et al. 2003b; Nakase et al. 2004; Siushansian et al. 2001). Because the C-terminal region is important in channel gating, this is a site of interest in elucidating Cx43’s neuroprotective mechanisms. The role of Cx43 in cellular damage is discussed in the following section.  1.4  Gap junctions and cellular injury  1.4.1 Gap junction proteins and cellular injury The role of gap junctions in cellular injury is often cited as “controversial” because several studies have shown that gap junctions are either protective or destructive. In astrocytes, Cx43 is the primary gap junction protein (Dermietzel et al. 1991; Dermietzel et al. 1989; Giaume et al. 1991; Yamamoto et al. 1990a). This particular protein has been shown to be protective in both in vivo and in vitro models (Blanc et al. 1998; Boengler et al. 2007; Giardina et al. 2007; Goubaeva et al. 2007; Heinzel et al. 2005; Lin et al. 2008; Lin et al. 2003; Nakase et al. 2004; Oguro et al. 2001; Ozog et al. 2002; Rodriguez-Sinovas et al. 2006; Siushansian et al. 2001; Theis et al. 2003). Our lab has shown that blockade of gap junctions in vitro enhanced neuronal death in an astrocyte neuron co-culture system (Ozog et al. 2002). Furthermore, our in vivo studies have revealed that mice with reduced Cx43 expression exhibit increased infarct volume and apoptosis (Nakase et al. 2003a; Nakase et al. 2003b; Nakase et al. 2004; Siushansian et al. 2001). In addition, it has been shown that Cx43 immunoreactivity increases after ischemic events in postmortem human brain tissue (Nakase et al. 2006), the significance of which is not understood. Although astrocyte Cx43 may be protective, the exact mechanism of action remains 23  to be determined. Gap junction intercellular communication may mediate neuroprotection during stroke by allowing metabolites (e.g. antioxidants, ATP, glucose) to move into areas of high energy demand, while also buffering cytotoxic levels of excitatory amino acids and ions (such as K+) through the astrocytic syncytium. Conversely, several studies have shown that gap junction proteins can promote cellular injury (de Pina-Benabou et al. 2005; Frantseva et al. 2002a; Lin et al. 1998; Nodin et al. 2005; Perez Velazquez et al. 2006; Rami et al. 2001; Rawanduzy et al. 1997; Thompson et al. 2006; Warner et al. 1995), possibly through intercellular propagation of cytotoxic substances, spreading depression-like depolarizations or by opening of neuronal hemichannels. It is known that gap junctions remain open during simulated ischemic conditions (e.g. calcium overload, oxidative stress or metabolic inhibition) (Cotrina et al. 1998; Lin et al. 1998), and much of the debate involves determining whether the propagation of harmful or protective molecules take precedence. However, it is now appreciated that connexins have gap junctionindependent functions, such as hemichannel activity. Connexins are not limited to glial cells, neurons also express connexins (and pannexins), adding to the complexity of determining a precise role of gap junction proteins in stroke. In the following sections I will discuss in detail the studies which support a protective or destructive role for connexins in cellular injury. These studies will be framed in the context of stroke and will build a platform for the hypotheses tested and the theory discussed in this thesis. Although on the surface it appears that the results of these studies are incongruent, I will discuss how differences in the experimental design and the type of connexin(s) or pannexin(s) under study may explain why gap junction proteins may have opposing roles. In the following sections I will also discuss the possibility that gap junction proteins may serve both a protective and destructive role in cellular injury and that gap junction injury studies are not necessarily contentious. 24  1.4.2 Gap junction proteins and cellular destruction As mentioned above, one theory of how gap junctions may be destructive is via the propagation of harmful molecules through gap junctions. One of the earliest and frequently cited studies suggesting gap junctions are destructive involved intraperitoneal (i.p.) injection with the gap junction blocker octanol, followed by middle cerebral artery occlusion (MCAO) and assessment of infarct volume 24 h later (Rawanduzy et al. 1997). Because gap junctions play a role in spreading depression (Nedergaard et al. 1995) these authors reasoned that gap junction blockade would reduce the expansion of infarct volume and block spreading depression-like depolarizations through gap junctions. Indeed, infarct volume was reduced and experimentallyinduced spreading depression was inhibited by octanol treatment in this study. These results contrast with experiments from our lab (see section 1.4.3) where infarct volume is increased following MCAO in mice with an astrocyte specific deletion of Cx43. Another group has also shown that experimentally-induced spreading depression velocity is increased when Cx43 is knocked out of astrocytes (Theis et al. 2003). Therefore it is likely that the Rawanduzy study may have revealed an astrocyte Cx43-independent pathway for protection that may or may not involve gap junctions. It is worth noting some potential flaws in the interpretation of the results from the original study. Since octanol was delivered i.p., it is not known whether this compound reached the brain at high enough concentrations to block gap junctions or if there were indirect actions of octanol on other organs (e.g. the heart). Octanol is also not specific for gap junctions, in fact octanol and many other anesthetics may exert neuroprotective properties though action on ion channels or neurotransmitter release (Matchett et al. 2009; Narahashi et al. 1998). It is now known that neuronal pannexin hemichannels open in response to ischemia (Thompson et al. 2006), and potentially octanol was acting on neuronal connexins or pannexins to reduce cell damage rather than acting on astrocyte gap junctions. 25  To test the hypothesis that gap junction communication plays a role in the propagation of harmful molecules, another study has also been performed with i.p. injection of octanol followed by transient global ischemia induced by occluding both carotid arteries (Rami et al. 2001). Hippocampal cell death was reduced in the octanol treatment group. Similar results have been found in a 4 vessel occlusion model with application of the gap junction blocker carbenoxolone (Perez Velazquez et al. 2006). In addition to the points raised for the Rawanduzy study, these studies used global ischemia (unlike our studies which use focal ischemia). Differences in stroke models are known to exist (Hossmann 2008; Traystman 2003), and are likely to contribute to the differences observed in our studies. Whereas focal occlusion is known to create a necrotic core and penumbral region, global ischemia tends to cause damage in brain regions susceptible to cell death, and by increasing the length of ischemia the number of affected regions increase. A global ischemia model may not be ideal to assess a beneficial role for gap junctions as there may not be the same gradients of healthy and non-healthy cells as created in stroke models with a penumbral region. However, global ischemic studies may have relevance to the events that occur in the necrotic core of a focal stroke, where gradients of healthy and unhealthy cells are unlikely to exist. One research group exposed hippocampal slices to oxygen-glucose deprivation and found that application of gap junction blockers reduced cell death at 48 hours (Frantseva et al. 2002b). This experimental setup again may not be ideal for studying an endogenous protective mechanism through gap junctions as OGD is applied uniformly to the hippocampal brain slice preparation, and domains of healthy cells may not exist (this model is beneficial in the study of states where blood supply is uniformly disrupted, for example systemic hypoperfusion). Interestingly, in this study reduction of Cx26 and Cx32, both neuronal connexins, reduced cell death suggesting neuronal connexins are important in mediating cell death. This also fits with 26  the experiments above describing a reduction in cell death with application of gap junction blockers (e.g. octanol) that may be acting on neuronal connexins (or pannexins). These authors also reduced Cx43 expression, which lead to a reduction in cell death. The authors note that hippocampal neurons may express Cx43, which could play a role in cellular destruction (Simburger et al. 1997). A similar experiment was performed in slice cultures, where it was found that gap junction blockade also decreased cell death at time points up to 24 hours (de PinaBenabou et al. 2005). This study also examined the effect of gap junction blockade on intrauterine hypoxia-ischemia, essentially finding the same result as in the slice cultures. These studies do not dispute our findings that astrocyte Cx43 plays a role in cellular protection in a focal ischemic model. Rather, they highlight the potential importance of neuronal connexins in cell death and suggest that the mechanisms at play in global ischemia are different. Culture models have also been used to study cell death. In one study, cultured hippocampal astrocytes were subjected to iodoacetate which depletes ATP levels (Nodin et al. 2005). It was found that gap junction blockade reduced cell death. Again, this could potentially be explained by the uniformly applied treatment. Another commonly cited paper which argues that gap junctions are destructive is a study that used a glioma cell line with overexpression of the human proto-oncogene bcl2 (Lin et al. 1998). These cells are more resistant to cell death. However, when these cells were coupled with cells that do not express this gene and then exposed to several types of insults (none of which were ischemic) the more resistant cells would die. The interpretation is that gap junction coupled cells that are susceptible to cell death can cause “bystander” cell death by propagating harmful molecules through gap junctions to cells that are normally protected from chemical insults. In this study the ratio of healthy to nonhealthy cells was important. When the number of resistant cells was increased, the bystander  27  cell death effect was abolished, indicating that too many injured cells may tip the balance toward cellular destruction.  1.4.3 Gap junction proteins and cellular protection In contrast to the above, numerous studies have suggested that gap junctions are protective. As in section 1.4.2, many different approaches have been used to examine the role of connexins, and several different subtypes of connexins have been explored. Not all of the studies listed below have direct relevance to the mechanisms at play in this thesis; however they are presented for completeness, since proponents for a role of gap junctions in cellular protection often cite these works. The underlying hypothesis for gap junction-based protection is that gap junction coupled astrocytes provide a conduit to buffer harmful ions or molecules and provide helpful molecules to an area of injury. Several studies from our lab have shown protection using a permanent MCAO focal stroke model when infarct volume was assessed 4 days after occlusion. When wild-type (Cx43+/+) and heterozygote (Cx43+/-) mice were subjected to MCAO, infarct volume was approximately 50% greater in the Cx43+/- mice (Siushansian et al. 2001). In a separate study it was again found that infarct volume was greater in Cx43+/- mice following MCAO (Nakase et al. 2003a). In this latter study it was found that apoptosis was greater in Cx43+/- mice in comparison to Cx43+/+ mice at 4 days, however at 1 day no differences in apoptosis were observed. Also, it was found that reactive gliosis was reduced in the Cx43+/- mice, suggesting Cx43 may be involved in astrocyte hypertrophy and proliferation. Furthermore, in order to specifically target the effect of Cx43 expression in astrocytes, a GFAP promoter cre-lox recombinase system was developed which selectively knocked Cx43 out of astrocytes (Cx43fl/fl/hGFAP-cre). The advantage of this system is that the role of Cx43 in astrocytes can be 28  assessed without altering Cx43 expression in other cell types. It was found that infarct volume was approximately 65% larger in Cx43fl/fl/hGFAP-cre mutants (Nakase et al. 2003b). Building on this study, it was shown that Cx43fl/fl/hGFAP-cre mice have greater apoptosis, reduced gliosis and increased inflammatory cell invasion following stroke (Nakase et al. 2004). These studies show that in the absence of Cx43, damage due to focal stroke is greater, suggesting a role for endogenous astrocyte Cx43 in cellular protection. One limitation of these studies is that mutant mice may have altered expression of many genes (Iacobas et al. 2003), which may be a confounding variable. Also, it is not clear from these studies how Cx43 is acting to protect cells. Although gap junction coupling is often thought of as a potential route for the delivery of protective molecules, recently it has been shown that hemichannels may play a role, as shown in a hypoxic preconditioning study. Preconditioning is explained in more detail in section 1.5.5. In brief, preconditioning involves sub-lethal exposure to a substance, which confers protection against subsequent insults when a larger dose is given. Exposure to a sub-lethal hypoxic stimulus increases ATP release via hemichannels (Lin et al. 2008). ATP is then converted to a neuroprotective molecule, adenosine, which was found to exert a protective effect following MCAO (Lin et al. 2008). The MCAO experiments in this thesis are performed on nonpreconditioned mice, so it is possible that hemichannel adenosine release may not be a factor in our work. However, this study did compare the effect of MCAO in non-preconditioned mice with a double deletion of astrocyte Cx43 and total Cx30 (Cx43fl/fl/hGFAP-cre/Cx30-/-). Cx30 is expressed in astrocytes to a lesser extent than Cx43, and unpublished observations in our lab indicate that infarct volumes following MCAO do not differ between Cx30+/+ and Cx30-/- mice. Therefore the mice used in the Lin (2008) study are comparable our experiments with Cx43fl/fl/hGFAP-cre mice (Nakase et al. 2003b). Consistent with our findings, infarct volume was  29  larger in the double mutant lacking both Cx30 and Cx43 in comparison to wild-type (Lin et al. 2008). The focus of this thesis is the exploration of astrocyte Cx43 in cellular injury, however for completeness I will mention that some studies have suggested that neuronal gap junction coupling is a protective mechanism. Although not tested explicitly, Cx32 and Cx36 are thought to be protective in the hippocampus following global ischemia because these connexins are expressed in interneurons and may play a role in synchronizing GABA release (Oguro et al. 2001). In this study, Cx32 knock-out mice were found to have more hippocampal CA1 neuron cell death, lending support to this theory. However, it is unclear why knock-down of Cx32 in the OGD hippocampal slice study described in the previous section (Frantseva et al. 2002b) resulted in neuroprotection. Several cell culture studies have also shown cellular protection during application of injurious molecules. Using a neuron-astrocyte co-culture model, our lab has shown that neuronal death is enhanced upon application of glutamate in the presence of gap junction blockers (Ozog et al. 2002). Glutamate was applied uniformly across the culture so gradients for protective molecules to pass along to areas of demand, as in a focal stroke model, are unlikely to exist. However, it is possible that protection could be explained by an action of adenosine as previously described (Lin et al. 2008), or by direct coupling between neurons and astrocytes, which is known to occur in cell culture (Froes et al. 1999). Similarly, in a co-culture system, it has been shown that blockade of gap junctions increases neuronal death when exposed to oxidative stress (Blanc et al. 1998). Although in this study neuron-astrocyte coupling was minimal, potentially other factors were at play (e.g. release of protective molecules or buffering of harmful molecules through hemichannels). The final cell culture study to be discussed is one in which Cx43, Cx32 or Cx40 over expression in C6 glioma cells conferred resistance to cell 30  death when exposed to a variety of non-ischemic insults (Lin et al. 2003). Although it may be difficult to compare glioma cell line studies to primary astrocytes studies, one interesting caveat to this work is that cellular protection was independent of gap junction communication, as isolated cells or cells expressing constructs which do not form gap junctions also displayed resistance to cell death. The study of another potential cellular protection mechanism is in its infancy. Cx43 is expressed in mitochondria and may play a role in modulating cell death. Mitochondrial blockade of Cx43 has been shown to release cytochrome c, an important trigger for apoptosis (Goubaeva et al. 2007). Also, mitochondrial Cx43 has been shown to be important in ischemic preconditioning (Heinzel et al. 2005; Rodriguez-Sinovas et al. 2006). Potential roles of mitochondrial Cx43 and its potential involvement cerebral ischemia are explored in subsequent sections of this thesis. Based on the above studies it is obvious that gap junction proteins play a diverse role in cellular injury. It is overly simplistic to state that these proteins as a group are either protective or destructive as it appears that protection or destruction depends on the nature of the insult and the type of connexin or pannexin involved. In the following section I will propose some ideas as to how gap junction proteins may serve this dual role.  1.4.4 Resolving controversies One of the main limitations in gap junction studies is the lack of selective blockers or enhancers for specific connexin or pannexin subtypes, and the lack of selectivity between hemichannels and gap junctions. It is also difficult to study the role of connexins in cellular damage because connexin functions are not limited to cell-cell communication and hemichannel  31  activity. As mentioned earlier, connexins also interact with at least 40 different proteins, which may have importance in signaling events that occur during cellular injury. Differing experimental findings in connexin studies may be due to the type of cells under study (e.g. neuronal, glial or cell line), experimental design, or the ratio of healthy to non-healthy cells (Rossi et al. 2007). Some general patterns emerge in the studies described above, however, at this time the story is far from complete. It appears that in studies of focal ischemia, where an ischemic core and penumbra are created, gap junctions tend to be protective, potentially because harmful compounds can be buffered away from the penumbra and protective molecules can move to the site of injury. Astrocyte hemichannels may also play a role in the release of protective molecules and potentially the buffering of toxic substances. The ratio of healthy to non-healthy cells may be more favorable in this situation as well. Whereas when global ischemia occurs, or if a brain slice is uniformly exposed to an ischemic insult, gap junctions tend to increase cell death potentially through spreading of toxic molecules and through the opening of neuronal hemichannels.  1.5  K+ handling by astrocytes  1.5.1 K+ changes during stroke Extracellular K+ concentration ([K+]o) is normally between 2.7 and 3.5 mM (Somjen 2002). A single action potential can increase [K+]o by 1 mM and intense neuronal stimulation can increase [K+]o by 5 mM (Adelman and Fitzhugh 1975; Heinemann and Lux 1977; reviewed by Walz 2000). Depending on the study, during ischemia and spreading depression, [K+]o can reach values up to 25 – 80 mM (see Hansen 1985; Nedergaard and Hansen 1993; Walz 2000). Large rises in K+ have subsequent effects on neuronal excitability, synaptic transmission, glutamate re-uptake and, ultimately, neuronal viability (Balestrino et al. 1986; Hablitz and 32  Lundervold 1981; Rausche et al. 1990; Somjen and Muller 2000; reviewed by Somjen 2002; Walz 2000). During ischemia rises in [K+]o have been described to occur in 3 phases. In the initial 2 min from the onset of ischemia there is a slow exponential rise in [K+]o to 10 mM. This is followed by a rapid rise in [K+]o from 10 mM to 60 mM which occurs in seconds. The final phase is a slower rise to 80 mM [K+]o over several minutes (Hansen 1985). These measurements were made in the cortex of animals exposed to global ischemia. K+ sensitive microelectrode recordings have also been performed in the rat cerebral cortex following MCAO. It was found that [K+]o declined along a gradient of approximately 60 mM to approximately 3 mM from the ischemic core to the furthest region recorded (likely representing normal cortex) (Sick et al. 1998). As astrocytes are thought to play a critical role in regulating [K+]o under both physiological and pathological conditions (reviewed by Leis et al. 2005; Walz 2000), their properties are discussed in the following section.  1.5.2 Astrocyte K+ uptake and buffering Astrocytes have been implicated in K+ buffering for decades (Ballanyi et al. 1987; D'Ambrosio et al. 1999; Holthoff and Witte 2000; Kofuji and Newman 2004; Lian and Stringer 2004; Meeks and Mennerick 2007; Orkand et al. 1966; Ransom and Sontheimer 1995; Schlue and Wuttke 1983; Simard and Nedergaard 2004; Walz 2000; Walz and Wuttke 1999). They clear K+ from the extracellular space in 2 basic ways; K+ uptake and spatial buffering (Leis et al. 2005). Astrocytes are equipped with a variety of plasmalemmal K+ uptake mechanisms, including the Na+,K+-ATPase, ion transporters (e.g. Na+-K+-2Cl- co-transport), and voltageactivated K+ channels (e.g. inward rectifier (Kir), delayed rectifier (Kdr), and A-type K+ channels) (Bordey and Sontheimer 1997; Djukic et al. 2007; Kofuji and Newman 2004; Kressin et al. 1995; Ljubkovic et al. 2006; Seifert et al. 2009; Walz 2000). As [K+]o is ~3 mM and [K+]i is 33  ~100 mM, K+ uptake is up a concentration gradient. The energy dependent Na+,K+-ATPase is an important route of K+ uptake (D'Ambrosio et al. 2002) as are Na+-K+-2Cl- channels at more elevated [K+]o concentrations (Leis et al. 2005; Walz 1987). Kir channels are also thought to be the most important for K+ uptake. Interestingly, these channels have a high open probability at resting membrane potentials, they conduct K+ better in the inward direction, and their conductivity increases when [K+]o is raised (Newman 1993; Ransom and Sontheimer 1995). Astrocytes are also extensively coupled by Cx43 based gap junctions, which may aid in the spatial redistribution of K+ from areas of high [K+]o/i to those of low [K+]o/i (Kofuji and Newman 2004; Leis et al. 2005; Orkand et al. 1966; Wallraff et al. 2006; Walz 2000; Xu et al. 2009; see also Appendix A, Kozoriz et al. 2006). However, some critics have argued that the short length constants of astrocytic processes may limit K+ redistribution in this manner (Chen and Nicholson 2000). In support of this, similar theories of K+ siphoning through glia have been challenged (Metea et al. 2007). Since astrocytes are well known to buffer rises in [K+]o, but there may be barriers to its distribution to neighbouring cells, this raises questions as to how K+ is handled once it enters the cell. The possibility that K+ is sequestered by intracellular organelles via a Cx43 dependent mechanism is discussed in sections 1.5.3 to 1.5.5. Under ischemic conditions it is thought that penumbral region astrocytes can maintain Na+,K+-ATPase activity, allowing for K+ uptake (Leis et al. 2005; Walz 1987), while the necrotic core is unable to take up K+ due to a loss of energy to support ion transport (Rossi et al. 2007). An interesting difference between astrocytes and neurons is that elevations of [K+]o increase pHi in astrocytes (Brookes and Turner 1994). This alkalinization allows for better glucose utilization, which can support further K+ uptake through the Na+,K+-ATPase (Leis et al. 2005). Also, alkalization increases Kir conductance, which may contribute to more K+ uptake (Deitmer and Rose 1996). Na+-K+-2Cl- channels are also postulated to contribute to K+ uptake at more 34  elevated [K+]o (Leis et al. 2005). K+ buffering through gap junctions has not been examined under ischemic conditions, but it is known that elevations of [K+]o are known to increase gap junction coupling (De Pina-Benabou et al. 2001). Whether this leads to clearance of K+ away from the ischemic region requires further study.  1.5.3 Intracellular ion sequestration Mechanisms for K+ uptake into the cell are well described, however how this ion is handled once it enters the cell is not completely understood. The exact K+ concentrations in organelles such as ER, Golgi and vesicles are not known, however, electron probe x-ray microanalysis of glia indicate that mitochondria are a cellular region that contains particularly elevated levels of K+ (Stys et al. 1997). Mitochondria have also been reported by others to contain elevated levels of [K+] (Garlid 1996; Kowaltowski et al. 2002). It is well established in multiple cell types that Ca2+ can be sequestered by mitochondria and the ER, and evidence also exists that mitochondria can sequester Na+ (Azarias et al. 2008; Fameli et al. 2007; Malli et al. 2003; Pivovarova et al. 1999; Poburko et al. 2004; Poburko et al. 2009; Reyes and Parpura 2008; reviewed by Garcia-Sancho and Verkhratsky 2008; Lee et al. 2002; Nicholls and Chalmers 2004; Pizzo and Pozzan 2007; Rizzuto and Pozzan 2006). From these studies it is known that Ca2+ entry through voltage gated Ca2+ channels, store operated channels or the Na+/Ca2+ exchanger (acting in reverse mode) creates subplasmalemmal [Ca2+]i microdomains. Ca2+ is then taken up by sarco/endoplasmic reticulum Ca2+-ATPase channels or the mitochondrial Ca2+ uniporter. Ca2+ can also exit the mitochondria via a mitochondrial Na+/Ca2+ exchanger (Palty et al. 2010) which is then taken up by sarco/endoplasmic reticulum Ca2+-ATPase channels (Fameli et al. 2007; Poburko et al. 2009). This serves to refill ER Ca2+ stores. Although less well described, Na+ entry into the mitochondria has been shown to involve 35  the Na+/Ca2+ exchanger and the Na+/H+ exchanger (Azarias et al. 2008). This thesis explores whether mitochondria sequester [K+]i via analogous mechanisms described for these other ions.  1.5.4 Mitochondria and K+ uptake Whether mitochondria sequester rises in [K+]o in situ is not known, although several lines of evidence suggest that mitochondria may be suited for K+ sequestration. K+ could enter through any of the plasma membrane K+ uptake routes described above, and mitochondria located in close proximity to the plasma membrane of astrocytes (Boengler et al. 2009; Golovina 2005; Grosche et al. 1999; Kolikova et al. 2006) could take up this K+. The outer mitochondrial membrane is permeable to large molecules and ions, and is not known to have specific K+ channels, although the voltage-dependant anion channels are known to be permeable to K+ (Colombini 1989; O'Rourke 2007). K+ uptake across the inner mitochondrial membrane could occur through a variety of K+ channels (mitoKATP, mitoKca and leak channels) and may be extruded by the K+/H+ exchanger (Garlid and Paucek 2003; Szabo et al. 2005). It is also known that the mitochondrial membrane potential is quite negative (~-180 mV) (Bernardi 1999), which creates a driving force for K+ across the inner membrane. Interestingly, mitochondrial matrix [K+] has been estimated to be as high as 150-180 mM (Garlid 1996; Kowaltowski et al. 2002), and astrocyte mitochondria are known to have higher [K+] than in the cytosol (Stys et al. 1997), suggesting mitochondria can concentrate K+ (cytosolic [K+]i in astrocytes is ~100 mM; Walz 1992). Cx43 is permeable to K+ (Wang and Veenstra 1997) and, as discussed in the following section, Cx43 has been found to localize to mitochondria in cardiomyocytes (Boengler et al. 2005). Cardiomyocyte mitochondrial Cx43 has recently been shown to play a role in K+ uptake in an isolated mitochondria preparation (Miro-Casas et al. 2009) lending further support to the 36  notion that this channel plays a role in mitochondrial K+ handling. Astrocytes sequester K+ and express Cx43. Whether astrocyte mitochondria sequester K+ in situ, and the role of mitochondrial Cx43 in this process is explored further in this thesis.  1.5.5 Mitochondrial connexins Some studies have found that protection from cellular injury by connexins is independent of gap junction intercellular communication (Lin et al. 2008; Lin et al. 2003), suggesting that other non-coupling related mechanisms may be involved. The existence of Cx43 on mitochondria is a relatively new discovery. Its presence was originally suggested using human umbilical vein endothelial cells (Li et al. 2002b) and has been shown more conclusively in cardiomyocytes (Boengler et al. 2005; Goubaeva et al. 2007; Halestrap 2006). Several techniques have been used to demonstrate the presence of Cx43 in heart mitochondria including: 1) flow cytometry, 2) confocal immunohistochemistry, 3) Western blot of mitochondrial membrane fractions, and 4) electron microscopy with immuno gold labeled antibodies (Boengler et al. 2005; Boengler et al. 2009; Goubaeva et al. 2007; Halestrap 2006; Miro-Casas et al. 2009; Rodriguez-Sinovas et al. 2006). In these studies several antibodies have been used to confirm the presence of Cx43 in mitochondria, and absence of staining has been shown in a knockout (Cx43Cre-ER(tetracycline)/fl) system. Because astrocytes express high levels of Cx43, this protein could potentially exist on the mitochondria of astrocytes. How Cx43 gets to the mitochondrial membrane has been studied (Rodriguez-Sinovas et al. 2006). It appears that Cx43 is imported via translocase of the outer membrane/translocase of inner membrane (TOM/TIM) mitochondrial protein import system as Cx43 has been shown to associate by co-immunoprecipitation with members of this system, such as TOM20 and heat shock protein 90 (Rodriguez-Sinovas et al. 2006). Furthermore, an inhibitor of protein import 37  through the TOM/TIM system, geldanamycin, halts the import of Cx43 into mitochondria (Rodriguez-Sinovas et al. 2006). The function and localization of mitochondrial Cx43 is not entirely clear. As mitochondria contain an inner and outer membrane, it is attractive to think of connexins as connectors of these 2 layers, much like between two adjacent plasma membranes. One group has found that Cx43 localizes primarily to the inner mitochondrial membrane (MiroCasas et al. 2009; Rodriguez-Sinovas et al. 2006) while another has determined its presence to be primarily on the outer membrane (Goubaeva et al. 2007). The reason for the difference has been suggested to be due to different subfractionation protocols (Ruiz-Meana et al. 2008). Gap junction-like structures appear to form between the outer mitochondrial membrane and ER in intrafusal muscle fibers (Ovalle 1971), although this observation remains unexplored. Recently it has been shown that Cx43 can form hemichannels on the inner mitochondrial membrane that are involved in K+ uptake (Miro-Casas et al. 2009). The existence of hemichannels may also be of interest as plasma membrane Cx43 has been suggested to release amino acids, ATP, glutamate and glutathione (Kang et al. 2008; Stridh et al. 2008; Ye et al. 2003), molecules which are also present in mitochondria. Given the importance of maintaining ion gradients for mitochondrial function, the presence of Cx43 hemichannels may be concerning (i.e. ion gradients could be dissipated by unregulated hemichannel openings). However, extrapolating from what is known about plasma membrane hemichannel opening, it is possible that mitochondrial Cx43 could be regulated by phosphorylation state, voltage, Ca2+ and pH (Contreras et al. 2003a; Retamal et al. 2007; Saez et al. 2005; Thompson et al. 2006) so that openings only occur in selected states. Mitochondrial Cx43 is thought to play a role in ischemic preconditioning in cardiomyocytes. Preconditioning was first demonstrated by Murry and Reimer (Murry et al. 1986), where brief ischemic episodes led to subsequent cardioprotection after exposure to a stimulus that would normally be lethal. Nearly any substance which causes cellular injury can 38  cause preconditioning when given at a sub-lethal dose (reviewed by Dirnagl et al. 2003). Several lines of evidence have pointed to mitochondria as a site important in preconditioning (reviewed by Murphy and Steenbergen 2008), but the specific mechanisms involved remain unclear. Cellular stress by homocysteine led to profound increases in mitochondrial Cx43 in human umbilical vein endothelial cells (Li et al. 2002b). Such increases in mitochondrial Cx43 were also observed in cardiomyocytes after preconditioning (Boengler et al. 2005). Gap junction blockers have been shown to block the effects of preconditioning (Li et al. 2002a) and preconditioning is not observed in Cx43+/- mice (Schwanke et al. 2002; Schwanke et al. 2003). Although it is difficult to dissociate the role of plasma membrane Cx43 and mitochondrial Cx43 in these studies, it is important to note that cell survival has been reported to occur independently of gap junction intercellular communication (Lin et al. 2003; Plotkin and Bellido 2001). Furthermore, mitochondrial ROS generation has been shown to play a role in preconditioning. In one study ROS generation was tested by using diazoxide-induced preconditioning, and it was found that Cx43-/- cardiomyocytes had reduced ROS generation and were not protected during reperfusion injury (Heinzel et al. 2005). Also, when mitochondrial Cx43 levels were reduced by application of geldanamycin, an inhibitor of the TOM/TIM pathway, protective effects of mitoKATP opening by diazoxide on ischemia-reperfusion were reduced (Rodriguez-Sinovas et al. 2006). The specific connection between mitochondrial Cx43 and ROS preconditioning pathways are not well defined (reviewed in Halestrap et al. 2007; Rodriguez-Sinovas et al. 2007). However, it may involve the Cx43-based mitochondrial K+ uptake described in this thesis, as other mitochondrial K+ channels are known to generate ROS and lead to preconditioning (Pain et al. 2000). Most of the work on mitochondrial Cx43 has been performed on cardiomyocytes, and most protection studies involved preconditioned hearts. Since astrocytes express Cx43 it is 39  possible that Cx43 localize to mitochondria in these cells, and it is possible that much of the work described above likely has direct relevance to astrocytes cellular protection. As mentioned previously, preconditioning decreases infarct volume following MCAO, an effect which is reduced in Cx43fl/fl/hGFAP-cre/Cx30-/- mice (Lin et al. 2008). Although the authors examined hemichannel activity, their cells also showed a dramatic increase in Cx43 expression, which may reflect increases in mitochondrial connexin described in the heart. In non-preconditioned heart mitochondria, K+ uptake occurs through mitochondrial Cx43 hemichannels. Mitochondrial K+ uptake has been shown to be protective (Hansson et al. 2010; Kowaltowski et al. 2001), and this thesis explores whether astrocytes can also take up K+ via a Cx43 dependent pathway.  1.6  Thesis hypothesis and manuscript chapter introduction Stroke is a devastating disease that has been studied for centuries. The cascade of events  that occur at the onset of stroke is complex and, although a vast amount of information is known about the biology of stroke, the story is not complete. Furthermore, effective treatment for this disease still remains elusive despite numerous promising studies in animal models. In this thesis I will expand upon the studies previously carried out in our lab, which have examined the role of astrocyte Cx43 in stroke. I will also explore an endogenous pathway that may play a role in sequestering rises in [K+]o that are known to occur during stroke. Previously our lab has shown that Cx43+/- mice and mice with an astrocyte specific deletion of Cx43 (Cx43fl/fl/hGFAP-cre) display enhanced infarct volume after MCAO (Nakase et al. 2003a; Nakase et al. 2003b; Nakase et al. 2004; Siushansian et al. 2001). One potential explanation for cellular protection is buffering of harmful substances away from, and passage of helpful molecules toward a site of injury through a gap junction coupled astrocyte syncytium. I was interested in studying astrocyte K+ buffering, as this ion is elevated during pathological 40  states such as stroke. How K+ is handled once it enters the cell is not well studied and I examined the role of mitochondria in taking up K+. In light of the recent finding of Cx43 on mitochondria I explored the role of mitochondrial Cx43 in K+ uptake in addition to other known K+ uptake pathways. Furthermore, it is known that the C-terminal region serves as a gate to the Cx43 pore and has many interacting molecules. I was interested in determining whether this region of Cx43 plays a role in stroke by using mutants where the C-terminal has been truncated (Maass et al. 2004). I also tested whether mitochondrial K+ sequestration was altered in these mutants. This thesis tests the hypothesis that mitochondrial Cx43 aids in buffering extracellular rises in K+ and that the C-terminal of Cx43 plays a role in stroke and mitochondrial K+ sequestration. The objectives of this thesis are addressed in 3 separate manuscript chapters. In chapter 2 I aimed to assess mitochondrial K+ uptake in cultured astrocytes using an intact cellular preparation. This led to the examination of mitochondrial Cx43 in buffering [K+]o. In chapter 3 I sought to characterize the protective properties of the C-terminal region of Cx43 in stroke using a MCAO model of focal stroke. Basic characterization of the cellular properties of this mutant was also explored. In chapter 4, I examined whether mitochondrial K+ sequestration is altered in these mutants. Cx43 has been shown to play a role in cell survival. Although much is known about the role of Cx43 in cellular injury, the exact mechanism(s) of protection await discovery. Mitochondrial Cx43 has recently been described, but its presence and function in the brain has not been explored. Furthermore, the C-terminal of Cx43 is a site of channel gating and my study aims to determine if the C-terminal is important in K+ buffering and ultimately stroke. This thesis furthers our understanding of gap junction proteins and cell survival.  41  Figure 1.1. Pathways of ischemic cell death. This is an overview of processes involved in ischemic cell death which illustrates major events that are hypothesized to contribute to cell death. Column 5 lists 5 principal morphological forms taken by dying or dead cells after an ischemic insult. Column 4 lists long-term functional or structural changes. Column 3 lists actions that are likely to cause long-term functional changes described in column 4. These are termed "perpetrators" because they are considered to be key damaging events in ischemic cell death. Columns 2 and 1 show changes in many variables, initiated by original inhibition of electron transport, whose most important end result is considered to be activation of perpetrators, but which may also have more direct effects on cell damage, that are not shown. Major outputs of these changes, lumped together as initiators and activators, are changes shown in column 2, which activate perpetrators. Interpreting the figure: as indicated above, colored boxes encompass different changes that cause change in variables represented by that color. Large numbers of positive-feedback cycles are readily seen. Depol, depolarization; pHi, intracellular pH; Nai, intracellular Na+; Cai, intracellular Ca2+; FFA, free fatty acids; PAF, platelet-activating factor; e Transport, electron transport. Used with permission from The American Physiological Society: Lipton P. 1999. Ischemic cell death in brain neurons. Physiol Rev 79(4):1431-568. The figure legend had been abridged.  42  Figure 1.2. Events in brain ischemia. (a) Schematic illustration of events that occur during severe brain ischemia, as occurs in global ischemia and in the core of focal ischemia. Interruption of ATP production leads to inhibition of the Na+-K+ ATPase and to a consequent decrease of the transmembrane ion gradients. The disruption of ion homeostasis depolarizes cells and causes a large release of glutamate into the extracellular space. Ischemia also leads to extracellular acidification. (b) Events in the penumbra of focal ischemia. Initially, the drop in ATP is less severe, but triggers repeated transient depolarizations and associated ion shifts, while [Glu]o rises slowly but steadily (left panel). If reperfusion is initiated soon enough, complete recovery or selective damage may occur (bottom). With increasing duration of ischemia and proximity to the core, the transient ischemic depolarizations may evolve into terminal depolarization and ionic disruption, which leads to pannecrosis and expansion of the infarct (right panel). Question marks next to transient ATP dips indicate our speculation that ischemic transient depolarizations may induce hypoxia, similarly to spreading depression in nonischemic brain tissue, and resultant transient enhanced ATP loss. Such an effect may act synergistically with ongoing partial ischemia to promote terminal disregulation and exacerbate damage Adapted by permission from Macmillan Publishers Ltd: Nat Neurosci. Rossi DJ, Brady JD, Mohr C. 2007. Astrocyte metabolism and signaling during brain ischemia. 10(11):1377-86.  43  Figure 1.3. Focal cerebral ischemia induces a complex series of mechanisms that damage brain cells. Brain tissue responds to most of the listed examples of noxious signals by inducing protective mechanisms. In infarcted tissue, destruction has overwhelmed protection, although tissue around the ischemic core might have been spared by restored substrate delivery (owing to delivery of collaterals, spontaneous or therapeutic reperfusion, or vasculogenesis) and cellular mechanisms of protection. In this cartoon, the x-axis reflects the evolution of the cascades over time and the y-axis illustrates the impact of each element of the destructive (top) and protective (bottom) cascades on final outcome. The red dotted lines envelop major pathophysiological entities of tissue destruction in stroke (grouped for the acute mechanisms of excitotoxicity and the delayed mechanisms of inflammation and apoptosis) and the green broken lines envelop corresponding protective tissue responses. Abbreviations: BM, bone marrow; COX-2, cyclooxygenase 2; EPO, erythropoietin; IL, interleukin; MMPs, matrix metalloproteinase; OFR, oxygen free radicals. Adapted by permission from Elsevier Limited: Trends Neurosci. Dirnagl U, Simon RP, Hallenbeck JM. 2003. Ischemic tolerance and endogenous neuroprotection. 26(5):248-54.  44  Figure 1.4. A schematic representation of gap junction formation between two cells. Six connexin monomers assemble to form connexons, which are also termed hemichannels. Hemichannels are transported to the plasma membrane where they adjoin to hemichannels on neighbouring cells to form gap junctions. These intracellular connections allow for the passage of ions and molecules.  45  Figure 1.5. Human Cx43 amino acid sequence and consensus interacting protein modules. Src homology 2 domain (SH2). SH2 domains characteristically bind phospho-tyrosine residues in a hydrophobic pocket. An additional pocket confers substrate specificity by binding amino acids more C-terminally located. Note that Tyr265 of Cx43 is target of Src phosphorylation. Src homology 3 domain (SH3). Aliphatic proline residues of the substrate located on one side of an -helix bind separately to hydrophobic pockets in the SH3 domain. The flanking region determines binding specificity. PSD95/disc large/ZO-1 homology domain (PDZ). PDZ domains form a hydrophobic pocket that binds the very C-terminal residues (-x-S/T/V/I/L-x-V/I/LCOOH) of their target proteins typically transmembrane proteins. Microtubules. Microtubule binding regions are often enriched in K, V, G, P and R. Adapted by permission from Oxford Press and Dr. Giepmans: Cardiovasc Res. Giepmans BN. 2004. Gap junctions and connexin-interacting proteins. 62(2):233-45. 46  1.7  References  Andriezen, W.L. 1893. The neuroglia elements of the brain. BMJ 2, 227–230. Adelman WJ, Jr., Fitzhugh R. 1975. Solutions of the Hodgkin-Huxley equations modified for potassium accumulation in a periaxonal space. Fed Proc 34(5):1322-9. Allen NJ, Barres BA. 2009. Neuroscience: Glia - more than just brain glue. Nature 457(7230):675-7. Astrup J, Siesjo BK, Symon L. 1981. Thresholds in cerebral ischemia - the ischemic penumbra. Stroke 12(6):723-5. Azarias G, Van de Ville D, Unser M, Chatton JY. 2008. Spontaneous Na+ transients in individual mitochondria of intact astrocytes. Glia 56(3):342-53. Bains JS, Oliet SH. 2007. Glia: they make your memories stick! Trends Neurosci 30(8):417-24. Balestrino M, Aitken PG, Somjen GG. 1986. The effects of moderate changes of extracellular K+ and Ca2+ on synaptic and neural function in the CA1 region of the hippocampal slice. Brain Res 377(2):229-39. Ballanyi K, Grafe P, ten Bruggencate G. 1987. Ion activities and potassium uptake mechanisms of glial cells in guinea-pig olfactory cortex slices. J Physiol 382:159-74. Barres BA. 2008. The mystery and magic of glia: a perspective on their roles in health and disease. Neuron 60(3):430-40. Bayley M, Lindsay P, Hellings C, Woodbury E, Phillips S. 2008. Balancing evidence and opinion in stroke care: the 2008 best practice recommendations. CMAJ 179(12):1247-9. Belliveau DJ, Naus CC. 1994. Cortical type 2 astrocytes are not dye coupled nor do they express the major gap junction genes found in the central nervous system. Glia 12(1):24-34. Bergoffen J, Scherer SS, Wang S, Scott MO, Bone LJ, Paul DL, Chen K, Lensch MW, Chance PF, Fischbeck KH. 1993. Connexin mutations in X-linked Charcot-Marie-Tooth disease. Science 262(5142):2039-42. Bernardi P. 1999. Mitochondrial transport of cations: channels, exchangers, and permeability transition. Physiol Rev 79(4):1127-55. Blanc EM, Bruce-Keller AJ, Mattson MP. 1998. Astrocytic gap junctional communication decreases neuronal vulnerability to oxidative stress-induced disruption of Ca2+ homeostasis and cell death. J Neurochem 70(3):958-70. Boengler K, Dodoni G, Rodriguez-Sinovas A, Cabestrero A, Ruiz-Meana M, Gres P, Konietzka I, Lopez-Iglesias C, Garcia-Dorado D, Di Lisa F and others. 2005. Connexin43 in cardiomyocyte mitochondria and its increase by ischemic preconditioning. Cardiovasc Res 67(2):234-44. 47  Boengler K, Konietzka I, Buechert A, Heinen Y, Garcia-Dorado D, Heusch G, Schulz R. 2007. Loss of ischemic preconditioning's cardioprotection in aged mouse hearts is associated with reduced gap junctional and mitochondrial levels of connexin43. Am J Physiol Heart Circ Physiol 292(4):H1764-9. Boengler K, Schulz R, Heusch G. 2006. Connexin43 signalling and cardioprotection. Heart 92(12):1724-7. Boengler K, Stahlhofen S, van de Sand A, Gres P, Ruiz-Meana M, Garcia-Dorado D, Heusch G, Schulz R. 2009. Presence of connexin43 in subsarcolemmal, but not in interfibrillar cardiomyocyte mitochondria. Basic Res Cardiol 104(2):141-7. Booher J, Sensenbrenner M. 1972. Growth and cultivation of dissociated neurons and glial cells from embryonic chick, rat and human brain in flask cultures. Neurobiology 2(3):97-105. Bordey A, Sontheimer H. 1997. Postnatal development of ionic currents in rat hippocampal astrocytes in situ. J Neurophysiol 78(1):461-77. Bouvier D, Spagnol G, Chenavas S, Kieken F, Vitrac H, Brownell S, Kellezi A, Forge V, Sorgen PL. 2009. Characterization of the structure and intermolecular interactions between the connexin40 and connexin43 carboxyl-terminal and cytoplasmic loop domains. J Biol Chem 284(49):34257-71. Brookes N, Turner RJ. 1994. K+-induced alkalinization in mouse cerebral astrocytes mediated by reversal of electrogenic Na+-HCO3-cotransport. Am J Physiol 267(6 Pt 1):C1633-40. Brown AM, Ransom BR. 2007. Astrocyte glycogen and brain energy metabolism. Glia 55(12):1263-71. Caplan L. 2009. Caplan's Stroke: A Clinical Approach, 4th ed. Philadelphia: Saunders. Caplan LR. 1992. Intracerebral haemorrhage. Lancet 339(8794):656-8. Chan B, Hayes B. 1998. Cost of stroke in Ontario, 1994/95. CMAJ 159(6 Suppl):S2-S8. Chan PH. 1996. Role of oxidants in ischemic brain damage. Stroke 27(6):1124-9. Chen KC, Nicholson C. 2000. Spatial buffering of potassium ions in brain extracellular space. Biophys J 78(6):2776-97. Chen Y, Swanson RA. 2003. Astrocytes and brain injury. J Cereb Blood Flow Metab 23(2):13749. Cheng YD, Al-Khoury L, Zivin JA. 2004. Neuroprotection for ischemic stroke: two decades of success and failure. NeuroRx 1(1):36-45. Cina C, Maass K, Theis M, Willecke K, Bechberger JF, Naus CC. 2009. Involvement of the cytoplasmic C-terminal domain of connexin43 in neuronal migration. J Neurosci 29(7):2009-21. 48  Colombini M. 1989. Voltage gating in the mitochondrial channel, VDAC. J Membr Biol 111(2):103-11. Contreras JE, Saez JC, Bukauskas FF, Bennett MV. 2003a. Functioning of Cx43 hemichannels demonstrated by single channel properties. Cell Commun Adhes 10(4-6):245-9. Contreras JE, Saez JC, Bukauskas FF, Bennett MV. 2003b. Gating and regulation of connexin43 (Cx43) hemichannels. Proc Natl Acad Sci U S A 100(20):11388-93. Cotrina ML, Kang J, Lin JH, Bueno E, Hansen TW, He L, Liu Y, Nedergaard M. 1998. Astrocytic gap junctions remain open during ischemic conditions. J Neurosci 18(7):252037. Cox AM, McKevitt C, Rudd AG, Wolfe CD. 2006. Socioeconomic status and stroke. Lancet Neurol 5(2):181-8. Cronier L, Crespin S, Strale PO, Defamie N, Mesnil M. 2009. Gap junctions and cancer: new functions for an old story. Antioxid Redox Signal 11(2):323-38. D'Ambrosio R, Gordon DS, Winn HR. 2002. Differential role of KIR channel and Na+/K+-pump in the regulation of extracellular K+ in rat hippocampus. J Neurophysiol 87(1):87-102. D'Ambrosio R, Maris DO, Grady MS, Winn HR, Janigro D. 1999. Impaired K+ homeostasis and altered electrophysiological properties of post-traumatic hippocampal glia. J Neurosci 19(18):8152-62. Dang X, Doble BW, Kardami E. 2003. The carboxy-tail of connexin43 localizes to the nucleus and inhibits cell growth. Mol Cell Biochem 242(1-2):35-8. De Pina-Benabou MH, Srinivas M, Spray DC, Scemes E. 2001. Calmodulin kinase pathway mediates the K+-induced increase in gap junctional communication between mouse spinal cord astrocytes. J Neurosci 21(17):6635-43. de Pina-Benabou MH, Szostak V, Kyrozis A, Rempe D, Uziel D, Urban-Maldonado M, Benabou S, Spray DC, Federoff HJ, Stanton PK and others. 2005. Blockade of gap junctions in vivo provides neuroprotection after perinatal global ischemia. Stroke 36(10):2232-7. Deitmer JW, Rose CR. 1996. pH regulation and proton signalling by glial cells. Prog Neurobiol 48(2):73-103. del Zoppo GJ. 2009. Relationship of neurovascular elements to neuron injury during ischemia. Cerebrovasc Dis 27 Suppl 1:65-76. Delmar M, Coombs W, Sorgen P, Duffy HS, Taffet SM. 2004. Structural bases for the chemical regulation of connexin43 channels. Cardiovasc Res 62(2):268-75. Dermietzel R, Gao Y, Scemes E, Vieira D, Urban M, Kremer M, Bennett MV, Spray DC. 2000. Connexin43 null mice reveal that astrocytes express multiple connexins. Brain Res Brain Res Rev 32(1):45-56. 49  Dermietzel R, Hertberg EL, Kessler JA, Spray DC. 1991. Gap junctions between cultured astrocytes: immunocytochemical, molecular, and electrophysiological analysis. J Neurosci 11(5):1421-32. Dermietzel R, Traub O, Hwang TK, Beyer E, Bennett MV, Spray DC, Willecke K. 1989. Differential expression of three gap junction proteins in developing and mature brain tissues. Proc Natl Acad Sci U S A 86(24):10148-52. Desplantez T, Halliday D, Dupont E, Weingart R. 2004. Cardiac connexins Cx43 and Cx45: formation of diverse gap junction channels with diverse electrical properties. Pflugers Arch 448(4):363-75. Dewey MM, Barr L. 1962. Intercellular Connection between Smooth Muscle Cells: the Nexus. Science 137(3531):670-672. Dirnagl U, Simon RP, Hallenbeck JM. 2003. Ischemic tolerance and endogenous neuroprotection. Trends Neurosci 26(5):248-54. Djukic B, Casper KB, Philpot BD, Chin LS, McCarthy KD. 2007. Conditional knock-out of Kir4.1 leads to glial membrane depolarization, inhibition of potassium and glutamate uptake, and enhanced short-term synaptic potentiation. J Neurosci 27(42):11354-65. Duffy HS, Sorgen PL, Girvin ME, O'Donnell P, Coombs W, Taffet SM, Delmar M, Spray DC. 2002. pH-dependent intramolecular binding and structure involving Cx43 cytoplasmic domains. J Biol Chem 277(39):36706-14. Duthe F, Dupont E, Verrecchia F, Plaisance I, Severs NJ, Sarrouilhe D, Herve JC. 2000. Dephosphorylation agents depress gap junctional communication between rat cardiac cells without modifying the connexin43 phosphorylation degree. Gen Physiol Biophys 19(4):441-9. Ekdahl CT, Kokaia Z, Lindvall O. 2009. Brain inflammation and adult neurogenesis: the dual role of microglia. Neuroscience 158(3):1021-9. Fameli N, van Breemen C, Kuo KH. 2007. A quantitative model for linking Na+/Ca2+ exchanger to SERCA during refilling of the sarcoplasmic reticulum to sustain [Ca2+] oscillations in vascular smooth muscle. Cell Calcium 42(6):565-75. Fauci AS. 2008. Harrison's principles of internal medicine 17th ed. New York: McGraw-Hill Medical. Feigin VL, Lawes CM, Bennett DA, Anderson CS. 2003. Stroke epidemiology: a review of population-based studies of incidence, prevalence, and case-fatality in the late 20th century. Lancet Neurol 2(1):43-53. Fields WS, Lemak, N.A. 1989. A history of stroke: Its recognition and treatment. New York: Oxford University Press. Fishman GI, Moreno AP, Spray DC, Leinwand LA. 1991. Functional analysis of human cardiac gap junction channel mutants. Proc Natl Acad Sci U S A 88(9):3525-9. 50  Frantseva MV, Kokarovtseva L, Naus CG, Carlen PL, MacFabe D, Perez Velazquez JL. 2002a. Specific gap junctions enhance the neuronal vulnerability to brain traumatic injury. J Neurosci 22(3):644-53. Frantseva MV, Kokarovtseva L, Perez Velazquez JL. 2002b. Ischemia-induced brain damage depends on specific gap-junctional coupling. J Cereb Blood Flow Metab 22(4):453-62. Froes MM, Correia AH, Garcia-Abreu J, Spray DC, Campos de Carvalho AC, Neto MV. 1999. Gap-junctional coupling between neurons and astrocytes in primary central nervous system cultures. Proc Natl Acad Sci U S A 96(13):7541-6. Fu CT, Bechberger JF, Ozog MA, Perbal B, Naus CC. 2004. CCN3 (NOV) interacts with connexin43 in C6 glioma cells: possible mechanism of connexin-mediated growth suppression. J Biol Chem 279(35):36943-50. Gabriele W, Renate S. 2009. Work loss following stroke. Disabil Rehabil 31(18):1487-93. Gaietta G, Deerinck TJ, Adams SR, Bouwer J, Tour O, Laird DW, Sosinsky GE, Tsien RY, Ellisman MH. 2002. Multicolor and electron microscopic imaging of connexin trafficking. Science 296(5567):503-7. Garcia-Sancho J, Verkhratsky A. 2008. Cytoplasmic organelles determine complexity and specificity of calcium signalling in adrenal chromaffin cells. Acta Physiol 192(2):263-71. Garlid KD. 1996. Cation transport in mitochondria--the potassium cycle. Biochim Biophys Acta 1275(1-2):123-6. Garlid KD, Paucek P. 2003. Mitochondrial potassium transport: the K+ cycle. Biochim Biophys Acta 1606(1-3):23-41. Giardina SF, Mikami M, Goubaeva F, Yang J. 2007. Connexin43 confers resistance to hydrogen peroxide-mediated apoptosis. Biochem Biophys Res Commun 362(3):747-52. Giaume C, Fromaget C, el Aoumari A, Cordier J, Glowinski J, Gros D. 1991. Gap junctions in cultured astrocytes: single-channel currents and characterization of channel-forming protein. Neuron 6(1):133-43. Giepmans BN. 2004. Gap junctions and connexin-interacting proteins. Cardiovasc Res 62(2):233-45. Giepmans BN, Hengeveld T, Postma FR, Moolenaar WH. 2001a. Interaction of c-src with gap junction protein connexin43. Role in the regulation of cell-cell communication. J Biol Chem 276(11):8544-9. Giepmans BN, Moolenaar WH. 1998. The gap junction protein connexin43 interacts with the second PDZ domain of the zona occludens-1 protein. Curr Biol 8(16):931-4. Giepmans BN, Verlaan I, Hengeveld T, Janssen H, Calafat J, Falk MM, Moolenaar WH. 2001b. Gap junction protein connexin43 interacts directly with microtubules. Curr Biol 11(17):1364-8. 51  Giffard RG, Monyer H, Choi DW. 1990. Selective vulnerability of cultured cortical glia to injury by extracellular acidosis. Brain Res 530(1):138-41. Giffard RG, Swanson RA. 2005. Ischemia-induced programmed cell death in astrocytes. Glia 50(4):299-306. Goldberg GS, Lampe PD, Sheedy D, Stewart CC, Nicholson BJ, Naus CC. 1998. Direct isolation and analysis of endogenous transjunctional ADP from Cx43 transfected C6 glioma cells. Exp Cell Res 239(1):82-92. Goldberg GS, Valiunas V, Brink PR. 2004. Selective permeability of gap junction channels. Biochim Biophys Acta 1662(1-2):96-101. Goldman SA, Pulsinelli WA, Clarke WY, Kraig RP, Plum F. 1989. The effects of extracellular acidosis on neurons and glia in vitro. J Cereb Blood Flow Metab 9(4):471-7. Goldstein LB, Adams R, Becker K, Furberg CD, Gorelick PB, Hademenos G, Hill M, Howard G, Howard VJ, Jacobs B and others. 2001. Primary prevention of ischemic stroke: A statement for healthcare professionals from the Stroke Council of the American Heart Association. Stroke 32(1):280-99. Golovina VA. 2005. Visualization of localized store-operated calcium entry in mouse astrocytes. Close proximity to the endoplasmic reticulum. J Physiol 564(Pt 3):737-49. Gordon GR, Choi HB, Rungta RL, Ellis-Davies GC, MacVicar BA. 2008. Brain metabolism dictates the polarity of astrocyte control over arterioles. Nature 456(7223):745-9. Goubaeva F, Mikami M, Giardina S, Ding B, Abe J, Yang J. 2007. Cardiac mitochondrial connexin43 regulates apoptosis. Biochem Biophys Res Commun 352(1):97-103. Grosche J, Matyash V, Moller T, Verkhratsky A, Reichenbach A, Kettenmann H. 1999. Microdomains for neuron-glia interaction: parallel fiber signaling to Bergmann glial cells. Nat Neurosci 2(2):139-43. Hablitz JJ, Lundervold A. 1981. Hippocampal excitability and changes in extracellular potassium. Exp Neurol 71(2):410-20. Halestrap AP. 2006. Mitochondria and preconditioning: a connexin connection? Circ Res 99(1):10-2. Hansen AJ. 1985. Effect of anoxia on ion distribution in the brain. Physiol Rev 65(1):101-48. Hansen AJ, Nedergaard M. 1988. Brain ion homeostasis in cerebral ischemia. Neurochem Pathol 9:195-209. Hansson MJ, Morota S, Teilum M, Mattiasson G, Uchino H, Elmer E. 2010. Increased potassium conductance of brain mitochondria induces resistance to permeability transition by enhancing matrix volume. J Biol Chem 285(1):741-50.  52  Haydon PG, Carmignoto G. 2006. Astrocyte control of synaptic transmission and neurovascular coupling. Physiol Rev 86(3):1009-31. He DS, Jiang JX, Taffet SM, Burt JM. 1999. Formation of heteromeric gap junction channels by connexins 40 and 43 in vascular smooth muscle cells. Proc Natl Acad Sci U S A 96(11):6495-500. Heinemann U, Lux HD. 1977. Ceiling of stimulus induced rises in extracellular potassium concentration in the cerebral cortex of cat. Brain Res 120(2):231-49. Heinzel FR, Luo Y, Li X, Boengler K, Buechert A, Garcia-Dorado D, Di Lisa F, Schulz R, Heusch G. 2005. Impairment of diazoxide-induced formation of reactive oxygen species and loss of cardioprotection in connexin43 deficient mice. Circ Res 97(6):583-6. Heiss WD. 1992. Experimental evidence of ischemic thresholds and functional recovery. Stroke 23(11):1668-72. Herve JC, Bourmeyster N, Sarrouilhe D, Duffy HS. 2007. Gap junctional complexes: from partners to functions. 29-65 p. Hertz L, Peng L, Lai JC. 1998. Functional studies in cultured astrocytes. Methods 16(3):293-310. Hodgson C. 1998. Prevalence and disabilities of community-living seniors who report the effects of stroke. CMAJ 159(6 Suppl):S9-S14. Holthoff K, Witte OW. 2000. Directed spatial potassium redistribution in rat neocortex. Glia 29(3):288-92. Hossmann KA. 1994. Viability thresholds and the penumbra of focal ischemia. Ann Neurol 36(4):557-65. Hossmann KA. 1996. Periinfarct depolarizations. Cerebrovasc Brain Metab Rev 8(3):195-208. Hossmann KA. 2006. Pathophysiology and therapy of experimental stroke. Cell Mol Neurobiol 26(7-8):1057-83. Hossmann KA. 2008. Cerebral ischemia: models, methods and outcomes. Neuropharmacology 55(3):257-70. Heart & Stroke Foundation of British Columbia & Yukon. 2005. British Colubmia Stroke Strategy. Report. http://www.canadianstrokestrategy.ca/eng/provincial/documents/BCStrokeStrategyFinal_ OCT07.pdf. Heart and Stroke Foundation of Canada. 2009. Stroke. Webpage. http://www.heartandstroke.com/site/c.ikIQLcMWJtE/b.3483933/k.CD67/Stroke.htm?src =home. Cited Dec. 1, 2009. Iacobas DA, Urban-Maldonado M, Iacobas S, Scemes E, Spray DC. 2003. Array analysis of gene expression in connexin43 null astrocytes. Physiol Genomics 15(3):177-90. 53  Kang J, Kang N, Lovatt D, Torres A, Zhao Z, Lin J, Nedergaard M. 2008. Connexin43 hemichannels are permeable to ATP. J Neurosci 28(18):4702-11. Kettenmann H, Verkhratsky A. 2008. Neuroglia: the 150 years after. Trends Neurosci 31(12):653-9. Kim DY, Kam Y, Koo SK, Joe CO. 1999. Gating connexin43 channels reconstituted in lipid vesicles by mitogen-activated protein kinase phosphorylation. J Biol Chem 274(9):55817. Kofuji P, Newman EA. 2004. Potassium buffering in the central nervous system. Neuroscience 129(4):1045-56. Kolikova J, Afzalov R, Giniatullina A, Surin A, Giniatullin R, Khiroug L. 2006. Calciumdependent trapping of mitochondria near plasma membrane in stimulated astrocytes. Brain Cell Biol 35(1):75-86. Kolliker, A. 1889. Handbuch der Gewebelehre des Menschen, Wilhelm Engelmann. Koulakoff A, Ezan P, Giaume C. 2008. Neurons control the expression of connexin30 and connexin43 in mouse cortical astrocytes. Glia 56(12):1299-311. Kowaltowski AJ, Cosso RG, Campos CB, Fiskum G. 2002. Effect of Bcl-2 overexpression on mitochondrial structure and function. J Biol Chem 277(45):42802-7. Kowaltowski AJ, Seetharaman S, Paucek P, Garlid KD. 2001. Bioenergetic consequences of opening the ATP-sensitive K+ channel of heart mitochondria. Am J Physiol Heart Circ Physiol 280(2):H649-57. Kozoriz MG, Bates DC, Bond SR, Lai CP, Moniz DM. 2006. Passing potassium with and without gap junctions. J Neurosci 26(31):8023-4. Kressin K, Kuprijanova E, Jabs R, Seifert G, Steinhauser C. 1995. Developmental regulation of Na+ and K+ conductances in glial cells of mouse hippocampal brain slices. Glia 15(2):173-87. Kwak BR, Hermans MM, De Jonge HR, Lohmann SM, Jongsma HJ, Chanson M. 1995. Differential regulation of distinct types of gap junction channels by similar phosphorylating conditions. Mol Biol Cell 6(12):1707-19. Kwak BR, Jongsma HJ. 1996. Regulation of cardiac gap junction channel permeability and conductance by several phosphorylating conditions. Mol Cell Biochem 157(1-2):93-9. Laird DW. 2006. Life cycle of connexins in health and disease. Biochem J 394(Pt 3):527-43. Laird DW. 2010. The gap junction proteome and its relationship to disease. Trends Cell Biol. 20(2): 92-101.  54  Lalancette-Hebert M, Gowing G, Simard A, Weng YC, Kriz J. 2007. Selective ablation of proliferating microglial cells exacerbates ischemic injury in the brain. J Neurosci 27(10):2596-605. Lampe PD, Lau AF. 2000. Regulation of gap junctions by phosphorylation of connexins. Arch Biochem Biophys 384(2):205-15. Lampe PD, Lau AF. 2004. The effects of connexin phosphorylation on gap junctional communication. Int J Biochem Cell Biol 36(7):1171-86. Lampe PD, TenBroek EM, Burt JM, Kurata WE, Johnson RG, Lau AF. 2000. Phosphorylation of connexin43 on serine368 by protein kinase C regulates gap junctional communication. J Cell Biol 149(7):1503-12. Lee CH, Poburko D, Kuo KH, Seow CY, van Breemen C. 2002. Ca2+ oscillations, gradients, and homeostasis in vascular smooth muscle. Am J Physiol Heart Circ Physiol 282(5):H157183. Leis JA, Bekar LK, Walz W. 2005. Potassium homeostasis in the ischemic brain. Glia 50(4):40716. Li G, Whittaker P, Yao M, Kloner RA, Przyklenk K. 2002a. The gap junction uncoupler heptanol abrogates infarct size reduction with preconditioning in mouse hearts. Cardiovasc Pathol 11(3):158-65. Li H, Brodsky S, Kumari S, Valiunas V, Brink P, Kaide J, Nasjletti A, Goligorsky MS. 2002b. Paradoxical overexpression and translocation of connexin43 in homocysteine-treated endothelial cells. Am J Physiol Heart Circ Physiol 282(6):H2124-33. Li X, Su V, Kurata WE, Jin C, Lau AF. 2008. A novel connexin43-interacting protein, CIP75, which belongs to the UbL-UBA protein family, regulates the turnover of connexin43. J Biol Chem 283(9):5748-59. Lian XY, Stringer JL. 2004. Astrocytes contribute to regulation of extracellular calcium and potassium in the rat cerebral cortex during spreading depression. Brain Res 1012(12):177-84. Lin JH, Lou N, Kang N, Takano T, Hu F, Han X, Xu Q, Lovatt D, Torres A, Willecke K and others. 2008. A central role of connexin43 in hypoxic preconditioning. J Neurosci 28(3):681-95. Lin JH, Weigel H, Cotrina ML, Liu S, Bueno E, Hansen AJ, Hansen TW, Goldman S, Nedergaard M. 1998. Gap-junction-mediated propagation and amplification of cell injury. Nat Neurosci 1(6):494-500. Lin JH, Yang J, Liu S, Takano T, Wang X, Gao Q, Willecke K, Nedergaard M. 2003. Connexin mediates gap junction-independent resistance to cellular injury. J Neurosci 23(2):430-41.  55  Lindsay P, Bayley M, McDonald A, Graham ID, Warner G, Phillips S. 2008. Toward a more effective approach to stroke: Canadian Best Practice Recommendations for Stroke Care. CMAJ 178(11):1418-25. Lipton P. 1999. Ischemic cell death in brain neurons. Physiol Rev 79(4):1431-568. Lipton SA, Nicotera P. 1998. Calcium, free radicals and excitotoxins in neuronal apoptosis. Cell Calcium 23(2-3):165-71. Liu S, Taffet S, Stoner L, Delmar M, Vallano ML, Jalife J. 1993. A structural basis for the unequal sensitivity of the major cardiac and liver gap junctions to intracellular acidification: the carboxyl tail length. Biophys J 64(5):1422-33. Liu X, Bolteus AJ, Balkin DM, Henschel O, Bordey A. 2006. GFAP-expressing cells in the postnatal subventricular zone display a unique glial phenotype intermediate between radial glia and astrocytes. Glia 54(5):394-410. Ljubkovic M, Marinovic J, Fuchs A, Bosnjak ZJ, Bienengraeber M. 2006. Targeted expression of Kir6.2 in mitochondria confers protection against hypoxic stress. J Physiol 577(Pt 1):17-29. Maass K, Chase SE, Lin X, Delmar M. 2009. Cx43 CT domain influences infarct size and susceptibility to ventricular tachyarrhythmias in acute myocardial infarction. Cardiovasc Res 84(3):361-7. Maass K, Ghanem A, Kim JS, Saathoff M, Urschel S, Kirfel G, Grummer R, Kretz M, Lewalter T, Tiemann K and others. 2004. Defective epidermal barrier in neonatal mice lacking the C-terminal region of connexin43. Mol Biol Cell 15(10):4597-608. Maass K, Shibayama J, Chase SE, Willecke K, Delmar M. 2007. C-terminal truncation of connexin43 changes number, size, and localization of cardiac gap junction plaques. Circ Res 101(12):1283-91. Malli R, Frieden M, Osibow K, Zoratti C, Mayer M, Demaurex N, Graier WF. 2003. Sustained Ca2+ transfer across mitochondria is Essential for mitochondrial Ca2+ buffering, soreoperated Ca2+ entry, and Ca2+ store refilling. J Biol Chem 278(45):44769-79. Martinez AD, Acuna R, Figueroa V, Maripillan J, Nicholson B. 2009. Gap-junction channels dysfunction in deafness and hearing loss. Antioxid Redox Signal 11(2):309-22. Martinez AD, Hayrapetyan V, Moreno AP, Beyer EC. 2003. A carboxyl terminal domain of connexin43 is critical for gap junction plaque formation but not for homo- or heterooligomerization. Cell Commun Adhes 10(4-6):323-8. Martins-Ferreira H, Nedergaard M, Nicholson C. 2000. Perspectives on spreading depression. Brain Res Brain Res Rev 32(1):215-34. Matchett GA, Allard MW, Martin RD, Zhang JH. 2009. Neuroprotective effect of volatile anesthetic agents: molecular mechanisms. Neurol Res 31(2):128-34. 56  McHenry LC, Jr. 1981. A history of stroke. Int J Neurol 15(3-4):314-26. Meeks JP, Mennerick S. 2007. Astrocyte membrane responses and potassium accumulation during neuronal activity. Hippocampus 17(11):1100-1108. Mennecier G, Derangeon M, Coronas V, Herve JC, Mesnil M. 2008. Aberrant expression and localization of connexin43 and connexin30 in a rat glioma cell line. Mol Carcinog 47(5):391-401. Metea MR, Kofuji P, Newman EA. 2007. Neurovascular coupling is not mediated by potassium siphoning from glial cells. J Neurosci 27(10):2468-71. Mies G, Iijima T, Hossmann KA. 1993. Correlation between peri-infarct DC shifts and ischaemic neuronal damage in rat. Neuroreport 4(6):709-11. Miro-Casas E, Ruiz-Meana M, Agullo E, Stahlhofen S, Rodriguez-Sinovas A, Cabestrero A, Jorge I, Torre I, Vazquez J, Boengler K and others. 2009. Connexin43 in Cardiomyocyte Mitochondria Contributes to Mitochondrial Potassium Uptake. Cardiovasc Res 83(4):747-56. Moore R, Mao Y, Zhang J, Clarke K. 1997. Economic burden of illness in Canada, 1993. Executive summary and recommendations. Chronic Dis Can 18(2):95-6. Moreno AP, Chanson M, Elenes S, Anumonwo J, Scerri I, Gu H, Taffet SM, Delmar M. 2002. Role of the carboxyl terminal of connexin43 in transjunctional fast voltage gating. Circ Res 90(4):450-7. Muller T, Moller T, Neuhaus J, Kettenmann H. 1996. Electrical coupling among Bergmann glial cells and its modulation by glutamate receptor activation. Glia 17(4):274-84. Murphy E, Steenbergen C. 2008. Mechanisms underlying acute protection from cardiac ischemia-reperfusion injury. Physiol Rev 88(2):581-609. Murray CJ, Lopez AD. 1997. Mortality by cause for eight regions of the world: Global Burden of Disease Study. Lancet 349(9061):1269-76. Murry CE, Jennings RB, Reimer KA. 1986. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation 74(5):1124-36. Musil LS, Cunningham BA, Edelman GM, Goodenough DA. 1990. Differential phosphorylation of the gap junction protein connexin43 in junctional communication-competent and deficient cell lines. J Cell Biol 111(5 Pt 1):2077-88. Nagy JI, Dudek FE, Rash JE. 2004. Update on connexins and gap junctions in neurons and glia in the mammalian nervous system. Brain Res Brain Res Rev 47(1-3):191-215. Nagy JI, Li X, Rempel J, Stelmack G, Patel D, Staines WA, Yasumura T, Rash JE. 2001. Connexin26 in adult rodent central nervous system: demonstration at astrocytic gap junctions and colocalization with connexin30 and connexin43. J Comp Neurol 441(4):302-23. 57  Nakase T, Fushiki S, Naus CC. 2003a. Astrocytic gap junctions composed of connexin43 reduce apoptotic neuronal damage in cerebral ischemia. Stroke 34(8):1987-93. Nakase T, Fushiki S, Sohl G, Theis M, Willecke K, Naus CC. 2003b. Neuroprotective role of astrocytic gap junctions in ischemic stroke. Cell Commun Adhes 10(4-6):413-7. Nakase T, Sohl G, Theis M, Willecke K, Naus CC. 2004. Increased apoptosis and inflammation after focal brain ischemia in mice lacking connexin43 in astrocytes. Am J Pathol 164(6):2067-75. Nakase T, Yoshida Y, Nagata K. 2006. Enhanced connexin43 immunoreactivity in penumbral areas in the human brain following ischemia. Glia 54(5):369-75. Narahashi T, Aistrup GL, Lindstrom JM, Marszalec W, Nagata K, Wang F, Yeh JZ. 1998. Ion channel modulation as the basis for general anesthesia. Toxicol Lett 100-101:185-91. National Institute of Neurological Disorders. 1995. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med 333(24):1581-7. Naus CC. 2002. Gap junctions and tumour progression. Can J Physiol Pharmacol 80(2):136-41. Nedergaard M, Astrup J. 1986. Infarct rim: effect of hyperglycemia on direct current potential and [14C]2-deoxyglucose phosphorylation. J Cereb Blood Flow Metab 6(5):607-15. Nedergaard M, Cooper AJ, Goldman SA. 1995. Gap junctions are required for the propagation of spreading depression. J Neurobiol 28(4):433-44. Nedergaard M, Dirnagl U. 2005. Role of glial cells in cerebral ischemia. Glia 50(4):281-6. Nedergaard M, Hansen AJ. 1993. Characterization of cortical depolarizations evoked in focal cerebral ischemia. J Cereb Blood Flow Metab 13(4):568-74. Newman EA. 1993. Inward-rectifying potassium channels in retinal glial (Muller) cells. J Neurosci 13(8):3333-45. Nicholls DG, Chalmers S. 2004. The integration of mitochondrial calcium transport and storage. J Bioenerg Biomembr 36(4):277-81. Nimmerjahn A, Kirchhoff F, Helmchen F. 2005. Resting microglial cells are highly dynamic surveillants of brain parenchyma in vivo. Science 308(5726):1314-8. Nishiyama A, Komitova M, Suzuki R, Zhu X. 2009. Polydendrocytes (NG2 cells): multifunctional cells with lineage plasticity. Nat Rev Neurosci 10(1):9-22. Nodin C, Nilsson M, Blomstrand F. 2005. Gap junction blockage limits intercellular spreading of astrocytic apoptosis induced by metabolic depression. J Neurochem 94(4):1111-23. O'Collins VE, Macleod MR, Donnan GA, Horky LL, van der Worp BH, Howells DW. 2006. 1,026 experimental treatments in acute stroke. Ann Neurol 59(3):467-77. 58  Oguro K, Jover T, Tanaka H, Lin Y, Kojima T, Oguro N, Grooms SY, Bennett MV, Zukin RS. 2001. Global ischemia-induced increases in the gap junctional proteins connexin 32 (Cx32) and Cx36 in hippocampus and enhanced vulnerability of Cx32 knock-out mice. J Neurosci 21(19):7534-42. Orkand RK, Nicholls JG, Kuffler SW. 1966. Effect of nerve impulses on the membrane potential of glial cells in the central nervous system of amphibia. J Neurophysiol 29(4):788-806. O'Rourke B. 2007. Mitochondrial ion channels. Annu Rev Physiol 69:19-49. Ovalle WK, Jr. 1971. Fine structure of rat intrafusal muscle fibers. The polar region. J Cell Biol 51(1):83-103. Ozog MA, Siushansian R, Naus CC. 2002. Blocked gap junctional coupling increases glutamateinduced neurotoxicity in neuron-astrocyte co-cultures. J Neuropathol Exp Neurol 61(2):132-41. Pain T, Yang XM, Critz SD, Yue Y, Nakano A, Liu GS, Heusch G, Cohen MV, Downey JM. 2000. Opening of mitochondrial KATP channels triggers the preconditioned state by generating free radicals. Circ Res 87(6):460-6. Palacios-Prado N, Bukauskas FF. 2009. Heterotypic gap junction channels as voltage-sensitive valves for intercellular signaling. Proc Natl Acad Sci 106(35):14855-60. Palty R, Silverman WF, Hershfinkel M, Caporale T, Sensi SL, Parnis J, Nolte C, Fishman D, Shoshan-Barmatz V, Herrmann S and others. NCLX is an essential component of mitochondrial Na+/Ca2+ exchange. Proc Natl Acad Sci 107(1):436-41. Panchin Y, Kelmanson I, Matz M, Lukyanov K, Usman N, Lukyanov S. 2000. A ubiquitous family of putative gap junction molecules. Curr Biol 10(13):R473-4. Paschen W, Mies G, Hossmann KA. 1992. Threshold relationship between cerebral blood flow, glucose utilization, and energy metabolites during development of stroke in gerbils. Exp Neurol 117(3):325-33. Paul DL. 1986. Molecular cloning of cDNA for rat liver gap junction protein. J Cell Biol 103(1):123-34. Paznekas WA, Boyadjiev SA, Shapiro RE, Daniels O, Wollnik B, Keegan CE, Innis JW, Dinulos MB, Christian C, Hannibal MC and others. 2003. Connexin43 (GJA1) mutations cause the pleiotropic phenotype of oculodentodigital dysplasia. Am J Hum Genet 72(2):408-18. Pekny M, Nilsson M. 2005. Astrocyte activation and reactive gliosis. Glia 50(4):427-34. Perea G, Navarrete M, Araque A. 2009. Tripartite synapses: astrocytes process and control synaptic information. Trends Neurosci 32(8):421-31. Perez Velazquez JL, Kokarovtseva L, Sarbaziha R, Jeyapalan Z, Leshchenko Y. 2006. Role of gap junctional coupling in astrocytic networks in the determination of global ischaemiainduced oxidative stress and hippocampal damage. Eur J Neurosci 23(1):1-10. 59  Pivovarova NB, Hongpaisan J, Andrews SB, Friel DD. 1999. Depolarization-induced mitochondrial Ca accumulation in sympathetic neurons: spatial and temporal characteristics. J Neurosci 19(15):6372-84. Pizzo P, Pozzan T. 2007. Mitochondria-endoplasmic reticulum choreography: structure and signaling dynamics. Trends Cell Biol 17(10):511-7. Plotkin LI, Bellido T. 2001. Bisphosphonate-induced, hemichannel-mediated, anti-apoptosis through the Src/ERK pathway: a gap junction-independent action of connexin43. Cell Commun Adhes 8(4-6):377-82. Poburko D, Lee CH, van Breemen C. 2004. Vascular smooth muscle mitochondria at the cross roads of Ca2+ regulation. Cell Calcium 35(6):509-21. Poburko D, Liao CH, van Breemen C, Demaurex N. 2009. Mitochondrial regulation of sarcoplasmic reticulum Ca2+ content in vascular smooth muscle cells. Circ Res 104(1):104-12. Rami A, Volkmann T, Winckler J. 2001. Effective reduction of neuronal death by inhibiting gap junctional intercellular communication in a rodent model of global transient cerebral ischemia. Exp Neurol 170(2):297-304. Ransom CB, Sontheimer H. 1995. Biophysical and pharmacological characterization of inwardly rectifying K+ currents in rat spinal cord astrocytes. J Neurophysiol 73(1):333-46. Rausche G, Igelmund P, Heinemann U. 1990. Effects of changes in extracellular potassium, magnesium and calcium concentration on synaptic transmission in area CA1 and the dentate gyrus of rat hippocampal slices. Pflugers Arch 415(5):588-93. Rawanduzy A, Hansen A, Hansen TW, Nedergaard M. 1997. Effective reduction of infarct volume by gap junction blockade in a rodent model of stroke. J Neurosurg 87(6):916-20. Rehncrona S, Kagstrom E. 1983. Tissue lactic acidosis and ischemic brain damage. Am J Emerg Med 1(2):168-74. Retamal MA, Schalper KA, Shoji KF, Bennett MV, Saez JC. 2007. Opening of connexin43 hemichannels is increased by lowering intracellular redox potential. Proc Natl Acad Sci 104(20):8322-7. Reyes RC, Parpura V. 2008. Mitochondria modulate Ca2+-dependent glutamate release from rat cortical astrocytes. J Neurosci 28(39):9682-91. Ribeiro Mde C, Hirt L, Bogousslavsky J, Regli L, Badaut J. 2006. Time course of aquaporin expression after transient focal cerebral ischemia in mice. J Neurosci Res 83(7):1231-40. Ridet JL, Malhotra SK, Privat A, Gage FH. 1997. Reactive astrocytes: cellular and molecular cues to biological function. Trends Neurosci 20(12):570-7. Rizzuto R, Pozzan T. 2006. Microdomains of intracellular Ca2+: molecular determinants and functional consequences. Physiol Rev 86(1):369-408. 60  Rodriguez-Sinovas A, Boengler K, Cabestrero A, Gres P, Morente M, Ruiz-Meana M, Konietzka I, Miro E, Totzeck A, Heusch G and others. 2006. Translocation of connexin43 to the inner mitochondrial membrane of cardiomyocytes through the heat shock protein 90dependent TOM pathway and its importance for cardioprotection. Circ Res 99:93-101. Rodriguez-Sinovas A, Cabestrero A, Lopez D, Torre I, Morente M, Abellan A, Miro E, RuizMeana M, Garcia-Dorado D. 2007. The modulatory effects of connexin43 on cell death/survival beyond cell coupling. Prog Biophys Mol Biol 94(1-2):219-32. Rossi DJ, Brady JD, Mohr C. 2007. Astrocyte metabolism and signaling during brain ischemia. Nat Neurosci 10(11):1377-86. Rouach N, Koulakoff A, Abudara V, Willecke K, Giaume C. 2008. Astroglial metabolic networks sustain hippocampal synaptic transmission. Science 322(5907):1551-5. Rubin JB, Verselis VK, Bennett MV, Bargiello TA. 1992. Molecular analysis of voltage dependence of heterotypic gap junctions formed by connexins26 and 32. Biophys J 62(1):183-93; discussion 193-5. Ruiz-Meana M, Rodriguez-Sinovas A, Cabestrero A, Boengler K, Heusch G, Garcia-Dorado D. 2008. Mitochondrial connexin43 as a new player in the pathophysiology of myocardial ischaemia-reperfusion injury. Cardiovasc Res 77(2):325-33. Saez JC, Retamal MA, Basilio D, Bukauskas FF, Bennett MV. 2005. Connexin-based gap junction hemichannels: gating mechanisms. Biochim Biophys Acta 1711(2):215-24. Salameh A, Dhein S. 2005. Pharmacology of gap junctions. New pharmacological targets for treatment of arrhythmia, seizure and cancer? Biochim Biophys Acta 1719(1-2):36-58. Scemes E, Giaume C. 2006. Astrocyte calcium waves: what they are and what they do. Glia 54(7):716-25. Schlue WR, Wuttke W. 1983. Potassium activity in leech neuropile glial cells changes with external potassium concentration. Brain Res 270(2):368-72. Schuier FJ, Hossmann KA. 1980. Experimental brain infarcts in cats. II. Ischemic brain edema. Stroke 11(6):593-601. Schwanke U, Konietzka I, Duschin A, Li X, Schulz R, Heusch G. 2002. No ischemic preconditioning in heterozygous connexin43-deficient mice. Am J Physiol Heart Circ Physiol 283(4):H1740-2. Schwanke U, Li X, Schulz R, Heusch G. 2003. No ischemic preconditioning in heterozygous connexin43-deficient mice--a further in vivo study. Basic Res Cardiol 98(3):181-2. Seifert G, Huttmann K, Binder DK, Hartmann C, Wyczynski A, Neusch C, Steinhauser C. 2009. Analysis of astroglial K+ channel expression in the developing hippocampus reveals a predominant role of the Kir4.1 subunit. J Neurosci 29(23):7474-88.  61  Shimada N, Graf R, Rosner G, Heiss WD. 1993. Ischemia-induced accumulation of extracellular amino acids in cerebral cortex, white matter, and cerebrospinal fluid. J Neurochem 60(1):66-71. Sick TJ, Feng ZC, Rosenthal M. 1998. Spatial stability of extracellular potassium ion and blood flow distribution in rat cerebral cortex after permanent middle cerebral artery occlusion. J Cereb Blood Flow Metab 18(10):1114-20. Siesjo BK, Elmer E, Janelidze S, Keep M, Kristian T, Ouyang YB, Uchino H. 1999. Role and mechanisms of secondary mitochondrial failure. Acta Neurochir Suppl 73:7-13. Silver IA, Deas J, Erecinska M. 1997. Ion homeostasis in brain cells: differences in intracellular ion responses to energy limitation between cultured neurons and glial cells. Neuroscience 78(2):589-601. Simard M, Nedergaard M. 2004. The neurobiology of glia in the context of water and ion homeostasis. Neuroscience 129(4):877-96. Simburger E, Stang A, Kremer M, Dermietzel R. 1997. Expression of connexin43 mRNA in adult rodent brain. Histochem Cell Biol 107(2):127-37. Sin WC, Tse M, Planque N, Perbal B, Lampe PD, Naus CC. 2009. Matricellular protein CCN3 (NOV) regulates actin cytoskeleton reorganization. J Biol Chem 284(43):29935-44. Siushansian R, Bechberger JF, Cechetto DF, Hachinski VC, Naus CC. 2001. Connexin43 null mutation increases infarct size after stroke. J Comp Neurol 440(4):387-94. Sofroniew MV. 2009. Molecular dissection of reactive astrogliosis and glial scar formation. Trends Neurosci 32(12):638-47. Sofroniew MV, Vinters HV. 2010. Astrocytes: biology and pathology. Acta Neuropathol 119(1):7-35. Sohl G, Willecke K. 2003. An update on connexin genes and their nomenclature in mouse and man. Cell Commun Adhes 10(4-6):173-80. Solan JL, Lampe PD. 2009. Connexin43 phosphorylation: structural changes and biological effects. Biochem J 419(2):261-72. Somjen GG. 2002. Ion regulation in the brain: implications for pathophysiology. Neuroscientist 8(3):254-67. Somjen GG, Muller M. 2000. Potassium-induced enhancement of persistent inward current in hippocampal neurons in isolation and in tissue slices. Brain Res 885(1):102-10. Spray DC, Ye ZC, Ransom BR. 2006. Functional connexin "hemichannels": a critical appraisal. Glia 54(7):758-73.  62  Stergiopoulos K, Alvarado JL, Mastroianni M, Ek-Vitorin JF, Taffet SM, Delmar M. 1999. Hetero-domain interactions as a mechanism for the regulation of connexin channels. Circ Res 84(10):1144-55. Stout CE, Costantin JL, Naus CC, Charles AC. 2002. Intercellular calcium signaling in astrocytes via ATP release through connexin hemichannels. J Biol Chem 277(12):104828. Stridh MH, Tranberg M, Weber SG, Blomstrand F, Sandberg M. 2008. Stimulated efflux of amino acids and glutathione from cultured hippocampal slices by omission of extracellular calcium: likely involvement of connexin hemichannels. J Biol Chem 283(16):10347-56. Struijs JN, van Genugten ML, Evers SM, Ament AJ, Baan CA, van den Bos GA. 2005. Modeling the future burden of stroke in The Netherlands: impact of aging, smoking, and hypertension. Stroke 36(8):1648-55. Stys PK, Lehning E, Saubermann AJ, LoPachin RM, Jr. 1997. Intracellular concentrations of major ions in rat myelinated axons and glia: calculations based on electron probe X-ray microanalyses. J Neurochem 68(5):1920-8. Swanson RA, Farrell K, Stein BA. 1997. Astrocyte energetics, function, and death under conditions of incomplete ischemia: a mechanism of glial death in the penumbra. Glia 21(1):142-53. Swanson RA, Ying W, Kauppinen TM. 2004. Astrocyte influences on ischemic neuronal death. Curr Mol Med 4(2):193-205. Symon L, Branston NM, Strong AJ, Hope TD. 1977. The concepts of thresholds of ischaemia in relation to brain structure and function. J Clin Pathol Suppl (R Coll Pathol) 11:149-54. Szabo I, Bock J, Jekle A, Soddemann M, Adams C, Lang F, Zoratti M, Gulbins E. 2005. A novel potassium channel in lymphocyte mitochondria. J Biol Chem 280(13):12790-8. Tagaya M, Haring HP, Stuiver I, Wagner S, Abumiya T, Lucero J, Lee P, Copeland BR, Seiffert D, del Zoppo GJ. 2001. Rapid loss of microvascular integrin expression during focal brain ischemia reflects neuron injury. J Cereb Blood Flow Metab 21(7):835-46. Takano T, Oberheim N, Cotrina ML, Nedergaard M. 2009. Astrocytes and ischemic injury. Stroke 40(3 Suppl):S8-12. Taylor TN, Davis PH, Torner JC, Holmes J, Meyer JW, Jacobson MF. 1996. Lifetime cost of stroke in the United States. Stroke 27(9):1459-66. Theis M, Jauch R, Zhuo L, Speidel D, Wallraff A, Doring B, Frisch C, Sohl G, Teubner B, Euwens C and others. 2003. Accelerated hippocampal spreading depression and enhanced locomotory activity in mice with astrocyte-directed inactivation of connexin43. J Neurosci 23(3):766-76.  63  Thompson JE. 1996. The evolution of surgery for the treatment and prevention of stroke. The Willis Lecture. Stroke 27(8):1427-34. Thompson RJ, MacVicar BA. 2008. Connexin and pannexin hemichannels of neurons and astrocytes. Channels 2(2):81-6. Thompson RJ, Zhou N, MacVicar BA. 2006. Ischemia opens neuronal gap junction hemichannels. Science 312(5775):924-7. Traystman RJ. 2003. Animal models of focal and global cerebral ischemia. Ilar J 44(2):85-95. van der Worp HB, van Gijn J. 2007. Clinical practice. Acute ischemic stroke. N Engl J Med 357(6):572-9. Von Lenhossek, M. 1893. Der feinere Bau des Nervensystems im Lichte neuester Forschung, Fischer’s Medicinische Buchhandlung H.Kornfield. Wallraff A, Kohling R, Heinemann U, Theis M, Willecke K, Steinhauser C. 2006. The impact of astrocytic gap junctional coupling on potassium buffering in the hippocampus. J Neurosci 26(20):5438-47. Wallraff A, Odermatt B, Willecke K, Steinhauser C. 2004. Distinct types of astroglial cells in the hippocampus differ in gap junction coupling. Glia 48(1):36-43. Walz W. 1987. Swelling and potassium uptake in cultured astrocytes. Can J Physiol Pharmacol 65(5):1051-7. Walz W. 1992. Mechanism of rapid K+-induced swelling of mouse astrocytes. Neurosci Lett 135(2):243-6. Walz W. 2000. Role of astrocytes in the clearance of excess extracellular potassium. Neurochem Int 36(4-5):291-300. Walz W, Wuttke WA. 1999. Independent mechanisms of potassium clearance by astrocytes in gliotic tissue. J Neurosci Res 56(6):595-603. Wang HZ, Veenstra RD. 1997. Monovalent ion selectivity sequences of the rat connexin43 gap junction channel. J Gen Physiol 109(4):491-507. Warner DS, Ludwig PS, Pearlstein R, Brinkhous AD. 1995. Halothane reduces focal ischemic injury in the rat when brain temperature is controlled. Anesthesiology 82(5):1237-45; discussion 27A. Westergaard N, Sonnewald U, Schousboe A. 1995. Metabolic trafficking between neurons and astrocytes: the glutamate/glutamine cycle revisited. Dev Neurosci 17(4):203-11. Wolf PA, D'Agostino RB, Belanger AJ, Kannel WB. 1991. Probability of stroke: a risk profile from the Framingham Study. Stroke 22(3):312-8. 64  Wozniak MA, Kittner SJ. 2002. Return to work after ischemic stroke: a methodological review. Neuroepidemiology 21(4):159-66. Xu L, Zeng LH, Wong M. 2009. Impaired astrocytic gap junction coupling and potassium buffering in a mouse model of tuberous sclerosis complex. Neurobiol Dis 34(2):291-9. Yamamoto T, Ochalski A, Hertzberg EL, Nagy JI. 1990a. LM and EM immunolocalization of the gap junctional protein connexin43 in rat brain. Brain Res 508(2):313-9. Yamamoto T, Ochalski A, Hertzberg EL, Nagy JI. 1990b. On the organization of astrocytic gap junctions in rat brain as suggested by LM and EM immunohistochemistry of connexin43 expression. J Comp Neurol 302(4):853-83. Yancey SB, John SA, Lal R, Austin BJ, Revel JP. 1989. The 43-kD polypeptide of heart gap junctions: immunolocalization, topology, and functional domains. J Cell Biol 108(6):2241-54. Ye ZC, Wyeth MS, Baltan-Tekkok S, Ransom BR. 2003. Functional hemichannels in astrocytes: a novel mechanism of glutamate release. J Neurosci 23(9):3588-96. Zahs KR, Kofuji P, Meier C, Dermietzel R. 2003. Connexin immunoreactivity in glial cells of the rat retina. J Comp Neurol 455(4):531-46. Zhang YW, Morita I, Yao XS, Murota S. 1999. Pretreatment with eicosapentaenoic acid prevented hypoxia/reoxygenation-induced abnormality in endothelial gap junctional intercellular communication through inhibiting the tyrosine kinase activity. Prostaglandins Leukot Essent Fatty Acids 61(1):33-40. Zigmond M. 1999. Fundamental neuroscience. San Diego: Academic Press.  65  Chapter 2: Temporary sequestration of potassium by mitochondria in astrocytes 3  3  A version of this chapter has been submitted for publication. Kozoriz, MG., Church, J., Ozog, MA., Naus, CC. and Krebs C. (2009) Temporary sequestration of potassium by mitochondria in astrocytes.  66  2.1  Introduction Changes in extracellular K+ concentration ([K+]o) can both reflect and, in turn, influence  neuronal function. Normally ~3 mM under resting conditions, [K+]o can increase to >10 mM during seizure activity, >25 mM during ischemia, and >50 mM during spreading depression, with consequent effects on neuronal excitability, synaptic transmission, neurotransmitter reuptake and, ultimately, neuronal viability (reviewed by Somjen 2002; Walz 2000). Astrocytes play a critical role in regulating [K+]o under both physiological and pathological conditions (reviewed by Leis et al. 2005; Walz 2000). Astrocytes are equipped with a variety of plasmalemmal K+ uptake mechanisms, including the Na+,K+-ATPase, ion transporters (e.g. Na+K+-2Cl- co-transport), and voltage-activated K+ channels (e.g. inward rectifier (Kir), delayed rectifier (Kdr), and A-type (Ka) K+ channels). Astrocytes are also extensively coupled by gap junctions which aid in the spatial redistribution of K+ from areas of high [K+]o/i to those of low [K+]o/i (Kofuji and Newman 2004; Leis et al. 2005; Orkand et al. 1966; Wallraff et al. 2006; Walz 2000). This spatial buffering is achieved when a local increase in [K+]o causes the K+ equilibrium potential (EK) to become more positive than the membrane potential, allowing K+ influx into an astrocyte. Meanwhile, at more distant sites, where EK remains more negative than the membrane potential, K+ leaves the syncytium. Another mechanism that could limit potentially detrimental increases in astrocytic [K+]i that might otherwise occur during K+ uptake is transient sequestration into intracellular stores. Mitochondria in close association with the plasmalemmal membrane have been shown to play an important role in internal Ca2+ and Na+ sequestration in a variety of cell types (Azarias et al. 2008; Murphy and Eisner 2009; Pivovarova et al. 1999; Poburko et al.; Poburko et al. 2007; Poburko et al. 2009; Reyes and Parpura 2008) and the potential contribution of analogous mechanisms to internal K+ handling are starting to be explored. 67  Several lines of evidence are consistent with a potential role for mitochondria in K+ sequestration. First, mitochondria are intimately associated with the plasma membrane (Golovina 2005; Kolikova et al. 2006), placing them in an appropriate location to sequester K+ that enters a cell across the plasma membrane. Second, the mitochondrial inner membrane is endowed with a variety of K+ channels and transporters that contribute to the regulation of the inner mitochondrial transmembrane potential (  m),  matrix volume, and oxidative  phosphorylation (Garlid and Paucek 2003). Third, mitochondria have a negative resting membrane potential, creating a driving force for K+ uptake (Bernardi 1999). Fourth, mitochondria maintain matrix [K+] at 150-180 mM (Garlid 1996; Kowaltowski et al. 2002), a higher concentration than [K+]i (~100 mM; Walz 1992). Finally, mitochondria in cardiomyocytes have been reported to act as sinks for K+o (Korge et al. 2005). In this study we investigated the role of mitochondria in the uptake of K+o by astrocytes. To avoid the difficulties and potential artifacts associated with isolated mitochondrial preparations (e.g. see Duchen 2004; Oliveira et al. 2007) we assessed mitochondrial K+ uptake in intact astrocytes. We determined that mitochondrial KATP channels (mitoKATP) and potentially mitochondrial connexin43 (Cx43), a protein that is abundantly expressed in astrocytes and that has recently been found to contribute to K+ uptake in isolated cardiac mitochondria (Miro-Casas et al. 2009), play a role in the temporary sequestration of K+ by astrocyte mitochondria.  2.2  Materials and methods All experiments were performed in accordance with the guidelines established by the  Canadian Council on Animal Care and were approved by The University of British Columbia Animal Care Committee.  68  Cortical astrocyte cultures. Cortical astrocytes from neonatal (1 - 2 d) CD-1 wild-type mice were cultured and plated, unless otherwise indicated, onto poly-D-lysine-coated glass coverslips in 24 well plates and maintained as previously described (Ozog et al. 2002). Primary astrocytes were used between 7 - 28 days in vitro and the results obtained were not dependent on the length of time that astrocytes had been maintained in culture. For a portion of the experiments, wild-type (Cx43+/+), heterozygous (Cx43+/-), and Cx43null (Cx43-/-) astrocytes were obtained from CD57/Bl6 mice generated by crossing heterozygous mice with the Gja1 null mutation (Reaume et al. 1995) using procedures as described above. Tissue from each newborn pup was genotyped by PCR using primers specific for the wild-type Cx43 and the disrupted Cx43 gene, as previously described (Naus et al. 1997). Solutions and test compounds. The standard perfusion medium contained (in mM): 136.5 NaCl, 3 KCl, 1.5 NaH2PO4, 1.5 MgSO4, 10 D-glucose, 2 CaCl2 and 10 HEPES (titrated to pH 7.35 with 10 M NaOH). Bicarbonate-containing perfusion medium contained (in mM): 117.5 NaCl, 3 KCl, 1.5 NaH2PO4, 1.5 MgSO4, 10 D-glucose, 2 CaCl2 and 29 NaHCO3, and was equilibrated with 5% CO2 in air (pHo 7.35). Media containing high-[K+] or the K+ channel blocking cocktail (3 mM BaCl2, 5 mM 4-AP and 1 mM TEA) were prepared by equimolar substitution for NaCl. In solutions containing BaCl2, NaH2PO4 was omitted and MgSO4 was replaced with MgCl2. Test compounds (bumetanide, carbenoxolone (CBX), carbonyl cyanide mchlorophenylhydrazone (CCCP), diazoxide, carbonyl cyanide ptrifluoromethoxyphenylhydrazone (FCCP), furosemide, glyburide, 18 -glycyrrhetinic acid (AGA), glycyrrhizic acid (GZA), 5-hydroxydecanoic acid (5-HD), oligomycin, omeprazole, ouabain, quinine, saponin and timolol) were obtained from Sigma-Aldrich Canada (ON) and were applied by superfusion. Experiments were performed at room temperature (20 - 22°C) unless otherwise indicated. 69  Measurement of cytoplasmic K+ concentration. Astrocytes were placed for 1 h at room temperature in standard HEPES-buffered medium containing 5 M of the K+-sensitive fluorophore potassium-binding benzofuran isophthalate acetoxymethyl ester (PBFI) and 0.05% pluronic F-127 (Molecular Probes Inc., Eugene, OR). Cells were then placed in fluorophore-free medium for 30 min to ensure de-esterification of the fluorophore and then mounted in a perfusion chamber on a Zeiss Axioskop2 FS Plus microscope (Carl Zeiss Canada Ltd., ON). Astrocytes were continuously superfused at ~2 mL/min throughout the course of an experiment. Measurements of [K+]i with PBFI were performed using the dual excitation ratio method. Fluorescence emissions >505 nm were captured by a 12-bit digital cooled CCD camera (Retiga EXi, QImaging, Burnaby, BC) from regions of interest placed on individual astrocytes. Raw emission intensity data at each excitation wavelength (340 and 380 nm; Lambda DG-5, Sutter Instrument Co., Novato, CA) were collected every 6 s, corrected for background fluorescence, and background-subtracted ratio pairs (BI340/BI380) were analyzed off-line. Unless otherwise indicated, fluorescence signals were not detectable from the test compounds used in this study. In addition, a subset of experiments similar to those shown in Fig. 2.3A were performed in the absence of PBFI. At the camera settings used in this study no signal was detected, indicating endogenous fluorescence signals are unlikely to affect our measurements. To determine the sensitivity of PBFI to K+ in our preparation of cultured astrocytes, in situ calibration experiments were performed in which astrocytes were exposed sequentially to media containing (in mM): 1.5 NaH2PO4, 1.5 MgSO4, 10 D-glucose, 2 CaCl2, 10 HEPES, four different [K+] values (range 80 - 140 mM), LiCl (as appropriate to balance osmolarity), and 10 M gramicidin (pH 7.35 with 1 M LiOH). Additionally, sensitivity to Na+ was assessed by performing calibrations in the presence of 65 mM NaCl. Sensitivity to pH was also assessed by adjusting the pH to either 6.5 or 8 with 1 M LiOH. PBFI sensitivity to 80, 100, and 140 mM K+ 70  was also assessed in the presence of 10 M CCCP. PBFI-derived BI340/BI380 ratio values were normalized to the BI340/BI380 ratio value obtained at [K+] = 100 mM, and the data points relating [K+] to the normalized, background-subtracted PBFI ratio values were fitted (r2 >0.99) by a three-parameter hyperbolic equation (Diarra et al. 2001). Measurement of mitochondrial K+ sequestration in situ by application of CCCP. Two paradigms were employed in this study. First, mitochondrial K+ was assessed using a protocol similar to that previously utilized in the examination of mitochondrial Ca2+ buffering (Duchen 2004; Murchison and Griffith 2000; Murchison et al. 2004). The mitochondrial membrane uncoupler CCCP (10 M) was applied for 2 min to reversibly collapse  m,  which caused efflux  of mitochondrial K+ that was detected by cytoplasmic PBFI (e.g. see Fig. 2.2A), giving a measure of mitochondrial K+ content. [K+]o was then increased (or decreased) for 5 min prior to the next 2 min application of CCCP (e.g. see Fig. 2.3A,B). The difference in the magnitude of response reflected mitochondrial K+ sequestration. For analysis, BI340/BI380 values obtained under resting conditions at the start of an experiment were subtracted from the BI340/BI380 values measured during the entire experiment. Changes in [K+]i evoked by brief (2 min) exposures to CCCP were quantified as the percent difference between the peak CCCP-induced BI340/BI380 transient measured after exposure to altered [K+]o and the peak CCCP-induced BI340/BI380 transient measured in the same cells during perfusion with normal [K+]o (3 mM). In the second paradigm, 10 M CCCP was applied for an extended period of time (~15 min) to uncouple the mitochondrial membrane thereby inhibiting K+ sequestration. [K+]o was altered in the continued presence of CCCP and [K+]i was monitored by changes in cytoplasmic PBFI ratio values. This paradigm is used to reflect changes in K+ entry across the plasma membrane in the absence of mitochondrial sequestration. For analysis, changes in [K+]i observed during changes in [K+]o were quantified by subtracting the steady-state BI340/BI380 71  value measured prior to altering [K+]o from the average BI340/BI380 value obtained over the final minute of exposure to altered [K+]o (e.g. see Fig. 2.3E,F). Measurement of cytoplasmic Na+ concentration. Measurements of [Na+]i in astrocytes exposed to CCCP were performed with sodium-binding benzofuran isophthalate (SBFI), as detailed previously (Diarra et al. 2001; Sheldon et al. 2004). Measurement of mitochondrial K+. Measurements of mitochondrial K+ were performed using a PBFI loading protocol which favors mitochondrial dye localization (Ljubkovic et al. 2006; Miro-Casas et al. 2009; Xu et al. 2002). In brief, astrocytes were placed for 1 h at 37 oC in standard HEPES-buffered medium containing 10 M PBFI and 0.05% pluronic F-127. Cells were then placed in fluorophore-free medium for 30 min. To permeabilize the plasma membrane and release PBFI from the cytoplasmic compartment cells were treated with 50 g/ml saponin for 5 min (Xu et al. 2002) in a mitochondria medium containing (in mM) 250 sucrose, 25 Tris, 3 EGTA, 5 MgCl2, 5 succinate and 5 glutamate (pH 7.3) (Ljubkovic et al. 2006). Preparations which were used to assess mitochondrial K+ uptake (e.g. see Fig. 2.6) were additionally incubated in a medium containing (mM) 200 sucrose, 50 TEA, 25 Tris, 3 EGTA, 5 MgCl2, 5 succinate and 5 glutamate (pH 7.3) for 20 min to deplete mitochondrial K+ (see Miro-Casas et al. 2009). Mitochondria from the permeabilized astrocytes were subsequently imaged in the mitochondria medium. K+ uptake into mitochondria was assessed by measuring the rate of change of PBFI derived ratio values upon exposure to 25 mM K+ (solution containing (mM) 200 sucrose, 25 KCl, 25 Tris, 3 EGTA, 5 MgCl2, 5 succinate and 5 glutamate; pH 7.3) applied at a bath perfusion rate of 4.2 mL/min. Assessment of changes in mitochondrial membrane potential. Astrocytes were placed in standard HEPES-buffered medium containing 5 M rhodamine-123 (Molecular Probes Inc.) for 15 min and then washed in standard HEPES medium for 30 min. The dye was excited at 495 nm 72  and emissions were collected at >505 nm. For analysis, F/F was measured by dividing rhodamine-123-derived BI495 values measured during exposure to high [K+]o by the average of BI495 values obtained during the control portion of an experiment. ATP measurements. Intracellular ATP content was determined by luciferin-luciferase luminescence using an ATP determination kit (Molecular Probes Inc.). Astrocytes grown in 6well plates were rinsed with standard perfusion medium. Cultures were then treated with either a) standard perfusion medium, b) 10 M CCCP for 2 min followed by a 10 min wash with standard perfusion medium, or c) 10 M CCCP for 10 min. Cells were then lysed with 80 µL of 10 mM Tris buffer, pH 7.5, 0.1 mM NaCl, 1 mM EDTA, and 0.01% Triton X-100 and a protease inhibitor mixture (Roche). Aliquots (10 µL) were then obtained and added to the luciferinluciferase luminescence reaction mixture and sample luminescence was detected with a Berthold LB9507 Lumat luminometer (Fisher Scientific, ON). ATP measurements were made in duplicate. Samples were also obtained for protein concentration determination using the BCA protein quantification kit (Pierce, IL). Cellular ATP content was estimated by assuming an astrocyte volume of 4.8 L/kg protein (Olson et al. 1986). Western blot. Whole cell fractions were obtained by lysing Cx43+/+ and Cx43-/astrocytes grown on 100 mm culture dishes in radioimmune precipitation lysis buffer containing phosphatase inhibitors (Sigma-Aldrich Canada) and protease inhibitors (Roche). The cells were then scraped and the solution was drawn up with a 22 ga needle and centrifuged at 10,000 g for 10 min. Protein concentrations were determined using the BCA protein quantification kit (Pierce). Whole cell fractions were boiled for 1 min in sodium dodecyl sulfate (SDS) sample buffer. Using a mitochondria isolation kit (Pierce), mitochondria were obtained according to the manufacturer’s protocol from sister cultures of Cx43+/+ and Cx43-/- astrocytes used in the whole cell fraction preparation. The final pellet was resuspended in 100 L SDS sample buffer and 73  boiled for 1 min. Samples were probed for the plasma membrane marker n-cadherin to ensure the purity of the mitochondrial fraction. Whole cell fractions containing 10 g protein and 15 L samples of the mitochondrial fraction were separated on a 12% polyacrylamide gel in parallel with molecular weight markers (Bio-Rad Laboratories, Hercules, CA). Proteins were transferred and probed according to standard protocols (Bates et al. 2007). The following primary antibodies at the dilutions indicated were applied overnight at 4oC: 1:8000 anti-Cx43 produced in rabbit (Sigma-Aldrich; catalogue #C6219), 1:1500 anti-OxPhos Complex V subunit produced in mouse (Molecular Probes Inc.) or 1:2500 anti-n-cadherin produced in mouse (BD Transduction Laboratories, ON). Horseradish peroxidase-tagged secondary antibody (Cedarlane Laboratories, ON) was applied for 1 h at a 1:20000 dilution and subsequently detected with Super-Signal chemiluminescent substrate (Pierce). The labeled blots were exposed to x-ray film to visualize antibody binding. As necessary, the blot was stripped with Restore Western blot stripping buffer (Pierce) and reprobed with a different primary antibody. To ensure equivalent loading of protein samples the blots were stripped and probed with 1:10000 anti-GAPDH produced in rabbit (Cedarlane Laboratories). Data analysis. Results are reported as means  S.E.M with the accompanying n value  referring to the number of cell populations (i.e. number of coverslips) analyzed under each experimental condition. In the imaging experiments, data were obtained simultaneously from 10-20 astrocytes per coverslip and averaged. Unless otherwise indicated, statistical comparisons were performed using Student's unpaired t-test, with a 95% confidence limit.  2.3  Results Measurement of cytoplasmic [K+] with PBFI in cultured astrocytes. Determination of  [K+]i with PBFI may be problematic because PBFI free acid in aqueous solution exhibits a 74  limited resolution for increases in [K+]i above the physiological range (Minta and Tsien 1989). Nevertheless, incorporated inside cells, PBFI has successfully been employed to measure increases in [K+]i from high resting [K+]i values (e.g. Froschauer et al. 2005; Kasner and Ganz 1992; Liu et al. 2003), findings which suggest that the K+ affinity of PBFI may become reduced in a cellular context (as is known to be the case for the related dye SBFI, the KD of which for Na+ displays a marked increase in cells compared to the dye in free solution; Levi et al. 1994). To examine this possibility, we performed in situ calibration experiments to examine the sensitivity of PBFI for [K+] over a physiologically relevant range ([K+]i = 80 to 140 mM). As shown in Fig. 2.1, increasing [K+]i from 80 to 140 mM in 20 mM steps led to detectible increases in PBFIderived ratio values, and the dye did not appear to saturate even at 140 mM [K+]. A second potential difficulty with the use of PBFI in the present experiments is its reported affinity for Na+ ions (Minta and Tsien 1989). To assess whether changes in [Na+]i could contribute to changes in PBFI fluorescence signals under the present experimental conditions, in situ calibration experiments (see Diarra et al. 2001) were performed by exposing astrocytes to either 1.5 or 65 mM [Na+]i (the latter approximating the rise in [Na+]i following 5 min exposure to CCCP; see Fig. B.1 – Appendix B). The calibration curves obtained at 1.5 and 65 mM Na+ did not differ (Fig. 2.1B; P>0.05, two-way ANOVA), indicating that the sensitivity of PBFI to [Na+] is weak under our experimental conditions and is unlikely to affect our measurements of [K+]i. In agreement, Kasner and Ganz (Kasner and Ganz 1992) found that [Na+] failed to alter the fluorescent properties of PBFI loaded into mesangial cells until [Na+]i was >75 mM. As protonophores, such as CCCP, are known to cause small changes in pH (Buckler and Vaughan-Jones 1998) we also performed calibration experiments at pH 6.5 and pH 8. Neither alteration caused a change in PBFI sensitivity for [K+] in comparison to the standard calibration curve obtained at pH 7.35 (Fig. 2.1B; P>0.05, two-way ANOVA). In addition, CCCP did not 75  alter the sensitivity of PBFI for [K+] (Fig. 2.1B; P>0.05, two-way ANOVA). Furthermore, PBFI is insensitive to changes in [Ca2+], [Mg2+] and osmolarity at physiological levels (Minta and Tsien 1989). Taken together these observations support the use of PBFI in the present study to measure increases in [K+]i above the resting level in astrocytes. Nevertheless, as discussed by Kasner and Ganz (Kasner and Ganz 1992), the hyperbolic form of the PBFI calibration curve underscores the difficulty of determining accurate absolute values for [K+]i, especially at high concentrations of K+. In the present study, therefore, changes in [K+]i are presented as changes in PBFI-derived BI340/BI380 ratio values. Finally, it is well-established that fluorescent [ion] indicators (e.g. fura-2, SBFI) loaded as AM esters can enter internal organelles as well as the cytoplasm. To examine the compartmentalization of PBFI in astrocytes under the standard (room temperature) loading conditions employed for the majority of the experiments reported here, astrocytes were treated with 0.05% saponin to release dye located in the cytoplasm. Saponin treatment reduced PBFIderived emission intensities at each excitation wavelength by >85% (data not shown), indicating that under our standard loading conditions the majority of PBFI was situated in the cytosolic compartment. The remaining PBFI-derived fluorescence signal presumably emanates from intracellular organelles, including mitochondria. With regard to the latter, it is important to note that if mitochondrial [K+] in situ is higher than cytosolic [K+] (Kowaltowski et al. 2002; Stys et al. 1997) any contribution from PBFI localized to mitochondria to an experimentally-induced change in the total (i.e. cytosolic + organellar) PBFI fluorescence signal will be relatively small (see the PBFI calibration curve presented in Fig. 2.1B). Furthermore, since mitochondrial [K+] was decreasing in response to the majority of our experimental maneuvers (e.g. see Fig. 2.2C), this will further decrease the contribution of any PBFI localized in mitochondria to the total 76  PBFI signal. These considerations suggest that, under the standard loading conditions employed, any PBFI fluorescence derived from mitochondria will not perturb our total PBFI signal. Consequently, we conclude that experimentally-induced changes in PBFI ratio values will largely reflect changes in cytoplasmic [K+] rather than [K+] within internal organelles. CCCP increases [K+] i in cultured astrocytes. We sought to determine whether astrocytic mitochondria in situ sequester K+ by exposing astrocytes to the uncoupling agent CCCP. As outlined in the Materials and methods section two paradigms were employed. First, as illustrated in Fig. 2.2A, a 2 min application of CCCP (10 M) under control (3 mM K+o) conditions was used to induce a rapid increase in the PBFI-derived BI340/BI380 ratio value (representing mitochondrial K+ efflux). Consistent with previous observations (Buckler and Vaughan-Jones 1998), the effect of a brief application of CCCP was reversible and washout of the protonophore was followed by a rapid fall in [K+]i to below resting levels, after which [K+]i slowly returned towards baseline. The undershoot of [K+]i after CCCP washout may represent the efflux of K+ through plasma membrane K+ channels (Liu et al. 2003) and/or the re-uptake of K+ into mitochondria. Additional 2 min applications of CCCP produced rises in [K+]i, the amplitudes of which remained unchanged (one way ANOVA, P>0.05) over successive (up to 4) CCCP applications (Fig. 2.2A). The increase in cytoplasmic [K+] detected by PBFI after uncoupling the mitochondrial membrane with CCCP is unlikely to reflect K+ entry across the plasma membrane as [K+]o is low in comparison to resting [K+]i (which in astrocytes has been estimated at ~100 mM; Walz 1992). Rather, because mitochondria maintain an elevated matrix [K+] (see Introduction), the release of K+ from mitochondria may be the source of the CCCP-induced increase in cytoplasmic [K+]. To examine this possibility directly, astrocytes were loaded with PBFI using an established protocol that promotes the localization of the dye in mitochondria (see Ljubkovic et al. 2006; Xu et al. 2002; Zoeteweij et al. 1994). Astrocytes were subsequently 77  exposed to 50 g/mL saponin to release any dye located in the cytoplasm (Fig. 2.2B). In separate experiments, dye localization to mitochondria was confirmed by co-labeling with the mitochondrial marker rhodamine-123 (data not shown; see Zoeteweij et al. 1994). The release of K+ from mitochondria induced by CCCP was then examined under Na+o- and K+o-free conditions to avoid the potentially confounding effects of Na+ and/or K+ entry across the plasma membrane. As shown in Fig. 2.2C, under these conditions the mitochondrial fluorescence signal slowly declined, and a 2 min application of CCCP caused a rapid further reduction in the fluorescence signal (n = 4) indicating the release of K+ from mitochondria (Liu et al. 2003). This is consistent with our suggestion that CCCP-induced increases in cytoplasmic [K+] emanate largely through K+ efflux from mitochondria. The continued decline of the PBFI signal after CCCP withdrawal seen in Fig. 2.2C reflects the fact that the experiment was conducted in saponin-treated cells in the absence of K+o, conditions that do not favour the refilling of mitochondria with K+. Next, using the standard PBFI loading protocol, astrocytes were subjected to the second paradigm. More prolonged applications of CCCP (>15 min) at 3 mM K+o evoked an initial rise in [K+]i that was followed by a slow decline to a plateau steady-state level in the continued presence of the protonophore (Fig. 2.2D). The rise in [K+]i, as discussed above, is due to mitochondrial K+ efflux, while the gradual decline in signal is suggestive of a gradual loss of K+ to the extracellular space. This paradigm can be used to assess plasma membrane K+ entry in the absence of mitochondrial K+ sequestration, as  m  is collapsed in the presence of CCCP (e.g.  Fig. 2.3E,F). Temporary sequestration of K+ in astrocyte mitochondria. A large body of evidence obtained in multiple cell types, including astrocytes, indicates that mitochondria are able to sequester Ca2+ (and Na+) ions entering across the plasma membrane and thereby limit potentially detrimental cytosolic [Ca2+] elevations that might otherwise occur upon Ca2+ entry into a cell 78  (reviewed by Poburko et al. 2004; Rizzuto and Pozzan 2006; see also Azarias et al. 2008; Poburko et al. 2009; Reyes and Parpura 2008). Therefore, we examined whether astrocyte mitochondria might act in an analogous manner to sequester K+ entering astrocytes. In these experiments, CCCP (10 M) was applied for 2 min (at [K+]o = 3 mM) to release K+ from mitochondria and, several minutes after CCCP washout, [K+]o was adjusted to 0 - 25 mM for 5 min, after which CCCP was again applied for 2 min at 3 mM [K+]o. As illustrated in Fig. 2.3A, the increase in [K+]i evoked by CCCP following exposure to elevated [K+]o was larger than the increase in [K+]i evoked by CCCP in the absence of a prior exposure to high [K+]o. Because the amplitudes of the CCCP-induced increases in PBFI-derived ratio values varied between different astrocyte cultures, peak amplitudes of the changes in BI340/BI380 ratio values evoked by CCCP following exposure to altered [K+]o were compared relative to the peak amplitudes of the responses obtained in the same population of astrocytes under control conditions (i.e. in the absence of a prior exposure to altered [K+]o). In many experiments, the PBFI signal in response to the initial application of CCCP failed to return completely to baseline prior to the second application of the uncoupling agent. Therefore, experiments were also performed in which elevated [K+]o was applied prior to the first application of CCCP (i.e. in the absence of an undershoot) rather than the second application of CCCP (i.e. during the undershoot after the first application). As illustrated in Fig. 2.3B, the increase in [K+]i evoked by CCCP continued to be greater following the prior application of raised [K+]o. The pooled data from these experiments are shown in Fig. 2.3C and indicate that CCCP-induced rises in [K+]i were significantly (P<0.05 in all cases) greater following 5 min exposures to 6.25 - 25 mM [K+]o than under control conditions; conversely, CCCP-induced rises in [K+]i were significantly (P<0.05) smaller when CCCP was applied after a 5 min exposure to K+-free medium. Quantitatively similar results  79  were obtained with another protonophore, FCCP (Fig. 2.3C), which is also known to elicit rises in [K+]i (Liu et al. 2003). The aforementioned data were obtained at room temperature in the absence of HCO3-. Therefore, additional experiments were performed under more physiological conditions, i.e. at 37oC or in the presence of HCO3-. At 37oC, the absolute magnitudes of the CCCP-induced increases in PBFI-derived ratio values were greater than those observed at room temperature; however, the relative increase in [K+]i evoked by CCCP following exposure to 12.5 mM K+o continued to be significantly larger than the increase in [K+]i evoked by CCCP in the absence of a prior exposure to high [K+]o and was not significantly different (P>0.05) to the result obtained at room temperature (Fig. 2.3C). In the presence of HCO3- the relative increase in [K+]i evoked by CCCP following exposure to 12.5 mM K+o continued to be significantly larger than the increase in [K+]i evoked by CCCP in the absence of a prior exposure to high [K+]o (Fig. 2.3C). [As reported previously, distinct increases in resting [K+]i were observed in the presence of HCO3- upon exposure to high [K+]o (Walz 1992), whereas high [K+]o evoked only small increases in resting [K+]i under HEPES-buffered conditions (see Fig. 2.3A,B and Fig. B.2 – Appendix B). A potential explanation for these observations is that Na+-HCO3- co-transporters may provide Na+ ions for Na+,K+-ATPase-dependent K+ uptake (Rose and Ransom 1996), hence overall K+ uptake may be reduced in HEPES based recording solutions.] In light of the fact that [ATP]i is known to decline in response to CCCP application (Budd and Nicholls 1996; Juthberg and Brismar 1997), additional experiments (n = 4) were performed in the presence of 20 M oligomycin, an ATP synthase inhibitor, in order to maintain ATP levels after the brief (2 min) application of CCCP (Budd and Nicholls 1996). No change in K+ sequestration was observed using this treatment, indicating that CCCP induced changes in [ATP] do not affect mitochondrial K+ sequestration (data not shown). 80  To examine the length of time that mitochondria are apparently able to sequester K+ in response to a period of elevated [K+]o, astrocytes were exposed to 12.5 mM [K+]o for 5 min, after which [K+]o was reduced to 3 mM for 30 or 60 s immediately prior to a 2 min application of 10 M CCCP also at 3 mM K+o. As illustrated in Fig. 2.3D, the relative increase in the CCCPinduced [K+]i transient elicited by pre-exposure to high [K+]o was abolished when CCCP was applied 60 s after the end of the 5 min pretreatment with 12.5 mM K+o (n = 5 in each case). To further assess the potential role of mitochondria in sequestering K+ during exposure to elevated [K+]o, [K+]o was altered during prolonged mitochondrial uncoupling with CCCP to completely dissipate  m.  As noted previously (Fig. 2.2D), under control conditions (3 mM  K+o), prolonged applications of CCCP elicited a rise in [K+]i, which then declined to a new steady-state, elevated level in the continued presence of CCCP. When [K+]o was then increased to concentrations varying from 6.25 to 25 mM during the plateau phase of the CCCP-induced [K+]i response, there was an additional rise in [K+]i (Fig. 2.3E,F). Conversely, when [K+]o was lowered to 0 mM a reduction in steady-state [K+]i was observed. These results contrast with the failure of raised [K+]o to substantially increase [K+]i under HEPES-buffered conditions in the absence of CCCP and are consistent with a role for mitochondria in sequestration of K+. Because CCCP is known to affect [Ca2+]i and mitochondrial Ca2+ handling (Murchison et al. 2004; Park et al. 2002), experiments similar to those illustrated in Fig. 2.3A and Fig. 2.3E were performed under external Ca2+-free conditions, which have previously been shown to nearly abolish CCCP-induced increases in [Ca2+]i (Park et al. 2002). Results similar to those observed in the presence of 2 mM Ca2+o were obtained in both transient (n = 4) and prolonged (n = 4) CCCP exposure protocols, effectively ruling out an influence of Ca2+ influx on K+ uptake under our experimental conditions (data not shown).  81  Effects of blocking plasmalemmal or mitochondrial K+ uptake mechanisms. A number of mechanisms have been suggested to contribute to the uptake of K+ across the plasma membrane of astrocytes, including the Na+,K+-ATPase, Na+-K+-2Cl- co-transport and a variety of voltageactivated K+ channels (Kofuji and Newman 2004; Ljubkovic et al. 2006; Walz 2000). The entry of K+ across the mitochondrial inner membrane into mitochondria occurs via a K+ leak pathway dependent on the  m  and a K+ uniporter, mitoKATP, whereas excess K+ can be removed by a  K+/H+ antiporter (Garlid and Paucek 2003); calcium sensitive mitochondrial potassium (mitoKCa) currents have also been described (Xu et al. 2002). To determine whether these mechanisms contribute to K+o uptake into astrocyte mitochondria, the effects of a variety of inhibitors were examined on the changes in [K+]i evoked by 5 min exposures to 12.5 mM K+o during prolonged mitochondrial uncoupling with CCCP, where mitochondrial K+ sequestration is impaired, and on the [K+]i transient induced by a 2 min application of CCCP immediately after a 5 min exposure to 12.5 mM K+o. Under conditions where mitochondrial sequestration mechanisms are impaired (i.e. prolonged mitochondrial uncoupling with CCCP), the effects of inhibiting established K+ entry pathways in the plasma membrane were examined on the change in [K+]i evoked by 12.5 mM K+o (Fig. 2.4A,B). Under these conditions the high [K+]o-evoked rise in [K+]i was reduced by inhibiting Kir with 50 M Ba2+, and was effectively blocked by a K+ channel blocking cocktail containing 3 mM Ba2+, 5 mM 4-AP and 1 mM TEA to inhibit Kir, Kdr and Ka channels (Fig. 2.4A,B; P<0.05 and n = 5 in each case). Although not typically described as a K+ uptake route in astrocytes (see Discussion), KATP channels were assessed by application of the inhibitor glyburide (1 M), which did not affect K+ entry (Fig. 2.4B; P>0.05 and n = 5). The Na+,K+ATPase inhibitor ouabain (1 mM) did not significantly affect the high [K+]o-evoked rise in [K+]i observed during prolonged exposures to CCCP (Fig. 2.4B, P>0.05; n = 6). Ouabain’s lack of 82  effect in this instance (compared to its effect to reduce rises in [K+]i evoked by 2 min applications of CCCP immediately after a 5 min exposure to 12.5 mM K+o, compared to control; see below) likely reflects the profound depletion of [ATP]i that occurs during prolonged applications of CCCP (Budd and Nicholls 1996; Juthberg and Brismar 1997). In support, 10 min applications of CCCP resulted in an 83% reduction in ATP levels (Fig. 2.4E). By calculating cellular ATP content at this time point we determined that astrocyte ATP concentration was 0.036 mM, corresponding to ~10% Na+,K+-ATPase transporter activity (Jewell and Lingrel 1991), indicating ouabain would have minimal effect on inhibiting K+ uptake under prolonged application of CCCP. Unfortunately, we were unable to examine the potential contribution of plasma membrane Na+-K+-2Cl- co-transport (Kofuji and Newman 2004; Walz 2000) as the inhibitors bumetanide and furosemide resulted in a marked reduction in resting [K+]i (see also Kasner and Ganz 1992), thereby confounding the interpretation of the K+ sequestration experiments. Also, quinine compounds could not be employed to block plasma membrane tandem pore domain K+ channels (K2P) because their intrinsic absorbance and fluorescence at the excitation wavelengths employed for PBFI markedly interfered with the PBFI signals. In sum, these results indicate that Kir, Kdr and Ka channels play a role in K+ uptake across the plasma membrane during exposure to high [K+]o. Next, to examine which plasma membrane K+ uptake routes play a role in mitochondrial K+ sequestration we tested the effects of the same inhibitors by adding them 3-5 min prior to elevating [K+]o to 12.5 mM for 5 min and then applying CCCP at 3 mM K+o for 2 min (Fig. 2.4C,D). The plasma membrane H+/K+-ATPase transporter was also examined. As illustrated in Fig. 2.4D, inhibition of inwardly rectifying K+ channels with 50 M Ba2+ (n = 5) significantly (P<0.05) reduced the amplitude of the [K+]i transient evoked by CCCP immediately after a 5 min exposure to 12.5 mM K+o, compared to the control response. A K+ channel blocking cocktail 83  containing 3 mM Ba2+, 5 mM 4-AP and 1 mM TEA (to block Kir, Kdr and Ka channels) also caused a reduction in the response (n = 6; P<0.05), although the reduction was not significantly different (P>0.05) to that evoked by 50 M Ba2+ alone. Inhibition of the H+/K+-ATPase exchanger with omeprazole (100 M) did not alter mitochondrial K+ sequestration (n = 5; P>0.05). Finally, compared to control, inhibition of the Na+,K+-ATPase with 1 mM ouabain (n = 6) significantly (P<0.05 in each case) reduced the amplitude of the [K+]i transient evoked by CCCP immediately after a 5 min exposure to 12.5 mM K+o, suggesting that ATP levels during a 2 min exposure to CCCP remain sufficient to maintain Na+,K+-ATPase activity (Juthberg and Brismar 1997). Indeed, we found that after a 2 min exposure to CCCP followed by a 12 min wash, ATP levels were reduced by 43% (cf the 83% reduction evoked by a 10 min application of CCCP; Fig. 2.4E); the calculated cellular ATP concentration was 0.12 mM, corresponding to ~35% Na+,K+-ATPase transporter activity (Jewell and Lingrel 1991). These results suggest that plasma membrane Kir, Kdr , Ka and Na+,K+-ATPase pathways contribute to K+ influx across the plasma membrane and subsequent mitochondrial sequestration. In conjunction, both CCCP paradigms are consistent with previous findings that the plasmalemmal Na+,K+-ATPase and inwardly rectifying K+ channels contribute to K+ accumulation by astrocytes (Kofuji and Newman 2004; Walz 2000), and suggest that these pathways contribute to K+ influx into astrocytes and subsequent mitochondrial K+ loading at elevated [K+]o. Nevertheless, K+ sequestration by mitochondria was not entirely blocked by the test compounds used, indicating that other mechanisms must also contribute. To examine the role of mitochondrial K+ channels in the sequestration of K+ by mitochondria, a variety of inhibitors were utilized under the brief 2 min CCCP application paradigm. The KATP channel blocker 5-HD was used at a concentration that inhibits mitoKATP channels and not plasma membrane KATP channels (500 M; Sato et al. 1998). Application of 584  HD 5 min prior to and during the application of 12.5 mM [K+]o reduced (P<0.05) the amount of K+ taken up into mitochondria, as assessed by a reduction in the [K+]i transient evoked by the subsequent 2 min application of CCCP at 3 mM K+o (n = 6; Fig. 2.4D). Consistent with previous findings (Liu et al. 2003), 5-HD alone had no apparent effect on resting PBFI-derived ratio values but did slightly (P>0.05) reduce the rise in [K+]i evoked by a 2 min application of CCCP under control (K+o = 3 mM) conditions in the absence of a prior application of elevated [K+]o. Application of glyburide, which inhibits both plasma membrane KATP channels and mitochondrial mitoKATP channels (Yang et al. 2009), also reduced mitochondrial K+ sequestration (n = 5; P<0.05; Fig. 2.4D). As noted above glyburide did not act on plasma membrane KATP channels indicating mitoKATP channels are the likely site of action. Application of the mitoKATP channel opener diazoxide at a concentration specific for mitochondrial channels (100 M; Garlid et al. 1997) failed to significantly augment the amplitude of the [K+]i transient induced by a 2 min application of CCCP after pretreatment with 12.5 mM K+o (n = 6; P>0.05; Fig. 2.4D), indicating that additional mitoKATP opening is not necessary for K+ sequestration by mitochondria. The effects of the K+/H+ exchange blocker, quinine, could not be assessed due to its autofluorescence and timolol, another K+/H+ exchange inhibitor, caused a large reduction in [K+]i under resting conditions (see also McLaughlin et al. 2001) making it difficult to interpret the effect of elevated [K+]o on K+ release evoked by CCCP. Other known mitochondrial K+ channel inhibitors could not be employed due to effects on plasmalemmal K+ channels and/or transporters. For example, compounds that block mitoKCa channels (Xu et al. 2002) also block plasma membrane KCa channels (Halestrap et al. 2007). Taken together, these results are consistent with previous findings that K+ is taken up across the plasma membrane of astrocytes by several routes and suggest that mitoKATP channels are an important route for the subsequent entry of K+ into mitochondria. 85  The contribution of Cx43 to mitochondrial K+ sequestration. Cx43 is highly expressed in astrocytes (Giaume et al. 1991), is permeable to K+ (Wang and Veenstra 1997), has been shown using a variety of techniques (including electron microscopy) to be present in mitochondria (Boengler et al. 2005; Boengler et al. 2009), and has been shown to be important in K+ uptake into isolated mitochondria (Miro-Casas et al. 2009). Therefore, we examined whether Cx43 plays a role in the temporary sequestration of K+ by astrocyte mitochondria. Initially, we examined whether Cx43 was present in astrocyte mitochondria. As illustrated in Fig. 2.5, Cx43 protein was observed in whole cell and mitochondrial fractions obtained from wild-type (Cx43+/+) but not Cx43-/- mice. These observations suggest that Cx43 is indeed present in astrocyte mitochondria and are consistent with similar findings in cardiac tissue and endothelial cells (Boengler et al. 2005; Goubaeva et al. 2007; Li et al. 2002). Before examining the role of gap junction proteins in K+ sequestration with the CCCP protocols described above, their involvement in mitochondrial K+ uptake was assessed directly using a protocol which preferentially loads PBFI into mitochondria followed by saponin permeabilization (Fig. 2.6A). As previously described in cardiomyocytes (Miro-Casas et al. 2009), application of 25 mM K+ caused an increase in mitochondrial PBFI-derived BI340/BI380 ratio values (Fig. 2.6B). The rate of mitochondrial K+ uptake was reduced in the presence of the gap junction blockers CBX (20 M; Fig. 2.6B) and AGA (10 M) and in Cx43-/- astrocytes (Fig. 2.6C). Additionally, the inactive analogue GZA (20 M) was without effect, and CBX and AGA were ineffective in Cx43-/- astrocytes (Fig. 2.6C). Next, the effects of the gap junction blockers CBX and AGA, and the inactive analogue GZA, were tested on mitochondrial K+ sequestration in situ. During prolonged exposures to CCCP (where  m  is completely dissipated and mitochondrial K+ sequestration is impaired),  CBX and AGA did not inhibit the rise in [K+]i normally evoked by exposure to 12.5 mM K+o 86  (Fig. 2.7A,B) suggesting, in turn, that connexon hemichannels do not play a role in K+ uptake at the level of the plasma membrane. In contrast, CBX and AGA significantly attenuated the expected augmentation of the rise in [K+]i induced by a brief (2 min) application of CCCP immediately following exposure to 12.5 mM K+o (Fig. 2.7C,D), suggesting that Cx43 plays a role in mitochondrial K+ uptake. The inactive analogue GZA did not affect (P>0.05) the temporary sequestration of K+ by mitochondria (Fig. 2.7D). In addition, CBX and AGA were without effect in Cx43-/- astrocytes (P>0.05; Fig. 2.8B), confirming the specificity of these compounds in our system. To examine if gap junction coupling played a role in mitochondrial K+ sequestration a subset of experiments were performed on solitary astrocytes (data not shown) and results were not significantly different from those obtained in non-solitary cultures. In support of these results, the rise in [K+]i evoked by 12.5 mM K+o in the continued presence of CCCP was not significantly different in Cx43-/- astrocytes compared to astrocytes obtained from wild-type (Cx43+/+) and heterozygous (Cx43+/-) littermates (Fig. 2.8A,B). However, compared to Cx43+/+ astrocytes, cells obtained from Cx43-/- mice exhibited reduced mitochondrial K+ sequestration (P<0.05; Fig. 2.8C,D). An intermediate response, which was not significantly different from that observed in Cx43+/+ and Cx43-/- astrocytes (P>0.05 in both cases), was observed in Cx43+/- astrocytes (Fig. 2.8D). In light of the finding that connexon hemichannels do not contribute to K+ uptake across the plasma membrane (see above), the observations that mitochondrial K+ sequestration is inhibited by gap junction blockers and in Cx43-null astrocytes suggest that connexins play a role in K+ sequestration by mitochondria downstream of the plasma membrane. The above results suggest that Cx43 represents a potential pathway for K+ influx into astrocyte mitochondria. To further assess this possibility, in the final series of experiments we compared the effects of high [K+]o on mitochondrial membrane potential (  m)  in Cx43+/+ and 87  Cx43-/- astrocytes loaded with the fluorescent voltage sensitive indicator rhodamine-123. Consistent with a contribution of Cx43 to K+ uptake by astrocyte mitochondria, mitochondrial depolarization in response to the application of 12.5 mM K+o was greater in Cx43+/+ compared to Cx43-/- astrocytes (Fig. 2.9A,B).  2.4  Discussion Astrocytes have been implicated in K+ buffering for decades (D'Ambrosio et al. 1999;  Holthoff and Witte 2000; Kofuji and Newman 2004; Lian and Stringer 2004; Orkand et al. 1966; Simard and Nedergaard 2004; Walz 2000). While the mechanisms by which astrocytes take up K+o have been extensively studied, how K+ is handled once it enters the cell remains unclear. Although K+ may be redistributed throughout the astroglial syncytium, at least in part via gap junction coupling (Wallraff et al. 2006; Xu et al. 2009), the short length constants of astrocytic processes may limit K+ redistribution (Chen and Nicholson 2000) and theories of K+ siphoning through glia have been challenged (Metea et al. 2007). The results presented here support a role for mitochondria in rapidly sequestering the extracellular K+ that is taken up across the plasma membrane of astrocytes. Technical considerations. To avoid the many potential shortcomings associated with the use of isolated mitochondrial preparations (Duchen 2004; Oliveira et al. 2007), we undertook the study of mitochondria in situ. Despite the potential limitations of PBFI, we found that the dye could be employed as a reliable indicator of changes in [K+]i in intact cells. Most importantly, although PBFI free acid in solution exhibits a low KD for K+ (Minta and Tsien 1989), in agreement with Kasner and Ganz (Kasner and Ganz 1992) we found that the K+ affinity of PBFI, like that of the related Na+-sensitive fluorophore SBFI (Levi et al. 1994), is increased when the dye is loaded into cells such that increases in [K+]i above resting levels evoked reproducible 88  increases in PBFI ratio values. Nevertheless, the resolution of PBFI for changes in [K+]i above physiological resting levels (~100 mM in astrocytes) is undoubtedly reduced and it is possible that small changes in [K+]i may have remained unresolved in the present experiments. In agreement with Kasner and Ganz (Kasner and Ganz 1992) we found no evidence to suggest that changes in [Na+]i within the range observed in the present experiments affected the sensitivity of PBFI to K+ (see also Zoeteweij et al. 1994). In addition, as noted in the Results, previous reports indicate that PBFI (both in vitro and in situ) is relatively insensitive to changes in [Ca2+], [H+] and osmolarity; in situ, for example, PBFI fluorescence signals are not effected when pHi is changed in the range 6.5 - 7.5 (see Kasner and Ganz 1992). However, since CCCP is known to cause changes in pH we were prudent to examine the sensitivity of PBFI to K+ at a pH of 6.5 and at pH 8 and we found PBFI signals were not altered under these conditions. Also, changes in [Mg2+]i during our experiential manipulations are unlikely to affect PBFI because protonophores cause modest rises in [Mg2+]i well below the threshold to alter the sensitivity of PBFI (Kubota et al. 2005; Minta and Tsien 1989). Temporary K+ sequestration by mitochondria was assessed by transiently uncoupling the mitochondria with 2 min applications of CCCP before and after altering [K+]o and the difference between the magnitudes of the CCCP-induced [K+]i transients reflected mitochondrial K+ sequestration. It is unlikely that the [K+]i transients induced by brief applications of CCCP arose from extracellular K+ entry because the [K+]i transients were still observed during the application of the K+ channel blocking cocktail, a finding that is consistent with previous reports which indicate that blockade of voltage-activated K+ channels does not alter protonophore-induced changes in plasma membrane currents (Juthberg and Brismar 1997; Park et al. 2002). Despite the possibility that a CCCP-induced disruption of the mitochondrial pH gradient could potentially limit K+ efflux from mitochondria via the mitochondrial K+/H+ exchanger, the 89  application of CCCP in situ (Fig. 2.2A) and in permeabilized astrocytes (Fig. 2.2B,C) caused significant mitochondrial efflux of K+. K+ handling by astrocytes. Using 2 CCCP application paradigms we examined plasma membrane K+ uptake or mitochondrial K+ sequestration across both plasma and mitochondrial membranes. Consistent with previous findings, blockade of Kir, Kdr and Ka channels and the Na+,K+-ATPase transporter significantly reduced high K+o-induced K+ flux into astrocytes although technical considerations precluded an examination of the potential involvement of other ion channels (e.g. K2P channels) and transport mechanisms (e.g. Na+-K+-2Cl- co-transport) (Kofuji and Newman 2004; Walz 2000). In contrast, K+ flux across the plasma membrane was not reduced by gap junction blockers or in Cx43-null astrocytes, suggesting that plasma membrane connexon hemichannels are not involved in K+ influx into astrocytes. Because application of CCCP alters [ATP]i, we also examined the possible role of plasma membrane KATP channels and the H+/K+-ATPase. However, it was found that these pathways do not contribute to plasma membrane K+ entry. Interestingly, we found that there was a larger increase in [K+]i when [K+]o was elevated prior to CCCP exposure and that when [K+]o was lowered to 0 mM there was a smaller CCCPinduced [K+]i response, consistent with the possibility that mitochondria help to maintain [K+]i at a constant level by temporarily sequestering K+ during local elevations in [K+]o/i and by releasing it during reductions in [K+]o/i. We also found that under HEPES-buffered recording conditions, rises in [K+]i were minimal upon exposure to elevated [K+]o yet a significant increase in mitochondrial K+ was observed. This is analogous to the well studied role of mitochondria in Ca2+ and Na+ sequestration from subplasmalemmal [ion]i microdomains that result from ion entry into a cell across the plasma membrane. In the case of Ca2+ ions, it is known that physical linkages exist 90  between subplasmalemmal mitochondria and the endo/sarcoplasmic reticulum, which ensures not only the efficient refilling of ER/SR Ca2+ stores but also acts to limit potentially detrimental cytosolic [Ca2+] elevations (reviewed by Poburko et al. 2004; Rizzuto and Pozzan 2006; see also Azarias et al. 2008; Poburko et al. 2009; Reyes and Parpura 2008). The entry of K+ into mitochondria appeared to be mediated, at least in part, by mitoKATP channels. Thus, the mitoKATP channel closer 5-HD blocked K+ sequestration, although it must be noted that 5-HD can be converted to 5-HD-CoA which can interfere with mitochondrial function (Hanley and Daut 2005). Therefore, we applied the mitochondrial mitoKATP inhibitor glyburide (which did not affect plasma membrane KATP channels), and found that mitochondrial K+ sequestration was reduced. This indicates that mitoKATP are a likely source of mitochondrial K+ entry. We also used diazoxide in an attempt to open mitoKATP channels; however, diazoxide did not promote additional K+ uptake, suggesting that these channels may have already been open under our experimental conditions. In this regard, it should be noted that the regulation of mitochondrial KATP in situ is complex and that at physiological ATP concentrations mitoKATP channels may not be closed because a variety of other mechanisms also play a role in regulating the opening mitochondrial KATP channels in situ (see Garlid and Paucek 2003). Alternatively, diazoxide can inhibit succinate dehydrogenase leading to depolarization of mitochondria (Halestrap et al. 2007) which in turn could reduce the driving force for K+ into the mitochondria. Although we attempted to examine the potential role of the mitochondrial K+/H+ antiporter (Garlid and Paucek 2003), the fluorescent properties of quinine precluded its use and timolol resulted in a loss of K+ from the cytosol, presumably reflecting its effects on other membrane ion transporters (McLaughlin et al. 2001). The role of Cx43 in mitochondrial K+ uptake. Given the abundance of gap junction proteins in astrocytes (Giaume et al. 1991), their permeability to K+ ions (Wang and Veenstra 91  1997), the reported presence of Cx43 on mitochondria (Boengler et al. 2005; Boengler et al. 2009; Goubaeva et al. 2007; Miro-Casas et al. 2009; Rodriguez-Sinovas et al. 2006) and their contribution to K+ uptake in isolated mitochondria (Miro-Casas et al. 2009), we employed pharmacological and genetic approaches to examine the potential role of Cx43 in mitochondrial K+ sequestration in situ. The combination of these approaches was designed to overcome the limitations of each method. Although the specificity of gap junction blockers remains an important consideration, we found that AGA and CBX (but not an inactive control) inhibited mitochondrial K+ uptake (although AGA had a much greater effect potentially reflecting its higher lipid solubility compared to CBX which in turn could allow for better permeation across the plasma and mitochondrial membranes). Also, CBX and AGA did not further enhance the effect in Cx43-/- astrocytes, suggesting these compounds are specific under our experimental conditions. The results obtained using the pharmacological approach were corroborated by a genetic loss-of-expression approach in Cx43+/- and Cx43-/- astrocytes, although this approach also has limitations as Cx43-null astrocytes exhibit altered expression of many genes (Iacobas et al. 2003). It is doubtful that K+ flux through plasmalemmal gap junctions between astrocytes could account for our results since K+o was uniformly applied to our populations of cultured astrocytes and PBFI-derived ratio values measured in individual astrocytes across a population were very similar, making it unlikely that domains of K+ existed to shuttle K+ to neighbouring cells connected by gap junctions. Furthermore, mitochondrial K+ sequestration was still observed in solitary astrocytes where cell-cell contacts are minimal. Our conclusion that Cx43 contributes to K+ influx into mitochondria was supported by the finding that Cx43 is present in astrocyte mitochondria, as previously reported in mitochondria from cardiac and endothelial tissue (Boengler et al. 2005; Goubaeva et al. 2007; Li et al. 2002). Also, a recent report using mitochondria from cardiomyocytes indicate that Cx43 hemichannels 92  locate to the inner mitochondrial membrane and contribute to K+ uptake (Miro-Casas et al. 2009). Physiological implications. It is well-established that mitochondria sequester Ca2+ and Na+, and the present results indicate that astrocytic mitochondria are also able to sequester K+. Although it is difficult to assess the amount of K+ that can be taken up by mitochondria, mitochondria occupy up to 7% of the cell volume in astrocytes (Pysh and Khan 1972) and can be recruited to lie in close proximity to the plasma membrane (Golovina 2005; Kolikova et al. 2006), suggesting that they could play an important role in buffering K+ influx from the extracellular space. In addition to sequestration of K+o described in this study, mitochondrial K+ uptake may also serve a beneficial role by facilitating respiration, improving ATP production, reducing mitochondrial permeability transition (Hansson et al. 2010; Kowaltowski et al. 2001; Rousou et al. 2004), and aiding in cellular protection (Korge et al. 2005; Rousou et al. 2004). It remains to be determined whether K+ uptake by mitochondrial Cx43 plays a role in cellular viability under pathological conditions, however it is known that disruption of Cx43 expression or function in astrocytes increases cellular injury (Nakase et al. 2004; Ozog et al. 2002).  93  Figure 2.1. In situ calibration of PBFI. A, A calibration experiment in which PBFI-loaded cultured astrocytes on a single coverslip were exposed to 10 M gramicidin-containing HEPES-buffered solutions at the [K+]o values (in mM) shown above the record. B, A plot of [K+] vs. PBFI-derived BI340/BI380 ratio values normalized to unity at [K+] = 100 mM from calibration experiments of the type shown in A. In comparison to the standard calibration curve (closed circle; n = 6), curves did not differ at [Na+]o = 65 mM (open circle; n = 5), nor when pH was altered to pH 6.5 (closed square; n = 5), pH 8 (open square; n = 5) or in the presence of CCCP (open triangle; n = 5). In all cases P>0.05 (two-way ANOVA).  94  Figure 2.2. CCCP evokes an increase in [K+]i in cultured astrocytes. A, Under standard PBFI loading conditions, superimposed records of the changes in [K+]i (represented by changes in cytosolic PBFI-derived BI340/BI380 ratio values) evoked by four consecutive 2 min applications of 10 M CCCP. For each response, BI340/BI380 ratio values were baseline-adjusted to the resting BI340/BI380 value observed immediately prior to the appropriate application of CCCP. The rises in [K+]i evoked by CCCP remained unaltered during repeated 2 min applications of CCCP (n = 5 preparations, P>0.05). B, An image of astrocytes (excitation wavelength 340 nm) after preferentially loading of PBFI into mitochondria and following 50 g/mL saponin treatment to release cytosolic PBFI. Under these conditions signals largely emanate from mitochondria. C, In saponin-treated astrocytes under external Na+- and K+-free recording conditions, mitochondrial-derived PBFI BI340/BI380 ratio values declined rapidly upon uncoupling with a 2 min application of 10 M CCCP indicating mitochondria release K+ which can be detected by cytosolic PBFI as indicated in A. D, Under standard PBFI loading conditions prolonged application of 10 M CCCP caused an immediate increase in [K+]i followed by a gradual decrease in [K+]i to a new steady state level. All records in A and D were obtained at [K+]o = 3 mM. 95  Figure 2.3. Effects of changes in [K+]o on CCCP-induced [K+]i rises. A, An initial 2 min application of CCCP (at [K+]o = 3 mM) evoked a [K+]i transient. Following the recovery of [K+]i to near resting levels, [K+]o was increased to 12.5 mM for 5 min, immediately after which CCCP was again applied at [K+]o = 3 mM. The rise in [K+]i evoked by the second application of CCCP (indicated by b) was larger than that evoked by the first application (indicated by a) of the protonophore. B, The same experiment as illustrated in A, except [K+]o was increased to 12.5 mM prior to the first application of CCCP. C, Summary of the results obtained in experiments of the type shown in A and B conducted under HEPES- (open bars) and HCO3-/CO2- (solid bars) buffered conditions. Also shown are results obtained under HEPES-buffered recording conditions at 37oC (hatched bars) and with FCCP (crosshatched bars). In all cases, changes in [K+]o (0 - 25 mM) were applied for 5 min immediately prior to the application of CCCP at [K+]o = 3 mM. Results are presented as percent changes in the peak amplitudes of CCCP-evoked PBFI-derived BI340/BI380 ratio value transients (representing [K+]i) after altered [K+]o, compared to control responses evoked by CCCP in the absence of a preceding change in [K+]o. In all cases, there was a significant change (*P<0.05) in the amplitude of [K+]i transients observed after pretreatment with altered [K+]o compared to those evoked under control conditions. Responses due to 12.5 mM [K+]o were significantly different under HEPES- vs. bicarbonate buffered conditions (P<0.05). Furthermore, responses to 12.5 mM [K+]o under HEPES based recording conditions were similar to the response in the FCCP and 37oC groups (P>0.05). D, Following an initial 2 min application of CCCP at [K+]o = 3 mM, astrocytes were exposed to 12.5 mM [K+]o for 5 min, after which [K+]o was reduced to 3 mM for 0, 30 or 60 s prior to the second application of CCCP (also at 3 mM K+o). The increase in the CCCP-induced [K+]i transient observed following pretreatment with 12.5 mM K+o was not significantly altered (P>0.05) at 30 s but was effectively abolished (*P<0.05) at 60 s after the end of the pretreatment (n = 5 in each case). E, Exposure to 12.5 mM [K+]o during mitochondrial uncoupling with prolonged CCCP treatment caused an increase in resting [K+]i. F, Summary of the results obtained in experiments of the type shown in E. Results are presented as the average amplitudes of the changes in PBFI-derived BI340/BI380 ratio values observed over the final 1 min of exposure to altered [K+]o after subtraction of the [K+]i value observed immediately prior to altering [K+]o. *P<0.05 in all cases.  96  97  Figure 2.4. K+ uptake mechanisms in cultured astrocytes. A, A representative trace of the effect of pretreatment with a K+ channel-blocking cocktail on the rise in [K+]i evoked by 12.5 mM K+o in the continuous presence of CCCP. B, Increases in [K+]i evoked by 5 min applications of 12.5 mM [K+]o during mitochondrial uncoupling with CCCP were significantly reduced in the presence of either 50 M Ba2+ or a cocktail containing 5 mM 4AP, 3 mM Ba2+ and 1 mM TEA (*P<0.05 in each case), while no reduction was observed in the presence of 1 mM ouabain or 1 M glyburide (P>0.05). C, A representative trace displaying the BI340/BI380 transients (representing [K+]i) evoked by 2 min applications of CCCP prior to and after 5 min pretreatment with 12.5 mM K+o. The K+ channel blocking cocktail significantly attenuated the expected increase in the magnitude of the [K+]i transient evoked by CCCP after exposure to 12.5 mM K+o (compare with Fig. 2.3A). D, The effects of ouabain, voltage-activated K+ channel inhibitors, omeprazole, glyburide, 5-HD and diazoxide in experiments of the type shown in C. All test compounds, except diazoxide and omeprazole, significantly reduced the 12.5 mM [K+]o-induced increase in the CCCP response (*P<0.05). E, The effect of CCCP application on intracellular ATP levels (nmol/mg protein normalized to control). A 2 min application of CCCP followed by a 12 min wash and a 10 min application of CCCP significantly reduced ATP levels (*P<0.05). 98  Figure 2.5. Cx43 association with astrocyte mitochondria. A representative immunoblot of Cx43 expression in whole cell (WC) and mitochondrial (Mito) fractions of Cx43+/+ (+/+) and Cx43-/- (-/-) astrocyte cultures. Also indicated are the plasma membrane marker n-cadherin, the mitochondrial marker OxPhos and GAPDH (loading control).  99  Figure 2.6. K+ uptake in mitochondria from cultured astrocytes. A. Mitochondria were preferentially loaded with PBFI and treated with saponin to remove cytosolic dye yielding signals largely derived from mitochondria. B. Mitochondria were initially perfused with a K+ free solution followed by application of a solution containing 25 mM K+. In Cx43+/+ cultures application of 25 mM K+ caused an increase in K+ in the mitochondria, an effect which was reduced by the Cx43 blocker CBX (20 M). C. A graph quantifying the changes in K+ uptake rate, defined by the change in ratio value over the first 100 s during the initial rise in PBFI ratio upon application of 25 mM K+. K+ uptake was significantly reduced with treatment of the Cx43 blockers CBX (20 M) and AGA (10 M), and was not reduced with the inactive analogue GZA (20 M). K+ uptake was also reduced in mitochondria obtained from Cx43-/- mice. Application of CBX or AGA was ineffective in further reducing K+ uptake in Cx43-/- mice. *P<0.05, n values in parentheses represent the number of individual coverslips tested.  100  Figure 2.7. Mitochondrial uptake of K+ is reduced in the presence of gap junction blockers. A, Pretreatment with the gap junction blocker CBX (20 M) did not attenuate the rise in [K+]i in response to the application of 12.5 mM K+o in the continued presence of CCCP. B, Changes in [K+]i evoked by 5 min applications of 12.5 mM K+o during mitochondrial uncoupling with CCCP were not reduced in the presence of 20 M CBX, 10 M AGA or 20 M GZA (P>0.05 in all cases, compared to control). C, A representative record displaying a reduction in the [K+]i transient evoked by a 2 min application of CCCP after 5 min pretreatment with 12.5 mM K+o in the presence of AGA. D, Using the same protocol illustrated in C, the effects CBX and GZA were also examined. All test compounds, except GZA, significantly attenuated (*P<0.05) the expected augmentation of CCCP-induced rises in [K+]i following exposure to 12.5 mM K+o.  101  Figure 2.8. Mitochondrial uptake of K+ is reduced in Cx43-/- astrocytes. A, A representative record of a prolonged CCCP application experiment depicting a rise in [K+]i in response to 12.5 mM K+o application in Cx43-/- astrocytes. B, Increases in [K+]i evoked by 5 min applications of 12.5 mM K+o during mitochondrial uncoupling with CCCP were not significantly different (P>0.05) in Cx43+/+, Cx43+/- or Cx43-/- astrocytes. C, A representative trace displaying the [K+]i transients evoked by 2 min applications of CCCP in the absence of and after 5 min pretreatment with 12.5 mM K+o in Cx43-/- astrocytes. D, Using the same protocol as in C, Cx43-/- astrocytes were found to have a reduced CCCP-induced [K+]i response after high [K+]o exposure in comparison to Cx43+/+ astrocytes (*P<0.05), while Cx43+/- astrocytes displayed an intermediate response. The response was also reduced in Cx43-/- astrocytes in the presence of CBX or AGA (*P<0.05), but no further reduction was observed with application of these compounds in comparison to Cx43-/- astrocytes in the absence of CBX or AGA (P>0.05).  102  Figure 2.9. Mitochondrial membrane potential during elevated [K+]o in Cx43+/+ and Cx43-/astrocytes. A, Changes in F/F values in rhodamine-123-loaded Cx43+/+ and Cx43-/- astrocytes upon exposure to 12.5 mM K+o. B, Bar graph summarizing the response to 12.5 mM [K+]o in Cx43+/+ and Cx43-/- astrocytes. The amplitudes of responses to elevated [K+]o were reduced in Cx43-/astrocytes (*P<0.05).  103  2.5  References  Azarias G, Van de Ville D, Unser M, Chatton JY. 2008. Spontaneous Na+ transients in individual mitochondria of intact astrocytes. Glia 56(3):342-53. Bates DC, Sin WC, Aftab Q, Naus CC. 2007. Connexin43 enhances glioma invasion by a mechanism involving the carboxy terminus. Glia 55(15):1554-64. Bernardi P. 1999. Mitochondrial transport of cations: channels, exchangers, and permeability transition. Physiol Rev 79(4):1127-55. Boengler K, Dodoni G, Rodriguez-Sinovas A, Cabestrero A, Ruiz-Meana M, Gres P, Konietzka I, Lopez-Iglesias C, Garcia-Dorado D, Di Lisa F and others. 2005. Connexin43 in cardiomyocyte mitochondria and its increase by ischemic preconditioning. Cardiovasc Res 67(2):234-44. Boengler K, Stahlhofen S, van de Sand A, Gres P, Ruiz-Meana M, Garcia-Dorado D, Heusch G, Schulz R. 2009. Presence of connexin43 in subsarcolemmal, but not in interfibrillar cardiomyocyte mitochondria. Basic Res Cardiol 104(2):141-7. Buckler KJ, Vaughan-Jones RD. 1998. Effects of mitochondrial uncouplers on intracellular calcium, pH and membrane potential in rat carotid body type I cells. J Physiol 513 (Pt 3):819-33. Budd SL, Nicholls DG. 1996. A reevaluation of the role of mitochondria in neuronal Ca2+ homeostasis. J Neurochem 66(1):403-11. Chen KC, Nicholson C. 2000. Spatial buffering of potassium ions in brain extracellular space. Biophys J 78(6):2776-97. D'Ambrosio R, Maris DO, Grady MS, Winn HR, Janigro D. 1999. Impaired K+ homeostasis and altered electrophysiological properties of post-traumatic hippocampal glia. J Neurosci 19(18):8152-62. Diarra A, Sheldon C, Church J. 2001. In situ calibration and [H+] sensitivity of the fluorescent Na+ indicator SBFI. Am J Physiol Cell Physiol 280(6):C1623-33. Duchen MR. 2004. Mitochondria in health and disease: perspectives on a new mitochondrial biology. Mol Aspects Med 25(4):365-451. Froschauer E, Nowikovsky K, Schweyen RJ. 2005. Electroneutral K+/H+ exchange in mitochondrial membrane vesicles involves Yol027/Letm1 proteins. Biochim Biophys Acta 1711(1):41-8. Garlid KD. 1996. Cation transport in mitochondria--the potassium cycle. Biochim Biophys Acta 1275(1-2):123-6. Garlid KD, Paucek P. 2003. Mitochondrial potassium transport: the K+ cycle. Biochim Biophys Acta 1606(1-3):23-41. 104  Garlid KD, Paucek P, Yarov-Yarovoy V, Murray HN, Darbenzio RB, D'Alonzo AJ, Lodge NJ, Smith MA, Grover GJ. 1997. Cardioprotective effect of diazoxide and its interaction with mitochondrial ATP-sensitive K+ channels. Possible mechanism of cardioprotection. Circ Res 81(6):1072-82. Giaume C, Fromaget C, el Aoumari A, Cordier J, Glowinski J, Gros D. 1991. Gap junctions in cultured astrocytes: single-channel currents and characterization of channel-forming protein. Neuron 6(1):133-43. Golovina VA. 2005. Visualization of localized store-operated calcium entry in mouse astrocytes. Close proximity to the endoplasmic reticulum. J Physiol 564(Pt 3):737-49. Goubaeva F, Mikami M, Giardina S, Ding B, Abe J, Yang J. 2007. Cardiac mitochondrial connexin43 regulates apoptosis. Biochem Biophys Res Commun 352(1):97-103. Halestrap AP, Clarke SJ, Khaliulin I. 2007. The role of mitochondria in protection of the heart by preconditioning. Biochim Biophys Acta 1767(8):1007-31. Hanley PJ, Daut J. 2005. KATP channels and preconditioning: a re-examination of the role of mitochondrial KATP channels and an overview of alternative mechanisms. J Mol Cell Cardiol 39(1):17-50. Hansson MJ, Morota S, Teilum M, Mattiasson G, Uchino H, Elmer E. 2010. Increased potassium conductance of brain mitochondria induces resistance to permeability transition by enhancing matrix volume. J Biol Chem 285(1):741-50. Holthoff K, Witte OW. 2000. Directed spatial potassium redistribution in rat neocortex. Glia 29(3):288-92. Iacobas DA, Urban-Maldonado M, Iacobas S, Scemes E, Spray DC. 2003. Array analysis of gene expression in connexin43 null astrocytes. Physiol Genomics 15(3):177-90. Jewell EA, Lingrel JB. 1991. Comparison of the substrate dependence properties of the rat Na,KATPase alpha 1, alpha 2, and alpha 3 isoforms expressed in HeLa cells. J Biol Chem 266(25):16925-30. Juthberg SK, Brismar T. 1997. Effect of metabolic inhibitors on membrane potential and ion conductance of rat astrocytes. Cell Mol Neurobiol 17(4):367-77. Kasner SE, Ganz MB. 1992. Regulation of intracellular potassium in mesangial cells: a fluorescence analysis using the dye, PBFI. Am J Physiol 262(3 Pt 2):F462-7. Kofuji P, Newman EA. 2004. Potassium buffering in the central nervous system. Neuroscience 129(4):1045-56. Kolikova J, Afzalov R, Giniatullina A, Surin A, Giniatullin R, Khiroug L. 2006. Calciumdependent trapping of mitochondria near plasma membrane in stimulated astrocytes. Brain Cell Biol 35(1):75-86.  105  Korge P, Honda HM, Weiss JN. 2005. K+-dependent regulation of matrix volume improves mitochondrial function under conditions mimicking ischemia-reperfusion. Am J Physiol Heart Circ Physiol 289(1):H66-77. Kowaltowski AJ, Cosso RG, Campos CB, Fiskum G. 2002. Effect of Bcl-2 overexpression on mitochondrial structure and function. J Biol Chem 277(45):42802-7. Kowaltowski AJ, Seetharaman S, Paucek P, Garlid KD. 2001. Bioenergetic consequences of opening the ATP-sensitive K+ channel of heart mitochondria. Am J Physiol Heart Circ Physiol 280(2):H649-57. Kubota T, Shindo Y, Tokuno K, Komatsu H, Ogawa H, Kudo S, Kitamura Y, Suzuki K, Oka K. 2005. Mitochondria are intracellular magnesium stores: investigation by simultaneous fluorescent imagings in PC12 cells. Biochim Biophys Acta 1744(1):19-28. Leis JA, Bekar LK, Walz W. 2005. Potassium homeostasis in the ischemic brain. Glia 50(4):40716. Levi AJ, Lee CO, Brooksby P. 1994. Properties of the fluorescent sodium indicator "SBFI" in rat and rabbit cardiac myocytes. J Cardiovasc Electrophysiol 5(3):241-57. Li H, Brodsky S, Kumari S, Valiunas V, Brink P, Kaide J, Nasjletti A, Goligorsky MS. 2002. Paradoxical overexpression and translocation of connexin43 in homocysteine-treated endothelial cells. Am J Physiol Heart Circ Physiol 282(6):H2124-33. Lian XY, Stringer JL. 2004. Astrocytes contribute to regulation of extracellular calcium and potassium in the rat cerebral cortex during spreading depression. Brain Res 1012(12):177-84. Liu D, Slevin JR, Lu C, Chan SL, Hansson M, Elmer E, Mattson MP. 2003. Involvement of mitochondrial K+ release and cellular efflux in ischemic and apoptotic neuronal death. J Neurochem 86(4):966-79. Ljubkovic M, Marinovic J, Fuchs A, Bosnjak ZJ, Bienengraeber M. 2006. Targeted expression of Kir6.2 in mitochondria confers protection against hypoxic stress. J Physiol 577(Pt 1):17-29. McLaughlin CW, Peart D, Purves RD, Carre DA, Peterson-Yantorno K, Mitchell CH, Macknight AD, Civan MM. 2001. Timolol may inhibit aqueous humor secretion by cAMPindependent action on ciliary epithelial cells. Am J Physiol Cell Physiol 281(3):C865-75. Metea MR, Kofuji P, Newman EA. 2007. Neurovascular coupling is not mediated by potassium siphoning from glial cells. J Neurosci 27(10):2468-71. Minta A, Tsien RY. 1989. Fluorescent indicators for cytosolic sodium. J Biol Chem 264(32):19449-57. Miro-Casas E, Ruiz-Meana M, Agullo E, Stahlhofen S, Rodriguez-Sinovas A, Cabestrero A, Jorge I, Torre I, Vazquez J, Boengler K and others. 2009. Connexin43 in Cardiomyocyte 106  Mitochondria Contributes to Mitochondrial Potassium Uptake. Cardiovasc Res 83(4):747-56. Murchison D, Griffith WH. 2000. Mitochondria buffer non-toxic calcium loads and release calcium through the mitochondrial permeability transition pore and sodium/calcium exchanger in rat basal forebrain neurons. Brain Res 854(1-2):139-51. Murchison D, Zawieja DC, Griffith WH. 2004. Reduced mitochondrial buffering of voltagegated calcium influx in aged rat basal forebrain neurons. Cell Calcium 36(1):61-75. Murphy E, Eisner DA. 2009. Regulation of intracellular and mitochondrial sodium in health and disease. Circ Res 104(3):292-303. Nakase T, Sohl G, Theis M, Willecke K, Naus CC. 2004. Increased apoptosis and inflammation after focal brain ischemia in mice lacking connexin43 in astrocytes. Am J Pathol 164(6):2067-75. Naus CC, Bechberger JF, Zhang Y, Venance L, Yamasaki H, Juneja SC, Kidder GM, Giaume C. 1997. Altered gap junctional communication, intercellular signaling, and growth in cultured astrocytes deficient in connexin43. J Neurosci Res 49(5):528-40. Oliveira JM, Jekabsons MB, Chen S, Lin A, Rego AC, Goncalves J, Ellerby LM, Nicholls DG. 2007. Mitochondrial dysfunction in Huntington's disease: the bioenergetics of isolated and in situ mitochondria from transgenic mice. J Neurochem 101(1):241-9. Olson JE, Sankar R, Holtzman D, James A, Fleischhacker D. 1986. Energy-dependent volume regulation in primary cultured cerebral astrocytes. J Cell Physiol 128(2):209-15. Orkand RK, Nicholls JG, Kuffler SW. 1966. Effect of nerve impulses on the membrane potential of glial cells in the central nervous system of amphibia. J Neurophysiol 29(4):788-806. Ozog MA, Siushansian R, Naus CC. 2002. Blocked gap junctional coupling increases glutamateinduced neurotoxicity in neuron-astrocyte co-cultures. J Neuropathol Exp Neurol 61(2):132-41. Park KS, Jo I, Pak K, Bae SW, Rhim H, Suh SH, Park J, Zhu H, So I, Kim KW. 2002. FCCP depolarizes plasma membrane potential by activating proton and Na+ currents in bovine aortic endothelial cells. Pflugers Arch 443(3):344-52. Pivovarova NB, Hongpaisan J, Andrews SB, Friel DD. 1999. Depolarization-induced mitochondrial Ca accumulation in sympathetic neurons: spatial and temporal characteristics. J Neurosci 19(15):6372-84. Poburko D, Lee CH, van Breemen C. 2004. Vascular smooth muscle mitochondria at the cross roads of Ca2+ regulation. Cell Calcium 35(6):509-21. Poburko D, Liao CH, Lemos VS, Lin E, Maruyama Y, Cole WC, van Breemen C. 2007. Transient receptor potential channel 6-mediated, localized cytosolic [Na+] transients drive Na+/Ca2+ exchanger-mediated Ca2+ entry in purinergically stimulated aorta smooth muscle cells. Circ Res 101(10):1030-8. 107  Poburko D, Liao CH, van Breemen C, Demaurex N. 2009. Mitochondrial regulation of sarcoplasmic reticulum Ca2+ content in vascular smooth muscle cells. Circ Res 104(1):104-12. Pysh JJ, Khan T. 1972. Variations in mitochondrial structure and content of neurons and neuroglia in rat brain: an electron microscopic study. Brain Res 36(1):1-18. Reaume AG, de Sousa PA, Kulkarni S, Langille BL, Zhu D, Davies TC, Juneja SC, Kidder GM, Rossant J. 1995. Cardiac malformation in neonatal mice lacking connexin43. Science 267(5205):1831-4. Reyes RC, Parpura V. 2008. Mitochondria modulate Ca2+-dependent glutamate release from rat cortical astrocytes. J Neurosci 28(39):9682-91. Rigby H, Gubitz G, Phillips S. 2009. A systematic review of caregiver burden following stroke. Int J Stroke 4(4):285-92. Rizzuto R, Pozzan T. 2006. Microdomains of intracellular Ca2+: molecular determinants and functional consequences. Physiol Rev 86(1):369-408. Rodriguez-Sinovas A, Boengler K, Cabestrero A, Gres P, Morente M, Ruiz-Meana M, Konietzka I, Miro E, Totzeck A, Heusch G and others. 2006. Translocation of connexin43 to the inner mitochondrial membrane of cardiomyocytes through the heat shock protein 90dependent TOM pathway and its importance for cardioprotection. Circ Res 99:93-101. Rose CR, Ransom BR. 1996. Intracellular sodium homeostasis in rat hippocampal astrocytes. J Physiol 491 (Pt 2):291-305. Rose CR, Ransom BR. 1997. Gap junctions equalize intracellular Na+ concentration in astrocytes. Glia 20(4):299-307. Rousou AJ, Ericsson M, Federman M, Levitsky S, McCully JD. 2004. Opening of mitochondrial KATP channels enhances cardioprotection through the modulation of mitochondrial matrix volume, calcium accumulation, and respiration. Am J Physiol Heart Circ Physiol 287(5):H1967-76. Sato T, O'Rourke B, Marban E. 1998. Modulation of mitochondrial ATP-dependent K+ channels by protein kinase C. Circ Res 83(1):110-4. Sheldon C, Diarra A, Cheng YM, Church J. 2004. Sodium influx pathways during and after anoxia in rat hippocampal neurons. J Neurosci 24(49):11057-69. Simard M, Nedergaard M. 2004. The neurobiology of glia in the context of water and ion homeostasis. Neuroscience 129(4):877-96. Somjen GG. 2002. Ion regulation in the brain: implications for pathophysiology. Neuroscientist 8(3):254-67.  108  Stys PK, Lehning E, Saubermann AJ, LoPachin RM, Jr. 1997. Intracellular concentrations of major ions in rat myelinated axons and glia: calculations based on electron probe X-ray microanalyses. J Neurochem 68(5):1920-8. Su G, Kintner DB, Flagella M, Shull GE, Sun D. 2002. Astrocytes from Na+-K+-Clcotransporter-null mice exhibit absence of swelling and decrease in EAA release. Am J Physiol Cell Physiol 282(5):C1147-60. Wallraff A, Kohling R, Heinemann U, Theis M, Willecke K, Steinhauser C. 2006. The impact of astrocytic gap junctional coupling on potassium buffering in the hippocampus. J Neurosci 26(20):5438-47. Walz W. 1992. Mechanism of rapid K+-induced swelling of mouse astrocytes. Neurosci Lett 135(2):243-6. Walz W. 2000. Role of astrocytes in the clearance of excess extracellular potassium. Neurochem Int 36(4-5):291-300. Wang HZ, Veenstra RD. 1997. Monovalent ion selectivity sequences of the rat connexin43 gap junction channel. J Gen Physiol 109(4):491-507. Xu L, Zeng LH, Wong M. 2009. Impaired astrocytic gap junction coupling and potassium buffering in a mouse model of tuberous sclerosis complex. Neurobiol Dis 34(2):291-9. Xu W, Liu Y, Wang S, McDonald T, Van Eyk JE, Sidor A, O'Rourke B. 2002. Cytoprotective role of Ca2+- activated K+ channels in the cardiac inner mitochondrial membrane. Science 298(5595):1029-33. Yang MK, Lee SH, Seo HW, Yi KY, Yoo SE, Lee BH, Chung HJ, Won HS, Lee CS, Kwon SH and others. 2009. KR-31761, a novel KATP-channel opener, exerts cardioprotective effects by opening both mitochondrial KATP and Sarcolemmal KATP channels in rat models of ischemia/reperfusion-induced heart injury. J Pharmacol Sci 109(2):222-32. Zoeteweij JP, van de Water B, de Bont HJ, Nagelkerke JF. 1994. Mitochondrial K+ as modulator of Ca2+-dependent cytotoxicity in hepatocytes. Novel application of the K+-sensitive dye PBFI K+-binding benzofuran isophthalate) to assess free mitochondrial K+ concentrations. Biochem J 299 (Pt 2):539-43.  109  Chapter 3: The connexin43 C-terminal region mediates neuroprotection during stroke.4  4  A version of this chapter has been published. Kozoriz, MG., Bechberger, JF., Bechberger, GR., Suen, MWH., Moreno, AP., Maass, K., Willecke, K. and Naus, CC. (2010) The connexin43 C-terminal region mediates neuroprotection during stroke. J Neuropathol Exp Neurol. 69(2): 196-206  110  3.1  Introduction Intercellular gap junction channels provide cytoplasmic continuity between adjacent cells  by allowing ions and molecules less than 1.8 kD to pass (Neijssen et al. 2005). The role of gap junctions in stroke is controversial. Some reports suggest that gap junctions allow passage of cytotoxic substances, while other authors suggest gap junctions pass required metabolites (e.g. antioxidants, adenosine triphosphate [ATP], glucose) to areas of demand while also buffering cytotoxic levels of excitatory amino acids and ions through the astrocytic syncytium (Contreras et al. 2004; Rossi et al. 2007). Both theories may be correct, and the ratio of healthy to unhealthy cells may determine whether gap junctions are protective or destructive (Rossi et al. 2007). We have previously shown that selectively knocking out the gap junction protein connexin43 (Cx43) in astrocytes results in larger infarct volumes in a middle cerebral artery occlusion (MCAO) permanent focal stroke mode (Nakase et al. 2003b), suggesting that astrocytic Cx43 limits damage caused by stroke. In addition to forming gap junctions, unopposed connexins may form hemichannels that are known to release substances such as ATP, glutamate and glutathione (Kang et al. 2008; Stridh et al. 2008). Hemichannel activity has been shown to play a protective role in hypoxic preconditioning by releasing ATP, which is subsequently converted to adenosine that may protect neighboring neurons (Lin et al. 2008). A site of interest in elucidating the neuroprotective role of Cx43 is its C-terminal (CT) region, which exhibits several phosphorylation sites that are believed to play a role in gap junction assembly and activity (Herve et al. 2007; Lampe and Lau 2004). Gating of the Cx43 channel caused by intracellular acid loads is also thought to involve the CT (Duffy et al. 2002). Moreover, the Cx43CT binds to other proteins such as -tubulin, -tubulin, CCN3, c-src and zona occludens-1 (Herve et al. 2007). Given the wide range of interactions, we were interested in determining whether truncation of the gate of Cx43 alters cellular injury during stroke. 111  In this study, we used mutant mice in which the CT of Cx43 has been truncated at amino acid residue 258 (Cx43 CT) (Maass et al. 2004). Truncation of this region removes casein kinase 1, protein kinase A, protein kinase C and a majority of the mitogen-activated protein kinase phosphorylation sites; the src homology 2, src homology 3, and the PDZ zona occludens 1 binding domains are also removed. Interaction sites for microtubules are preserved in this mutant, as are a p34cdc2, mitogen-activated protein kinase and src phosphorylation site (Giepmans et al. 2001; Giepmans and Moolenaar 1998; Herve et al. 2007; Kieken et al. 2009; Lampe and Lau 2000; Lampe et al. 2000; Zucker and Nicholson 2002). Cells with the Cx43 CT mutation are known to electrically couple, but single channel recordings in cardiomyocytes reveal a lack of the 30 pS sub-state and prolonged channel opening times (Maass et al. 2007). Similar truncations of the CT are also known to cause disruption of channel gating (Moreno et al. 2002). Because of the broad functions attributed to Cx43, in this initial study we sought to determine the importance of the CT region of Cx43 in stroke by examining several processes relevant to stroke. We tested the susceptibility to cellular injury, astrogliosis and inflammatory cell invasion in Cx43 CT mice following MCAO. Furthermore, we explored the expression and functional properties of Cx43 CT in adult cortices and cultured astrocytes.  3.2  Materials and methods All experiments were performed in accordance with the guidelines established by the  Canadian Council on Animal Care and were approved by The University of British Columbia Animal Care Committee. Knockout mice and MCAO. Mice from a 129/Ola/C57BL/6 genetic background in which Cx43 has been replaced by Cx43K258stop (Cx43  CT  ) were used (Maass et al. 2004). Breeding 112  and genotyping of Cx43+/+, Cx43+/-, Cx43  CT/+  , Cx43  CT/-  and Cx43  CT/ CT  mice have been  previously described (Maass et al. 2004; Maass et al. 2007; Naus et al. 1997). Cx43  CT/ CT  mice  are not viable due to an epidermal barrier defect (Maass et al. 2004), but it is possible to culture Cx43  CT/ CT  astrocytes near birth. Details of the MCAO surgeries are described in our previous  studies (Nakase et al. 2003b; Nakase et al. 2004; Siushansian et al. 2001). In brief, 8- to 10week-old mice were anaesthetized with sodium pentobarbital (65 mg/kg intraperitoneally [i.p.]; MTC Pharmaceuticals, Cambridge, Ontario, Canada). Once they were unconscious, hair over the surgery site was cut with an electric razor, and then the mice were secured to a 37oC heating pad and placed in a stereotaxic frame. A skin incision was made, and the squamosal bone was exposed by retracting the temporalis muscle. A 3 mm burr hole was made over the MCA with a Dermel drill using a no. 105 engraving cutter drill bit (Robert Bosch Tool Corporation, Mount Prospect, IL). The dura mater was pierced with a needle, and the MCA was occluded above and below the rhinal fissure by electrocautery. The skin incision was closed with sutures, and the mice were kept in a cage warmed on a 37oC heating pad until they regained consciousness. Infarct volume. At 4 days after MCAO, the mice were anesthetized with sodium pentobarbital (70 mg/kg i.p.) and were transcardially perfused with phosphate-buffered saline (PBS) followed by perfusion with 10% formalin (Sigma-Aldrich, Oakville, Canada). Brains were removed and placed in 10% formalin overnight then stored in a 10% formalin-30% sucrose solution. Using a cryostat (HM 505E; Micron, Walldorf, Germany), 20- m-thick brain slices were collected every 100 m and mounted sequentially on glass microscope slides. To measure infarct sizes, sections were stained with 0.125% thionin (Fisher Scientific, Ottawa, Ontario, Canada). Quantification of the infarct volume consisted of adding together the measured lesion area in each section as previously described using Scion Image Software (Scion Corporation, Frederick, MD) (Nakase et al. 2004; Siushansian et al. 2001). 113  Examination of brain vasculature. To visualize the brain vasculature, 8- to 10-week-old mice were anesthetized with sodium pentobarbital (65 mg/kg i.p.) and were transcardially perfused with PBS then 10% formalin followed by 3 to 5 mL of a staining solution containing 3.5 g gelatin (Fisher Scientific) dissolved in 60 mL water, plus 10 mL of 25% mannitol (SigmaAldrich) and 20 mL of India ink (Speedball Art Company, Houston, TX) (Berry et al. 1975). The mice were stored at 4oC for 24 hours to allow for solidification of the gelatin, after which the brains were extracted and stored in 10% formalin. Dark ink filled blood vessels were apparent using this protocol, and brains were photographed using a digital camera (Canon Powershot SD750; Cannon, Mississauga, Ontario, Canada). To examine the microvasculature in these mice 50- m coronal sections were obtained using a vibratome (Campden Instruments Ltd., Lafayette, IN). Sections were mounted on glass slides and stained with 1% neutral red (Sigma-Aldrich) for 5 minutes followed by washes in water, 50 % ethanol, 70% ethanol, 100% ethanol (twice) and 100% xylene (twice) (Hekmatpanah 2007). Slides were mounted using Cryoseal XYL (RichardAllan Scientific, Kalamazoo, MI), and images were obtained on a Leica microscope (Leica Microsystems, Richmond Hill, Ontario, Canada). Astrocyte cultures. Neonatal (0 - 1 day) brains were removed from mice and placed in PBS. Neocortices were isolated, freed of meninges, and placed in high-glucose Dulbecco's modified Eagle medium (DMEM; Invitrogen Corp., Burlington, Ontario, Canada) supplemented with 10% fetal bovine serum (FBS; HyClone, Logan, UT), 10 U/mL penicillin, and 10 U/mL streptomycin (Invitrogen). Tissue was mechanically disrupted and passed through a 70- m cell strainer (BD Falcon, Bedford, MA). Cells were resuspended in high-glucose DMEM supplemented with 10% FBS, 10 U/mL penicillin, and 10 U/mL streptomycin, and maintained in a humidified incubator in 95% air/5% CO2 at 37°C. A fresh medium change was performed  114  every 4 days (Ozog et al. 2002). Confluent cells were passed onto glass coverslips or plastic culture dishes and astrocytes were used between 14 and 21 days in vitro. Immunohistochemistry. Immunohistochemistry was performed as previously described on brain tissue obtained 4 days after MCAO and after fixation with 4% paraformaldehyde (Sigma-Aldrich) (Nakase et al. 2003a; Nakase et al. 2003b; Nakase et al. 2004). In brief, 10- m glass-mounted sections were rinsed in PBS and exposed to 0.3% Triton X-100 (Fisher Scientific) in PBS for 2 minutes. After a brief rinse in PBS, cells were blocked in 5% normal goat serum (NGS; Sigma-Aldrich) for 30 minutes. Sections were then incubated overnight in 1.5% NGSPBS and primary antibody. Sections were then washed 3 times for 10 minutes with PBS and subsequently incubated for 1 hour with the secondary antibody. After incubation, slides were rinsed 3 times for 10 minutes with PBS, dipped in water, and mounted with ProLong Gold antifade reagent with 4',6-diamidino-2-phenylindole (DAPI; Molecular Probes, Eugene, OR). Control experiments were performed by omitting the primary antibodies and by incubating the primary antibody with the opposite secondary. Primary antibodies were used at the concentrations indicated: anti-Cx43 C-terminal (amino acid residues 363-382) raised in rabbit (1:2000 dilution, catalogue no. C6219, Sigma-Aldrich), anti-Cx43 cytoplasmic loop (amino acids residues 131-142) raised in mouse (1:50, catalogue no. 05-763, Upstate, Lake Placid, NY), antiglial fibrillary acidic protein (GFAP) raised in mouse (1:1000, catalogue no. G3893, SigmaAldrich) and anti-ionized calcium-binding adaptor molecule 1 (IBA-1) produced in rabbit (1:500, catalogue no. 019-19741, Wako Pure Chemical Industries, Richmond, VA). Secondary antibodies consisted of a 1:500 dilution of highly cross-adsorbed goat anti-rabbit or anti-mouse IgG conjugated to Alexa Fluor 488 (catalogue no. A-11029 and no. A-11034) or Alexa Fluor 568 (catalogue #A-11031 and #A-11036) used as appropriate (Molecular Probes). All images were  115  obtained on a Zeiss Axioplan2 fluorescence microscope (Carl Zeiss Ltd, Toronto, Ontario, Canada). Immunocytochemistry on astrocyte cultures was performed as described previously using the same concentrations of Cx43 antibodies and secondary antibodies (Bates et al. 2007). In brief, astrocytes were rinsed in PBS, fixed in 4% paraformaldehyde, briefly rinsed in PBS and then exposed to 0.1% Triton X-100 PBS (Fisher Scientific) for 2 minutes. After a brief rinse, cells were blocked in 5% normal goat serum NGS for 20 minutes. Astrocytes were then incubated for 1 hour in PBS containing 1.5% NGS and primary antibody. Astrocytes were then washed 3 times for 10 minutes with PBS and subsequently incubated for 1 hour with a secondary antibody. After incubation, astrocytes were rinsed 3 times for 10 minutes with PBS, dipped in water, and mounted with ProLong Gold with DAPI. Western Blot. Adult brain samples and astrocyte cultures were prepared and probed as previously described (Maass et al. 2007). In brief, tissue was lysed in Laemmli buffer supplemented with complete mini (Roche, Indianapolis, IN) and phosphatase inhibitors (SigmaAldrich), and the samples were drawn up by a syringe through a 26-gauge needle. Samples were then sonicated and centrifuged at 10,000 g for 10 minutes. Protein concentrations were determined using a BCA protein quantification kit (Pierce, Rockford, IL). Protein samples (30 g) were heated at 37oC for 30 minutes in Laemmli buffer and loaded onto a 12% polyacrylamide gel for separation. Protein was then transferred to a nitrocellulose membrane. The membrane was blocked in 5% milk in Tris-buffered saline with Tween (TBS-T) for 1 hour followed by incubation in a primary antibody in 5% milk and TBS-T overnight. The blot was then washed 3 times in TBS-T for 10 minutes and was then incubated in secondary antibody. Next, the blot was washed 3 times in TBS-T for 10 minutes and was placed in TBS for 5 minutes. Protein was detected by exposing the blot to Super-Signal chemiluminescent substrate 116  (Pierce). Note that when the anti-Cx43 N-terminal antibody was used, all incubations were in 1% milk and in the absence of Tween. Primary antibodies were used at the following dilutions: anti-Cx43 C-terminal (1:8000), anti-Cx43 N-terminal clone P1E11 (1:300; amino acid residues 1-19; Fred Hutchington Cancer Research Center, Seattle, WA), anti-Cx43 cytoplasmic loop (1:200), Cx30 (1:250, catalogue no. 71-2200, Zymed, San Francisco, CA) and antiglyceraldehyde 3-phosphate dehydrogenase (GAPDH; 1:10,000; HyTest, Turku, Finland; catalogue #5G4 MAb 6C5). The secondary antibodies, goat anti-mouse or anti-rabbit horseradish peroxidase-tagged were used at a 1:3000 dilution (catalogue no. A6154 and no. A8924, Sigma). To ensure equal loading of protein samples, the blots were stripped and reprobed for GAPDH. Astrogliosis and inflammatory cell measurements. Ten-micrometer-thick sections from mice subjected to MCAO were probed with a low concentration of anti-GFAP antibody to stain reactive astrocytes (Nakase et al. 2003a; Nakase et al. 2004). Micrographs were obtained along the lateral border of the section under a 16x objective from the edge of necrosis to the edge of astrogliosis, where reactive astrocyte staining was similar to that of the rest of the section (Nakase et al. 2003a; Nakase et al. 2004); each 16x field corresponded to 550 m x 550 m area. For each mouse, the extent of astrogliosis was determined by averaging 5 measurements from the necrotic edge to the edge of astrogliosis in 3 separate sections. To quantify the extent of inflammatory cells present in the peri-infarct region of the stroke, 10- m-thick sections were labeled with anti-IBA-1 to detect microglia and infiltrated blood-derived macrophages. Sections (n = 3 per mouse) were examined under 16x magnification. In each section, cells were counted in 3 randomly chosen fields in the region adjacent to the edge of necrosis, and data from each field were averaged (Nakase et al. 2004).  117  Gap junction coupling. To assess the extent of gap junction coupling, cells were incubated with 50 L of gap junction-permeable carboxyfluorescein (0.1%) and gap junctionimpermeable dextran-rhodamine (0.1%) (Molecular Probes) in PBS. Cells were scraped with a scalpel blade and allowed to incubate for 2 minutes after which excess dye was washed off with PBS. A gap junction blocker, carbenoxolone (CBX; 20 mol/L; Sigma-Aldrich), was then added to inhibit further coupling. The distance carboxyfluorescein traveled from the source cells was taken as a measure of the extent of coupling. Electrophysiology. A dual whole-cell voltage clamp technique was applied to cell pairs to measure their junctional conductance (gj). The patch solution consisted of (in mmol/L): 130 CsCl2; 0.5 CaCl2; 2 Na2ATP; 3 MgATP; 10 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES); 10 EGTA (pH 7.2). The recording solution consisted of (in mmol/L): 130 NaCl, 7 CsCl2, 2.0 CaCl2, 0.6 MgCl2, 10 HEPES, pH 7.4. Single channel currents were measured by using freshly plated astrocytes on glass coverslips. Unitary junction currents were observed after 30-second superfusion of 2 mmol/L halothane and digitized during long-voltage steps applied to one of the cells. Amplitudes of unitary opening or closing current events were digitized through a HEKA 9000 amplifier (HEKA, Lambrecht, Germany) filtered at 200 Hz. Probability distribution histograms of the events and Gaussian distribution best fits for each peak were calculated for each experiment (Origin 7.0; OriginLab Corporation, Northampton, MA). Single channels current traces were digitized at 5 kHz, filtered at 100 Hz and used to generate all point histograms with a bin resolution of 0.05-0.025 pS. Hemichannel assay. Hemichannel activity was assessed by propidium iodide uptake (Lin et al. 2008). Cells were washed with Hanks buffered salt solution (HBSS; Invitrogen Corp.) with Ca2+ and then incubated in HBSS with or without Ca2+containing 1.5 mmol/L propidium iodide for 15 minutes. Cells were then washed three times with HBSS and were fixed and mounted on 118  glass slides for subsequent analysis. CBX was also applied 5 minutes before and during the experiment in a portion of cells to test the specificity of hemichannel openings. Calcium imaging. Confluent astrocytes were placed in imaging buffer (in mmol/L): 136.5 NaCl, 3.0 KCl, 1.5 NaHCO3, 1.5 MgSO4, 10.0 D-glucose, 2.0 CaCl2 and 10.0 HEPES with 5 mol/L fura-2 AM (Molecular Probes) for 1 hour at room temperature. After washing the cells were mounted in a perfusion chamber on a Zeiss Axioskop2 FS Plus microscope (Carl Zeiss Canada). Measurement of [Ca2+]i was performed using the dual excitation ratio method (340 and 380 nm; Lambda DG-5, Sutter Instrument Co., Novato, CA) under a 20x objective, and fluorescence emissions greater than 505 nm were captured by a 12-bit digital cooled CCD camera every 5 seconds (Retiga EXi, QImaging, Burnaby, Ontario, Canada) and analyzed off line using SlideBook (version 4.1 Intelligent Imaging Innovations, Denver, CO). Calcium waves were generated by mechanical stimulation (Naus et al. 1997; Venance et al. 1995). The following three parameters were measured: the distance of Ca2+ wave propagation (measured as the distance between the stimulated cell and the furthest activated cell); the wave velocity (calculated by dividing the furthest distance by the time interval taken to reach that astrocyte); and the efficacy of the Ca2+ spread (calculated as the number of activated cells divided by the total number of cells). Data analysis. Results are reported as means  SE. Statistical comparisons were  performed using a one-way analysis of variance with a Student-Newman-Keuls multiple comparisons test, with a 95% confidence limit unless otherwise indicated.  3.3  Results Given the importance of gap junctions in brain cellular injury (Talhouk et al. 2008) and the  role of the CT region of Cx43 in channel gating, we sought to determine the significance of the 119  Cx43 CT mutation during stroke after MCAO. In comparison with wild-type mice, infarct volumes were greater in Cx43+/-, Cx43  CT/+  and Cx43  CT/-  mutant mice (P<0.05) (Fig 3.1).  Moreover, infarct volumes from Cx43+/+ mice were compared with Cx43 mice were compared with Cx43  CT/-  CT/+  mice, and Cx43+/-  mice, and in both instances, the presence of the Cx43 CT  mutation conferred a larger infarct volume (Fig. 3.1), and no differences were observed between Cx43  CT/+  and Cx43  CT/-  mice (P>0.05). The vasculature of each genotype was examined by  filling blood vessels with India ink. Although brain vascularization patterns were variable within each genotype at the gross and microvasculature levels, no obvious abnormalities were detectable across the 4 genotypes (Fig. C.1 – Appendix C). Astrocytes increase their GFAP expression and interweave to form a gliotic scar after a stroke (Pekny and Nilsson 2005), and Cx43 is known to be increased in reactive and proliferating astrocytes after stroke (Haupt et al. 2007). We previously showed that Cx43+/- mice exhibit reduced astrogliosis after MCAO (Nakase et al. 2003a; Nakase et al. 2004). In Cx43+/+ mice, the region of gliosis extended 891 ± 69 m (n = 6) from the necrotic core. As in our previous studies, the gliotic region was significantly reduced to 591 ± 77 m in Cx43+/- mice (n = 6, P<0.05). Gliosis was further reduced to 558 ± 98 m and 500 ± 31 m in Cx43 Cx43  CT/-  CT/+  and  mice respectively (n = 6, P<0.05; Fig. 3.2A,B), suggesting that disruption of the CT  region of Cx43 interferes with glial scar formation. Microglial/macrophage responses in infarcts can be beneficial or harmful depending on the inflammatory context (Ekdahl et al. 2009). Ablation of proliferating microglia is known to increase ischemic injury (Lalancette-Hebert et al. 2007), and microglia are known to effect gap junctional communication (Kieken et al. 2009; Meme et al. 2006; Retamal et al. 2007). Furthermore, microglia express Cx43 and Cx36 (Dobrenis et al. 2005; Eugenin et al. 2001). Cx43 CT mice are known to have increased inflammation and neutrophil recruitment in an 120  inflammatory lung model (Sarieddine et al. 2009), and previous studies have shown that microglial invasion is increased in the penumbral region after MCAO in Cx43+/- mice (Nakase et al. 2004). Here, we found that the Cx43 CT mutation leads to changes in microglia/macrophage (IBA-1 labeled cells) in the peri-infarct region following stroke. In wild-type brains, there were 27 ± 3.9 inflammatory cells per 16x field view (n = 6). Inflammatory cell invasion was greater in Cx43+/- (43 ± 5.3 cells; n = 6, P<0.05), Cx43  CT/+  (50 ± 7.0 cells; n = 6, P<0.05) and Cx43  CT/-  brain sections (52 ± 3.1 cells; n = 6, P<0.05), Fig. 3.3A,B. Microglia/macrophages in the periinfarct region had morphology consistent with those of activated microglia (Cuadros and Navascues 1998; Walz 1999). Immunostaining of astrocyte cultures (Fig. 3.4A,B) and brain sections (Fig. 3.4C,D) using primary antibody raised against the C-terminal or cytoplasmic loop of Cx43 revealed typical membrane localization in Cx43+/+ cells and less expression in Cx43+/- cells, as previously reported (Naus et al. 1997). Membrane localization was also observed in Cx43  CT/+  astrocytes  with both antibodies (Fig. 3.4), but more distinct immunolabeled gap junction plaques were observed, and cytoplasmic staining remained sparse. Plaques were also observed in Cx43  CT/-  using the cytoplasmic loop antibody, but expression seemed reduced overall (Fig. 3.4B,D). No staining was observed using the C-terminal antibody in these sections because the corresponding epitope is lacking (Fig. 3.4A,C). Therefore, the use of the anti-C-terminal antibody in Cx43  CT/ CT  cultures resulted in no signal, whereas the cytoplasmic loop antibody revealed  overall reduced Cx43 expression, but plaque formation was still maintained (Fig. 3.4B). To characterize Cx43 expression further, Western blots were performed on samples from astrocyte cultures and cortical brain tissue (Fig. 3.5). The concentration of Cx43 protein correlated with the intensity of staining observed in Fig. 3.4. Bands representing Cx43 were observed in Cx43+/+, Cx43+/- and Cx43  CT/+  samples when probed with the anti-C-terminus Cx43 121  antibody. Furthermore, the use of an antibody raised against the N-terminal of Cx43 resulted in additional bands at approximately 30 kDa in the Cx43  CT/+  , Cx43  CT/-  and Cx43  CT/ CT  cells,  representing the truncated form of Cx43. Cx30 staining was also performed. Staining was observed in adult cortical brain samples, whereas Cx30 was absent in cultured astrocytes as previously reported (Fig. 3.5). When the density of the Cx30 band was normalized to GAPDH (n=4) and subsequently normalized to the Cx30 band obtained from Cx43+/+ samples, it was found that Cx30 levels were unchanged in Cx43+/- mice (1.004 ± 0.09; P>0.05). Non-significant decreases in Cx30 were observed in Cx43  CT/+  (0.71 ± 0.15; P>0.05) and Cx43  CT/-  (0.76 ± 0.13;  P >0.05) tissue. Because gap junctional intercellular communication has been proposed to play a role in stroke (Contreras et al. 2004; Rossi et al. 2007), confluent cultures of astrocytes were subjected to scrape load assays. In comparison with Cx43+/+ cultures, where dye passage extended 160 ± 8 m from the scrape edge, Cx43+/- and Cx43  CT/+  cells exhibited reduced levels of coupling (100  ± 4 m and 138 ± 10 m, respectively). Coupling was further reduced in Cx43 Cx43  CT/ CT  CT/-  and  astrocytes (74 ± 13 m and 72 ± 18 m, respectively) (Fig. 3.6A).  Double whole-cell voltage clamp was used to study the electrophysiological properties of Cx43 and Cx43 CT. Representative current traces from astrocytes expressing full length Cx43 are shown in Fig. 3.6B1,B2. The difference in current calculated from the mean of the peak currents indicate that the total opening of the channel was 110 pS with a residual state of 20pS. Astrocytes expressing Cx43 CT formed gap junction channels with different electrophysiological properties. As shown in the trace in Fig 3.6B3,B4 both recorded channel openings show a conductance of 108 and 113 pS with no 20 pS residual conductance. Moreover, the open time of these channels is much longer, indicating a reduced sensitivity to transjunctional voltage. These are properties previously reported for C-terminal truncated Cx43 channels in 122  cardiomyocytes and transfected mammalian tumour cells (Maass et al. 2007; Moreno et al. 2002). To determine the extent of hemichannel activity, astrocytes were exposed to the hemichannel-permeable dye propidium iodine under normal and Ca2+-free conditions. Dye uptake was minimal in the presence of Ca2+, as extracellular Ca2+ blocks hemichannels (Fig. 3.6C). To determine if uptake was specific for hemichannels and not a result of cell death, the gap junction blocker CBX was applied. Dye uptake was largely reduced in the presence of this blocker (Fig. 3.6C) suggesting that the observed responses are mediated by hemichannels. For analysis, the average number of dead cells (i.e. labeled cells under Ca2+-free conditions with CBX) was subtracted from the average number of labeled cells under Ca2+-free conditions in the absence of CBX. Hemichannel activity was greater in Cx43+/ and Cx43  CT/-  CT  astrocytes and less in Cx43+/-  astrocytes in comparison to wild-type astrocytes. No further change in activity  was observed in Cx43  CT/ CT  astrocytes, relative to Cx43  CT/-  astrocytes.  Ca2+ signaling is important in astrocyte signaling and blood vessel regulation (Mulligan and MacVicar 2004). Ca2+ waves were generated in culture by mechanical stimulation. The distance of wave travel was reduced in Cx43  CT/-  and Cx43  CT/ CT  cultures (Fig. 3.6D1). The  fraction of cells activated by mechanical stimulation was reduced in Cx43+/-, Cx43 Cx43  CT/ CT  CT/-  and  cultures (Fig. 3.6D2), and wave velocity was unchanged in all genotypes tested  (Scemes et al. 1998) (Fig. 3.6D3). Representative figures of Ca2+ waves in Cx43+/- and Cx43  CT/-  3.4  Discussion  cultures are shown in Fig. 3.6E1,E2.  Astrocytes are known to protect neurons during brain ischemia (Rossi et al. 2007). During MCAO a necrotic core develops which is surrounded by a penumbral region. It is 123  thought that astrocytes limit the expansion of cell death in the penumbra by buffering cytotoxic substances such as free radicals, glutamate and K+, and by providing nutrients such as antioxidants, glucose and ATP in an attempt to reduce cellular damage (Contreras et al. 2004; Rossi et al. 2007). It has been suggested that astrocyte Cx43 plays a role in reducing lesion expansion during stroke (Nakase et al. 2003a; Nakase et al. 2004). The C-terminal of Cx43 is a site of many interacting molecules and is known to be important in channel gating (Herve et al. 2007; Lampe and Lau 2004; Maass et al. 2007; Moreno et al. 2002). We report that Cx43 CT mice display increased stroke damage, suggesting disruption of the terminal portion of Cx43 significantly alters the ability of astrocytes to protect the brain, and that full-length Cx43 is important in attenuating brain damage. During stroke one of the initial events in neurons is a decrease in ATP levels that leads to disruption of ion gradients, including rises in extracellular K+, intracellular Na+, intracellular Ca2+ and a decrease in pH. Because astrocytes maintain ATP levels longer, they are less susceptible to ionic disruption (Rossi et al. 2007; Silver et al. 1997), but this balance is shortlived when ischemia persists. A reduction in pH is known to close gap junctions, and this gating involves the C-terminal region (Duffy et al. 2004; Duffy et al. 2002; Liu et al. 1993; Morley et al. 1996). During stroke in wild-type mice, astrocytes would have a decrease in pH that would lead to reduced coupling. This could be beneficial as the passage of cytotoxic molecules to neighboring cells would be reduced (Lin et al. 1998), but channel gating is impaired in the Cx43 CT mice. Interestingly, in this study, the infarct volume in Cx43 to that in the Cx43  CT/-  CT/+  mice was equivalent  mice, suggesting that the Cx43 CT mutation has a dominant negative  role in ischemic damage. Therefore, passage of cytotoxic molecules could have lead to enhanced infarct volume observed in these mutants. This scenario does not necessarily negate our previous findings, where reduced expression of Cx43 (hence, less passage of cytotoxic 124  molecules) has been shown to be destructive during MCAO (Nakase et al. 2003b; Nakase et al. 2004; Siushansian et al. 2001). It is possible that in the early stages of ischemia, gap junction coupling is important for aiding cell survival by allowing the passage of protective molecules such as glucose and antioxidants. It is likely that the ratio of healthy to non-healthy cells determines whether gap junctions are protective (Rossi et al. 2007), and in the initial stages of stroke, there are more healthy astrocytes in comparison with later stages. It is unlikely that changes in blood vessel structure contributed to the differences in infarct sizes in this study because there were no obvious vessel abnormalities or differences observed when the vessels were filled with India ink. Furthermore, the brain tissue in each of the genotypes under study appeared well perfused by microvessels. In addition, another study has shown that after MCAO, there were no differences in cerebral blood flow between Cx30–/– Cx43fl/fl:hGFAP–Cre knockout and wild-type mice (Lin et al. 2008). Infarct sizes were also unlikely to have been affected by migration of new neural tissue elements at the early time points studied because under conditions that promote generation of new cortical tissue following stroke, NeuN and BrdU labeled cells were not detectable in infarcts at 3 days after and were not significantly appreciable until approximately 11 days after stroke (Kolb et al. 2007). Our results indicate that in the mutants under study, microglia are increased in the periinfarct region of the stroke and that astrogliosis is reduced. Shortly after stroke, inflammatory processes and astrogliosis begin. Cx43 expression is high in infarct regions (Nakase et al. 2006), and it is known to increase in reactive and proliferating astrocytes after stroke (Haupt et al. 2007). The exact role of Cx43 in reactive gliosis is not well described, but it could potentially be involved in coordination of scar formation or through interaction with purinergic receptors involved in cell proliferation and migration (Haupt et al. 2007; Scemes 2008; Scemes et al. 2003). Cx43-null mice are known to have reduced expression of P2Y1 receptors (Scemes et al. 125  2003), and this interaction is believed to involve the C-terminal domain of Cx43 (Scemes 2008). Given the importance of C-terminal involvement, it is not surprising that reduced astrogliosis was found in both Cx43 knockouts (Nakase et al. 2003a; Nakase et al. 2004) and in the Cx43 CT mutants used in this study. Activated microglia are known to decrease astrocytic gap junction intercellular communication and increase hemichannel activity (Kielian and Esen 2004; Meme et al. 2006; Retamal et al. 2007). Because microglial responses were increased in Cx43 CT mutants, it is possible that altered communication in the Cx43 CT mutants caused a failure to control microglial recruitment that might lead to aggravation of brain tissue injury. Recently, in a lung inflammation model it has been shown that Cx43 CT mice have increased inflammation and neutrophil recruitment (Sarieddine et al. 2009). Interestingly, activated microglia are known to decrease the GFAP expression (Rohl et al. 2007) which correlates with our present findings. Furthermore, microglia themselves are also known to express Cx43 and to couple (Eugenin et al. 2001). Therefore, any disruption of Cx43 (i.e. either reduction of Cx43 or alteration in the C-terminal region) could alter microglia-microglia or microglia-astrocyte signaling processes, thereby resulting in an altered astroglial response. The exact mechanisms involving a Cx43 C-terminal domain interaction require further study. Full-length Cx43 was found to be reduced in Cx43 Cx43  CT/ CT  and Cx43  CT/+  and absent in Cx43  brain tissue. Cx43 CT protein was abundantly expressed in Cx43  CT/ CT  CT/-  CT/+  and  , Cx43  CT/-  tissue, but despite Cx43 CT robust expression (as detected by Western blot),  gap junction plaque formation was reduced. A change in localization and reduction in plaque formation in the heart has also been described in Cx43  CT/-  mice (Maass et al. 2007). The larger  plaques are likely by a longer Cx43 half-life (Maass et al. 2004), potentially because of altered degradation, as the C-terminal has been shown to interact with the ubiquitin degradation system (Li et al. 2008). 126  Our data indicates that coupling is reduced in Cx43 CT-expressing astrocytes. Although electrophysiological recordings indicate that Cx43 CT has a prolonged opening time, the reduced coupling likely reflects a decrease in Cx43 at the plasma membrane. It is important to note that even though Cx43 CT expression is reduced in the heart in these mutants, whole-cell conduction between cardiomyocytes is not significantly altered in Cx43 CT mutants, indicating that there is a large safety factor for proper cell-cell communication under normal conditions (Maass et al. 2007). In addition, heart rates do not differ between Cx43+/- and Cx43  CT/-  mice,  also indicating that electrical conduction is not altered in these mutants (Maass et al. 2004). Astrocyte Ca2+ signaling is known to regulate arteriole diameter (Mulligan and MacVicar 2004). This may be important in vivo for perfusion of the penumbral region of a stroke. Our data indicate that in Cx43  CT/-  astrocytes, Ca2+ wave distance and efficacy were reduced.  Conversely, the Ca2+ wave properties studied were unchanged in Cx43  CT/+  astrocytes. Ca2+  wave propagation depends on both hemichannel activity and gap junction communication (Scemes et al. 1998). Hence the reduction in coupling observed in these cells may be compensated for by an increase in hemichannel activity. Hemichannel activity after preconditioning has been proposed to be protective in MCAO, however, in the absence of a preconditioning stimulus (as in the present study), hemichannel activity is low (Lin et al. 2008). In Ca2+-free conditions, only a small proportion of astrocytes displayed hemichannel activity, whereas Cx43  CT/+  astrocytes displayed a slight increase in  activity. Notably, at physiological Ca2+ concentrations, hemichannel activity was negligible (Saez et al. 2005). Extrapolating to our stroke data, it is unlikely that increased hemichannel activity plays a role in Cx43 CT stroke size as increased activity would be expected to decrease infarct volume (Lin et al. 2008) rather than increase it as we found.  127  In summary, we found that truncation of the C-terminal of Cx43 increases brain damage caused by experimental MCAO. The consequences of this mutation were broad and included alteration of gliosis, microglial/macrophage response, coupling, channel gating, hemichannel activity, Ca2+ waves and Cx43 expression and localization. Based on comparisons between Cx43+/+ and Cx43-/- astrocytes (Iacobas et al. 2003), it is also likely that changes in gene expression occur as well. Determining which of these processes is most important in stroke pathology is complex, and there are differences between the in vitro and in vivo systems under study (i.e. Cx30 expression in the latter). Unlike our studies with Cx43+/- mice (Siushansian et al. 2001), however, truncation of the C-terminal involves disruption of the gating region of Cx43 without the approximately 50% reduction in Cx43 expression observed in the Cx43+/- studies. This CT mutant allows us to study whether the pore of Cx43 is sufficient for protection or whether the C-terminal region is necessary for protection. Gap junction activity and/or interactions at the CT may affect the ultimate outcome of a stroke.  128  Figure 3.1. The Cx43 CT mutation results in increased infarct size after MCAO. Infarct volumes after MCAO were increased in Cx43+/-, Cx43 CT/+ and Cx43 CT/- mice compared with Cx43+/+ mice (*P<0.05). Infarcts were also larger in Cx43 CT/+ mice compared with Cx43+/+ mice (†P<0.05); and Cx43 CT/- infarcts were larger when compared with Cx43+/-infarcts (‡P<0.05). Numbers of mice in each group are indicated. Representative coronal sections showing infarcts corresponding to each group are shown above the graph.  129  Figure 3.2. Reactive astrogliosis is decreased in Cx43 CT mice. A, Staining for GFAP revealed a decrease in the extent of astrogliosis in the peri-infarct region of Cx43+/-, Cx43 CT/+ and Cx43 CT/- mice. B, A graph quantifying the degree of astrogliosis indicated by the distance of the gliosis (extent) from the necrotic core in each of the genotypes tested (*P<0.05 in comparison to Cx43+/+). There was no significant difference between Cx43 CT/+ and Cx43 CT/- mice, P>0.05). Numbers of animals in each group are indicated.  130  Figure 3.3. The inflammatory reaction is increased in Cx43 CT mice. A, IBA-1 staining (red) revealed an increase in microglia/macrophages in the peri-infarct region of Cx43+/-, Cx43 CT/+ and Cx43 CT/- mice compared to control mice. Nuclei are labeled with DAPI (blue). B, A graph quantifying the number of microglia/macrophages in each of the genotypes tested (*P<0.05 in comparison to Cx43+/+. There was no significant difference observed between Cx43 CT/+ and Cx43 CT/- mice, P>0.05). Numbers of mice in each group are indicated.  131  Figure 3.4. Immunohistochemistry for Cx43 in Cx43 CT mutants. A, Cultured astrocytes from mice with the indicated genotypes were stained for Cx43 using an antibody raised against the C-terminal of Cx43 (green). Nuclei are labeled with DAPI (blue). Staining was absent in Cx43 CT/- and Cx43 CT/ CT astrocytes because of lack of the C-terminal. B, Cultured astrocytes were also stained with an antibody against the cytoplasmic loop of Cx43. Cx43 plaques appeared to form in all genotypes, but expression was reduced in Cx43+/-, Cx43 CT/+, Cx43 CT/- and Cx43 CT/ CT genotypes. C, Brain sections from the cortical region in the distribution of the MCA from the indicated genotypes were stained for Cx43 (green) using an antibody raised against the C-terminal of Cx43 and the intracellular loop (D), staining appeared similar to that observed in culture. Because Cx43 CT/ CT mice are not viable, only cultured astrocytes could be evaluated.  132  Figure 3.5. Western blot of adult cortex and astrocyte culture tissue. Samples obtained from the cortex of adult mice or cultured astrocytes were stained with the antiCx43 C-terminal, anti-Cx43 N-terminal, anti-Cx30 or anti-GAPDH antibodies. Staining with the C-terminal antibody was observed in Cx43+/+, Cx43+/- and Cx43 CT/+ samples and absent in Cx43 CT/- samples. Using the N-terminal Cx43 antibody, staining of full length (*) and truncated (**) Cx43 could be observed. In adults, Cx30 expression was similar across the genotypes tested and absent in cultured tissue. GAPDH served as a loading control.  133  Figure 3.6. Gap junction coupling, electrophysiology, hemichannel activity and Ca2+ waves in culture. A1, Gap junctional communication was measured by the distance of dye passage through scraped confluent cultures. A2, A graph quantifying the distance of dye passage, gap junction coupling was reduced in each of the genotypes tested compared with wild-type (*P<0.05). B1, Junctional currents during an 80 mV pulse for 10 seconds from astrocytes expressing full-length Cx43. B2, Digitized points obtained from the region indicated by the bar in B1 were grouped in 0.025 bins and plotted in the histogram. The main peaks were fitted to Gaussian equations and the differences between the means indicated a channel total conductance of 110 pS and a residual of 20pS. B3. Junctional currents in Cx43 CT tissue as recorded in B1. B4, Digitized points obtained from the region indicated by the bar in B3 were grouped in 0.05 bins and plotted in the histogram. The main peaks were fitted to Gaussian equations and the differences between the means indicated channel total conductances of 108 and 113 pS and no residual currents were recorded. C, In comparison to Cx43+/+ astrocytes hemichannel activity is increased in Cx43 CT/+ cells (*P<0.05) under Ca2+ free conditions (-Ca2+). Under -Ca2+ conditions Cx43+/- and Cx43 CT/- had reduced hemichannel activity (*P<0.05), whereas hemichannel activity was not altered in Cx43 CT/ CT astrocytes (P>0.05). Specificity of dye uptake through hemichannels was tested using the gap junction blocker CBX. Data under Ca2+ conditions (+Ca2+) are also plotted. D1, The distance of Ca2+ wave travel was less in Cx43+/-, Cx43 CT/- and Cx43 CT/ CT cultures; D2, The efficacy of Ca2+ wave generation caused by mechanical stimulation was greater in Cx43+/+ than in Cx43 CT/- and Cx43 CT/ CT cultures; D3, The velocity of Ca2+ wave propagation remained unchanged in each of the genotypes. E1, A representative figure of a Ca2+ wave experiment in Cx43+/- astrocyte cultures 20 seconds after stimulation at the site indicated (*). The dashed line indicates the region of astrocytes activated at this time point. E2, A similar figure to E1 showing a Ca2+ wave in Cx43 CT/- astrocytes.  134  135  3.5  References  Bates DC, Sin WC, Aftab Q, Naus CC. 2007. Connexin43 enhances glioma invasion by a mechanism involving the carboxy terminus. Glia 55(15):1554-64. Berry K, Wisniewski HM, Svarzbein L, Baez S. 1975. On the relationship of brain vasculature to production of neurological deficit and morphological changes following acute unilateral common carotid artery ligation in gerbils. J Neurol Sci 25(1):75-92. Contreras JE, Sanchez HA, Veliz LP, Bukauskas FF, Bennett MV, Saez JC. 2004. Role of connexin-based gap junction channels and hemichannels in ischemia-induced cell death in nervous tissue. Brain Res Brain Res Rev 47(1-3):290-303. Cuadros MA, Navascues J. 1998. The origin and differentiation of microglial cells during development. Prog Neurobiol 56(2):173-89. Dobrenis K, Chang HY, Pina-Benabou MH, Woodroffe A, Lee SC, Rozental R, Spray DC, Scemes E. 2005. Human and mouse microglia express connexin36, and functional gap junctions are formed between rodent microglia and neurons. J Neurosci Res 82(3):30615. Duffy HS, Ashton AW, O'Donnell P, Coombs W, Taffet SM, Delmar M, Spray DC. 2004. Regulation of connexin43 protein complexes by intracellular acidification. Circ Res 94(2):215-22. Duffy HS, Sorgen PL, Girvin ME, O'Donnell P, Coombs W, Taffet SM, Delmar M, Spray DC. 2002. pH-dependent intramolecular binding and structure involving Cx43 cytoplasmic domains. J Biol Chem 277(39):36706-14. Ekdahl CT, Kokaia Z, Lindvall O. 2009. Brain inflammation and adult neurogenesis: the dual role of microglia. Neuroscience 158(3):1021-9. Eugenin EA, Eckardt D, Theis M, Willecke K, Bennett MV, Saez JC. 2001. Microglia at brain stab wounds express connexin43 and in vitro form functional gap junctions after treatment with interferon-gamma and tumor necrosis factor-alpha. Proc Natl Acad Sci U S A 98(7):4190-5. Giepmans BN, Hengeveld T, Postma FR, Moolenaar WH. 2001. Interaction of c-src with gap junction protein connexin43. Role in the regulation of cell-cell communication. J Biol Chem 276(11):8544-9. Giepmans BN, Moolenaar WH. 1998. The gap junction protein connexin43 interacts with the second PDZ domain of the zona occludens-1 protein. Curr Biol 8(16):931-4. Haupt C, Witte OW, Frahm C. 2007. Up-regulation of connexin43 in the glial scar following photothrombotic ischemic injury. Mol Cell Neurosci 35(1):89-99. Hekmatpanah J. 2007. Cerebral microvessel perfusion and pathologic alteration of the brain during drowsiness and coma caused by brain tumor: a laboratory study on rats. Surg Neurol 67(6):564-71 136  Herve JC, Bourmeyster N, Sarrouilhe D, Duffy HS. 2007. Gap junctional complexes: from partners to functions. Prog Biophys Mol Biol 94(1-2):29-65. Iacobas DA, Urban-Maldonado M, Iacobas S, Scemes E, Spray DC. 2003. Array analysis of gene expression in connexin43 null astrocytes. Physiol Genomics 15(3):177-90. Kang J, Kang N, Lovatt D, Torres A, Zhao Z, Lin J, Nedergaard M. 2008. Connexin43 hemichannels are permeable to ATP. J Neurosci 28(18):4702-11. Kieken F, Mutsaers N, Dolmatova E, Virgil K, Wit AL, Kellezi A, Hirst-Jensen BJ, Duffy HS, Sorgen PL. 2009. Structural and molecular mechanisms of gap junction remodeling in epicardial border zone myocytes following myocardial infarction. Circ Res 104(9):110312. Kielian T, Esen N. 2004. Effects of neuroinflammation on glia-glia gap junctional intercellular communication: a perspective. Neurochem Int 45(2-3):429-36. Kolb B, Morshead C, Gonzalez C, Kim M, Gregg C, Shingo T, Weiss S. 2007. Growth factorstimulated generation of new cortical tissue and functional recovery after stroke damage to the motor cortex of rats. J Cereb Blood Flow Metab 27(5):983-97. Lalancette-Hebert M, Gowing G, Simard A, Weng YC, Kriz J. 2007. Selective ablation of proliferating microglial cells exacerbates ischemic injury in the brain. J Neurosci 27(10):2596-605. Lampe PD, Lau AF. 2000. Regulation of gap junctions by phosphorylation of connexins. Arch Biochem Biophys 384(2):205-15. Lampe PD, Lau AF. 2004. The effects of connexin phosphorylation on gap junctional communication. Int J Biochem Cell Biol 36(7):1171-86. Lampe PD, TenBroek EM, Burt JM, Kurata WE, Johnson RG, Lau AF. 2000. Phosphorylation of connexin43 on serine368 by protein kinase C regulates gap junctional communication. J Cell Biol 149(7):1503-12. Li X, Su V, Kurata WE, Jin C, Lau AF. 2008. A novel connexin43-interacting protein, CIP75, which belongs to the UbL-UBA protein family, regulates the turnover of connexin43. J Biol Chem 283(9):5748-59. Lin JH, Lou N, Kang N, Takano T, Hu F, Han X, Xu Q, Lovatt D, Torres A, Willecke K and others. 2008. A central role of connexin43 in hypoxic preconditioning. J Neurosci 28(3):681-95. Lin JH, Weigel H, Cotrina ML, Liu S, Bueno E, Hansen AJ, Hansen TW, Goldman S, Nedergaard M. 1998. Gap-junction-mediated propagation and amplification of cell injury. Nat Neurosci 1(6):494-500. Liu S, Taffet S, Stoner L, Delmar M, Vallano ML, Jalife J. 1993. A structural basis for the unequal sensitivity of the major cardiac and liver gap junctions to intracellular acidification: the carboxyl tail length. Biophys J 64(5):1422-33. 137  Maass K, Ghanem A, Kim JS, Saathoff M, Urschel S, Kirfel G, Grummer R, Kretz M, Lewalter T, Tiemann K and others. 2004. Defective epidermal barrier in neonatal mice lacking the C-terminal region of connexin43. Mol Biol Cell 15(10):4597-608. Maass K, Shibayama J, Chase SE, Willecke K, Delmar M. 2007. C-terminal truncation of connexin43 changes number, size, and localization of cardiac gap junction plaques. Circ Res 101(12):1283-91. Meme W, Calvo CF, Froger N, Ezan P, Amigou E, Koulakoff A, Giaume C. 2006. Proinflammatory cytokines released from microglia inhibit gap junctions in astrocytes: potentiation by beta-amyloid. Faseb J 20(3):494-6. Moreno AP, Chanson M, Elenes S, Anumonwo J, Scerri I, Gu H, Taffet SM, Delmar M. 2002. Role of the carboxyl terminal of connexin43 in transjunctional fast voltage gating. Circ Res 90(4):450-7. Morley GE, Taffet SM, Delmar M. 1996. Intramolecular interactions mediate pH regulation of connexin43 channels. Biophys J 70(3):1294-302. Mulligan SJ, MacVicar BA. 2004. Calcium transients in astrocyte endfeet cause cerebrovascular constrictions. Nature 431(7005):195-9. Nakase T, Fushiki S, Naus CC. 2003a. Astrocytic gap junctions composed of connexin43 reduce apoptotic neuronal damage in cerebral ischemia. Stroke 34(8):1987-93. Nakase T, Fushiki S, Sohl G, Theis M, Willecke K, Naus CC. 2003b. Neuroprotective role of astrocytic gap junctions in ischemic stroke. Cell Commun Adhes 10(4-6):413-7. Nakase T, Sohl G, Theis M, Willecke K, Naus CC. 2004. Increased apoptosis and inflammation after focal brain ischemia in mice lacking connexin43 in astrocytes. Am J Pathol 164(6):2067-75. Nakase T, Yoshida Y, Nagata K. 2006. Enhanced connexin43 immunoreactivity in penumbral areas in the human brain following ischemia. Glia 54(5):369-75. Naus CC, Bechberger JF, Zhang Y, Venance L, Yamasaki H, Juneja SC, Kidder GM, Giaume C. 1997. Altered gap junctional communication, intercellular signaling, and growth in cultured astrocytes deficient in connexin43. J Neurosci Res 49(5):528-40. Neijssen J, Herberts C, Drijfhout JW, Reits E, Janssen L, Neefjes J. 2005. Cross-presentation by intercellular peptide transfer through gap junctions. Nature 434(7029):83-8. Ozog MA, Siushansian R, Naus CC. 2002. Blocked gap junctional coupling increases glutamateinduced neurotoxicity in neuron-astrocyte co-cultures. J Neuropathol Exp Neurol 61(2):132-41. Pekny M, Nilsson M. 2005. Astrocyte activation and reactive gliosis. Glia 50(4):427-34. Retamal MA, Froger N, Palacios-Prado N, Ezan P, Saez PJ, Saez JC, Giaume C. 2007. Cx43 hemichannels and gap junction channels in astrocytes are regulated oppositely by 138  proinflammatory cytokines released from activated microglia. J Neurosci 27(50):1378192. Rohl C, Lucius R, Sievers J. 2007. The effect of activated microglia on astrogliosis parameters in astrocyte cultures. Brain Res 1129(1):43-52. Rossi DJ, Brady JD, Mohr C. 2007. Astrocyte metabolism and signaling during brain ischemia. Nat Neurosci 10(11):1377-86. Saez JC, Retamal MA, Basilio D, Bukauskas FF, Bennett MV. 2005. Connexin-based gap junction hemichannels: gating mechanisms. Biochim Biophys Acta 1711(2):215-24. Sarieddine MZ, Scheckenbach KL, Foglia B, Maass K, Garcia I, Kwak BR, Chanson M. 2009. Connexin43 modulates neutrophil recruitment to the lung. J Cell Mol Med. Scemes E. 2008. Modulation of astrocyte P2Y1 receptors by the carboxyl terminal domain of the gap junction protein Cx43. Glia 56(2):145-53. Scemes E, Dermietzel R, Spray DC. 1998. Calcium waves between astrocytes from Cx43 knockout mice. Glia 24(1):65-73. Scemes E, Duval N, Meda P. 2003. Reduced expression of P2Y1 receptors in connexin43-null mice alters calcium signaling and migration of neural progenitor cells. J Neurosci 23(36):11444-52. Silver IA, Deas J, Erecinska M. 1997. Ion homeostasis in brain cells: differences in intracellular ion responses to energy limitation between cultured neurons and glial cells. Neuroscience 78(2):589-601. Siushansian R, Bechberger JF, Cechetto DF, Hachinski VC, Naus CC. 2001. Connexin43 null mutation increases infarct size after stroke. J Comp Neurol 440(4):387-94. Stridh MH, Tranberg M, Weber SG, Blomstrand F, Sandberg M. 2008. Stimulated efflux of amino acids and glutathione from cultured hippocampal slices by omission of extracellular calcium: likely involvement of connexin hemichannels. J Biol Chem 283(16):10347-56. Talhouk RS, Zeinieh MP, Mikati MA, El-Sabban ME. 2008. Gap junctional intercellular communication in hypoxia-ischemia-induced neuronal injury. Prog Neurobiol 84(1):5776. Venance L, Piomelli D, Glowinski J, Giaume C. 1995. Inhibition by anandamide of gap junctions and intercellular calcium signalling in striatal astrocytes. Nature 376(6541):590-4. Walz W. 1999. Cerebral Ischemia: Molecular and Cellular Pathophysiology. Totowa, New Jersey: Humana Press. Zucker SN, Nicholson BJ. 2002. Mutagenic approaches to modifying gap junction phenotype. Curr Drug Targets 3(6):441-53. 139  Chapter 4: Truncation of the C-terminal region of connexin43 and mitochondrial potassium sequestration.5  5  A version of this chapter is nearly ready to be submitted for publication in a refereed academic journal. Kozoriz, MG. and Naus, CC. (2010) Truncation of the C-terminal region of connexin43 and mitochondrial potassium sequestration.  140  4.1  Introduction In the brain, baseline extracellular K+ concentration ([K+]o) is ~3 mM, which can increase  to 25 - 80 mM during ischemia and spreading depression (Hansen 1985; Walz 2000). Rises in [K+]o alter neuronal excitability, synaptic transmission and neurotransmitter re-uptake, which may have pathological consequences (reviewed by Somjen 2002; Walz 2000). Astrocytes are known to play an important role in handling [K+]o (reviewed by Leis et al. 2005; Walz 2000). These cells take up K+ across the plasma membrane via Na+,K+-ATPase, Na+-K+-2Cl- cotransport and voltage-activated K+ channel (Kir, Kdr and Ka) pathways (Bordey and Sontheimer 1997; Djukic et al. 2007; Kofuji and Newman 2004; Kressin et al. 1995; Ljubkovic et al. 2006; Seifert et al. 2009; Walz 2000). Once inside the cell, K+ can be sequestered by mitochondria (see chapter 2), and it is also thought that K+ is buffered to areas of lower K+ via gap junction coupled astrocytes (Kofuji and Newman 2004; Leis et al. 2005; Orkand et al. 1966; Wallraff et al. 2006; Walz 2000; Xu et al. 2009; also see Appendix A - Kozoriz et al. 2006). Connexin43 (Cx43) is the primary gap junction protein in astrocytes (Dermietzel et al. 1991; Dermietzel et al. 1989; Yamamoto et al. 1990). In addition to its plasma membrane localization, Cx43 localizes to mitochondria (Boengler et al. 2005; Boengler et al. 2007; Boengler et al. 2009; Goubaeva et al. 2007; Li et al. 2002; Miro-Casas et al. 2009; RodriguezSinovas et al. 2006; also see chapter 2) and is important in mitochondrial K+ sequestration (see chapter 2). Although the study of mitochondrial Cx43 is in its infancy, Cx43 has been shown to form hemichannels in the inner mitochondrial membrane that are involved in K+ uptake (MiroCasas et al. 2009). Disruption of Cx43 expression is known to increase infarct volume following middle cerebral artery occlusion (MCAO) (Nakase et al. 2003; Nakase et al. 2004; Siushansian et al. 2001), and truncation of the C-terminal region of Cx43 also increases infarct volume (see chapter 141  3). The C-terminal region of Cx43 is an important site for channel activity regulation (Duffy et al. 2002; Herve et al. 2007; Lampe and Lau 2004). Given the role of Cx43 in mitochondrial K+ sequestration, we were interested in determining if truncation of the gating region of Cx43 alters mitochondrial K+ sequestration. In this study, we compare mitochondrial K+ sequestration in cultured astrocytes obtained from wild-type (Cx43+/+), heterozygote (Cx43+/-), and mice expressing a Cx43-null mutation and a truncated Cx43 allele, where the C-terminal region had been truncated at amino acid residue 258 (Cx43  CT/-  ) (Maass et al. 2004; Maass et al. 2007). In comparison to Cx43+/+ preparations,  mitochondrial K+ sequestration was similarly reduced in Cx43+/- and Cx43  CT/-  astrocytes. As  mitochondrial K+ uptake is dependent on the mitochondria membrane potential ( voltages were assessed in each genotype. No differences in  m  m),  baseline  were observed suggesting  voltage differences do not explain why K+ uptake is altered in the mutants. A reduction in K+ sequestration in Cx43+/- and Cx43  CT/-  astrocytes could, in part, explain  why stroke is larger in these mutants in comparison to Cx43+/+ mice. However, stroke is larger in Cx43  CT/-  mice compared to Cx43+/- mice (see chapter 3), suggesting other pathways must be  important in determining the overall extent of infarct volume. Based on our current understanding of the role of Cx43 in cellular injury, a mechanism involving gap junction coupling and K+ sequestration is proposed.  4.2  Materials and methods All experiments were performed in accordance with the guidelines established by the  Canadian Council on Animal Care and were approved by The University of British Columbia Animal Care Committee.  142  Knockout mice and cortical astrocyte culture. Mice from a 129/Ola/C57BL/6 genetic background in which Cx43 has been replaced by Cx43K258stop (referred to here as Cx43 CT) were used in this study (Maass et al. 2004). Breeding and genotyping of Cx43+/+, Cx43+/- and Cx43  CT/-  mice have been previously described (Maass et al. 2004; Maass et al. 2007; Naus et al.  1997). Brains from 0-1 day old pups were placed in phosphate-buffered saline. Neocortices were isolated, freed of meninges and placed in high glucose Dulbecco modified Eagle medium (DMEM; Invitrogen, Burlington, Canada) supplemented with 10% fetal bovine serum (FBS; HyClone, Logan, UT), 10 U/mL penicillin, and 10 U/mL streptomycin (Invitrogen). A portion of brain tissue was also collected for genotyping. Cortical tissue was mechanically disrupted and passed through a 70 m cell strainer (BD Falcon, Bedford, MA). Cells were re-suspended in high glucose DMEM supplemented with 10% FBS, 10 U/mL penicillin and 10 U/mL streptomycin, and maintained in a humidified incubator in 95% air/5% CO2 at 37°C. A fresh medium change was performed every 4 days (Ozog et al. 2002). Confluent cells were passed onto poly-D-lysine-coated glass coverslips in 24 well plates and were fed every 4 days. Experiments were performed at 14-21 days in vitro. Measurement of cytoplasmic K+ concentration. Standard perfusion medium contained (in mM): 136.5 NaCl, 3 KCl, 1.5 NaH2PO4, 1.5 MgSO4, 10 D-glucose, 2 CaCl2 and 10 HEPES (titrated to pH 7.35 with 10 M NaOH), and perfusion media containing 12.5 mM K+ were prepared by equimolar substitution for NaCl. Astrocytes were placed for 1 hour at room temperature (20 - 22°C) in standard perfusion medium containing 5 M of the K+-sensitive fluorophore potassium-binding benzofuran isophthalate acetoxymethyl ester (PBFI) and 0.05% pluronic F-127 (Molecular Probes, Eugene, OR). Cells were then placed in fluorophore-free medium for 30 min to ensure de-esterification of the fluorophore and then mounted in a perfusion chamber on a Zeiss Axioskop2 FS Plus microscope (Carl Zeiss Canada, ON). 143  Astrocytes were continuously superfused at ~2 mL/min at room temperature throughout the course of an experiment unless otherwise indicated. Measurements of [K+]i with PBFI were performed using the dual excitation ratio method. Fluorescence emissions >505 nm were captured by a 12-bit digital cooled CCD camera (Retiga EXi, QImaging, Burnaby, BC) from regions of interest placed on individual astrocytes. Raw emission intensity data at each excitation wavelength (340 and 380 nm; Lambda DG-5, Sutter Instrument Co., Novato, CA) were collected every 6 s, corrected for background fluorescence, and background-subtracted ratio pairs (BI340/BI380) were analyzed off-line. Measurement of mitochondrial K+ sequestration in situ by application of CCCP. To examine mitochondrial K+ content, the mitochondrial membrane uncoupler carbonyl cyanide mchlorophenylhydrazone (CCCP, 10 M; Sigma-Aldrich Canada, ON) was applied for 2 min to reversibly collapse  m.  This caused efflux of K+ from mitochondria that was detected by  cytoplasmic PBFI (see chapter 2). [K+]o was then increased to 12.5 mM for 5 min, followed by another 2 minute application of CCCP. The increase in magnitude of CCCP-induced [K+]i transient following application of elevated [K+]o compared to control CCCP-induced [K+]i transient reflected an increase in mitochondrial K+ sequestration (see chapter 2). Results for these experiments are reported as percent change in peak response between the two conditions (i.e. control vs. 12.5 mM K+ treatment). Measurement of mitochondrial K+. Measurements of mitochondrial K+ were performed using a PBFI loading protocol which favors mitochondrial dye localization (Ljubkovic et al. 2006; Miro-Casas et al. 2009; Xu et al. 2002; also see chapter 2). In brief, astrocytes were placed for 1 hour at 37 oC in standard perfusion medium with 10 M PBFI and 0.05% pluronic F-127. Cells were then placed in fluorophore-free medium for 30 min. To permeabilize the plasma membrane and release PBFI from the cytoplasmic compartment, cells were treated with 144  50 g/ml saponin (Sigma) for 5 min (Xu et al. 2002) in a mitochondria medium containing (in mM): 250 sucrose, 25 Tris, 3 EGTA, 5 MgCl2, 5 succinate and 5 glutamate (pH 7.3) (Ljubkovic et al. 2006). Preparations were then incubated in a medium containing (in mM): 200 sucrose, 50 TEA, 25 Tris, 3 EGTA, 5 MgCl2, 5 succinate and 5 glutamate (pH 7.3) for 20 min to deplete mitochondrial K+ (Miro-Casas et al. 2009). Mitochondria from the permeabilized astrocytes were subsequently imaged in the mitochondria medium. K+ uptake into mitochondria was assessed by measuring the rate of change of PBFI derived ratio values upon exposure to 25 mM K+ solution containing (in mM): 200 sucrose, 25 KCl, 25 Tris, 3 EGTA, 5 MgCl2, 5 succinate and 5 glutamate (pH 7.3) applied at a bath perfusion rate of 4.2 mL/min. Measurement of mitochondrial membrane potential. Astrocytes were placed in standard perfusion medium containing 5 M rhodamine-123 (R123; Molecular Probes) for 15 min and then washed in fluorophore free medium for 30 min. The dye was excited at 495 nm and emissions were collected at 515 - 565 nm. Mitochondrial voltage was assessed by completely depolarizing the  m  with 10 M CCCP. This caused release of dye from the mitochondria that  could be observed in the cytosolic region above the cell nucleus (Kahlert et al. 2005). For analysis, peak BI495 after CCCP application was subtracted from BI495 values obtained during the control portion of an experiment. As R123 may be extruded from astrocytes, in a subset of experiments mitochondria were also loaded with mitotracker red (Molecular Probes) (Buckman et al. 2001). To control for differences in R123 loading or extrusion, in these experiments R123 signals were normalized to the mitotracker red signal. Data analysis. Results are reported as means  S.E.M with the accompanying n value  referring to the number of cell populations (i.e. number of coverslips) analyzed under each experimental condition. Data were obtained simultaneously from 10-20 regions of interest per  145  coverslip and averaged. Unless otherwise indicated, statistical comparisons were performed using a one-way ANOVA with a 95% confidence limit.  4.3  Results Mitochondrial Cx43 is known to contribute to mitochondrial K+ uptake (Miro-Casas et al.  2009; also see chapter 2). As the C-terminal region of Cx43 is important in channel gating, we decided to determine if mitochondrial K+ uptake is altered in Cx43 CT astrocytes. Mitochondrial K+ uptake was assessed directly using a protocol that preferentially loads PBFI into mitochondria followed by saponin permeabilization to remove cytosolic dye (Miro-Casas et al. 2009; also see chapter 2) (Fig. 4.1A). Application of 25 mM K+ caused an increase in mitochondrial PBFI-derived BI340/BI380 ratio values in each of the genotypes (Fig. 4.1B). In comparison to Cx43+/+ astrocytes, K+ uptake rate was reduced in Cx43+/- astrocytes (Fig. 4.1C), which is consistent with reports that Cx43 is important in mitochondrial K+ uptake (Miro-Casas et al. 2009; also see chapter 2). To determine the effect of the Cx43 CT mutation, K+ uptake rates were compared between Cx43+/- and Cx43 reduced in Cx43  CT/-  CT/-  mutants. Although K+ uptake rate was  astrocytes in comparison to Cx43+/+ astrocytes, the uptake rate did not  differ from Cx43+/- astrocytes (Fig. 4.1C). Because results from isolated mitochondrial preparations may differ from experiments performed in situ (Duchen 2004; Oliveira et al. 2007), we assessed mitochondrial K+ sequestration in situ (see chapter 2). Astrocytes were loaded with PBFI using a protocol that favors dye localization to the cytosol. To assess mitochondrial K+, CCCP (10 M) was applied for 2 min (at [K+]o = 3 mM) to collapse  m  and release K+ from mitochondria into the cytosol  (Fig. 4.2A). This was compared to the amplitude of response generated by 2 min application of CCCP (again applied at [K+]o = 3 mM) when [K+]o was elevated to 12.5 mM in the preceding 5 146  min. In Cx43+/+ astrocytes, application of elevated [K+]o resulted in an increase in K+ sequestered by mitochondria (Fig. 4.2B). Using the same paradigm, K+ sequestration was found to be reduced in Cx43+/- in and Cx43  CT/-  mutants (Fig. 4.2B), and no differences in K+  sequestration were observed between Cx43+/- and Cx43  CT/-  astrocytes.  The above results suggest that disruption of Cx43 results in a reduction of mitochondrial K+ sequestration, and that truncation of the C-terminal region does not result in further changes of K+ sequestration. Because mitochondrial K+ uptake depends in part on the whether there were differences in resting  m  in Cx43+/+, Cx43  and Cx43  m, CT/-  Mitochondrial membrane potential was assessed by loading astrocytes with the dye R123 (Fig. 4.3A) and uncoupling  m  and Cx43  CT/-  astrocytes. m  sensitive  with 10 M CCCP (Fig. 4.3B). Signals were  quantified as the peak increase in BI495. No difference was found in Cx43  we tested  m  between Cx43+/+,  astrocytes (Fig. 4.3C). Because R123 is known to leak out of astrocytes a  subset of experiments was also performed by co-loading cells with mitotracker red and normalizing data to this more stable dye. Again no difference was found between Cx43+/+, Cx43  and Cx43  CT/-  astrocytes (Fig. 4.3D), suggesting that changes in  m  do not explain the  reduction in K+ sequestration in these mutants.  4.4  Discussion Astrocytes are known to play an important role in handling rises in [K+]o. We have  previously shown that mitochondria sequester rises in [K+]o through mitochondrial Cx43 channels (chapter 2). Because the C-terminal region of Cx43 is involved in channel gating we examined whether astrocytes from Cx43  CT/-  mutants had altered mitochondrial K+ sequestration.  In comparison to Cx43+/+ astrocytes it was found that K+ sequestration was similarly reduced in Cx43  and Cx43  CT/-  astrocytes. Since truncation of the C-terminal region is known to prolong 147  channel opening time (Moreno et al. 2002), K+ uptake would presumably be greater in these mutants. Contrary to this assumption, K+ uptake was similar in Cx43 One explanation is that Cx43 channels are maximally open in Cx43  and Cx43  CT/-  astrocytes.  astrocytes, hence, C-  terminal gating may not be relevant in this scenario. It is currently not known how mitochondrial Cx43 channels are regulated, but based on the wealth of data regarding plasma membrane Cx43 hemichannels, it is possible to speculate how these channels behave in mitochondria. Plasma membrane Cx43 channels are orientated with the C-terminal facing the cytosol and open at membrane potentials above +60 mV (Contreras et al. 2003). In a recent study, Cx43 was found to localize to the inner mitochondrial membrane with its C-terminal facing the inner membrane space in the direction of the cytosol (Miro-Casas et al. 2009). Mitochondria have a negative resting membrane potential, but because of their orientation, mitochondrial Cx43 hemichannels face a relative positive voltage that exceeds the +60 mV threshold described for plasma membrane hemichannels. Therefore, it is entirely possible that Cx43 hemichannels are open under resting conditions. This explanation is likely an oversimplification since mitochondrial Cx43 channel activity regulation is probably complex and likely involves numerous regulators such as pH, phosphorylation and interacting proteins. We also examined whether differences in Cx43  and Cx43  CT/-  m  could explain why K+ uptake was less in  astrocytes. Another study has shown that  m  in Cx43 knock-out  cardiomyocytes does not differ from wild-type cells (Miro-Casas et al. 2009). In line with this observation no differences in  m  were found in astrocytes from Cx43 , Cx43  astrocytes. This suggests differences in  m  and Cx43  CT/-  do not explain why K+ sequestration is reduced in  these mutants. Our lab has shown that infarct volume following MCAO is larger in Cx43+/- mice (Siushansian et al. 2001). One potential explanation is that in Cx43+/+ mice gap junctions allow 148  for the passage of protective molecules from healthy cells to the penumbral stroke region. Also, harmful molecules may be moved away from this region through gap junctions (Fig 4.4A). In Cx43+/- mice gap junction coupling is reduced and potentially these mice cannot fully benefit from gap junction communication during stroke (Fig 4.4B). As mitochondria sequester rises in [K+]o through mitochondrial Cx43 channels (chapter 2), it is possible that this also contributes to the difference in infarct volume observed in Cx43  and Cx43  mice. Impairment of  mitochondrial sequestration may result in larger rises in [K+]o, which could lead to glutamate release and neuronal death, potentially increasing infarct volume (see Fig 4.4A,B). With respect to Cx43 CT mutants, it is known that infarct volume is larger in Cx43 to Cx43  CT/-  mice in comparison  mice (chapter 3), yet K+ sequestration is similarly reduced in both of these mutants.  Although a reduction in K+ sequestration may account for some of the damage in Cx43  CT/-  mutants (Fig 4.4C), another explanation is needed to address why stroke is larger in these mutants. It is possible that changes in gap junction communication may account for the larger infarct volume in Cx43  CT/-  mice. During ischemic insults gap junctions initially remain open  and close in the later stages of cell death (Cotrina et al. 1998). During early ischemia in Cx43  CT/-  mutants, gap junction coupling may provide limited protection as in Cx43+/- mice (Fig  4.4C). During late ischemia where there are more damaged cells, Cx43 CT channels that have altered gating may allow for the propagation of harmful molecules to neighbouring healthy cells leading to larger infarct volumes (Fig 4.4D). Hemichannel activity has been shown to play a role in MCAO infarct volume in some situations. For example, hypoxic preconditioning increases Cx43 hemichannel activity resulting in neuroprotective adenosine release (Lin et al. 2008). Hemichannel activity is controversial (see Spray et al. 2006) as channels are gated by physiological levels of extracellular Ca2+. In the presence of extracellular Ca2+, Cx43 , Cx43  and Cx43  CT/-  astrocytes had nominal 149  hemichannel activity (see chapter 3). When Ca2+ was removed from the extracellular solution hemichannel activity was slightly increased in these genotypes, but Cx43 astrocytes had comparatively less activity than Cx43  and Cx43  CT/-  astrocytes. In culture, Cx43  hemichannels open in dose-dependent manner in response to reductions in [Ca2+]o, and it has been shown that hemichannel activity is maximal at 10 M [Ca2+]o (Li et al. 1996). It is possible that a decline in [Ca2+]o during ischemia (see Fig. 1.2) is sufficient to open hemichannels. A potential role for hemichannels in ischemic damage is also summarized in Fig 4.4.  150  Figure 4.1. K+ uptake in Cx43 +, Cx43 /- and Cx43 CT/- mitochondria. A, Mitochondria were preferentially loaded with PBFI and treated with saponin to remove cytosolic dye yielding signals a largely derived from mitochondria. Cx43 CT/- astrocytes are shown in this example. B, Mitochondria were initially perfused with a K+ free solution followed by application of a solution containing 25 mM K+. In Cx43+/+ cultures, application of 25 mM K+ caused an increase in K+ in mitochondria, an effect which was reduced in Cx43 /- and Cx43 CT/mitochondria. C, A graph quantifying the changes in K+ uptake rate, defined by the change in ratio value over the first 100 s during the initial rise in PBFI ratio upon application of 25 mM K+. K+ uptake was significantly reduced in Cx43 /- and Cx43 CT/- mitochondria. *P<0.05, n values in parentheses represent the number of individual coverslips tested.  151  Figure 4.2. Mitochondrial K+ sequestration is reduced in Cx43 /- and Cx43 CT/- astrocytes. A, A record displaying [K+]i transients evoked by 2 min applications of CCCP in Cx43 CT/astrocytes. Only a small increase in [K+]i is observed after 5 min pretreatment with 12.5 mM K+o. B, Using the same protocol illustrated in A, K+ sequestration was compared in Cx43 /+, Cx43 /- and Cx43 CT/- astrocytes. Compared to Cx43 /+ astrocytes, K+ sequestration was reduced in Cx43 /- and Cx43 CT/- astrocytes (*P<0.05). This reduction did not differ between Cx43 /- and Cx43 CT/- astrocytes (P>0.05). Respective n values are in parentheses.  152  + /Figure 4.3. Resting and Cx43 CT/- astrocytes. m does not differ between Cx43 , Cx43 CT/A, A representative image of R123 loaded Cx43 astrocytes showing dye localization to mitochondria. B, Application of 10 M CCCP caused release of dye into the cytosol. To avoid artifacts caused by differing numbers of mitochondria in a field, regions of interest were placed in the area above the nucleus where mitochondria were seldom observed. As shown, application of CCCP causes dye release into the cytosol of this area. C, When peak signals after CCCP application (reflecting m) were compared, no differences were observed amongst the genotypes under study (P>0.05). D, Because R123 is known to leak out of astrocytes, a subset of experiments were performed with the addition of dye mitotracker red. R123 fluorescence values signals were normalized to the mitotracker red fluorescence values. Similarly, no differences in m were observed (P>0.05). Respective n values are in parentheses.  153  154  Figure 4.4. Schematic representation of the potential roles of Cx43 during stroke. A, In Cx43+/+ mice, astrocyte mitochondrial Cx43 may sequester rises in [K+]o to limit the damaging effects of this ion. Healthy cells which surround the penumbral region of the stroke can buffer toxic substances and allow for passage of protective molecules through gap junctions. Hemichannels may release ATP that is converted to adenosine which can protect neurons. B, In Cx43+/- mice K+ sequestration and gap junction coupling is less, potentially explaining why infarct volume following MCAO is larger in these mice. Less protective adenosine may be released in these mice. C, In Cx43 CT/- mice infarct volume is larger than in Cx43+/- mice. Potentially this is due to the additive effects of reduced K+ sequestration, hemichannel activity and coupling in early ischemia followed by continued coupling and hemichannel activity due to disruption of the C-terminal region in the later stages of ischemia (D). At these later time points the ratio of healthy to unhealthy cells is likely reduced potentially favouring propagation of toxic substances.  155  156  4.5  References  Boengler K, Dodoni G, Rodriguez-Sinovas A, Cabestrero A, Ruiz-Meana M, Gres P, Konietzka I, Lopez-Iglesias C, Garcia-Dorado D, Di Lisa F and others. 2005. Connexin43 in cardiomyocyte mitochondria and its increase by ischemic preconditioning. Cardiovasc Res 67(2):234-44. Boengler K, Konietzka I, Buechert A, Heinen Y, Garcia-Dorado D, Heusch G, Schulz R. 2007. Loss of ischemic preconditioning's cardioprotection in aged mouse hearts is associated with reduced gap junctional and mitochondrial levels of connexin43. Am J Physiol Heart Circ Physiol 292(4):H1764-9. Boengler K, Stahlhofen S, van de Sand A, Gres P, Ruiz-Meana M, Garcia-Dorado D, Heusch G, Schulz R. 2009. Presence of connexin43 in subsarcolemmal, but not in interfibrillar cardiomyocyte mitochondria. Basic Res Cardiol 104(2):141-7. Bordey A, Sontheimer H. 1997. Postnatal development of ionic currents in rat hippocampal astrocytes in situ. J Neurophysiol 78(1):461-77. Buckman JF, Hernandez H, Kress GJ, Votyakova TV, Pal S, Reynolds IJ. 2001. MitoTracker labeling in primary neuronal and astrocytic cultures: influence of mitochondrial membrane potential and oxidants. J Neurosci Methods 104(2):165-76. Contreras JE, Saez JC, Bukauskas FF, Bennett MV. 2003. Gating and regulation of connexin43 (Cx43) hemichannels. Proc Natl Acad Sci U S A 100(20):11388-93. Cotrina ML, Kang J, Lin JH, Bueno E, Hansen TW, He L, Liu Y, Nedergaard M. 1998. Astrocytic gap junctions remain open during ischemic conditions. J Neurosci 18(7):252037. Dermietzel R, Hertberg EL, Kessler JA, Spray DC. 1991. Gap junctions between cultured astrocytes: immunocytochemical, molecular, and electrophysiological analysis. J Neurosci 11(5):1421-32. Dermietzel R, Traub O, Hwang TK, Beyer E, Bennett MV, Spray DC, Willecke K. 1989. Differential expression of three gap junction proteins in developing and mature brain tissues. Proc Natl Acad Sci U S A 86(24):10148-52. Djukic B, Casper KB, Philpot BD, Chin LS, McCarthy KD. 2007. Conditional knock-out of Kir4.1 leads to glial membrane depolarization, inhibition of potassium and glutamate uptake, and enhanced short-term synaptic potentiation. J Neurosci 27(42):11354-65. Duchen MR. 2004. Mitochondria in health and disease: perspectives on a new mitochondrial biology. Mol Aspects Med 25(4):365-451. Duffy HS, Sorgen PL, Girvin ME, O'Donnell P, Coombs W, Taffet SM, Delmar M, Spray DC. 2002. pH-dependent intramolecular binding and structure involving Cx43 cytoplasmic domains. J Biol Chem 277(39):36706-14.  157  Goubaeva F, Mikami M, Giardina S, Ding B, Abe J, Yang J. 2007. Cardiac mitochondrial connexin43 regulates apoptosis. Biochem Biophys Res Commun 352(1):97-103. Hansen AJ. 1985. Effect of anoxia on ion distribution in the brain. Physiol Rev 65(1):101-48. Herve JC, Bourmeyster N, Sarrouilhe D, Duffy HS. 2007. Gap junctional complexes: from partners to functions. Prog Biophys Mol Biol 94(1-2):29-65. Kahlert S, Zundorf G, Reiser G. 2005. Glutamate-mediated influx of extracellular Ca2+ is coupled with reactive oxygen species generation in cultured hippocampal neurons but not in astrocytes. J Neurosci Res 79(1-2):262-71. Kofuji P, Newman EA. 2004. Potassium buffering in the central nervous system. Neuroscience 129(4):1045-56. Kozoriz MG, Bates DC, Bond SR, Lai CP, Moniz DM. 2006. Passing potassium with and without gap junctions. J Neurosci 26(31):8023-4. Kressin K, Kuprijanova E, Jabs R, Seifert G, Steinhauser C. 1995. Developmental regulation of Na+ and K+ conductances in glial cells of mouse hippocampal brain slices. Glia 15(2):173-87. Lampe PD, Lau AF. 2004. The effects of connexin phosphorylation on gap junctional communication. Int J Biochem Cell Biol 36(7):1171-86. Leis JA, Bekar LK, Walz W. 2005. Potassium homeostasis in the ischemic brain. Glia 50(4):40716. Li H, Brodsky S, Kumari S, Valiunas V, Brink P, Kaide J, Nasjletti A, Goligorsky MS. 2002. Paradoxical overexpression and translocation of connexin43 in homocysteine-treated endothelial cells. Am J Physiol Heart Circ Physiol 282(6):H2124-33. Li H, Liu TF, Lazrak A, Peracchia C, Goldberg GS, Lampe PD, Johnson RG. 1996. Properties and regulation of gap junctional hemichannels in the plasma membranes of cultured cells. J Cell Biol 134(4):1019-30. Lin JH, Lou N, Kang N, Takano T, Hu F, Han X, Xu Q, Lovatt D, Torres A, Willecke K and others. 2008. A central role of connexin43 in hypoxic preconditioning. J Neurosci 28(3):681-95. Ljubkovic M, Marinovic J, Fuchs A, Bosnjak ZJ, Bienengraeber M. 2006. Targeted expression of Kir6.2 in mitochondria confers protection against hypoxic stress. J Physiol 577(Pt 1):17-29. Maass K, Ghanem A, Kim JS, Saathoff M, Urschel S, Kirfel G, Grummer R, Kretz M, Lewalter T, Tiemann K and others. 2004. Defective epidermal barrier in neonatal mice lacking the C-terminal region of connexin43. Mol Biol Cell 15(10):4597-608. Maass K, Shibayama J, Chase SE, Willecke K, Delmar M. 2007. C-terminal truncation of connexin43 changes number, size, and localization of cardiac gap junction plaques. Circ Res 101(12):1283-91. 158  Miro-Casas E, Ruiz-Meana M, Agullo E, Stahlhofen S, Rodriguez-Sinovas A, Cabestrero A, Jorge I, Torre I, Vazquez J, Boengler K and others. 2009. Connexin43 in Cardiomyocyte Mitochondria Contributes to Mitochondrial Potassium Uptake. Cardiovasc Res 83(4):747-56. Moreno AP, Chanson M, Elenes S, Anumonwo J, Scerri I, Gu H, Taffet SM, Delmar M. 2002. Role of the carboxyl terminal of connexin43 in transjunctional fast voltage gating. Circ Res 90(4):450-7. Nakase T, Fushiki S, Sohl G, Theis M, Willecke K, Naus CC. 2003. Neuroprotective role of astrocytic gap junctions in ischemic stroke. Cell Commun Adhes 10(4-6):413-7. Nakase T, Sohl G, Theis M, Willecke K, Naus CC. 2004. Increased apoptosis and inflammation after focal brain ischemia in mice lacking connexin43 in astrocytes. Am J Pathol 164(6):2067-75. Naus CC, Bechberger JF, Zhang Y, Venance L, Yamasaki H, Juneja SC, Kidder GM, Giaume C. 1997. Altered gap junctional communication, intercellular signaling, and growth in cultured astrocytes deficient in connexin43. J Neurosci Res 49(5):528-40. Oliveira JM, Jekabsons MB, Chen S, Lin A, Rego AC, Goncalves J, Ellerby LM, Nicholls DG. 2007. Mitochondrial dysfunction in Huntington's disease: the bioenergetics of isolated and in situ mitochondria from transgenic mice. J Neurochem 101(1):241-9. Orkand RK, Nicholls JG, Kuffler SW. 1966. Effect of nerve impulses on the membrane potential of glial cells in the central nervous system of amphibia. J Neurophysiol 29(4):788-806. Ozog MA, Siushansian R, Naus CC. 2002. Blocked gap junctional coupling increases glutamateinduced neurotoxicity in neuron-astrocyte co-cultures. J Neuropathol Exp Neurol 61(2):132-41. Rodriguez-Sinovas A, Boengler K, Cabestrero A, Gres P, Morente M, Ruiz-Meana M, Konietzka I, Miro E, Totzeck A, Heusch G and others. 2006. Translocation of connexin43 to the inner mitochondrial membrane of cardiomyocytes through the heat shock protein 90dependent TOM pathway and its importance for cardioprotection. Circ Res 99:93-101. Seifert G, Huttmann K, Binder DK, Hartmann C, Wyczynski A, Neusch C, Steinhauser C. 2009. Analysis of astroglial K+ channel expression in the developing hippocampus reveals a predominant role of the Kir4.1 subunit. J Neurosci 29(23):7474-88. Siushansian R, Bechberger JF, Cechetto DF, Hachinski VC, Naus CC. 2001. Connexin43 null mutation increases infarct size after stroke. J Comp Neurol 440(4):387-94. Somjen GG. 2002. Ion regulation in the brain: implications for pathophysiology. Neuroscientist 8(3):254-67. Spray DC, Ye ZC, Ransom BR. 2006. Functional connexin "hemichannels": a critical appraisal. Glia 54(7):758-73.  159  Wallraff A, Kohling R, Heinemann U, Theis M, Willecke K, Steinhauser C. 2006. The impact of astrocytic gap junctional coupling on potassium buffering in the hippocampus. J Neurosci 26(20):5438-47. Walz W. 2000. Role of astrocytes in the clearance of excess extracellular potassium. Neurochem Int 36(4-5):291-300. Xu L, Zeng LH, Wong M. 2009. Impaired astrocytic gap junction coupling and potassium buffering in a mouse model of tuberous sclerosis complex. Neurobiol Dis 34(2):291-9. Xu W, Liu Y, Wang S, McDonald T, Van Eyk JE, Sidor A, O'Rourke B. 2002. Cytoprotective role of Ca2+- activated K+ channels in the cardiac inner mitochondrial membrane. Science 298(5595):1029-33. Yamamoto T, Ochalski A, Hertzberg EL, Nagy JI. 1990. LM and EM immunolocalization of the gap junctional protein connexin43 in rat brain. Brain Res 508(2):313-9.  160  Chapter 5: Conclusion  161  5.1  Contribution to novel knowledge The work presented in this thesis offers several novel contributions. Although it has been  appreciated for decades that mitochondria can accumulate potassium against a concentration gradient (reviewed in O'Rourke 2007), it is not known whether mitochondria within intact cells play a role in extracellular K+ handling. The results presented in this thesis demonstrate that astrocyte mitochondria sequester rises in extracellular K+. This effect may be important in pathological states, such as ischemia, where K+ is known to increase. Another novel finding is that the gap junction protein Cx43 is present in the mitochondria of astrocytes and contributes to mitochondrial K+ sequestration. The study of mitochondrial K+ in situ is difficult because of technological limitations. However, a novel technique for indirectly examining mitochondrial K+ levels is described in this thesis. By uncoupling the mitochondrial membrane potential and detecting K+ release into the cytosol with PBFI, mitochondrial K+ could be assessed. Lastly, building on previous studies in our lab that show Cx43 is important in ischemic damage, this thesis presents the novel finding that the C-terminal region of Cx43 is important in stroke. Mutation of the C-terminal region resulted in alteration of several process including mitochondrial K+ sequestration, gap junction coupling, reactive gliosis and inflammatory cell invasion. This study lays the groundwork for determining which specific regions of the Cterminal region are involved in neuroprotection.  5.2  Mitochondrial Cx43 based K+ sequestration and the Cx43 C-terminal region’s role in  neuroprotection The gap junction field is faced with the problem of reconciling the reasons why in some incidences connexins are protective and in others they are destructive. As outlined in the 162  introduction chapter of this thesis, connexin research is complex because there are at least 21 different connexin subtypes which are differentially expressed in various body tissues. Connexin subtypes are not exclusive in the formation of hemichannels and gap junctions, as channels can be composed of different connexin subtypes. This arrangement potentially creates numerous types of channels with a wide range of properties. Furthermore, in addition to their traditional role of providing an intercellular conduit for passage of electrical activity and molecules, connexins are now recognized to have several other cellular functions. Connexins form hemichannels, which allow for direct communication with the extracellular space (Contreras et al. 2003; Spray et al. 2006). Based on studies of tumor cells, connexins are found to be involved in cell growth, proliferation and invasion (Bates et al. 2007; Fu et al. 2004; Hirschi et al. 1996; Loewenstein 1979; Zhang et al. 2003), and although the pathways of this action are not entirely clear it is possible that connexins act in the nucleus to regulate cell growth (Dang et al. 2003). Cx43 is expressed in mitochondria (Boengler et al. 2005; Boengler et al. 2007; Boengler et al. 2009; Goubaeva et al. 2007; Li et al. 2002; Rodriguez-Sinovas et al. 2006), and has been shown to play a role in K+ uptake (Miro-Casas et al. 2009). Connexins are known to interact either directly or indirectly with at least 40 proteins (reviewed by Laird 2010), and preliminary mass spectrometry data from our lab has identified over 150 potential interactions with Cx43. These interactions may serve to alter basic functions of connexins such as channel localization, activity and degradation, and may form a platform for cellular signaling (Ai and Pogwizd 2005; Hunter et al. 2005; Leithe and Rivedal 2004; Lurtz and Louis 2007; Shaw et al. 2007; reviewed by Laird 2010). The study of connexins is further complicated by the fact that pharmacological agents that block or enhance specific connexin actions are not available. However, newly developed peptide compounds show promise in blocking hemichannel activity and will likely be commercially available in the near future. 163  Because of the range of functions attributed to connexins, it is not surprising that a clear understanding of the role of connexins in cellular injury has remained elusive. Many papers state the role of connexins in cellular injury is controversial, yet to the best of my knowledge, no two published studies have used a similar design to study the role of connexins in cellular injury. Given the range of roles for connexins it is quite possible that connexins are involved in both cellular protection and cellular destruction, and it is the experimental context which dictates cellular fate. It is possible that differences in results may be due to the type of connexin(s) or pannexin(s), type of cells used (e.g. neuronal, glial or cell line), experimental design, or the ratio of healthy to non-healthy cells. Some general patterns emerge in the connexin literature (see chapter 1). In studies of focal ischemia (Lin et al. 2008; Nakase et al. 2003a; Nakase et al. 2003b; Nakase et al. 2004; Siushansian et al. 2001), where an ischemic core and penumbra are created, gap junctions tend to be protective, potentially because harmful compounds can be buffered away from the penumbra and protective molecules can move to the site of injury. Astrocyte hemichannels may also play a role in the release of protective molecules and may buffer toxic substances (Lin et al. 2008). The ratio of healthy to non-healthy cells may also be more favorable in this situation. Whereas in global ischemia, or if a brain slice is uniformly exposed to an ischemic insult, gap junctions tend to increase cell death, potentially through spreading of toxic molecules and through the opening of neuronal hemichannels (de Pina-Benabou et al. 2005; Frantseva et al. 2002; Perez Velazquez et al. 2006; Rami et al. 2001; Simburger et al. 1997). An exception to this situation is if connexins in interneurons are disrupted, which may affect protective GABA release (Oguro et al. 2001). Also, because pannexin1 channels can be blocked by most traditional connexin blockers, previous studies must be reinterpreted taking into account that neuronal pannexin1 channels open during ischemic  164  conditions (Thompson et al. 2006). This may explain why “gap junctions” were found to be destructive in these studies. The results presented in chapter 2 indicate that astrocyte mitochondrial Cx43 plays a role in sequestering rises in [K+]o. Although not tested explicitly, this could serve a protective role in the initial stages of ischemia in both global and focal ischemic models. In the early stages of stroke mitochondrial Cx43-based K+ uptake may curb excessive [K+]o accumulation, which could protect against excessive neuronal depolarization and neurotoxic glutamate release. In the penumbral region created by focal stroke, mitochondrial K+ clearance could act synergistically with other astrocyte Cx43-based protective mechanisms like gap junction communication to allow for further clearance of K+ and other toxic molecules. These results appear to fit with our previous studies showing Cx43 serves a protective role following stroke (Nakase et al. 2003a; Nakase et al. 2003b; Nakase et al. 2004; Siushansian et al. 2001). Mitochondrial K+ uptake likely has its limits, and excessive mitochondrial K+ accumulation will most likely result in mitochondrial swelling and cell death; this may be particularly pronounced in regions closer to the necrotic core where [K+]o levels are higher. Astrocytes have long been known to play a role in extracellular K+ uptake (Orkand et al. 1966; reviewed by Kofuji and Newman 2004; Leis et al. 2005; Walz 2000) and it is thought that gap junctions play a role in buffering this ion through an astrocyte coupled syncytium (Wallraff et al. 2006; Xu et al. 2009; reviewed by Kofuji and Newman 2004; Leis et al. 2005; Walz 2000). Spatial buffering of K+ is not thought to lead to K+ accumulation within astrocytes (discussed in Walz 2000). However, the results presented in this thesis challenge that notion by demonstrating a role for mitochondria in extracellular K+ sequestration. Some studies have shown that disruption of gap junction coupling leads to rises in [K+]o (Wallraff et al. 2006; Xu et al. 2009), but this does not invalidate our results. Rises in [K+]o could also be explained, at least in part, by 165  impairment of Cx43-based mitochondrial K+ sequestration. In addition, the distribution of K+ to neighbouring glial cells through gap junctions is also problematic because modeling studies suggest that astrocyte processes are not suited for distributing K+, even over short distances (Chen and Nicholson 2000). This increases the possibility that intracellular ion sequestration is playing an important role. Mitochondria are known to accumulate K+ against a concentration gradient (reviewed in O'Rourke 2007), making this organelle an ideal site for K+ sequestration. K+ is known to enter the mitochondria via mitoKATP channels, mitoKca channels and K+ leak channels. K+ uptake is counterbalanced by extrusion by the K+/H+ exchanger (reviewed by Garlid 1996; Garlid and Paucek 2001; Garlid and Paucek 2003; O'Rourke 2007). Another pathway for K+ uptake involving mitochondrial Cx43 is described in this thesis. Mitochondrial K+ buffering through any of the channels described above may be problematic. At this time, the exact K+ buffering capacity of mitochondria is not known, and likely there a point at which K+ uptake becomes harmful. Undoubtedly, too much K+ uptake would lead to excessive mitochondrial swelling and subsequent cell death (reviewed by Garlid 1996; Garlid and Paucek 2001; Garlid and Paucek 2003; O'Rourke 2007). However, in the brain, the extracellular space is small (Sykova and Nicholson 2008) and the number of astrocytes (and mitochondria) is substantial. Therefore, in vivo the amount of K+ entering individual mitochondria from the extracellular space may be manageable and could have a minimal impact on mitochondrial swelling. Mitochondrial K+ channel regulation is complex (reviewed by Garlid 1996; Garlid and Paucek 2001; Garlid and Paucek 2003), and the existence of a Cx43-based K+ uptake route adds to this complexity. The conditions and modulators which regulate mitochondrial Cx43 channel activity require further study (discussed below), but likely involve the same factors which regulate plasma membrane hemichannels (e.g. membrane voltage, pH, Ca2+ and channel 166  phosphorylation). In addition, studies of the physiological importance of Cx43-based K+ uptake are sparse. However, mitochondrial K+ uptake play a role in facilitating respiration, improving ATP production, reducing mitochondrial permeability transition (Hansson et al. 2010; Kowaltowski et al. 2001; Rousou et al. 2004), and aiding in cellular protection (Korge et al. 2005; Rousou et al. 2004), and mitochondrial Cx43 could serve a similar role. Interestingly, gap junction blockers are known to increase release of the pro-apoptotic factor cytochrome c (Goubaeva et al. 2007), supporting a role of mitochondrial Cx43 in cellular protection. In chapter 3, it was found that the C-terminal region of Cx43 is important in cellular protection during stroke, and in chapter 4 K+ buffering is shown to be impaired in astrocytes expressing this mutation. The finding that the C-terminal region of Cx43 is protective fits with a recent study describing increased ischemic heart damage in these mutants (Maass et al. 2009). As discussed in chapter 3 and 4, in the early stages of focal stroke astrocyte Cx43 may be important in aiding cell survival by buffering toxic molecules and providing nutrients to vulnerable cells in the penumbral region, while in the later stages of ischemia the ratio of healthy to unhealthy cells declines, and potentially a consequent shift occurs towards propagation of toxic molecules. When full length Cx43 is expressed, the presence of a gate which can respond to a drop in pH can reduce the propagation of these molecules (see Fig. 4.4A). However, in the absence of the C-terminal region, gating may be ineffective and could result in amplification of cellular injury (see Fig. 4.4D). The reduction in [K+]o sequestration by mitochondria, as shown in chapter 4, could also increase damage in the early stages of stoke creating more cellular damage (see Fig. 4.4C). In conjunction, this could explain why stroke is larger in the presence of Cx43 CT when compared to Cx43+/- mice.  167  5.3  Strengths and weaknesses There are several strengths to the studies presented in this thesis. Mitochondria are often  studied in isolation, which may be problematic because this organelle is known to behave differently in situ (Duchen 2004; Oliveira et al. 2007; Scott and Nicholls 1980). Using an intact cellular system, mitochondrial K+ handling was assessed by uncoupling the mitochondrial membrane and measuring the resulting release of K+ with cytosolic PBFI. Using this method, it was found that rises in [K+]o lead to increases in mitochondrial K+. Mitochondria are known to take up K+, but the finding that rises in [K+]o can be sequestered by mitochondria is novel. Cx43 has been shown to localize to mitochondria, but its functional significance is not exactly known. The results presented in this thesis show that Cx43 provides an avenue for mitochondrial K+ uptake, a result supported by recent similar findings in isolated cardiomyocyte mitochondria (Miro-Casas et al. 2009). There are also some weaknesses to the material presented in this thesis. Although astrocyte culture systems have the advantage of easily studying cellular pathways, the behavior of astrocyte mitochondrial K+ dynamics in the intact brain requires further study. In particular, it is important to determine whether mitochondrial K+ uptake is indeed beneficial in early ischemia and harmful in late ischemia. Also, it is unknown if mitochondria contribute in other states where K+ increases, such as during neuronal activity, epilepsy and spreading depression. As discussed in chapter 2, the dye used to detect K+ (PBFI) has its limitations. PBFI derived ratio values saturate at elevated [K+] and this dye is known to be sensitive to Na+. Although several control experiments were performed to deal with these limitations, the study of intracellular K+ would benefit from the creation of dyes that detect K+ at physiological levels in the absence of non-specific ion interference. Mitochondrial K+ was indirectly assessed in most of the K+ sequestration experiments; future studies would benefit from the creation of dyes that 168  can directly assess mitochondrial K+ like CoroNa Red for assessment of mitochondrial [Na+] (Azarias et al. 2008). In chapters 3 and 4 the role of the C-terminal region was examined using a mutant mouse which expressed a truncated form of Cx43. Mutant mice are a powerful tool, but their use is complicated by the possibility of altered gene expression (Iacobas et al. 2003) and that the expression of other proteins may be altered to compensate. Also it is possible that a truncated protein has trafficking and expression profiles that are different than native protein. Although our study shows that Cx43 CT is expressed at the cell membrane and that coupling is still present, these mutants have obvious alterations in expression pattern. It appears that Cx43 content is reduced and that gap junction plaques appear larger, possibly due to impaired plaque degradation (Leykauf et al. 2006; Maass et al. 2004). The consequences of these changes in stroke are currently not known. Moreover, even though a disruption in channel gating is a potential mechanism for the increase in infarct volume, the C-terminal region is a site of many interacting molecules (reviewed by Laird 2010) and the exact sites of interest await discovery. Using a mutagenic approach, future studies in our lab will explore which regions of the Cterminal are critical for neuroprotection. As mentioned above, future studies would benefit from the development of blockers that can distinguish between gap junctions, hemichannels and the specific subtypes of connexins and pannexins which could help determine which channels play a role in protection or destruction. Although mechanisms were proposed as to why stroke is larger in Cx43 mutants, ultimately these blockers could provide a clear picture of which pathways are involved. 5.4  Future directions In this study injury was examined 4 days following MCAO. It is important to understand  the role of connexins across a range of time points as it is entirely possible that the beneficial or 169  harmful effects of connexins depend on the phase of stroke, and this knowledge could be beneficial in developing therapeutic treatment strategies. Also, since thrombolytic agents are used in the treatment of ischemic stroke, it would be useful to study the role of connexins in models where ischemia is transient and reperfusion occurs. As mentioned, the C-terminal region of Cx43 is an interaction site for many proteins and further research will explore the specific regions of Cx43 that are important in neuroprotection. Additionally, recent data from our lab using mass spectrometry has identified over 150 proteins that may interact with Cx43 either directly or indirectly. Confirming these interactions and determining which may play a role in cellular protection or destruction could provide an avenue for therapeutic targets. Intriguingly, many of the potential interacting proteins are mitochondrial (shown in Table D.1; Appendix D) opening the door for many theories of the role of Cx43 in mitochondria. This promises to be a fruitful area of future research. Using coimmunoprecipitation we have already confirmed some of these interactions. One protein of interest is the outer mitochondrial membrane voltage-dependent anion channel (VDAC), which has been shown to co-immunoprecipitate with Cx43 in (Fig. E.1 - Appendix E). The role of protein interactions with Cx43 in cellular damage is also an area of future study, since a particularly relevant question to this thesis is whether interactions with Cx43 can modulate mitochondrial K+ uptake. Future studies could also focus on the regulation of mitochondrial Cx43 hemichannels. Plasma membrane hemichannel activity under resting conditions is controversial, as it seems that unregulated channel openings would allow for disruption of ion gradients, which could be detrimental (reviewed by Spray et al. 2006). However under pathological conditions it appears that plasma membrane hemichannels are active and do not necessarily lead to cell death (Lin et al. 2008; Retamal et al. 2007a; Retamal et al. 2007b; Thompson et al. 2006). Mitochondrial ion 170  gradient regulation is complex and as discussed in chapter 4, hemichannels may have a propensity to be open based on the mitochondrial membrane voltage. However, openings may be counteracted by the lower pH in the mitochondrial intermembrane space, matrix Ca2+ and channel phosphorylation. It is important to determine exactly what conditions and factors open and close hemichannels, and to put Cx43 channel activity into context with other known ion transport mechanisms. Another area of future research is determining which functions of connexins are important during ischemia. The advent of blockers against hemichannels could serve as an excellent tool for determining their role in stroke. As mentioned, development and testing of gap junction blockers and openers for the specific subtypes of connexins would be a boon for stroke research. Our lab is starting a collaboration with Zealand Pharma to explore the role of a gap junction opener rotigaptide in stroke (Dhein et al. 2009). Also, technologies now exist to monitor stroke in vivo (Sigler et al. 2008). This could be adapted for connexin research to examine stroke in knockouts or in the presence of gap junction blockers to determine the effect of connexins on stroke in vivo in real time.  5.5  Overall significance and potential applications of findings There are numerous studies outlining the role of connexins in pathological conditions.  During ischemia gap junctions have been shown to have either protective or destructive properties. Although several reasons for this discrepancy have been described, one limitation is likely our incomplete understanding of all of the roles connexins may have. This thesis describes an additional role of connexins: an avenue for sequestration of rises of [K+]o into mitochondria. Although the utility of this pathway in providing protection or destruction was not assessed in this study, once established, this could be an important target for therapeutic strategies. 171  Compounds or manipulations which increase mitochondrial K+ have showed therapeutic promise (Garlid et al. 2009; Hansson et al. 2010; Ljubkovic et al. 2006; Xu et al. 2002). In addition, this thesis shows that a lack of the C-terminal region of Cx43 increases damage due to stroke. Potentially, agents that target the C-terminal region could be applied during late ischemia to reduce damage in stroke patients by providing additional channel gating when gap junction communication is harmful. The time course of exogenous C-terminal application is likely important, because in early ischemia blockade of gap junctions may inhibit their beneficial properties. Understanding the role of connexins in cellular pathology is also important because many compounds to which patients are regularly exposed are known to block gap junctions. For example, alcohol and many anesthetics routinely used in hospital are known to block gap junctions (Wentlandt et al. 2004; Wentlandt et al. 2006). In the future, it is possible that clinical decisions regarding treatment of patients with brain injury could take into account exposure to these gap junction blockers.  172  5.5  References  Ai X, Pogwizd SM. 2005. Connexin43 downregulation and dephosphorylation in nonischemic heart failure is associated with enhanced colocalized protein phosphatase type 2A. Circ Res 96(1):54-63. Azarias G, Van de Ville D, Unser M, Chatton JY. 2008. Spontaneous Na+ transients in individual mitochondria of intact astrocytes. Glia 56(3):342-53. Bates DC, Sin WC, Aftab Q, Naus CC. 2007. Connexin43 enhances glioma invasion by a mechanism involving the carboxy terminus. Glia 55(15):1554-64. Boengler K, Dodoni G, Rodriguez-Sinovas A, Cabestrero A, Ruiz-Meana M, Gres P, Konietzka I, Lopez-Iglesias C, Garcia-Dorado D, Di Lisa F and others. 2005. Connexin43 in cardiomyocyte mitochondria and its increase by ischemic preconditioning. Cardiovasc Res 67(2):234-44. Boengler K, Konietzka I, Buechert A, Heinen Y, Garcia-Dorado D, Heusch G, Schulz R. 2007. Loss of ischemic preconditioning's cardioprotection in aged mouse hearts is associated with reduced gap junctional and mitochondrial levels of connexin43. Am J Physiol Heart Circ Physiol 292(4):H1764-9. Boengler K, Stahlhofen S, van de Sand A, Gres P, Ruiz-Meana M, Garcia-Dorado D, Heusch G, Schulz R. 2009. Presence of connexin43 in subsarcolemmal, but not in interfibrillar cardiomyocyte mitochondria. Basic Res Cardiol 104(2):141-7. Chen KC, Nicholson C. 2000. Spatial buffering of potassium ions in brain extracellular space. Biophys J 78(6):2776-97. Contreras JE, Saez JC, Bukauskas FF, Bennett MV. 2003. Gating and regulation of connexin43 (Cx43) hemichannels. Proc Natl Acad Sci U S A 100(20):11388-93. Dang X, Doble BW, Kardami E. 2003. The carboxy-tail of connexin43 localizes to the nucleus and inhibits cell growth. Mol Cell Biochem 242(1-2):35-8. de Pina-Benabou MH, Szostak V, Kyrozis A, Rempe D, Uziel D, Urban-Maldonado M, Benabou S, Spray DC, Federoff HJ, Stanton PK and others. 2005. Blockade of gap junctions in vivo provides neuroprotection after perinatal global ischemia. Stroke 36(10):2232-7. Dhein S, Hagen A, Jozwiak J, Dietze A, Garbade J, Barten M, Kostelka M, Mohr FW. 2009. Improving cardiac gap junction communication as a new antiarrhythmic mechanism: the action of antiarrhythmic peptides. Naunyn Schmiedebergs Arch Pharmacol. In Press. Duchen MR. 2004. Mitochondria in health and disease: perspectives on a new mitochondrial biology. Mol Aspects Med 25(4):365-451. Frantseva MV, Kokarovtseva L, Perez Velazquez JL. 2002. Ischemia-induced brain damage depends on specific gap-junctional coupling. J Cereb Blood Flow Metab 22(4):453-62.  173  Fu CT, Bechberger JF, Ozog MA, Perbal B, Naus CC. 2004. CCN3 (NOV) interacts with connexin43 in C6 glioma cells: possible mechanism of connexin-mediated growth suppression. J Biol Chem 279(35):36943-50. Garlid KD. 1996. Cation transport in mitochondria--the potassium cycle. Biochim Biophys Acta 1275(1-2):123-6. Garlid KD, Costa AD, Quinlan CL, Pierre SV, Dos Santos P. 2009. Cardioprotective signaling to mitochondria. J Mol Cell Cardiol 46(6):858-66. Garlid KD, Paucek P. 2001. The mitochondrial potassium cycle. IUBMB Life 52(3-5):153-8. Garlid KD, Paucek P. 2003. Mitochondrial potassium transport: the K+ cycle. Biochim Biophys Acta 1606(1-3):23-41. Goubaeva F, Mikami M, Giardina S, Ding B, Abe J, Yang J. 2007. Cardiac mitochondrial connexin43 regulates apoptosis. Biochem Biophys Res Commun 352(1):97-103. Hansson MJ, Morota S, Teilum M, Mattiasson G, Uchino H, Elmer E. 2010. Increased potassium conductance of brain mitochondria induces resistance to permeability transition by enhancing matrix volume. J Biol Chem. 285(1):741-50. Hirschi KK, Xu CE, Tsukamoto T, Sager R. 1996. Gap junction genes Cx26 and Cx43 individually suppress the cancer phenotype of human mammary carcinoma cells and restore differentiation potential. Cell Growth Differ 7(7):861-70. Hunter AW, Barker RJ, Zhu C, Gourdie RG. 2005. Zonula occludens-1 alters connexin43 gap junction size and organization by influencing channel accretion. Mol Biol Cell 16(12):5686-98. Iacobas DA, Urban-Maldonado M, Iacobas S, Scemes E, Spray DC. 2003. Array analysis of gene expression in connexin-43 null astrocytes. Physiol Genomics 15(3):177-90. Kofuji P, Newman EA. 2004. Potassium buffering in the central nervous system. Neuroscience 129(4):1045-56. Korge P, Honda HM, Weiss JN. 2005. K+-dependent regulation of matrix volume improves mitochondrial function under conditions mimicking ischemia-reperfusion. Am J Physiol Heart Circ Physiol 289(1):H66-77. Kowaltowski AJ, Seetharaman S, Paucek P, Garlid KD. 2001. Bioenergetic consequences of opening the ATP-sensitive K+ channel of heart mitochondria. Am J Physiol Heart Circ Physiol 280(2):H649-57. Laird DW. 2010. The gap junction proteome and its relationship to disease. Trends Cell Biol 20(2):92-101. Leis JA, Bekar LK, Walz W. 2005. Potassium homeostasis in the ischemic brain. Glia 50(4):40716. 174  Leithe E, Rivedal E. 2004. Ubiquitination and down-regulation of gap junction protein connexin43 in response to 12-O-tetradecanoylphorbol 13-acetate treatment. J Biol Chem 279(48):50089-96. Leykauf K, Salek M, Bomke J, Frech M, Lehmann WD, Durst M, Alonso A. 2006. Ubiquitin protein ligase Nedd4 binds to connexin43 by a phosphorylation-modulated process. J Cell Sci 119(Pt 17):3634-42. Li H, Brodsky S, Kumari S, Valiunas V, Brink P, Kaide J, Nasjletti A, Goligorsky MS. 2002. Paradoxical overexpression and translocation of connexin43 in homocysteine-treated endothelial cells. Am J Physiol Heart Circ Physiol 282(6):H2124-33. Lin JH, Lou N, Kang N, Takano T, Hu F, Han X, Xu Q, Lovatt D, Torres A, Willecke K and others. 2008. A central role of connexin43 in hypoxic preconditioning. J Neurosci 28(3):681-95. Ljubkovic M, Marinovic J, Fuchs A, Bosnjak ZJ, Bienengraeber M. 2006. Targeted expression of Kir6.2 in mitochondria confers protection against hypoxic stress. J Physiol 577(Pt 1):17-29. Loewenstein WR. 1979. Junctional intercellular communication and the control of growth. Biochim Biophys Acta 560(1):1-65. Lurtz MM, Louis CF. 2007. Intracellular calcium regulation of connexin43. Am J Physiol Cell Physiol 293(6):C1806-13. Maass K, Chase SE, Lin X, Delmar M. 2009. Cx43 CT domain influences infarct size and susceptibility to ventricular tachyarrhythmias in acute myocardial infarction. Cardiovasc Res 84(3):361-7. Maass K, Ghanem A, Kim JS, Saathoff M, Urschel S, Kirfel G, Grummer R, Kretz M, Lewalter T, Tiemann K and others. 2004. Defective epidermal barrier in neonatal mice lacking the C-terminal region of connexin43. Mol Biol Cell 15(10):4597-608. Miro-Casas E, Ruiz-Meana M, Agullo E, Stahlhofen S, Rodriguez-Sinovas A, Cabestrero A, Jorge I, Torre I, Vazquez J, Boengler K and others. 2009. Connexin43 in Cardiomyocyte Mitochondria Contributes to Mitochondrial Potassium Uptake. Cardiovasc Res 83(4):747-56. Nakase T, Fushiki S, Naus CC. 2003a. Astrocytic gap junctions composed of connexin43 reduce apoptotic neuronal damage in cerebral ischemia. Stroke 34(8):1987-93. Nakase T, Fushiki S, Sohl G, Theis M, Willecke K, Naus CC. 2003b. Neuroprotective role of astrocytic gap junctions in ischemic stroke. Cell Commun Adhes 10(4-6):413-7. Nakase T, Sohl G, Theis M, Willecke K, Naus CC. 2004. Increased apoptosis and inflammation after focal brain ischemia in mice lacking connexin43 in astrocytes. Am J Pathol 164(6):2067-75. 175  Oguro K, Jover T, Tanaka H, Lin Y, Kojima T, Oguro N, Grooms SY, Bennett MV, Zukin RS. 2001. Global ischemia-induced increases in the gap junctional proteins connexin 32 (Cx32) and Cx36 in hippocampus and enhanced vulnerability of Cx32 knock-out mice. J Neurosci 21(19):7534-42. Oliveira JM, Jekabsons MB, Chen S, Lin A, Rego AC, Goncalves J, Ellerby LM, Nicholls DG. 2007. Mitochondrial dysfunction in Huntington's disease: the bioenergetics of isolated and in situ mitochondria from transgenic mice. J Neurochem 101(1):241-9. Orkand RK, Nicholls JG, Kuffler SW. 1966. Effect of nerve impulses on the membrane potential of glial cells in the central nervous system of amphibia. J Neurophysiol 29(4):788-806. O'Rourke B. 2007. Mitochondrial ion channels. Annu Rev Physiol 69:19-49. Perez Velazquez JL, Kokarovtseva L, Sarbaziha R, Jeyapalan Z, Leshchenko Y. 2006. Role of gap junctional coupling in astrocytic networks in the determination of global ischaemiainduced oxidative stress and hippocampal damage. Eur J Neurosci 23(1):1-10. Rami A, Volkmann T, Winckler J. 2001. Effective reduction of neuronal death by inhibiting gap junctional intercellular communication in a rodent model of global transient cerebral ischemia. Exp Neurol 170(2):297-304. Retamal MA, Froger N, Palacios-Prado N, Ezan P, Saez PJ, Saez JC, Giaume C. 2007a. Cx43 hemichannels and gap junction channels in astrocytes are regulated oppositely by proinflammatory cytokines released from activated microglia. J Neurosci 27(50):1378192. Retamal MA, Schalper KA, Shoji KF, Bennett MV, Saez JC. 2007b. Opening of connexin43 hemichannels is increased by lowering intracellular redox potential. Proc Natl Acad Sci U S A 104(20):8322-7. Rodriguez-Sinovas A, Boengler K, Cabestrero A, Gres P, Morente M, Ruiz-Meana M, Konietzka I, Miro E, Totzeck A, Heusch G and others. 2006. Translocation of connexin43 to the inner mitochondrial membrane of cardiomyocytes through the heat shock protein 90Dependent TOM pathway and its importance for cardioprotection. Circ Res 99:93-101. Rossi DJ, Brady JD, Mohr C. 2007. Astrocyte metabolism and signaling during brain ischemia. Nat Neurosci 10(11):1377-86. Rousou AJ, Ericsson M, Federman M, Levitsky S, McCully JD. 2004. Opening of mitochondrial KATP channels enhances cardioprotection through the modulation of mitochondrial matrix volume, calcium accumulation, and respiration. Am J Physiol Heart Circ Physiol 287(5):H1967-76. Scott ID, Nicholls DG. 1980. Energy transduction in intact synaptosomes. Influence of plasmamembrane depolarization on the respiration and membrane potential of internal mitochondria determined in situ. Biochem J 186(1):21-33.  176  Shaw RM, Fay AJ, Puthenveedu MA, von Zastrow M, Jan YN, Jan LY. 2007. Microtubule plusend-tracking proteins target gap junctions directly from the cell interior to adherens junctions. Cell 128(3):547-60. Sigler A, Goroshkov A, Murphy TH. 2008. Hardware and methodology for targeting single brain arterioles for photothrombotic stroke on an upright microscope. J Neurosci Methods 170(1):35-44. Simburger E, Stang A, Kremer M, Dermietzel R. 1997. Expression of connexin43 mRNA in adult rodent brain. Histochem Cell Biol 107(2):127-37. Siushansian R, Bechberger JF, Cechetto DF, Hachinski VC, Naus CC. 2001. Connexin43 null mutation increases infarct size after stroke. J Comp Neurol 440(4):387-94. Spray DC, Ye ZC, Ransom BR. 2006. Functional connexin "hemichannels": a critical appraisal. Glia 54(7):758-73. Sykova E, Nicholson C. 2008. Diffusion in brain extracellular space. Physiol Rev 88(4):1277340. Thompson RJ, Zhou N, MacVicar BA. 2006. Ischemia opens neuronal gap junction hemichannels. Science 312(5775):924-7. Wallraff A, Kohling R, Heinemann U, Theis M, Willecke K, Steinhauser C. 2006. The impact of astrocytic gap junctional coupling on potassium buffering in the hippocampus. J Neurosci 26(20):5438-47. Walz W. 2000. Role of astrocytes in the clearance of excess extracellular potassium. Neurochem Int 36(4-5):291-300. Wentlandt K, Kushnir M, Naus CC, Carlen PL. 2004. Ethanol inhibits gap-junctional coupling between P19 cells. Alcohol Clin Exp Res 28(9):1284-90. Wentlandt K, Samoilova M, Carlen PL, El Beheiry H. 2006. General anesthetics inhibit gap junction communication in cultured organotypic hippocampal slices. Anesth Analg 102(6):1692-8. Xu L, Zeng LH, Wong M. 2009. Impaired astrocytic gap junction coupling and potassium buffering in a mouse model of tuberous sclerosis complex. Neurobiol Dis 34(2):291-9. Xu W, Liu Y, Wang S, McDonald T, Van Eyk JE, Sidor A, O'Rourke B. 2002. Cytoprotective role of Ca2+- activated K+ channels in the cardiac inner mitochondrial membrane. Science 298(5595):1029-33. Zhang YW, Kaneda M, Morita I. 2003. The gap junction-independent tumor-suppressing effect of connexin43. J Biol Chem 278(45):44852-6.  177  Appendices  178  Appendix A: Journal club – passing potassium with and without gap junctions  179  Adapted by permission from The Society for Neuroscience: J Neurosci. Kozoriz MG, Bates DC, Bond SR, Lai CP, Moniz DM. 2006. Passing potassium with and without gap junctions. 26(31):8023-4.  180  Appendix B: Supplemental figures for chapter 2  Figure B.1. Comparison of the [Na+]i and [K+]i transients evoked by CCCP in cultured astrocytes. To directly assess the potential impact of changes in [Na+]i to the CCCP-evoked increases in PBFI ratio values observed under our experimental conditions, we loaded astrocytes with SBFI and measured the changes in [Na+]i evoked by 2 and 5 min applications of CCCP. The application of CCCP for 2 min causes a rapid and robust increase in [K+]i (as represented by PBFI-derived BI340/380 ratio values; open circles). In a parallel experiment in SBFI-loaded astrocytes (closed circles), the application of CCCP for 2 min caused a more gradual rise in [Na+]i, the peak of which was observed after the [K+]i transient had returned to baseline. The [Na+]i response to a subsequent 5 min application of CCCP is also shown. Applications of CCCP for 2 and 5 min increased [Na+]i to a maximum of 42.0 3.3 and 66.7 5.6 mM, respectively (n = 4 coverslips, a total of 62 astrocytes). Also, the kinetics of the [Na+]i and [K+]i responses to 2 min applications of CCCP were markedly different. In the case of PBFI-derived signals, the application of CCCP resulted in an immediate increase in BI340/BI380 ratio values which rapidly declined when CCCP was withdrawn. In contrast, [Na+]i rose slowly to reach 18.2 2.8 mM (n = 4) at the end of the 2 min application of CCCP (i.e. ~5 mM above resting [Na+]i). The maximum increase in [Na+]i did not occur until ~5.5 min after CCCP washout, a time at which PBFI-derived signals had returned to baseline, further supporting the use of PBFI as an indicator of [K+]i in the present experiments. Consistent with previous reports in astrocytes (Rose and Ransom 1996; Rose and Ransom 1997; Su et al. 2002), resting [Na+]i was 13.0 0.9 mM (n = 4 coverslips).  181  Figure B.2. Applications of 12.5 mM K+o under HEPES and bicarbonate recording conditions. The application of 12.5 mM K+o caused a small increase in PBFI-derived BI340/BI380 ratio values under HEPES-buffered recording conditions. In comparison, the response to 12.5 mM K+ was greater in 5% CO2/95% air equilibrated bicarbonate-based recording solutions (pH was 7.35 throughout).  182  Appendix C: Supplemental figure for chapter 3  A  B  Figure C.1. Brain vasculature in Cx43+/+, Cx43+/-, Cx43 CT/+ and Cx43 CT/- mice. A, Mice brains from each genotype under study showing the vascular pattern of the region served by the MCA. Blood vessels in 5 mice from each genotype were labeled by transcardial perfusion of India ink. Vascular patterns were variable within each genotype, but no gross abnormalities were observed upon comparing each genotype. B, Micrographs of neutral red stained sections obtained with a 20x objective in the approximate region served by the MCA showing that India ink-labeled microvessels are similar in each of the genotypes.  183  Appendix D: Mitochondria protein interactions with Cx43 as identified by mass spectrometry  Table D.1: Potential mitochondrial protein interactions with Cx43 Using mass spectrometry for Cx43-protein interactions in C6 glioma cells, a database of over 150 potential interacting proteins has been identified (data courtesy of Dr. Vincent Chen). Mitochondrial proteins that interact with Cx43 are categorized based on their location within the mitochondria. Proteins involved in transport of peptides to the mitochondria are also shown. As mass spectrometry provides information about potential interactions, it is necessary to perform additional experiments to confirm the validity of these interactions (see Appendix E).  Matrix  Outer membrane Inner membrane  - ornithine aminotransferase - glutamic-oxaloacetic transaminasemitochondrial (aspartate aminotransferase2) - acyl-coenzyme a long chain - heat shock 60 kDa protein 1 - dihydrolipoamide dehydrogenase - prohibitin - voltage-dependent anion channel (VDAC)  - malate dehydrogenase - trifunctional protein-beta subunit - trifunctional protein-alpha subunit - glycine c-acetyltransferase - phospholipid hydroperoxide glutathione peroxidase - solute carrier family 25 (mitochondrial carrier phosphate carrier) member 3 - solute carrier family 25 (mitochondrial carrier adenine nucleotide translocator) member 4 - heat shock 60 kDa protein 1 - glutamate dehydrogenase 1 - NADH dehydrogenase Fe-S protein isoform CRA_A - ATP-H+ mitochondrial f1alpha subunit - ubiquinol-cytochrome c reductase core protein dihydrolipoamide s-succinyltransferase (e2 component of 2-oxo-glutarate complex) - ATP/H+ transporting mitochondrial f1 beta subunit - inner membranemitochondrial protein - thioredoxin Mitochondria - heat shock protein alpha class a member 1. - heat shock 70 kDa protein 9 (protein folding) transport - heat shock protein 90 kDa alpha class b member 1 (protein folding)  184  Appendix E: Co-immunoprecipitation of VDAC with Cx43  Figure E.1. Cx43 co-immunoprecipitates with VDAC in wild-type astrocytes. Astrocyte lysate or lysate incubated with the pull down antibodies anti-VDAC or anti-ANT were probed for Cx43 to determine if these proteins co-immunoprecipitate. Cx43 is present in astrocyte lysate and not control samples incubated with IgG antibody. As indicated by the black arrow, the outer mitochondrial membrane protein VDAC pulls down Cx43, while the mitochondrial protein ANT is not pulled down. The ~50 kDa bands in the VDAC lane likely represents staining of the rabbit anti-VDAC antibody with the anti-rabbit secondary antibody used to detect Cx43. Other bands may represent non-specific interactions or degradation products.  185  Appendix F: UBC animal care certificate  186  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0069988/manifest

Comment

Related Items