UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Functional studies of Calbindin-D28K and its role in intracellular calcium homeostasis Rintoul, Gordon Leslie 2000

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

Item Metadata


831-ubc_2001-611655.pdf [ 11.22MB ]
JSON: 831-1.0090479.json
JSON-LD: 831-1.0090479-ld.json
RDF/XML (Pretty): 831-1.0090479-rdf.xml
RDF/JSON: 831-1.0090479-rdf.json
Turtle: 831-1.0090479-turtle.txt
N-Triples: 831-1.0090479-rdf-ntriples.txt
Original Record: 831-1.0090479-source.json
Full Text

Full Text

Functional Studies of Calbindin-D 28K and its Role in Intracellular Calcium Homeostasis by Gordon Leslie Rintoul B.Sc. Biology, University of Western Ontario, 1990 Degree in Hons. Standing, Cell Biology, University of Western Ontario, 1992 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Physiology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A October 2000 © Gordon Rintoul, 2000. In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of / A y S i £>'O y S i 6>1 O^j y The University of British Columbia Vancouver, Canada Date /UoJe^J^A z ? . ^ 2&DQ Abstract Calbindin-D28k (CaBP) is a 28 kD calcium-binding protein found in specific neuronal populations in the mammalian brain. The hypothesized Ca2+-buffering action of CaBP is the basis of suggestions that this protein may serve to protect neurons against cell death mediated by large or prolonged increases in intracellular free Ca 2* concentration. However, to date, there is little direct evidence to support this hypothesis. To address this question directly, we have examined Ca -buffering by CaBP in stably transfected H E K 293 and HeLa cell lines. A variety of methods were employed to induce calcium transients, including transfection of N M D A receptors followed by activation with glutamate. In all experiments there was evidence of CaBP-mediated Ca2+-buffering. Moreover, when N M D A R transfected cells were exposed to excitotoxic concentrations of glutamate, cells expressing CaBP exhibited enhanced survival over controls. CaBP was unable to prevent acute necrotic cell death but significantly protected cells from delayed, presumably apoptotic cell death. To examine the potential influence of CaBP upon intracellular Ca -oscillations, stably-transfected HeLa cells were treated with histamine, while measuring intracellular Ca 2 + . The observation that CaBP flattened the profile of component C a 2 + peaks, coupled with data from H E K cell lines, provides unequivocal evidence that CaBP can act to buffer increases in intracellular Ca 2 + . Utilizing a novel method for resolving intracellular C a 2 + waves, it was found that transfection with CaBP, or loading with artificial C a 2 + buffers, attenuated the velocity of C a 2 + waves. The scope of attenuation appeared to be a function of the buffer binding kinetics. The rate of Ca2 +-binding by CaBP was apparently too slow to influence Ca 2 + -interaction between the closely situated IP3 receptors in initiation sites, where the faster on-Ill rate buffer B A P T A exerted a significant effect. However, both CaBP and B A P T A had significant effects upon events which were more distal to the source of C a 2 + release, including effects between the more sparsely distributed IP3 receptors involved in the propagation of C a 2 + waves, and global changes in Ca 2 + . In view of the fact that C a 2 + waves and oscillations have been shown to modulate neuron development and gene expression, it is possible that the effects of CaBP may include influencing these processes. iv Table of Contents Page Abstract ii Table of Contents iv List of Tables and Figures viii Acknowledgements x Chapter 1 General Introduction 1 Intracellular Calcium Homeostasis 2 Influx of Extracellular C a 2 + 3 N M D A Receptors 5 Release of C a 2 + from Intracellular Pools 6 IP3 Mediated Calcium Signaling 8 Calbindin-D28K 9 Distribution of CaBP in the CNS and Electrophysiological Correlates 13 Does CaBP have a Neuroprotective Role in the CNS ? 14 Studies in Human Disease IV Functional Studies of CaBP in the CNS and Model Cell Lines 17 Excitotoxicity 19 Rationale 21 Hypotheses 22 Chapter 2 Materials and Methods 23 Cloning and Expression of the Human Recombinant CaBP 24 Polymerase Chain Reaction 24 DNA Sequence Analysis 27 Protein Expression 28 Generation and Characterization of Stable Cell Lines 32 Mammalian Cell Culture 32 Stable Expression of CaBP 33 Estimation of CaBP Concentration in Transfected Cells 35 V Transient Expression of the N M D A Receptor 38 Measurement of Intracellular Calcium : NR1/NR2A Transfected Cells 39 Measurement of Intracellular Calcium : Flash Photolysis 40 Imaging Calcium Oscillations in HeLa Cells 41 Calcium Wave Propagation Analysis in HeLa Cells 41 Image Analysis & Statistical Methods 42 Excitotoxicity Studies 42 2+ Chapter 3 Effect of Recombinant CaBP on Intracellular Ca Transients in Stably-transfected HEK 293 Cells 44 Chapter Summary 45 Introduction 47 Results 49 Characterization of Ca Transients in NR1/NR2A Transfected HEK 293 Cultures49 Effect of CaBP on Glutamate Treated HEK Cells Transfected with NR1/NR2A 50 Modulation of Br-A23187 Mediated Ca Transients by CaBP 53 Effect of Stably Expressed CaBP in HEK Cells Following Flash Photolysis of NP-EGTA 57 Ca2+ Imaging of Stably transfected HEK cells treated with ATP 62 Effect of CaBP Upon Capacitative Ca2+ Influx 64 Effect of CaBP on Cell Survival Following Excitotoxic Activation of NMDA Receptors 64 Discussion 66 Effect of CaBP on NR1/NR2A Mediated Ca2+ Transients 68 4-Br-A23187 71 ATP-Induced Release of Ca2+from the ER 72 NP-EGTA 73 Capacitative Ca2+ Entry 74 Overview of Ca2+ Imaging Experiments 75 Excitotoxicity 78 Summary/ Conclusions 80 Chapter 4 Effect of Recombinant CaBP on IP3 Mediated Calcium Oscillations 82 Chapter Summary 83 Introduction 84 Results 86 Effects of Indicator Dyes on Ca2+ Oscillations 86 Effect of Artificial Calcium Buffers on the Oscillatory Behavior ofHeLa Cells 88 Effect Of CaBP on Calcium Oscillations 90 Calcium Wave Propagation in Stable HeLa Cell Lines 96 Effect of Artificial Calcium Buffers and CaBP upon Calcium Wave Propagation. 102 Discussion 104 Calcium Oscillations in HeLa Cells 104 Effects of Indicator Dyes 105 Ca2+ Diffusion, Fixed and Mobile Ca2+ Buffers 106 Ca2+ Oscillations and Artificial Ca2+ Buffers in HeLa Cells 106 Modulation of [Ca2+]j Oscillations by CaBP 110 Calcium Waves I l l Mechanism of the Attenuation of Ca2+ Wave Velocity 112 Summary / Conclusions 114 Chapter 5 General Discussion & Conclusions 117 CaBP and C a 2 + Transport 118 CaBP as a Trigger Protein 118 CaBP as a C a 2 + Buffer 120 Possible Consequences of C a 2 + Buffering by CaBP 122 CaBP, Mitochondria and Cell Death 124 CaBP as Modulator of Intracellular Signaling 125 Future Studies 127 Studies in Transfected HEK Cells 127 Studies in Primary Hippocampal Cultures 128 In Vivo Studies 129 Conclusions and General Summary 130 Abbreviations 133 References 134 vii List of Tables and Figures Page Table 1 Constants (Kd, k o n and k0ff) of C a 2 + chelators, Ca2+-indicators and 9-1-Ca binding proteins 11 Table 2 Summary of Effect of Stably transfected CaBP in H E K 293 Cells 76 Figure 1 Major routes of C a 2 + uptake and release involved in intracellular C a 2 + homeostasis 4 Figure 2. PCR of Human CaBP cDNA 26 Figure 3. Purification of recombinant human CaBP 30 Figure 4. Spectrophotometric studies of recombinant human calbindin > 2 + demonstrating a conformational change upon binding Ca 31 Figure 5. Expression of CaBP in a stable H E K cell line 34 Figure 6. Quantification of CaBP Expression in Stably Transfected Cells 37 Figure 7. Response of NR1/NR2A transfected H E K 293 cells to glutamate 51 Figure 8. Response of CaBP and control stably-transfected cells transiently transfected with NR1/NR2A to treatment with glutamate 52 Figure 9. Pooled data of recovery of stable H E K 293 cells transiently transfected with NR1/NR2A and stimulated with glutamate 54 Figure 10. Effect of CaBP upon 4-Br-A23187-induced C a 2 + transients in HEK293 cells 55 viii Figure 11. Pooled calcium imaging data from stable H E K 293 cells treated with 4-Br-A3187... 56 Figure 12. Effect of CaBP on recovery following flash photolysis of NP-EGTA in fluo-3 loaded stable H E K cells 58 Figure 13. Effect of CaBP on recovery following flash photolysis of NP-EGTA in fura-2 loaded stable H E K cells 60 Figure 14 Pooled data of recovery from NP-EGTA released C a 2 + in fura-2 loaded stable H E K cell lines 61 Figure 15. Effect of CaBP upon ATP induced C a 2 + release in stable H E K 293 cells 63 Figure 16. Effect of CaBP upon capacitative influx of Ca in stably transfected H E K 293 cells 65 Figure 17. Effect of CaBP upon NRl/NR2A-mediated cytotoxicity in stably-transfected H E K 293 cells 67 Figure 18 Classification of Calcium Oscillatory Types in HeLa cells 87 Figure 19. Effect of Calcium buffers on Oscillatory Behaviour of HeLA Cells 89 Figure 20. Effect of recombinant CaBP on calcium oscillatory pattern in stable HeLa cell lines 91 Figure 21. Influence of CaBP upon individual calcium transients in oscillating stable HeLa cells 93 ix Figure 22 The effect of CaBP upon rate of C a 2 + increase in stable HeLa cells treated with 100 uM histamine 94 Figure 23. Effect of CaBP on response of HeLa cells loaded with fura-2 and treated with 1 u M histamine... 95 9-1-Figure 24. Effect of recombinant CaBP upon Ca influx in fura-2 loaded HeLa cells treated with histamine 97 Figure 25. Example determination of wave velocity for an individual HeLa cell treated with 100 u M histamine 99 Figure 26 Resolution of two calcium waves, initiated at distinct sites in an individual HeLa cell treated with 100 u M histamine 100 Figure 27 Summary of resolved calcium wave velocities in HeLa cells 103 X Acknowledgements I am greatly indebted to my supervisor, Ken Baimbridge for allowing me to carry out these studies in his lab. I wish to express my sincere gratitude for his support, guidance and encouragement, well above and beyond the call of duty, throughout all stages of my degree. I am particularly grateful for his many suggestions and seemingly infinite patience during the preparation of this thesis. I would also like to thank Stella Atmadja for her kindness and assistance in the lab. Many members of the Baimbridge Lab, past and present, have been very supportive. In particular, I would like to thank Khaled Abdel-Hamid and Claudia Krebs who were very encouraging and provided much valued and insightful discussion in the initial and final stages of my degree respectively. I would also like to thank the many previous members of the K G B lab, too numerous to list here, both graduate students and 4 t h year honours Physiology students. Two standouts that I would like to mention are; Sebastian Howie, who, in the dark ages before automated sequencing, assisted me in manually sequencing many clones of the Calbindin gene and Adam Sidky who got me started with the NP-EGTA experiments. I would like to thank Dr. Ross MacGillivray for allowing me to work in his lab and for his gargantuan generosity. I was welcomed into the MacGillivray lab as one of their own. I am indebted to JP Heale, who, while pursuing his own experiments, took time to patiently teach me most of the molecular biological techniques described in this thesis. In addition, JP's humour, companionship and (loud) musical selections helped me through many late nights in the MacGillivray lab. I would also like to thank Jeff Hewitt for helping me design & synthesize the Calbindin primers and for his assistance in many of my experiments. Thanks also go to Bea Tarn, Hung Vo, Leisa Steinbergs and Mark Brown who were particularly kind and helpful during my stay in the MacGillivray lab. The H E K cell experiments described in this thesis are a result of our collaboration with Dr. Lynn Raymond, who provided invaluable guidance. The assistance of Nansheng Chen and Chris Price in the Raymond lab in developing the transfection of H E K cultures was much appreciated. I am also very grateful to Dr. Raymond for her critical reading of this thesis and many helpful suggestions. In my many years at U B C , I have encountered no department which equals the Physiology Department for it's friendly and social atmosphere. This is largely due to the tireless efforts and generosity of Dr. Ray Pederson. I would like to thank both Ray and Margaret for many fond (albeit sometimes blurry) memories of "Grad Retreat" on Mayne Island. The family feel to the department was also contributed to by many grad students; I cannot thank them all here but I would like to say thanks to Val Smith, Claire Sheldon, Chris Brett, Tony Kelly, Herman Fernandes, Gabi Weichert and K i m Barker. In the final phases of the preparation of this thesis, Pauline Dan and Nathalie Gaudreault were particularly encouraging and helped to keep both my spirits and blood caffeine levels high. Special thanks also go to Neville Rising, Dave Leggett, Jeannine Ray, Bryce Pasqualotto and Rick Gelling for providing the occasional, much needed distraction from my studies. Their companionship and encouragement helped to entertain and sustain me throughout this degree. Many other members of the Physiology Department went out of their way to help and encourage me along the way. I would like to thank Tony Pearson and Kenny Kwok for their many words of wisdom and good humour. The staff members of the department were also very kind to me; Zaira, Nancy, Marie, John, Jack, Joe and Dave helped to make the Physiology department a great place to work. I am also very appreciative of the efforts of my supervisory committee members. My chair, Dr. David Mathers, I thank for critically reading my thesis and making many helpful suggestions. I would like to thank all my committee members; Dr. Rob Douglas, Dr. John Church and Dr. Ed Moore for providing much appreciated guidance. I would like to express my appreciation to my examination committee, Dr. Cassey van Breemen, Dr. Ron Reid, Dr. Ross MacGillivray, Dr. Jim McLarnon and Dr. Marco Celio for their many suggestions for improving this thesis. Finally, I would like to thank my family: my brothers David and Peter, and my parents, Gord Sr and Mary. Far from being the so-called "forgotten middle child", from them I have received limitless love, support and encouragement which made this thesis possible. Dedication To my parents, Gord Sr & Mary Rintoul. Who taught me all the important stuff. This thesis is a direct result of your love and support Thanks Mum. Thanks Dad. Chapter 1 General Introduction 2 Intracellular Calcium Homeostasis Within the cells of all organisms, the level of the divalent cation calcium (Ca 2 + ) is maintained at a concentration greater than 10,000-fold lower than that of the extracellular environment. As early forms of life evolved utilizing a energy currency involving phosphates, the maintenance of a low intracellular calcium concentration ([Ca 2 +]i) was critical to cell survival, as high concentrations would result in phosphate precipitation. It seems likely that a corollary of this safety mechanism was the evolution of systems which utilized controlled calcium flux to transduce signals of internal and external origin. This change in [Ca2+]j could be detected by calcium physically binding to target molecules or by the resultant charge migration. Thus the maintenance of a low [Ca ]i became necessary not only for cell survival but also to enable signal transduction; only against a precisely regulated background of resting [Ca ]i are cells able to sense and respond to influx or release of Ca (Williams, 1990). The low resting intracellular levels of Ca 2 + , usually less than lOOnM, are maintained through active processes which balance the slow leak of C a 2 + through the plasma membrane. These active processes result in C a 2 + being transported back across 2_|_ the plasma membrane or sequestered in intracellular organelles. Active extrusion of Ca 2"f" is largely handled by the high affinity, low capacity Ca -ATPase in the plasma membrane while active sequestration is accomplished by similar ATP-hydrolysing pumps in the membrane of the endoplasmic (or sarcoplasmic) reticulum (ER / SR). 74-When a high capacity for Ca extrusion from the cytoplasm is required, active transport 3 o f C a 2 + may also be carried out by the low affinity plasmalemal Na + -Ca 2 + exchanger (Fig-1). 9+ Against this background of dynamic, regulated [Ca ]i, the intracellular release or influx of C a 2 + from the external environment is able to mediate a myriad of cellular functions, either through 2 n d messenger systems or effects on membrane potential. The physiological roles of the intracellular Ca signal are very diverse, including gene transcription, contraction, cell division, neurotransmitter release (or secretion), metabolism, and learning and memory (for a recent, general review of functions see Berridge, 1998). 9-+-Influx of Extracellular Ca Changes in [Ca2 +]i can be achieved by the regulated movement of C a 2 + from the external environment into the cytoplasm through three classes of Ca channels. The first are voltage gated channels which respond to membrane depolarization (reviewed by 9+ Armstrong and Hille, 1998). The second are mechanoreceptors, that allow influx of Ca in response to cell membrane stretch or tension (Gotoh and Takahashi, 1999). The third are ligand-gated channels that are dependent upon the binding of other molecules for their activity including the N-methyl-D-aspartate (NMDA) subclass of glutamate receptors (reviewed by Hollmann and Heinemann, 1994). 4 Figure 1. Major routes of Ca 2 + uptake and release involved in intracellular Ca2" homeostasis. A l l arrows indicate directional movement of C a 2 + unless otherwise indicated. ER - Endoplasmic Reticulum; Mito - Mitochondrion; C R A C - Calcium Release Activated Channel; P M C A - Plasma Membrane Ca 2 +-ATPase; ROCC -Receptor-operated C a 2 + channel; V O C C - Voltage-operated C a 2 + channel; RyR -Ryanodine Receptor; IP3R - IP3-Receptor; SERCA - Sarco/Endoplasmic Reticulum Ca 2 +-ATPase; PTP - Mitochondrial Permeability Transition Pore 5 NMDA Receptors Glutamate is the most prevalent excitatory neurotransmitter in the mammalian brain. A major class of glutamate receptors are those which function as ion channels. These ionotropic receptors have been classified according to their pharmacological and electrophysiological properties: a-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA) receptors, kainate (KA) receptors, and N M D A receptors. The N M D A receptor, named for its selective activation by N M D A , has been the subject of intense scientific investigation, due largely to its suggested role in long-term potentiation (LTP) (Bliss 2_|_ , and Collingridge, 1993) and Ca -mediated cell death (reviewed by Choi, 1994). N M D A receptors, like other ionotropic glutamate receptors, are permeable to N a + and K + . However, they differ from other receptor subtypes in this family in that they are also permeable to C a 2 + (MacDermott et al., 1986), are blocked by M g 2 + in a voltage dependent fashion (Nowak et al., 1984) and require glycine as a co-agonist (Johnson and Ascher, 1987). This channel therefore integrates chemical and electrical signals: in order for it to conduct ions it must bind glutamate and glycine and the Mg block must be removed by membrane depolarization. N M D A receptors consist of multiple subunits. Functional receptors all require NR1, a 97 kDa protein which, when expressed in Xenopus oocytes, conducts currents which are small relative to those measured in neurons, but nonetheless exhibits M g 2 + -dependent block and C a 2 + conductance (Moriyoshi et al., 1991). Fully functional N M D A receptors are likely hetero-oligomers of NR1 and one of four related subunits (NR2A-D). In experiments where NR1 and one of NR2A-2D proteins are co-expressed, 6 currents 5-60 times the amplitude of those measured in cells transfected with NR1 alone are recorded but no measurable currents are produced in experiments where NR2 subunits are expressed alone (Monyer et al., 1992). Functional heterogeneity of N M D A receptors is achieved through differential expression of the NR2 subunits in the mammalian CNS. When combined with NR1, different NR2 subunits confer distinct pharmacological and ion channel properties to the N M D A receptor. The NR1 receptor is expressed widely in the brain (Petralia et al., 1994). However, expression of the various NR2 subunits varies with respect to the region of the CNS and stage of development (Watanabe et al., 1993, Monyer et al., 1992). This spatial and temporal variation in subunit expression provides for receptor diversity and therefore fine tuning of receptor activation and functional effects. Functional roles have been attributed to the N M D A R based upon its ability to detect co-incidence of glutamate receptor binding and membrane depolarization. N M D A R participation in learning and memory, presumably through a mechanism related to LTP, has been extensively studied (for a current review see Malenka and Nicoll, 1999). The potential pathophysiological role of the N M D A R has also received much attention, being implicated in acute and degenerative neuronal cell death (Choi, 1992, 1995 ; Bonfoco et al., 1995). Release of Ca 2 + from Intracellular Pools The ER of most cells constitutes a major intracellular store of Ca 2 + . Uptake of C a 2 + from the cytoplasm occurs through active transport by sarco(endo) plasmic 7 reticulum ATPases (SERCA pumps) in the ER membrane. This family of ATPases consists of three members, SERCA 1-3. SERCA 1 pumps are localized almost exclusively in fast-twitch skeletal muscle. SERCA 2 pumps exist in two isoforms as a result of alternative splicing: SERCA 2a is expressed in cardiac and smooth muscle and SERCA 2b appears to be the major form expressed in brain tissue. SERCA 3 is expressed in a variety of tissues, including the cerebellum (Wu et al, 1995). A n understanding of the cellular role of SERCA pumps has been facilitated by the use of the tumor promoting agent thapsigargin which specifically blocks the action of SERCA pumps and causes depletion of ER Ca stores (Thastrup et al., 1990). Release of C a 2 + from the ER occurs through one of two related types of channels. In non-excitable cells, release of C a 2 + from the ER occurs through a receptor activated by the intracellular second messenger inositol (1,4,5) trisphosphate (IP3). In excitable cells such as neurons, in addition to IP3 receptors (IP3R), the ER membrane contains channels which are directly activated by Ca 2 + . These are called ryanodine receptors (RyR), due to their sensitivity to the plant alkaloid, ryanodine (reviewed by McPherson and Campbell, 1993) Mitochondria constitute the second major intracellular store of Ca . Uptake of C a 2 + by mitochondria occurs via a C a 2 + uniporter, driven by the electrochemical gradient across the inner mitochondria membrane. Efflux of C a 2 + from mitochondria occurs through a Na + /Ca 2 + exchanger or the high conductance mitochondrial permeability transition (MPT) pore (Hunter and Haworth, 1979; Crompton et al., 1987). 2_[_ The MPT pore can be activated by high levels of intramitochondrial Ca and has been suggested to play a role in apoptotic cell death (Lemasters et al., 1998). 8 IP3 Mediated Calcium Signaling HeLa cells, a prototypical non-excitable cell line, respond to application of 9+ histamine with IP3 mediated intracellular Ca release (Bootman and Berridge, 1996). Histamine initiates phospholipase C-mediated hydrolysis of the membrane phospholipid phophatidylinositol 4,5-bisphosphate (PIP2) to D A G and IP3. IP3 diffuses rapidly in the cytosol and activates the IP3-receptor (IP3R) located on the surface of the ER. The 9+ activity of this channel is also modulated by cytosolic Ca . This modulatory effect exhibits a bell shaped sensitivity, with a peak synergistic effect occurring at 200-300nM [Ca 2 +] i (Bezprozvanny et al., 1991; Finch et al., 1991). This biphasic sensitivity of the receptor is thought to underlie the generation and propagation of C a 2 + waves and oscillations. Oscillations of [Ca2+]j have captured the attention of many researchers, as it has become apparent that oscillatory frequency can encode activation of various cellular processes such as gene expression (Berridge, 1997). As such, factors which can modulate calcium oscillations may play a critical role in defining how a particular cell type responds to external signals which induce intracellular calcium release. 9-1-Individual Ca spikes within an oscillatory response are not simply global increases in [Ca 2 + ] i . Each spike can be resolved temporally and spatially. Rises in cell C a 2 + begin in a defined region and spread throughout the cell. The defined "front" of this C a 2 + release has been described as a C a 2 + wave. It has been hypothesized that the basis of such waves involves a feed-forward mechanism, in which IP3 acts upon IP3 receptors 9 on the ER, releasing calcium which in turn acts synergistically with IP3 to activate adjacent IP3 receptors (Bootman et al., 1997). In excitable cells, ryanodine receptors co-exist with IP3 receptors on the surface of the ER. These receptors open in response to an increase in [Ca ]\, and hence may play a part in the generation of Ca waves and oscillations in these tissues. Therefore, in either excitable or non-excitable tissue, a major determinant of [Ca2+]j dynamics may involve C a 2 + feedback or feed-forward mechanisms. In this context, it is probable that any factor which may compete with IP 3 or ryanodine receptors for calcium, or which may influence the diffusion of calcium, may modulate the generation and/or propagation of calcium waves and oscillations. Calbindin-D28K Another possible fate of cytoplasmic Ca is that of binding to intracellular Ca -binding proteins. Currently these proteins are classified according to'their putative function: either "trigger" or "buffer" (Baimbridge et a l , 1992). Trigger proteins such as the ubiquitous calmodulin are components of second messenger pathways mediating many cellular functions. The function of buffer proteins is to bind excess Ca z as [Ca z +]i increases, although it is possible that these proteins have as yet undiscovered targets. One member of this protein family is calbindin-D28K (CaBP). Previously known as vitamin D-dependent calcium binding protein, CaBP is a 28kD intracellular calcium binding protein (Wasserman and Taylor, 1966). Examination 10 ofthe amino acid sequence of CaBP reveals six regions which encode the consensus sequence for the helix-loop helix structure of the "EF-hand" motif responsible for high-affinity Ca -binding (Parmentier et al., 1987). However, under physiological conditions 9+ only four of these Ca -binding sites appear to be functional (K d =513 nM) (I. Mody, Personal communication; Bredderman and Wasserman, 1974; Cheung et al., 1993) (Table 1). In the calcium binding protein superfamily, trigger proteins such as troponin C or calmodulin are components of second messenger cascades, undergoing 9+ conformational changes upon binding of Ca and subsequently modulating the activity of target molecules. These proteins are not uniformly distributed within cells but are often concentrated in specific regions such as the post-synaptic density where they bind to intracellular cytoskeletal or contractile elements. CaBP is generally considered to be a member of the "buffer" category of calcium binding proteins along with parvalbumin and calretinin. These proteins are distinguished by the fact that they appear to be uniformly distributed within the cytoplasm of cells that express them (Baimbridge et al., 1992). Although the hypothesized role of CaBP is to act as a buffer of intracellular Ca , it has yet to be demonstrated that Ca2+-buffering is its exclusive role. Indeed, recent structural (Berggard et al., 2000) and experimental (Bellido et al., 2000) observations lend support to the idea that CaBP may not only function as an intracellular Ca buffer, but may interact with target proteins in the cell. CaBP was originally isolated from chick epithelial tissue by Wasserman and Taylor (1966), and subsequently shown to be found in many other Ca2+-transporting epithelia, including the distal tubule ofthe kidney and the chicken oviduct. In C a 2 + transporting cells, the synthesis of CaBP is regulated by the active metabolite of vitamin 11 (xlO6 M ' Y 1 ) k0ff (S"1) Kd (nM) BAPTA 600 [1, 2] 300 [3] 500 [1] EGTA 1.5 [1] 0.45 [4] 860 (pH7.0) [1] 220 (pH 7.3) [1] FLUO-3 920 [5] 424 [5] 462 [5] FURA-2 760 [5] 109 [5] 144 [5] CaBP-9k 20 [6] 8.6 [6] 430 [6 ] CaBP-28K 77 [7] 39.5 [7] 513 human [7] 300 chick [8] Table 1. Constants (Kd, kon and koir) of Ca chelators, Ca -indicators and Ca binding proteins. Constants were determined at 22°C unless otherwise indicated. References: (1) Adler et al., 1991. Reported values were measured under physiological conditions. (2) Tsien, 1980. (3) Pethig et al., 1989. (4) Harafuji and Ogawa, 1980. (5) Lattanzio and Bartschat, 1991. (6) Feher et al., 1992. (7) I. Mody and U.V. Nagerl, U C L A , Personal Communication .Unpublished results of experiments employing recombinant protein generated as described in materials and methods. (8) Cheung et al., 1993. 12 D, 1,25-dihydroxy-cholecalciferol in a typical steroid-like manner. Expression of CaBP is not limited to Ca -transporting tissues. It has been detected in many vertebrate tissues with notable exceptions being striated muscle and mammalian intestinal epithelium (Wasserman and Fullmer, 1982 ). In vertebrates, CaBP is found in both central and peripheral nervous systems. CaBP expression in the nervous system differs from expression in Ca -transporting tissues in that it is not dependent on vitamin D (Baimbridge and Parkes, 1981; Varghese et al., 1988). The concentration of CaBP within rat cerebellar Purkinje cells has been estimated to be about 0.1-0.2 mM, and since each molecule can bind up to four Ca 2 + , the total Ca2+-buffering capacity is therefore in the order of 0.4-0.8 mmol/litre cytoplasm (Baimbridge et al., 1982). This extraordinarily large buffering capacity is similar to that found in the absorptive cells of the chick gut (Feher et al., 1992), and a similar concentration has been reported for rat cerebellar Purkinje cells using a quantitative immunohistochemical procedure (Davenport et al., 1990) and a fura-2 titration method (Fierro and Llano, 1996). In the absence of evidence for a "trigger" function it has been speculated that this protein may form a major intraneuronal C a 2 + buffering system, (Baimbridge and Parkes, 1981; Baimbridge et a l , 1982; Jande et al., 1981; Mody et al., 1987). 13 Distribution of CaBP in the CNS and Electrophysiological Correlates Within the nervous system CaBP is found in distinct neuronal populations such as cerebellar Purkinje cells and the principal cells of the dentate gyrus and CA1/CA2 sub-fields of the hippocampal formation (Baimbridge and Miller, 1982; Buchan and Baimbridge, 1988; Celio,1990; Jande et al., 1981). The distribution studies have described the presence of CaBP in a variety of neurons with different functions, neurotransmitter content or receptor profiles. For example, CaBP is found in major projecting pathways that are either inhibitory (Purkinje cells) or excitatory (hippocampal dentate granule and CA1 pyramidal neurons), as well as local circuit inhibitory intemeurons in the cortex. CaBP is found in some, but not all neurons with a high density of N M D A receptors or IP3 receptors, and examples could be given of CaBP associated with almost all of the known classical neurotransmitters (Baimbridge et al., 1992). The intra-neuronal distribution of CaBP appears to be largely cytosolic, with an immunohistochemical staining pattern that is even in intensity throughout the cell volume as observed with light microscopy, and no evidence of association with specific membranes or organelles in Purkinje neurons observed at the E M level (Pasteels et al., 1986). Kawaguchi and Kubota (1993) examined the electrophysiological properties of rat frontal cortex neurons and correlated their response with the presence of either CaBP or parvalbumin (PV). PV was present in neurons which showed fast-spiking characteristics in response to either intracellular current injection or synaptic stimulation. This result was consistent with findings in rat hippocampal CA1 intemeurons (Kawaguchi et al., 1987) and a general association of PV with neurons that are functionally more active (Celio, 1990). CaBP-positive neurons had low threshold spikes and a distinct morphology compared with the PV-positive 14 neurons, but both CaBP-positive and -negative neurons in the rat striatum show similar low threshold spikes (Kawaguchi et al., 1987). In an interesting report, L i et al (1995) determined that introduction of CaBP into rat supraoptic neurons using patch clamp technology, suppressed Ca2+-dependent depolarizing afterpotentials and changed activity from phasic to continuous firing. However, Baimbridge et al (1991) failed to show an association between the presence or absence of CaBP and either non-accommodating or bursting responses to current injection in rat CA1 pyramidal neurons, and others failed to show any relationship between CaBP and high threshold Ca 2 + conductances in guinea-pig lateral septal neurons (Doutrelant et al, 1993). Does CaBP have a Neuroprotective Role in the CNS ? Mattson et al (1991) reported a correlation between the presence of CaBP and a relative sparing of cultured hippocampal neurons to excitotoxic levels of glutamate. However, since CaBP-positive neurons represent a distinct phenotype, it is certainly conceivable that other factors may have contributed to neuron sparing. For example, there may be reduced numbers of E A A receptors or voltage operated Ca2+-channels (VOCCs) which could result in a reduced Ca 2 + entry during exposure to excitotoxins. These result were not supported by Mockel and Fisher (1994) who found no significant correlation between the presence of CaBP and sensitivity to excitoxicity in similar cultures. The effects of artificial Ca 2 + buffers on changes in E A A and high K+-evoked increases in [Ca2+]j in cultured hippocampal neurons and their effects upon susceptibility to 15 excitotoxicity have been examined (Abdel-Hamid and Baimbridge, 1997). Preloading neurons with l,2-bis(2-ammophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA) resulted in an effective buffering of increases in [Ca2+]j induced by transient (<25 sec) exposure to N M D A or 50 mM K + as was evident by the reduction in the rate of rise and the peak [Ca2+]t response, together with a significant prolongation of the time to recover to prior resting [Ca2+]j levels. In addition, an enhanced susceptibility to excitotoxicity in BAPTA-loaded neurons has been reported (Dubinsky, 1993; Abdel-Hamid and Baimbridge, 1997). More recently, it has been shown that neurons naturally expressing CaBP have larger [Ca2+]j responses, but, like BAPTA, these responses are prolonged and excitotoxicity is enhanced (Abdel-Hamid et al, Unpublished Results). As with Mattson et al (1991) the problem of the correlative nature of the data arises, and alternative explanations are possible. Evidence supporting a neuroprotective role of CaBP has been reported by Goodman et al (1993) in which only those later-developing granule cells close to the hilus (which are devoid of CaBP-irnmunoreactivity; Baimbridge, 1992) were vulnerable to anoxia-induced cell death in 7 day old rat pups. In a similar model it has been shown that the non-CaBP containing neurons in the dorso-lateral quadrant of the striatum are more sensitive to anoxia-induced cell death than the CaBP-positive neurons in the ventro-medial quadrant (Burke and Baimbridge, 1993). Additionally Sloviter (1989), using a perforant path stimulation model of epilepsy, has suggested that hilar neurons devoid of either CaBP or PV immunoreactivity, were selectively vulnerable, an argument elegantly supported by Scharfman and Schwartzkroin (1989) who showed that these vulnerable neurons could be protected by the presence of BAPTA. A further model of cell death that has supported a neuroprotective role of CaBP, and one which has been reproduced in a number of laboratories, is the selective 16 destruction of dopamine neurons in the substantia nigra pars compacta (SNc) following treatment with the neurotoxic agent, l-Methyl-4-phenyl-l,2,3,64etrahydropyridine (MPTP) (German et al., 1992; Iacopino et al., 1992; Lavoie et al., 1991). The essential finding is that the dorsal tier of dopamine neurons in the SNc which contain CaBP (Gerfen et al., 1985, 1987) are relatively spared from cell death induced by MPTP treatment. This data is also entirely correlative and there are other possible explanations. For example, the CaBP containing dorsal tier of dopaminergic neurons in SNc have lower levels of the dopamine transporter which is required for MPTP uptake (Sanghera, 1994). Evidence against a neuroprotective role of CaBP can be found by a careful examination of the effects of global forebrain ischemia in the rat. This results in the complete destruction of the CA1 pyramidal neurons in the hippocampal formation, regardless of whether or not they contain CaBP (Freund et a l , 1990; Mudrick and Baimbridge, 1989). CA3 neurons are resistant to ischemia-induced cell death and yet these neurons contain neither CaBP, PV or Calretinin, another neuron specific calcium binding protein with a high homology with CaBP but distinct neuronal distribution (Rogers, 1987). This lack of correlation between calcium-binding protein content and vulnerability to ischemia is in fact not just restricted to the hippocampus but extends to the neocortex, thalamus and striatum (Freund et al., 1990). Further evidence against a neuroprotective role has been obtained from studies of the effects of ischemia on CaBP "knockout" mice (CB -/-) in which less damage occurred in the CA1 region when compared to control (CB +/+) littermates with their normal complement of CaBP (Klapstein et al., 1998). 17 Studies in Human Disease CaBP has been detected immunohistochemically in distinct populations of human neurons, the loss of which has been clearly identified with a particular disease. For example CaBP is present in the cholinergic neurons of the nucleus basalis of Meynert (Celio and Norman, 1985), and some of the cortical neurons which are lost in Alzheimer's disease (Ichimiya et al., 1988; Jande et al., 1981). Similarly, CaBP is present in a very large percentage of the striatonigral projecting neurons (Gerfen et al., 1985,1987) which are lost in Huntington's disease (Seto-Oshima et al., 1988). Seto-Oshima et al (1988) have proposed that the normal presence of CaBP in the neostriatal neurons, which receive a major excitatory input from the cortex, may serve to prevent excitotoxic effects and that the selective loss of these neurons in Huntington's disease may reflect a failure of the normal Ca2+-buffering capability. In contrast, there is a relative sparing of the CaBP containing neurons in the dorsal tier of dopamine neurons in the human SNc when examined in post-mortem samples from Parkinson's disease patients (Yamada et al., 1990). Further aspects of this topic have been reviewed (Heizmann and Braun, 1992). Functional Studies of CaBP in the CNS and Model Cell Lines A specific reduction of CaBP has been reported in the dentate granule cells of the rat hippocampal formation following kindling (Baimbridge and Miller, 1984; Baimbridge et a l , 1985; Mody et al., 1987) and a similar loss of CaBP in dentate granule cells from human 18 temporal lobe epilepsy has been recently reported (Magloczky et a l , 1997). Using the kindling model, Kohr et al (1991) have associated this loss of CaBP with a reduced Ca 2 + entry through voltage-operated Ca 2 + channels. They interpreted this data as indicating that the loss of CaBP allows for a more efficient Ca 2 + dependent inactivation of the Ca 2 + channels, and this was supported by the ability of the presence of BAPTA in the recording electrode to reverse this effect. It should be noted that the implication of this result is that the presence of a Ca 2 + -buffer will increase the net flux of Ca 2 + into a neuron, consistent with our own observations (Abdel-Hamid et al, Unpublished Results), and the effects of BAPTA upon excitotoxicity reported by Abdel-Hamid and Baimbridge (1997). However, Chard et al (1993) did not observe any effect of CaBP on Ca 2 + channel inactivation following its direct injection into rat dorsal root ganglion neurons. They did find an 8-fold decrease in the rate of rise in [Ca2+]j, a reduction in the peak [Ca2+]j and an altered kinetics of decay of [Ca2+]j to a single slow component, following a brief depolarization. Thus they concluded that CaBP could act as a Ca 2 + buffer but that this effect had no influence on Ca 2 + channel kinetics. Transfection of the gene for CaBP has been achieved in a number of different cells. In GH3 cells CaBP overexpression was associated with a reduced Ca 2 + entry through VOCCs and a reduction in [Ca2+]; transients evoked by depolarization (Lledo et al., 1992). In lymphocytes, overexpression of CaBP was associated with an increased survival in the presence of the Ca 2 + ionophore A23187 and a decrease in the apoptotic effects of dexamethasone and forskolin (Dowd et al., 1992). In cultured rat neurons, transfection of CaBP using adenovirus was associated with a loss of post-tetanic potentiation (Chard et al., 1995). In motoneuron hybrid cells, transfection of CaBP with a retrovirus prevented amyotrophic lateral sclerosis IgG-mediated cytotoxicity (Ho et al., 1996). Despite these 19 interesting findings it is clear that the results of these transfection studies, and those using other means of experimentally manipulating levels of CaBP, have not resolved the function of CaBP. Indeed, in some case, the data is entirely conflicting (Kohr et al.,1991; Lledo et al., 1992). Excitotoxicity The concept of "excitotoxicity" was introduced by Olney et al (1971) to describe the deleterious effects of excessive stimulation of neurons with E A A . The essential role of C a 2 + in cell death was suggested by Schanne et al (1979). Their "calcium hypothesis" of cell injury proposed that a massive influx of C a 2 + mediates, or initiates a "final common pathway" to cell death. Influx of C a 2 + through N M D A R has been implicated in numerous neuropathologies, largely due to the apparent ability of N M D A R antagonists to ameliorate glutamate mediated neurotoxicity (reviewed by Choi, 1988). The central role of C a 2 + in excitotoxicity has been underscored in more recent reports which have attempted to correlate induced increases in [Ca ]i and subsequent cell death. Studies employing the fluorescent indicator fura-2 reported no significant correlation between the magnitude of N M D A R mediated [Ca 2 +]i increase and cell death (Tymiyanski et al., 1993a; Dubinski and Rothman, 1991). On the basis of these results, it was hypothesized that the route of C a 2 + entry may be a more important determinant of cell survival following C a 2 + influx (Tymianski et al., 1993a). This hypothesis has 2_|_ received support through claims of "privileged access" of Ca from N M D A R to 20 mitochondria, the reported mediators of apoptotic cell death (Peng and Greenamyre, 1998). However, studies which examined 4 5 C a 2 + influx in response to N M D A R activation did report a correlation between amount of C a 2 + influx and cell death (Eimerl and Schramm, 1994; Hartley et al., 1993; Lu et al., 1996). This apparent discrepancy may be explained by recent reports which have examined the accuracy of high affinity calcium indicators (such as fura-2) in [Ca2+]j determination. Hyrc et al (1997) demonstrated that fura-2 selectively underestimates the [Ca 2 +]i associated with influx through N M D A R . Stout and Reynolds (1999) confirmed these findings and reported that when low affinity calcium indicators were employed, there was indeed a correlation between [Ca ], and cytotoxicity. Interestingly, transfection and activation of the cloned subunits of the N M D A R into non-neuronal cell types can produce "acquired excitoxicity": co-expression of the 2_j_ N R l and NR2A/B subunits has been shown to mediate increases in [CaZT]i (Grant et al., 1997) and leads to excitotoxic cell death (Cik et al., 1993; Anegawa et al., 1995; Boekman and Aizenman, 1996; Raymond et al., 1996). 21 Rationale On the basis of its known Ca2 +-binding properties, and in view of the lack of evidence to support a modulatory role for this protein, it has been speculated that CaBP functions primarily as an intracellular Ca2+"buffer that may serve to protect neurons from the potentially harmful effects of large and/or prolonged increases in [Ca 2 +]i. However, evidence in the literature provides support both for and against such a function, possibly in part due the many different experimental approaches that have sought to elucidate such a function, both in vivo and in vivo. The objective of the present studies was to utilize simple cellular systems to examine both the proposed Ca -buffering and protective functions of CaBP. The use of stably transfected cell lines ensured that the only variable altered between control and experimental cells would be the expression of CaBP. In addition, much of the interest in the function of CaBP has been focussed on its potential role under pathophysiological situations with little attention being placed upon its function(s) under physiological conditions. Therefore, a further objective of the present studies was to investigate the influence of CaBP upon intracellular Ca 2 + -signalling events such as Ca oscillations and waves. 22 Hypotheses 1) Transfection of CaBP into non-neuronal cell lines will enhance the cell's endogenous Ca -buffering capacity. 2) Increased Ca2+-buffering in H E K 293 cells by transfected CaBP will enhance cell survival following activation of recombinant N M D A receptors. 3) Transfected CaBP will modulate the temporal and spatial characteristics of IP3-induced Ca release from intracellular stores in HeLa cells, decreasing the frequency of C a 2 + oscillations and attenuating the velocity of C a 2 + waves. Chapter 2 Materials and Methods 24 Cloning and Expression ofthe Human Recombinant CaBP The molecular biology techniques employed in this study were based on those described in Molecular Cloning, A Laboratory Manual by Sambrook et al., 1989 and Current Protocols in Molecular Biology, edited by Frederick M . Ausubel, 1987. Additional details are provided where modifications were made to these procedures. Polymerase Chain Reaction The human CaBP gene was cloned from a human fetal brain cDNA library (Promega) via polymerase chain rection (PCR) amplification. Primers were designed to incorporate restriction sites which would facilitate cloning of the gene into the expression vectors (Fig. 2A). Design and analysis of oligonucleotide primers was carried out utilizing Oligo 4.0 Primer Analysis Software (Wojciech Rychlik). Restriction analysis of all D N A sequences was carried out using the computer program PC/Gene (Intelligenetics Inc). Primers were synthesized on an Applied Biosystems 380A D N A synthesizer. Oligonucleotides were eluted from the synthesis column with 2 mL of 8 M N H 4 O H . Following overnight incubation in a 55°C water bath, the ammonium hydroxide was removed by evaporation in a Sorvall Speed Vac Centrifuge for 4 hours. The oligonucleotide pellet was resuspended in 200 uL of distilled water and quantified by 25 measuring the absorbance at 260 nm. (lunit A260 = 33 ug/mL of single stranded DNA). The optimal quantity of the amplified cDNA library template was determined empirically by diluting 1 to 10,000 fold. PCR reactions consisted of various dilutions of the cDNA library template DNA, 100 ng of each oligonucleotide, 0.05 m M dNTPs (Pharmacia), I X Buffer E (67 m M Tris Base pH 9.01, 1.5 m M magnesium sulfate, 166 m M ammonium sulfate, 10 m M (3-mercaptoethanol), and 5 units of T. aquaticus polymerase in a total volume of 50 uL. Reaction mixtures were overlayed with 3 drops of mineral oil, then cycled on a Perken Elmer thermocycler as follows: 1 minute at 94 °C, 1 minute at 50 °C, and 1 minute at 72 °C, for 25 cycles. Amplified fragments were separated by electrophoresis in a 1% agarose gel in the presence of lpg/mL ethidium bromide. D N A fragments were visualized by ultraviolet light illumination and photographed with a Polaroid camera. A n ~800bp band, corresponding to the coding region of CaBP was excised from the agarose gel and eluted using a "Geneclean" kit (Bio 101) according to the manufactures instructions (Fig. 2B). Eluted fragments were cut with the restriction enzymes EcoRV and Xbal (Life Technologies) and ligated to corresponding restriction sites in the cloning vector pBluescript II (Stratagene). 26 A H5E1: 5'- A T C GAT ATC A T G G C A G A A TCC C A C C T G CAG-3' H3X1: 5'-ATC TCT AGA G T G GTT G C G GCC A C C A A C T C T A - 3 ' Figure 2. PCR of Human CaBP cDNA. (A). Primers employed to amplify human CaBP cDNA. H5E1 incorporates the EcoRVrestriction site (bold) along with the start codon (underlined) and the 5' end of the CaBP gene. H3X1 targets the stop codon (underlined), and a segment of the 3'-untranslated region. It incorporates a recognition sequence for the restriction endonuclease Xba\ (bold). (B) Ethidium bromide stained agarose gel of D N A electrophoresis of human CaBP gene PCR amplification product. 27 DNA Sequence Analysis D N A sequence analysis was carried out using a T7-Sequencing Kit (Pharmacia) as per the manufacturer's instructions, with the following modifications. D N A template (10 pg), combined with 100 ng of a sequencing primer was denatured with 0.2 N sodium hydroxide in a volume of 20 uL in a boiling water bath for 2 minutes, followed by snap cooling on ice. The D N A was subsequently precipitated with 0.1 volumes of 3 M sodium acetate pH 5.6, and 2.5 volumes of 95% ethanol. Precipitated D N A was collected by centrifugation in an Eppendorf micro-centrifuge for 15 minutes at 14,000 rpm. The D N A pellet was then resuspended in 10 uL of I X Annealing Buffer (40 m M Tris base pH 7.5, 20 m M magnesium chloride, 50 mM sodium chloride). To the D N A template and annealed primer solution was added 1 uL of 0.1 M di-thiothreitol, 2uL I X labelling mix, 5 pCi ct 3 5S-dATP, 2 uL of a 1:8 dilution of Sequenase enzyme, and this mixture was left to incubate at room temperature for 2 minutes. In a 60 well micro-titre plate (Nunc), 2.5 uL of 80 uM dideoxyguanosine, dideoxyadenosine, dideoxythymidine, and dideoxycytidine in 50 m M sodium chloride were added to separate wells and warmed to 37°C on a heating block. The above "labelling reaction" (3.5 uL) was added to each of the wells containing the respective dideoxy-nucleotides and incubated at 37 °C for 5 minutes. A termination solution (4uL, 95% formamide, 20 m M EDTA, 0.05% bromophenol blue, and 0.05%) xylene cyanol FF) was then added to terminate the reaction. Samples were boiled in a boiling water bath for 2 minutes, then immediately transferred to ice. 28 Samples were subjected to electrophoresis in a 10% acrylamide, 8M urea, I X T B E (0.89 M Tris base, 0.89 M boric acid, 25 mM EDTA; pH 8.0) gel on a Model S2 sequencing apparatus (Life Technologies) for up to 5 hours. Following electrophoresis of the sequencing reactions, gels were dried under vacuum with a Bio-Rad gel drier at 80°C for 1 hour and the labeled nucleic acid fragments were visualized by autoradiography using film from Island Scientific.The CaBP D N A sequence was subsequently verified by automated D N A sequencing (Nucleic Acid and Protein Sequencing Unit, UBC). Protein Expression The coding region of CaBP was inserted, in frame, into the bacterial expression vector pMAL-C2 (New England Biolabs) to allow the expression of large quantities of recombinant protein. The entire CaBP coding region was excised from pBluscript II with the restriction endonucleases EcoRV and Xbal, and inserted into the XmnI and Xbal restriction sites of the pMal-C2 vector. The DH5-a strain of E.coli was transformed with this construct and production of the full-length fusion protein was confirmed by SDS-PAGE and western blot detection of CaBP. The resultant protein from this expression system is a fusion protein of CaBP and Maltose Binding Protein (MBP), separated by a Factor Xa protease recognition site. Protein expression and purification was carried out according to the manufacturer's instructions. Briefly, following confirmation of correct insertion of the CaBP sequence into the p M A L vector 29 by restriction analysis, the expression vector was transformed into E.coli and grown at 37°C in 1 liter of L B broth to an OD of 0.5 (measured at 600 nm). The expression of the MBP-CaBP fusion protein was induced by addition of 2.5 m M IPTG, and the culture was allowed to incubate for a further two hours at 37°C with constant agitation. Bacteria were collected by centrifugation, washed, and resuspended in a Tris buffer (20 m M Tris pH 7.4, 200 m M NaCI, 1 m M EDTA). Lysis of the resuspended bacteria was carried out by sonication at 4°C. Cell membranes and denatured proteins were removed by centrifugation. The crude soluble protein fraction was then passed through a maltose column which separated the fusion protein from the soluble bacterial lysate. The fusion protein was eluted from the column with 10 mM maltose in Tris buffer. The fusion protein was cleaved by digestion with bovine protease Factor Xa (0.1 % w/v of total fusion protein). The cleaved protein mixture was passed through the maltose column once again to remove the MBP. The CaBP protein was concentrated with a mini-Amicon concentrator and a 10,000 kDa cutoff filter. The correct size of the purified protein was confirmed by SDS-PAGE. Western blot analysis of the purified protein revealed that the protein was recognized by an anti-chick CaBP antibody, which has been shown previously to cross react with human CaBP (Fig. 3). Further characterization of the recombinant CaBP was performed in the lab of Dr. Les Burtnick (Department of Chemistry, UBC). It was confirmed that the recombinant CaBP binds Ca 2 + , and undergoes a conformational change upon binding C a 2 + similar to that observed for avian CaBP (Gross et al., 1987) (Fig. 4). Mass Spectroscopy of the purified protein, performed by Dr. Michael Murphy (Department of Microbiology, UBC), yielded an interesting result. A protein of double the molecular 30 Xmn I fccoK 1 Xbal Psti ATC GAG GGA A G G ATT TCA GAA TTC GGA TCC TCT AGA GTC GAC CTG C A G G C A AGC TTG .... Promoter (Maltose Binding Protein) Figure 3. Purification of recombinant human CaBP. A) pMal Expression vector construct. B) SDS P A G E of the stages of purification. Shown are the fusion protein from the bacterial lysate (Fusion Protein), the result of cleavage of the fusion protein with Factor Xa (Cut Fusion) and the purified CaBP protein. 31 MAX-300.0 C MIN=0.0 300 310 320 330 340 350 360 370 380 390 400 Wavelength (nm) Wavelength (nm) Figure 4. Spectrophotometric studies of recombinant human CaBP demonstrating a conformational change upon binding C a 2 + . A. Fluorescence emission spectra for human CaBP in the presence and absence of Ca 2 + ; (A) 0 mM C a 2 + (B) 0.2 mM C a 2 + (C) 0.6 m M C a 2 + . The observed changes in fluorescence suggest a conformational change upon C a 2 + binding, causing a tryptophan residue to move into a more hydrophobic region. B. Difference spectrum of C a 2 + bound recombinant human CaBP. The reference cell contained Calbindin in buffer with no Ca 2 + . The trace depicts the result of adding (A) 1 mM and (B) 1.8 mM Ca 2 + . The results indicate that C a 2 + causes a conformational change, moving aromatic chromophores to a more hydrophobic environment. C. Near U V circular dichroism spectra of recombinant human calbindin in (A) 0 mM C a 2 + ,(B) 1.8 mM C a 2 + and (C) 2.6 mM Ca 2 + . The changes depicted in the trace are a result of changing environments of aromatic chromophores incurred by protein conformational changes (Spectrophotometric studies carried out by Seung Min Cha in the Lab of Dr. Les Burtnick, Department of Chemistry, UBC). 32 weight (minus 2 hydrogens) calculated from the amino acid sequence (30,025 kDa) was detected. This apparent dimerization was confirmed by SDS-PAGE, and was usually observed in the absence of DTT or other reducing agents, suggesting that dimerization was the result of a disulfide linkage. In addition, the degree of dimerization increased significantly i f solutions of CaBP were concentrated above lmg/mL. Care was therefore taken to limit all CaBP containing solutions to less than lmg/mL. Disulfide-mediated dimerization of CaBP has been reported by Berggard et al, (2000) who speculate that this is not a physiologically relevant phenomenon, owing to the reducing environment of the cell cytosol. Generation and Characterization of Stable Cell Lines Mammalian Cell Culture H E K 293 (Quantum Biotechnologies) and HeLa (A gift from Dr. Alison Buchan, Department of Physiology, UBC) cultures were maintained in Dulbecco's minimal essential medium (DMEM) containing 5% fetal bovine serum (Life Technologies) at 37 °C in an atmosphere of 95% air, 5% C 0 2 . 33 Stable Expression of CaBP Expression vector D N A for transfection was prepared using a Qiagen Maxi-prep kit according to the manufacturer's instructions. The human CaBP gene was inserted into the pCDNA3 expression vector (Invitrogen; Carlsbad, CA), which contains the neomycin (G418) resistance gene. H E K 293 and HeLa cultures were transfected using standard calcium phosphate precipitation transfection techniques with lOpg of expression vector DNA. Briefly, D N A was diluted in a 40mM CaCl2 solution in a total volume of 250 uL. This solution was added drop-wise to 250 uL of 2xHBS. After a 1 minute incubation at room temperature this transfection solution was added to a 10 cm plate of cells which had been plated the previous day at 106 cells per plate. Following overnight incubation, cells were washed twice with fresh media, then split between two 10 cm plates and incubated in the presence of G418 (Life Technologies; 1 mg/mL G418 for H E K 293 cells; 0.5 mg/mL for HeLa cells). Control cell lines were generated by stably transfecting with the empty pCDNA3 vector. Individual colonies of cells resistant to G418 were isolated and amplified. Expression (or absence, in control cultures) of CaBP was verified by immunohistochemical staining of cultures as previously described (Buchan and Baimbridge, 1988) and western blot analysis (Fig. 5). One H E K 293 clone (HEK-CB12) was selected for these studies on the basis of its relatively high intensity of staining observed by immunohistochemistry (IHC). A control cell line (HEK-p3) was also generated that was stably transfected with the pCDNA3 vector alone. In all HeLa cell lines isolated, expression of CaBP as assessed by IHC (and in comparison to the HEK-CB12 cell lines) was very low. Therefore, an alternate expression vector with a 34 A Figure 5. Expression of CaBP in an H E K cell line. A) Control cell (HEK-p3) line stably transfected with the pCDNA3 vector. Cells were stained for CaBP and counter-stained with cresyl violet. No expression of CaBP was evident in control or untransfected cell lines. B) Photomicrograph of HEK-CB12 cell line. Cells were fixed and stained for CaBP by IHC. C) Western blot of soluble protein extracts from HEK-CB12 and HEK-p3 cell lines. 35 modified C M V promoter was employed. The CaBP gene was inserted into the Xho l and Xbal sites of the mammalian expression vector pCINeo (Promega). Generation of clonal lines of HeLa cells containing the pCINeo-CaBP construct or the empty pCINeo vector (HeLa- pCINeo) was carried out as described above. Expression of CaBP from the pCINeo vector in HeLa cells was better than that observed for the pCDNA3 vector. As assessed by the intensity of IHC staining, the highest expressing clone (HeLa-CaBP), was selected for subsequent studies. Estimation of CaBP Concentration in Transfected Cells The concentration of CaBP in transfected cells was estimated using a western slot-blot method (Hersham et a l , 1993). Transfected cells, grown on 10-cm tissue culture plates, were dislodged by trypsinization in 10 m M phosphate-buffered saline (pH 7.4). Live cells were counted using trypan blue exclusion, and replicate volumes containing at least 106 cells were centrifuged at 600 g for 3 min at room temperature. Cell pellets were stored at -20°C. For slot-blot analysis, cell pellets were resuspended in 500 uL of 50 m M Tris-buffered saline (TBS) containing 1 mM EDTA, boiled for 5 min, and centrifuged in a microfuge for 10 min. The supernatant was collected and the concentration of total soluble protein (TSP) was determined by A280 and A 2 6o measurements. Correction for nucleic acid contamination was accomplished through use of the equation (Peterson, 1983): [Protein] (mg/mL) = (A 2 8o x 1.55)-(A26o x 0.76) 36 Aliquots (200 uL) containing from 0.05-20 ug of TSP were applied to a nitocellulose membrane using a Bio-Rad slot-blot apparatus according to the manufacturer's instructions. For standards, we used the recombinant human CaBP (see above) in the range of 0-1000 ng in addition to TSP extracted from a rat cerebellar homogenate. After nitrocellulose membranes had been blocked overnight at 4°C in TBS containing 2% skim milk, 50% horse serum, and 0.2% Tween-20, they were incubated with polyclonal antibodies against CaBP (raised in rabbits against purified bovine cerebellar CaBP; Buchan and Baimbridge, 1988) at a dilution of 1:1,000 in TBS-Tween for 1 h at room temperature. Blots were rinsed twice in TBS-Tween (5 min each), and the secondary antibody (donkey anti-rabbit IgG conjugated with horseradish peroxidase; Amersham, Arlington Heights, IL, U.S.A.) was applied at a dilution of 1:2,000 for 1 h at room temperature. After two additional washes, Amersham's enhanced chemiluminescence system was used to visualise the CaBP protein. The amount of CaBP in cell extracts was determined according to the method of Hersham et al. (1993). Autoradiographs were scanned on an Eagleeye imaging system (Stratagene) and densitometry was carried out with Stratagene EagleSight software (Fig. 6). The concentration of protein for HEK-CB12 and HeLa-CaBP cell lines were estimated to be 18 and 3.5 ug/mg TSP respectively. The rat cerebellar standard yielded a concentration of 24 ug/mg TSP. In the rat cerebellum, only Purkinje cells and climbing fibers contain CaBP, therefore this value is likely an underestimation (in proportion to the relative volume of Purkinje cells and climbing fibers to total cerebellar volume). The average expression level of CaBP in the transfected cell lines is therefore less than that 37 28 k D a - ^ B 125.0 62.5 31.2 15.7 7.8 3.9 20.0 < 10.0 • 5.0' 2.5 0.6 0.3 Figure 6. Quantification of CaBP Expression in Stably Transfected Cells. A) Western blot analysis of soluble protein extracted from HeLa-CaBP and HeLa-pCINeo cell cultures. B) Quantitative E C L slot blot analysis of HeLa-CaBP cells cultures. Recombinant CaBP standards were blotted with TSP extracts from HeLa-CaBP cells and (for comparison) with equivalent quantities of rat cerebellum TSP. C) Standard curve generated from densitometry of recombinant CaBP standards. 38 expressed naturally in Purkinje cells, but probably equivalent to or higher than levels expressed in other neurons (Baimbridge et al., 1992). Transient Expression ofthe N M D A Receptor Two separate constructs were employed to co-express the NR1 and NR2A subunits of the N M D A receptor (Raymond et al., 1996), both under control of the strong C M V promoter. Eight hours prior to transfection, cells were plated at a density of l x l 0 6 cells /10 cm plate. Transient transfection with 3ug of each N M D A subunit was carried out using Lipofectamine (Life Technologies) according the manufacturer's instructions. Following a 5 hour transfection period cells were washed twice with medium and replated on poly-D-lysine coated 22 mm circular glass coverslips. Cultures were incubated for a further 24h in the presence of 1 mM DL-2-Amino-5-phosphonopentanoic acid (APV) (Precision Neurochemicals; North Vancouver, BC) to protect transfected cells from excitotoxic cell death (Raymond et al., 1996), prior to loading with fura-2 and imaging. 39 Measurement of Intracellular Calcium : NR1/NR2A Transfected Cells • 2"f" Two intracellular Ca indicators were employed in these experiments; fura-2 and fluo-3. Fura-2 has the advantage of being a ratiometric dye and therefore facilitates 2_|_ estimation of intracellular [Ca ]j . However, the quantum yield of fura-2 is much lower than that of fluo-3 (a non-ratiometric dye) and therefore necessitated higher loading concentrations. Therefore in experiments which were sensitive to the presence of high levels of Ca2+-indicators (discussed in Chapter 4), fluo-3 was used instead of fura-2. Transiently transfected cells were loaded with fura-2 by incubation at room temperature for 30 minutes in a balanced salt solution (BSS; NaCI, 139 mM; KCI, 3.5 mM; Na 2 HP0 4 , 3 mM; NaHC0 3 , 2 mM; HEPES acid, 6.7 mM; HEPES-Na, 3.3 mM; D-glucose, 11 mM; C a C l 2 ; 1.8 mM; M g C l 2 j 1 mM; Na pyruvate, 5 mM; glycine 2 uM; pH 7.35) containing 6 u M fura-2-AM (Molecular Probes), 0.05% bovine serum albumin and 1 m M A P V . Fura-2 imaging was performed using a Zeiss-Attofluor™ digital fluorescence imaging system, as previously described (Abdel-Hamid and Baimbridge, 1997). Coverslips of transfected cells were mounted in a heated chamber and superfused with BSS containing lOOuM A P V at 37°C for 15 min prior to any data collection. The ratio of fluorescence of fura-2 at 334/380 nm excitation (corrected for background fluorescence) was determined at a rate of 1 s"1 using a 63x objective in individual H E K cells, before, during, and for 10 minutes following superfusion of the cells with BSS containing 500 uM glutamate. Parameters of the glutamate-induced C a 2 + transient examined were the rate of rise, peak response and rate of recovery. Fura-2 measurements are expressed in terms of fluorescence ratios. When calibrated by the 40 methods of Grynkiewicz et al. (1985), ratios of 0.5, 1.0, 2.0 and 3.0 were calculated to correspond to [Ca2+]i values of 80.8 nM, 280.9 nM, 783.8 nM and 1.4 u M respectively. Measurement of Intracellular Calcium : Flash Photolysis In another series of experiments, HEK-CB12 and HEK-p3 cells were loaded with the fluorescent indicator fluo-3 or fura-2 and the caged calcium compound NP-EGTA (Ellis-Davies and Kaplan, 1994) by the use of their respective cell permeant acetoxy-methyl ester derivatives (Molecular Probes) as previously described (Sidky and Baimbridge, 1997). Briefly, individual coverslips plated with HEK-CB12 or HEK-p3 cultures were incubated for 1 hour at room temperature in the dark in BSS containing 3.1 u M fluo-3-AM or 6 uM fura-2-AM and 8 uM N P - E G T A - A M . They were then transferred to fresh BSS and incubated at room temperature for at least 10 minutes prior to mounting on the imaging chamber. Flash photolysis of NP-EGTA, achieved by a 0.4 s exposure to unfiltered U V light, results in a rapid and transient increase in [Ca 2 +]i which is relatively uniform throughout the cytoplasmic volume (Ellis-Davies and Kaplan, 1994). Cells which did not display at least a two-fold increase in background-corrected fluorescence intensity (fluo-3) in response to flash photolysis were excluded from further analysis. Fluo-3 fluorescence intensity (in the range of 0-255 arbitrary units) was determined after flash photolysis in individual cells (up to 99 in a single field of view) for up to 10 minutes at a rate of 1 s"1 for the first 3 minutes and thereafter at a rate of 0.5 s"1. For each cell, the background-corrected fluorescence intensity was normalized 41 between a value of 0 and 1 with 0 being the baseline value to which fluo-3 fluorescence decayed, and 1 being the fluorescence at the first time-point immediately (~ 0.6 s) following photolysis. Data collection was carried out in a similar fashion for fura-2 loaded cells. Imaging Calcium Oscillations in HeLa Cells To capture Ca oscillations, HeLa cells were loaded with fura-2 or fluo-3 as described above, and imaged at a rate of 1 s"1 using a 63x objective. Individual cells (up to 50 in a single field of view) were monitored before during and after a 300 s superfusion with BSS containing 1 u M histamine for 300 s. These experiments were all performed at room temperature. Calcium Wave Propagation Analysis in HeLa Cells To investigate calcium wave propagation velocity, individual elongated HeLa cells were selected for analysis and imaged using a lOOx objective. A single pixel wide line was selected along the length of the HeLa cell and the fluorescence along that line was measured at a rate of 15 images s"1 for approximately 30s. Fluorescence intensity values were corrected for background and converted to %AF/F. A criterion of an 42 increase in fluorescence of 50% of the maximum value was selected as the threshold for the determination of the time course of a calcium wave along the scanned region. Wave velocity was then determined by plotting the time point at which specific pixels along the scanned line reached the 50%> AF/F value. Image Analysis & Statistical Methods A l l imaging data analysis was carried out in Microsoft Excel employing custom designed macros written in Visual Basic™ by this author. Statistical comparisons between treatment groups were performed using unpaired t-tests. Excitotoxicity Studies H E K stable cell lines were transfected with NR1/NR2A and the marker plasmid pCMV(3 (Clontech) as described above. Following removal of the transfection mixture, each 10cm plate of transfected cells (either HEK-CB12 or HEK-p3) were split evenly among the wells of a 6-well tissue culture plate. Cells were allowed to attach to the plates for 18 hrs in D M E M containing ImM A P V prior to excitotoxic treatments. Cultures were washed twice with a bicarbonate-buffered salt solution, (BBSS; 140 m M NaCI, 1.4 m M CaCl 2 , 5.4 mM KCI, 1.2 m M NaH 2 P0 4 , 21 m M Glucose, 26 m M 43 N a H C 0 2 ; pH 7.4), then incubated in BBSS containing 200 u M N M D A at 37° in an atmosphere of 5% C 0 2 for 1 hour. In control cultures (3 wells of the six well plate) ImM A P V was substituted for N M D A , but they were otherwise treated identically. Following treatment, cells were washed twice with BBSS then incubated for a further 6 or 24 hrs in fresh D M E M with ImM A P V (37° 5% C0 2 ) . Cells were then fixed (4% formaldehyde, 0.2% glutaraldehyde in PBS; 5 min room temperature) and stained for p-galactosidase activity. Surviving cells which had been successfully transfected (as assessed by the presence of P-galactosidase staining) were counted by examining 21 fields (20x objective) across each well of the tissue culture plates. In excitotoxicity experiments, to assess cell death, remaining cells were counted. This relies upon the assumption that dead cells will detach from plates. This assumption was made based upon initial observations of plates co-transfected with NR1, NR2A and the marker plasmid pCMVP, in which no N M D A receptor inhibitor was added. In these experiments, no cells positively stained with the P-gal marker plasmid remained attached to the plate. 44 Chapter 3 Effect of Recombinant CaBP on Intracellular Ca 2 + Transients in Stably-transfected HEK 293 Cells 45 Chapter Summary 1. The effect of stable CaBP expression in H E K 293 cultures on changes in [Ca2+]j induced by a variety of stimuli was investigated. 2. H E K 293 cells expressing the NR1/NR2A subunits of the N M D A receptor responded to brief glutamate stimulation with C a 2 + transients that were prolonged in cells expressing CaBP. 3. C a 2 + transients induced by treatment with the ionophore 4-Br-A23187 displayed an 9+ attenuated rate of rise and peak [Ca ]j and a prolonged recovery in cells expressing CaBP. 4. Flash Photolysis of NP-EGTA resulted in a rapid release of intracellular C a 2 + which required longer periods of time to dissipate in cells expressing CaBP. 5. Exposure of H E K 293 stable cell lines to ATP resulted in Ca transients in which the 9+ • rate of rise and peak [Ca ]\ was reduced in cells expressing CaBP. 6. Capacitative C a 2 + influx was induced by treatment with thapsigargin and modulation of external Ca 2 + . The resulting rate of rise in [Ca2+]j was reduced in cells expressing CaBP. 7. In studies of delayed, Ca -dependent cell death, CaBP transfected cells exhibited enhanced survival 24 hours after a 1 hour exposure to 200 u M N M D A . However, acute, necrotic cell death observed after the first 6 hours was not prevented by the presence of CaBP. 8. These results therefore provide direct evidence for a Ca -buffering effect of CaBP which serves to protect cells from death mediated most likely by apoptosis. 47 Introduction It has been hypothesized that CaBP acts as an intracellular C a 2 + buffer, and that one of the consequences of this action is to enhance cell survival in C a 2 + mediated cytotoxicity. To address this hypothesis, a number of studies have examined the effect of CaBP on C a 2 + transients utilizing whole cell C a 2 + imaging or electrophysiological 2+ 2+ techniques. Reductions in peak [Ca ]\ and rate of decay of Ca transients induced by depolarization have been reported in neurons in which CaBP was injected by patch pipette (Chard et al., 1993) and in retrovirally transfected G H 3 cells (Lledo et al, 1992). In CaBP null mutant mice studies, peak [Ca ]\ was lower and recovery of synaptically evoked postsynaptic C a 2 + transients were slower in wild-type (CaBP+) mice (Airaksinen et al, 1997). Peak [Ca response reduction has also been reported in glutamate-stimulated neurons transfected with CaBP using a viral vector (Meier et al, 1998) and in CaBP-transfected PC 12 cells exposed to ATP (McMahon, 1998). Consistent with many of the studies cited above, the decay from the peak [Ca 2 +]j was also prolonged in neurons naturally expressing CaBP (Abdel-Hamid et al, In Preparation), however these neurons displayed an increased peak [Ca2 +]i in response to N M D A . The experiments described in this chapter systematically examined the effects of CaBP on C a 2 + transients. The objective was to examine the effect of stably-transfected CaBP upon intracellular C a 2 + transients originating from different sources (intracellular vs extracellular) and having different temporal and spatial characteristics (Ca gradient vs global release). 48 Initial experiments were an extension of our recent electrophysiological investigation of the effect of CaBP upon transfected N M D A receptor activity (Price et al., 1999). It was found that in cells transfected with CaBP there was a significant slowing of the development of Ca -dependent rundown of peak glutamate-evoked current. This effect was presumed to be due to the buffering action of CaBP, although a protein-protein interaction cannot be ruled out as a possible mechanism for modulation of C a 2 + influx. To generate an influx of Ca with similar dynamics but utilizing a structurally unrelated molecule, stable cells were exposed to the ionophore 4-Br-A23187. To assess C a 2 + buffering on Ca2+-transients with different dynamics, flash photolysis of the caged C a 2 + compound, NP-EGTA, was employed. Use of NP-EGTA allowed us to examine the effect of a global, nearly instantaneous release of Ca 2 + , independent of internal or external stores. To examine the possible effects of CaBP upon release of internal stores o fCa 2 + , ATP was used to activate the PLC-IP3 pathway, and subsequently release Ca from the ER. This likely produces gradients of C a 2 + across the cell due to sequential activation of IP 3 receptors. Finally, many recent studies have focused on C a 2 + influx stimulated by emptying of internal stores or "capacitative" Ca influx (see reviews by Putney, 1999; Berridge, 1995). The mechanism through which this operates has yet to be elucidated, but likely involves a plasma membrane channel. By emptying intracellular 2_|_ stores with thapsigargin, we examined the effect of CaBP on capacitative Ca entry. 49 Results Characterization of Ca Transients in NR1/NR2A Transfected HEK 293 Cultures To examine the effect of transfected CaBP upon N M D A receptor mediated changes in [Ca ]•„ HEK-CB12 and HEK-p3 cells were transiently transfected with the NR1 and NR2A receptor subunits. While measuring changes in [Ca2+]j with fura-2 or fluo-3, cells were exposed for 20s to 500uM glutamate, followed by rapid washout. Following transfection with the NR1/NR2A subunits, cells which successfully expressed functional N M D A receptors were identified by ratiometric imaging on the basis of their response to 500uM glutamate. R was observed that variable magnitudes of C a 2 + transients were generated in response to glutamate, likely due to differing levels of N M D A receptor subunit expression (Raymond et al., 1996). As described in previous studies of N M D A receptor transfected H E K 293 cells, the continuous presence of an N M D A channel antagonist in the culture medium was necessary following transfection to prevent cytotoxic cell death by inappropriate channel activation (Raymond et al., 1996). This is likely due to the fact that in neurons, N M D A receptor operated channels demonstrate a voltage-dependent block by M g 2 + , whereas the resting membrane potential of H E K cells are in the range where M g 2 + blockade is relieved (ie < -50mV). In accord with this observation, during Ca2 +-imaging experiments, a slow increase in [Ca2+]j was observed when no antagonist was present (data not shown). Therefore during all imaging experiments, 100 p M of the competitive antagonist A P V was present throughout. Under these conditions, short exposure (<20s) to glutamate resulted in a 50 rapid increase in [Ca ]i to a peak value followed by a slow recovery upon washout. It was observed that maximal increase in [Ca2+]j was achieved with doses of glutamate between 500 u M and ImM (Fig. 7). Effect of CaBP on Glutamate Treated HEK Cells Transfected with NR1/NR2A From transfection to transfection and from cell to cell, variable magnitudes of Ca transients were produced in response to glutamate, presumably due to differing levels of N M D A receptor subunit expression. Similar variations in CaBP expression levels were noted in transiently transfected CaBP in H E K cells. For the following experiments therefore, we utilized stable cell lines to alleviate this variability. Fig. 8 depicts two representative results from a typical C a 2 + imaging experiment involving a 20s, 500 u M glutamate treatment of HEK-p3 and HEK-CB12 cells which have been transiently transfected with the NR1/NR2A subunit combination. Responses from a total of 107 control HEK-p3 cells and 104 HEK-CB12 cells were analyzed. No significant difference was observed in the rate of 334/380 ratio increase (HEK-p3: 0.117±0.01; HEK-CB12: 0.128±0.012 ratio units s"1). The peak ratios ofthe HEK-p3 (2.46±0.13) and HEK-CB12 (2.41 ±0.09) cell lines were not significantly different. However, in the recovery to baseline levels the two cell types differed significantly. When the C a 2 + response of each cell was normalized to the peak and baseline value, the recovery times to 75%, 50% and 25% of the maximum were found to be significantly longer in the CaBP expressing cells (HEK-CB12: 75%=31.4±3.2 s, 50%=111±10.8 s, 51 2.0 | A f B | C o.o ] , , , , , , — 0 200 400 600 800 1000 1200 Time (s) Figure 7 Response of NR1/NR2A transfected HEK 293 cells to glutamate. Mean [Ca 2 +]i response of 4 H E K 293 cells transiently transfected with NR1/NR2A subunits of the N M D A receptor and briefly exposed to (A) 200uM (B) 500uM and (C) ImM glutamate. [Ca 2 +]i is expressed as the ratio of fura-2 fluorescence at 334 and 380 nm excitation. 52 2.5 0 100 200 300 400 500 600 Time (s) Figure 8 Response of HEK-CB12 and HEK-p3 cells transiently transfected with NR1/NR2A to treatment with glutamate. Ratio of ftira-2 fluorescence in H E K 293 cells transiently transfected with NR1/NR2A. Shown is a representative response of an HEK-p3 cell compared with that of an HEK-CB12 cell. Cells were superfused for 20s with 500 p M glutamate at 37° 53 25%=220.3±14.9 sN=104; HEK-p3: 75%=17.7±1.5 s, 50%= 57±7.9 s, 25%=134.1±11.7 sN=107) (Fig. 9). Modulation of Br-A23187 Mediated Ca Transients by CaBP To eliminate the high variability of Ca -responses due to the levels of N M D A receptor expression, the ionophore 4-Br-A23187 was applied to the stable H E K cell lines. 4-Bromo-A23187 is a non-fluorescent C a 2 + ionophore which has been used extensively to increase intracellular C a 2 + levels. Upon perifusion of fura-2 loaded H E K -293 stable cell lines with 2.5 uM 4-Br-A23187 for 100s, [Ca2 +]i increased rapidly to a peak which recovered towards baseline values following washout (Fig 10). No effect upon [Ca 2 +]i was observed in control (0.75% DMSO) treated cells. When compared to HEK-p3 cells (N=53), HEK-CB12 cells (N=41) displayed a significant attenuation in the rate of rise (slope calculated between 25% and 75% of maximum ratio; HEK-p3 = 0.085±0.005 ratio units s'\ HEK-CB12 =0.058±0.003 ratio units s"1; pO.Ol) , peak ratio (Fig. 11 A) (HEK-p3 Ratio m a x = 4.2±0.25, HEK-CB12 Ratio m a x =3.1+0.16; pO.OOl), and decay (Fig. 1 IB) (HEK-p3 = -0.0117±0.0008 ratio units s"1, HEK-CB12 =-0.0095 ±0.0005 ratio units s"1) 54 Figure 9 Pooled data of recovery of stable H E K 293 cells transiently transfected with NR1/NR2A and stimulated with glutamate. Bars indicate mean time to recovery to the percentage of maximal response indicated. Error bars indicate S E M . *** p<0.001. HEK-p3, N=107; HEK-CB12, N=104. 55 4.0 -, 0 50 100 150 200 250 300 350 400 Time (s) Figure 10 Effect of CaBP upon 4-Br-A23187-induced C a 2 + transients in HEK293 cells. Mean [Ca2+]j responses of H E K cells from a single coverslip of either HEK-p3 (N=18) or HEK-CB12 (N=22) cells. Cells were exposed to a 100 s supervision of 2.5 u M 4-Br-A23187 in BSS at 37°. 56 o oo CO C O C O re B a s e l i n e Max B 300 250 ^ 200 a> E % 150 > 8 100 a) CC 50 • C o n t r o l • C a B P 75% 50% % of P e a k R e s p o n s e 25% Figure 11 Pooled calcium imaging data from stable H E K 293 cells treated with 4-Br-A3187. A) Basal levels prior to, and mean maximum ratios following treatment with 4-Br-A23187. B) A total of 18 HEK-p3 cells were compared to 22 HEK-CB12 cells. The Y-axis denotes mean time to recovery to the percentage of the maximum response indicated on the X-axis. Error bars indicate SEM. * p<0.05. 57 Effect of Stably Expressed CaBP in HEK Cells Following Flash Photolysis of NP-EGTA Both of the above experiments involved influx of extracellular Ca 2 + , resulting in a gradient of C a 2 + extending from the channels/pores. NP-EGTA was utilized as a method of globally and nearly instantaneously releasing C a 2 + intracellularly. Initial experiments with this caged Ca compound were carried out with the non-ratiometric dye fluo-3 to avoid inappropriate photolysis by U V wavelengths of light necessary for fura-2 ratiometric imaging. Since non-photolysed NP-EGTA can act as a Ca buffer, near-complete photolysis by a 0.4s flash of unfiltered U V light was confirmed in control experiments involving sequential U V flash exposures, in agreement with Sidky and Baimbridge (1997). In all experiments with NP-EGTA and fluo-3 loaded stable H E K 293 cells, flash photolysis of NP-EGTA resulted in an increase in fluo-3 fluorescence that was already declining from the time of the first measurement taken (~0.6s). A l l cells demonstrated a rapid initial recovery phase followed by a slower recovery to baseline. In HEK-CB12 cells, the initial fast-recovery phase was unaffected, however a pronounced "plateau" phase was evident in the initial phase of the slow recovery. Full recovery to baseline [Ca 2 +]i was observed in both HEK-p3 and HEK-CB12 cell lines after about 250s (Fig. 12). An obvious drawback of the above method is the non-quantitative nature of fluo-3 imaging. Attempts were therefore made to utilize fura-2, while minimizing unwanted photolysis of the caged C a 2 + compound. In preliminary experiments, it was found that a 58 0 50 100 150 200 250 300 Time (s) Figure 12 Effect of CaBP on recovery following flash photolysis of NP-EGTA in fluo-3 loaded stable HEK cells. Mean normalized intensity of fluo-3 fluorescence following flash photolysis of NP-EGTA in HEK-p3 cells (N=61) and HEK-CB12 cells (N=80). 59 single exposure of NP-EGTA loaded cells to the wavelengths of U V light necessary for fura-2 excitation (334 and 380 nm) resulted in minimal photolysis of NP-EGTA. Therefore, in fura-2 experiments, fields of cells loaded with NP-EGTA were limited to a single ratio collection (single exposure to 334 and 380nm light) prior to flash photolysis with unfiltered U V . The pattern of the [Ca2+]j response to NP-EGTA photolysis in fura-2 loaded cells (Fig. 13) was essentially the same as that seen with fluo-3. However, fura-2 loaded cells required longer to recover to baseline than fluo-3 loaded cells. This may be a result of additional buffering by fura-2, as cells were loaded with more fura-2 (6 u M loading concentration) than fluo-2 (3.1 u M loading concentration). In comparing HEK-CB12 and HEK-p3 cells loaded with fura-2, peak ratios achieved upon uncaging were not statistically different (p>0.05). Average peak ratios were 2.28±0.16, N=8 and 2.54±0.17, N=9 (334/380 ratio units) for HEK-p3 and HEK-CB12 cultures respectively. Using the calibration described in Chapter 2 these ratios would correspond to 1.04 u M and 1.07 u M [Ca2 +]i respectively. As with the fluo-3 loaded cells there was a significant prolongation of [Ca2 +]i in the recovery phase (Fig. 14), with the time to decay to 50% and 25% of the maximum ratio significantly longer in the HEK-CB12 cells (44.72±10.9 s and 191.1+18.4 s, N=8) as compared to the HEK-p3 cells (14.1+3.9 s and 134.3±16.5 s, N=9). 60 100 200 300 Time (s) 400 500 Figure 13 Effect of CaBP on recovery following flash photolysis of NP-EGTA in fura-2 loaded stable HEK cells. A single ratio acquisition was followed by a 0.4s exposure to unfdtered U V light. Shown are representative responses of single HEK-p3 and HEK-CB12 cells. Figure 14 Pooled data of recovery from N P - E G T A released Ca in fura-2 loaded stable H E K cell lines. Bars indicated mean time to recovery to the percentage of maximal response indicated. Error bars indicate SEM. * p<0.05. 62 Ca2+ Imaging of Stably transfected HEK cells treated with ATP To determine the effect of CaBP on C a 2 + released from internal stores, [Ca2 +]i was monitored with fiira-2 in stable H E K cells treated with ATP. ATP likely acts by binding to a P-type purinergic receptor which subsequently activates the PLC-InsP3 pathway, inducing release of C a 2 + through IP 3 receptors in the ER (Bischof et al., 1997). HEK-CB12 and HEK-p3 cells exposed to a continuous superfusion of lOuM ATP exhibited immediate increases in [Ca2 +]i, to a peak followed by a rapid recovery to basal levels, despite the continued presence of the agonist. Representative results of such 94-an experiment are shown in Fig. 15. Unlike [Ca ]j responses to N M D A receptor activation, ATP induced a marked [Ca ]\ oscillatory response (Fig. 15A inset) which 2_[_ persisted until the removal of the agonist. In averaged Ca responses of ATP treated cells (Fig. 15 B&C) , significant reductions in the peak [Ca2+]j value (HEK-p3 mean ratio: 2.88±0.09, N=44; HEK-CaBP mean ratio:2.51±0.07, N=49) and rate of rise (HEK-p3: 0.40+0.018 ratio units s"1, N=44; HEK-CaBP 0.28±0.014 ratio units s"1, N=49) were evident. In the initial recovery phase, HEK-p3 cells were significantly faster to recover, requiring less time (8.9±1.42 s N=44) than HEK-CB12 cells (24.5±3.16 s N=49) to recover to 75% of the maximum ratio. No significant differences were found in the later stages of recovery, however, the oscillatory patterns in [Ca 2 +]i , precluded a meaningful analysis of recovery times. o CO CO CO CO A 3.0 2.5 2.0 63 .2 1.5 CO DH CN CO 1.0 0.5 0.0 Control CaBP 50 B 100 Time (s) c 150 200 CO co CO co CO a. co o> Q_ 0) O) co > < o ra L t 4-o ra a. 0.40 I • Control • CaBP 0.35 ** 0.30 0.25 0.20 0.15 0.10 0.05 0.00 Figure 15 Effect of CaBP upon A T P induced C a 2 + release in stable H E K 293 cells. A) Average fura-2 fluorescence ratios of stably-transfected H E K 293 cell perfused with 10 u M ATP. Mean values of 44 HEK-p3 cells and 49 HEK-CB12 cells are shown. The horizontal bar indicates the 300s treatment period. Inset: Representative trace of a single H E K cell demonstrating C a 2 + oscillations.B) Mean maximum fura-2 ratios of stably-transfected H E K 293 cells in response to perfusion with lOuM ATP. C) Rate of increase in fura-3 fluorescence in stably transfected cells (HEK-p3 N=44, HEK-CB12 N=49) upon perfusion with 10 u M ATP. Error bars indicate SEM. * p<0.05. *** pO.OOl 64 Effect of CaBP Upon Capacitative Ca Influx A novel route of C a 2 + entry has recently been proposed in which the emptying of internal stores results in a so-called capacitative Ca 2 +-influx. We were able to induce this form of Ca influx, and determine the effect of CaBP, by the use of thapsigargin. Thapsigargin treatment results in a transient rise in [Ca ]j which returns to baseline over 10 min (data not shown). Perfusion of thapsigargin pre-treated cells with C a 2 + -free (200 u M EGTA) BSS followed by re-introduction of 1.8mM external C a 2 + resulted in a large sustained increase in fura-2 ratios which gradually returned towards baseline levels. When compared to HEK-p3 cultures (Fig. 16), HEK-CB12 cells exhibited a slower rate of rise in [Ca2 +]i (HEK-p3: 0.040±0.004 ratio units s"1, N=8; HEK-CB12: 0.028±0.03 ratio units s_1,N=9). No significant differences were observed in peak [Ca 2 +]i levels. Effect of CaBP on Cell Survival Following Excitotoxic Activation of NMDA Receptors Given the modulatory effect of CaBP upon C a 2 + transients in the above experiments, studies of cell survival following N M D A channel mediated C a 2 + influx were carried out. To assess the protective (or deleterious) effects of CaBP in C a 2 + mediated cell death, HEK-CB12 and HEK-p3 cells were transiently transfected with NR1/NR2A and 65 A 1.6 -r 450 500 550 600 650 700 750 800 Time (s) Figure 16 Effect of CaBP upon capacitative influx of Ca + in stably transfected HEK 293 cells. A) Representative responses of HEK-CaBP cells and HEK-p3 cells following induction of capacitative calcium influx. Cells pre-treated with thapsigargin were perfused with 200pM EGTA followed by reintroduction of 1.8mM C a 2 + (represented by the horizontal bar). B) Mean slopes of the rate of increase in 9+ 9+ [Ca ]i during capacitative Ca influx. Slopes of 9 HEK-p3 cells were compared to slopes of 8 HEK-CaBP cells. Error bars indicate S E M . * p<0.05. 66 the marker pCMVB. Transfected cultures were treated with 200 u M N M D A in BBSS for one hour (37° 5% C O 2 ), then returned to the original media containing ImM A P V . Control cultures were treated with ImM A P V replacing N M D A . Surviving cells were fixed 6 or 24 hours following recovery and stained for P-galactosidase. Fig. 17 illustrates the mean percentage survival of successfully transfected (as assessed by P-galactosidase staining) cells at 6 and 24 hours after treatment. At 6 hours, approximately half of the transfected cells remained, with no significant difference in survival between CaBP-CB12 and HEK-p3 cell lines (HEK-p3: 46.6+4.2 %; HEK-CB12: 44.5±5.5 %). However, at 24 hours, further significant cell death occurred in the control cultures, while the percentage of surviving HEK-CB12 cells was not significantly different from that observed at 6 hours (HEK-p3: 16.4+2.6%; HEK-CB12: 34.3+4.9%). Discussion Previous studies which examined the effect of CaBP on Ca transients and neuronal survival in the face of C a 2 + -mediated cytotoxicity have produced contradictory results (Mattson et al., 1991; Abdel-Hamid et al., In Preparation). However, these studies examined differences between phenotypically different neurons, which could conceivably express different levels of other proteins involved in Ca influx/homeostasis. For example, if CaBP expressing neurons also exhibited differential expression of N M D A receptors, the magnitudes of Ca transients and therefore cell survival could be affected. In the present study therefore, we have employed a cellular 67 Figure 17 Effect of CaBP upon NRl/NR2A-mediated cytotoxicity in stably-transfected H E K 293 cells. Cell cultures were co-transfected with NR1/NR2A and pCMVp then exposed for 1 h to 200 uM N M D A . Cells were fixed at 6 (six separate experiments) or 24 (7 separate experiments) hours post-treatment. The bars represent the mean percentage of surviving P-gal positive cells compared to A P V treated controls for each cell line. Error bars indicate SEM. *p<0.05. 68 system to examine Ca transients and cell survival, in which the only variable between control and experimental groups is the expression of CaBP. Effect of CaBP on NR1/NR2A Mediated Ca2+ Transients Prior electrophysiological studies have demonstrated that transfection of NR1/NR2A subunits results in the formation of functional N M D A receptors that are similar to those found in neurons (Raymond et al, 1996). We have examined the effect of recombinant CaBP upon whole-cell intracellular C a 2 + transients generated by these transfected receptors. The major effect of the presence of CaBP was a significant prolongation of the recovery to baseline values. This observation is consistent with studies of transfected or naturally occurring CaBP in other cell types (Airakisinen et al, 1997; Chard et al 1993; Lledo et al, 1992; Abdel-Hamid et al., In Preparation). However, in our experiments we did not observe a significant reduction (Chard et al 1993; Lledo et al, 1992; Airakisinen et al, 1997; Meier et al, 1998; McMahon, 1998) or increase (Abdel-Hamid et al., Unpublished Results) in peak [Ca2+]j values, nor did we observe any significant effect of CaBP upon the rate of rise in [Ca ]\ induced by N M D A receptor stimulation. In considering the potential effects of CaBP, the speed of binding of Ca to 2+ CaBP could be a critical factor, especially in relation to the speed of binding of Ca to the indicator dye or other potential Ca2+-buffering sites in the cell (see Table 1). As the concentration of C a 2 + ions increases within a cell, C a 2 + has the potential to bind to a 69 number of possible Ca2+-buffers (including the indicator dye used, and CaBP) depending upon both the K d of the buffer and the on-rate(s) for the high affinity binding site(s). The K d will determine the number of available sites at any particular [Ca2+][ . The structure of fura-2 is based upon that of B A P T A and has a very rapid on rate for C a 2 + binding (k o n = 7.6 x 108 M"V), in the order of 10 x faster than CaBP (ko„ = 0.77 x 108 M-V) . It has been previously shown that B A P T A loading of cultured hippocampal neurons results in significant alterations in depolarization or excitatory amino acid-induced C a 2 + transients (Abdel-Hamid and Baimbridge, 1997). Short (20-25s) exposures of BAPTA-loaded neurons to glutamate resulted in transients that had lower magnitudes, rates of rise and decay. Longer exposures (3-30 min) resulted in transients that were not lower in magnitude but still displayed reduced rates of rise and recovery. In the present study, during N M D A receptor mediated C a 2 + transients, it is likely that CaBP becomes saturated with C a 2 + during the period of continuous C a 2 + entry. On removal of glutamate, Ca2+-entry will soon terminate and the normal Ca2+-homeostatic mechanisms will begin to remove C a 2 + from the cell cytoplasm. For example, as the plasma membrane Ca-ATPase pumps C a 2 + out of the cell, a local reduction in [Ca 2 +]i will be 2_j_ created near the plasma membrane that will favour the release of Ca bound to CaBP into the cytoplasm thereby maintaining an elevated [Ca ]\. Essentially CaBP is acting as a Ca2+-buffer and will continue to do so until its store of bound C a 2 + is depleted. This may take some time since the off-rate of C a 2 + bound to CaBP is slow, (K 0 f f = 39.5 s"1). This process is not unlike the mechanisms that have been proposed to account for the transcellular transport of C a 2 + across enterocytes in the gut (or other Ca 2 + -70 transporting epithelial cells) (Feher et al., 1992). In this process, CaBP serves as a large capacity Ca2+-buffer that is in equilibrium with the prevailing [Ca2+]j . As [Ca2+]j rises, C a 2 + will bind to the vacant high-affinity Ca2 +-binding sites of CaBP and the increase in [Ca ]\ will be limited to a value close to the K d for CaBP (513 nM) (I. Mody, Personal Communication), provided that the total net influx of C a 2 + does not exceed the Ca -binding capacity of CaBP, (determined by the intracellular concentration of CaBP and the volume of the cell). One possible alternative mechanism by which CaBP may prolong the recovery of [Ca 2 +]i in the transiently transfected cells, is by buffering C a 2 + and thereby modulating the Ca2+-dependent desensitization of the N M D A receptor (Legendre et al., 1993; Tong and Jahr, 1994). In effect, CaBP will increase the net influx of C a 2 + and thereby increase the total amount of C a 2 + that subsequently must be removed from the cell. Supporting this possibility are our observations that the presence of CaBP results in a significant delay in the onset of Ca -dependent rundown of N M D A currents recorded from HEK-293 transfected with the NR1 and NR2A subunits of the N M D A receptor (Price et al., 1999). In the experiments described above, no significant differences in either the peak C a 2 + or the rate of rise of [Ca2+]j were observed when comparing HEK-CB12 cells with HEK-p3 cells. A possible limitation inherent in these experiments was the high degree of variability in the magnitudes of the C a 2 + transients, likely a result of differential transfection of the NR1/NR2A subunits. In addition, recent studies have demonstrated that fura-2 selectively underestimates the rise in [Ca2 +]i associated with the influx of C a 2 + through N M D A receptors (Hyrc et al., 1997; Stout and Reynolds, 1999). Given the 71 latter observations, it is quite possible that the increase in [Ca2+]j induced by N M D A receptor stimulation, measured in these experiments with fura-2, were seriously underestimated. A very large influx of C a 2 + could have overwhelmed any Ca2+-buffering capacity of CaBP, thus explaining the apparent lack of effect of this protein on the rate 2_|_ of rise and peak [Ca ]j. In order to explore this possibility further, and to minimize the cell-to-cell variance in experimentally-induced changes in [Ca 2 +]i, additional experiments were conducted on the stable H E K cell lines utilizing methods which would presumably provide for a more modest and consistent stimulation of Ca transients from cell to cell. 4-Br-A23187 Application of the ionophore 4-Br-A23187 resulted in Ca transients which were much larger in magnitude than those observed in other Ca2 +-imaging experiments. CaBP clearly reduced the maximum attained [Ca 2 +]i, as well as the rates of C a 2 + increase and decay. This finding is significant in the light of studies which have shown that exposure to A23187 induces apoptotic cell death in neurons (Gwag et al., 1999) and that apoptotic cell death in lymphocytes is reduced in the presence of transfected CaBP (Dowd et al., 1992). Of additional interest in the report by Gwag et al (1999) is the finding that the type of cell death was dependent upon the magnitude of change in [Ca 2 +]i. High (2 uM) doses of A23187 resulted in necrotic cell death, whereas more moderate (250 nM) doses (with correspondingly lower fura-2 measurements) resulted in 72 apoptotic cell death. They suggest that this is an example of how amplitude modulation o f C a 2 + can have differential cellular effects. Our experiments therefore suggest that CaBP may be able to similarly modulate cellular processes by reducing peak [Ca 2 +]i responses. ATP-InducedRelease of Ca2+from the ER To assess the possible buffering effects of CaBP upon C a 2 + released from internal stores, C a 2 + transients in stable H E K cell lines were induced by exposure to ATP. ATP induces release of C a 2 + from the ER via the PLC-IP3 pathway. Clearly 2_|_ transfected CaBP was able to attenuate the resultant rise in [CaZ T]i- CaBP may also influence this pathway by another mechanism. The IP3R exhibits a Ca feedback mechanism; at low [Ca 2 +]i levels, C a 2 + acts synergistically with IP3 to activate receptors and high [Ca2+]j inhibits IP3R activation (Bezprozvanny et al., 1991; lino and Endo, 1992). (This bi-phasic sensitivity to C a 2 + of the IP3R is fundamental to generation of C a 2 + waves which will be discussed in the following chapter.) Therefore, at low [Ca ]•„ CaBP may compete with IP 3 receptors for Ca 2 + , inhibiting channel activation. At high [Ca 2 +]i, CaBP may also compete for Ca 2 + , therefore preventing inactivation, although the K d of CaBP suggests that it may be saturated at high [Ca2+]j and therefore more effective at inhibiting channel activation at low [Ca 2 +]j. Further aspects of the effect of CaBP upon IP3-mediated C a 2 + release will be discussed in Chapter 4. 73 NMDA-subunit-mediated and ATP-mediated C a 2 + release share a similar mode of change in [Ca 2 + ]i : both allow external C a 2 + to flow down its concentration gradient into the cell, in turn creating a gradient of Ca extending from the channels to deeper levels within the cell. To examine the effect of CaBP upon C a 2 + transients with different dynamic properties, we utilized a "caged" Ca compound. NP-EGTA The use of the caged C a 2 + compound NP-EGTA allows for a global, rapid increase in [Ca 2 +]i that is independent of the activation of plasma membrane and intracellular receptors or channels (Ellis-Davies and Kaplan, 1994). The release of C a 2 + from NP-EGTA is extremely rapid (in the order of a few us; Ellis-Davies and Kaplan, 1994) and would essentially be complete between the time at which exposure to U V light is terminated and the time of the initial measurement. This method also avoids the generation of steep gradients of [Ca2 +]i and the subsequent recovery from the increase in [Ca 2 +]i can be observed in the absence of continued C a 2 + entry. Under these conditions we observed clear evidence of a C a 2 + buffering effect of CaBP. An initial, very rapid decline in C a 2 + was followed, only in the presence of CaBP, by a pronounced plateau phase and then a prolongation of the slow phase of recovery ofthe Ca -transient. Similar patterns of recovery were observed in experiments employing fluo-3 and fura-2 fluorochromes. The peak C a 2 + response following photolysis of NP-EGTA approached l u M , well above the estimated Kd for CaBP (513 nM I. Mody, Personal 74 Communication) suggesting that the protein is likely to be saturated with C a 2 + soon after uncaging. In the absence of CaBP there is a prolonged fast component of the recovery phase from an increase in C a 2 + induced by photolysis of NP-EGTA, suggesting that, as the cell's normal C a 2 + homeostatic mechanisms remove excess Ca 2 + , the component bound to the indicator dye is released relatively rapidly. Indeed, fluo-3 or fura-2 will act as a Ca2+-buffer that will be in equilibrium with the prevailing [Ca2+]j determined by a 9+ combination of the cell's normal Ca -homeostatic mechanisms, (and the contribution of CaBP, i f it is present). The most prominent feature of the effect of CaBP on the recovery is a pronounced plateau phase. We suggest that this plateau phase is sustained by the 9+ 2+ slow release of Ca from the binding sites on CaBP. Essentially, as Ca is removed from the cell it will be replaced by C a 2 + previously bound to CaBP until such time as all of this source of C a 2 + is depleted and baseline levels of [Ca2 +]i are restored. This 2+ mechanism is similar to that proposed to account for the role of CaBP in Ca transport (Feher et al., 1992). Out findings lend support to this mechanism in H E K cells, as the "plateau phase" in recovery occurs at a [Ca2 +]i of approximately 500 n M (334/380 ratio 1.5), very close to the reported Kd (513 nM) for CaBP. Capacitative Ca2+ Entry 9+ To examine the effect of CaBP upon capacitative Ca entry, internal stores were emptied of C a 2 + utilizing thapsigargin, an inhibitor of ER C a 2 + -ATPase (Thastrup et al. 1990). Removal and subsequent re-introduction of external Ca resulted in a sustained * • 2+ rise in [Ca ]j, the rate of which was decreased in the presence of CaBP (Fig. 14). The mechanism(s) responsible for this route of C a 2 + entry have yet to be elucidated, although it has been speculated that a so-called C a 2 + influx factor (CIF) is released by 2+ the ER when internal Ca stores are depleted, and this factor in some manner induces 2+ Ca -entry from across plasma membrane (for a review see Putney and McKay, 1999) Whatever the mechanism, CaBP had a significant effect on the rate of [Ca2+]j increase, consistent with the action of a C a 2 + buffer, and similar to that observed for ATP and 4-Br-A23187 stimulated C a 2 + transients. Overview of Ca2+ Imaging Experiments Stable H E K 293 cell lines were employed as a simple test system for the intracellular C a 2 + modulating potential of recombinant CaBP. Table 2 summarizes the C a 2 + -imaging observations made in the H E K stable cell lines. A fast, high capacity C a 2 + buffer would be expected to reduce the rate of [Ca ]\ rise, lessen the peak [Ca ] h and prolong the recovery. In all experiments CaBP met at least one of these criteria. In experiments where ATP or 4-Br-A23187 were employed to induce Ca2+-transients, all of the criteria described above were met. Of particular interest is the consistent finding that CaBP slows the recovery phase of C a 2 + transients regardless of the source of the Ca 2 + . This is also in agreement with previous reports that used a wide variety of cell models 2+ • and methods of inducing Ca transients. 76 Table 2 Summary of Effect of Stably transfected CaBP in H E K 293 Cells NR1/NR2A Transfection + Glutamate ATP 4-Br-A23187 NP-EGTA Flash Photolysis Capacitative C a 2 + Influx Rate of Rise = I i NA I Peak [Ca 2 + ] i = I = = Recovery Rate I I I i NA I Increased relative to controls ^ Decreased relative to controls = No Significant difference between experimental and controls NA No data available The lack of a consistent effect of CaBP upon peak [Ca2+]j and the rate of rise in [Ca ]i induced by the different stimuli used in the above experiments could be explained by a number of factors. First, in the case of N M D A receptor stimulation, it is likely, on the basis of recent evidence (Hyrc et al, 1997; Stout and Reynolds, 1999), that the use of fura-2 may have underestimated the rates of rise and magnitudes of the increases in [Ca2+]j and that the actual values were sufficiently large to overwhelm any Ca2+-buffering effect of CaBP. Second, flash photolysis of NP-EGTA results in an almost instantaneous increase in [Ca2+]i increase and as such competition between the indicator dyes uses (fluo-3 and fura-2) and CaBP would be heavily in favour of binding preferentially to the former since their on-rates for Ca are an order of magnitude faster than CaBP. Indeed, under these conditions it is likely that binding of C a 2 + to CaBP occurred subsequent to its release from the indicator dyes. Third, it was notable that a small but significant effect of CaBP upon the rate of rise of [Ca2+]i was observed with the use of ATP, capacitative Ca2+-entry, and 4-Br-A23187. In each case, the stimulus 9+ used resulted in slow rate of rise of [Ca ]\ relative to that seen with N M D A or NP-EGTA. Overall our data support a Ca2+-buffering role for CaBP but also provide evidence that any Ca2+-buffering effect of CaBP can be overwhelmed by very large and/or very rapid increases in [Ca2 +]j. The consistent effect of CaBP in prolonging the rate of recovery of increases in [Ca 2 +]i, no matter the route of increase, would suggest that CaBP is more effective at buffering C a 2 + as it diffuses away from the regions of very high [Ca2+]j that occur close to the source of Ca2+-entry, i.e. close to N M D A receptor-operated channels in the plasma membrane or IP3 and ryanodine receptors in 78 the endoplasmic reticulum. In this context CaBP may, as a result of its particular binding kinetics and mobility, be very effective at limiting the magnitude of the rise in [Ca2 +]i in areas of the cell away from the points of Ca2+-entry. A similar action of CaBP has been proposed by others (Chard et al 1995; Roberts, 1994; Feher, 1983). Excitotoxicity To examine the effect of CaBP upon cell death induced by N M D A receptor stimulation of HEK-CB12 and HEK-p3 cells, we used a protocol modified from Raymond et al., (1996). Cells stably expressing CaBP, and transfected with the NR1/NR2A subunits of the N M D A channel, showed no difference in the degree of cell death after 6h, but a significant reduction in further cell death between 6-24h. Since greater than 50% of the cells died in the first 6h of these experiments it is likely that these cells expressed the highest levels of functional N M D A receptors, resulting in magnitudes of C a 2 + influx that would overwhelm any Ca2+-buffering capacity of the cells, regardless of their CaBP content, and that those more moderately transfected cells survived the initial treatment, allowing detection of CaBP-enhanced survival. The early cell death observed with excessive N M D A receptor stimulation, is likely necrotic in nature and results mostly from the influx of large amounts of sodium, chloride and water, rather than the influx of Ca 2 + . This type of cell death has been observed both in the H E K 293 cell model system (Raymond et al., 1996), and in cultured neurons (Rothman, 1985; Olney et al. 1986; Choi, 1987), and it is not perhaps surprising that the presence or absence of CaBP has little or no effect. On the contrary, 2+ delayed cell death in models of excitotoxicity have been correlated with Ca entry 79 during and after the period of stimulation, and is observed when cells are capable of recovering from the early sodium, chloride and water influx (Rothman et al 1987; Choi 1987; Randall and Thayer; 1992; Manev et al 1989). A pattern of early necrotic and delayed apoptotic cell death in neuronal cultures that is dependent upon the concentration of the agonist used, has been reported for the Ca2+-ionophore, Br-A23187, (Gwag et al 1999), and for glutamate and N M D A (Ankarcrona et al 1995; Bonfoco et al 1995). In all cases high concentrations of the agonist induced necrotic cell death whereas lower concentrations resulted in apoptotic cell death. A protective effect of for CaBP against excitotoxicity has been previously shown in a variety of models of cell death (Iacopino et al., 1992; Mattson et al, 1991, Burke and Baimbridge 1993; Goodman et al; 1996; Ho et al., 1996; Meier et al., 1997, 1998; Roy et al., 1998; McMahon, 1998; Phillips et al, 1999), including two other reports that suggested a specific protection by CaBP against apoptotic cell death (Dowd et al, 1992; Diop et al 1995). Our data suggest that the failure of CaBP to protect against cell death either in vitro (Mockel and Fischer, 1994) or in vivo (Freund et al 1990) to be due to excessive stimulation leading to necrotic, rather than apoptotic death. The results of the present studies suggest an apparent discordance between the Ca2+-buffering effects of CaBP and its ability to protect cells from Ca2+-mediated cell death. The lack of an effect of CaBP on peak [Ca2+]j stimulated by N M D A receptor activation, and the prolongation of the recovery of Ca -transient to baseline values, would not immediately suggest that CaBP would protect against Ca2+-mediated cell death. A dissociation between measurements of [Ca ]\ and cell death has been previously reported, (Michaels and Rothman, 1990) although more recent reports 80 suggest that this may have been due to the underreporting by the use of fura-2 of increases in [Ca2+]j induced specifically by N M D A activation (Hyrc et al 1997; Stout and Reynolds 1999). However, recent reports of the role of mitochondria in cell death (Schinder et al 1996; Green and Reed, 1998; Duchen, 1999; Vergun et al, 1999; Nicholls and Budd, 2000), and in particular the observation that blockade of mitochondrial uptake can prevent neuronal cell death. (Stout et al., 1998), may provide an alternate 2~b explanation. We speculate that, through its ability to bind Ca , CaBP could limit mitochondrial sequestration of Ca , and hence influence cell survival following influx of excitotoxic levels of Ca 2 + . This possibility is of interest considering the work published on the role of CaBP in Ca 2 + transporting tissue such as the gut, where this protein was originally discovered (Wasserman and Taylor, 1966). By comparing the relative affinities for Ca 2 + of the plasma membrane Ca2+-ATPase, CaBP and mitochondria, Rassmusen and Gustin (1978) suggested 2+ 2+ that the preferential binding of Ca to CaBP may prevent Ca -uptake into mitochondria during Ca2+-transport; a potentially essential function considering that mitochondrial Ca 2 + -uptake occurs at a cost to ATP generation at a time when significant amounts of ATP are required to pump out Ca 2 + at the serosal membrane (Rassmusen and Gustin, 1978). Summary / Conclusions The use of stable H E K cell lines has allowed for unequivocal assessment of the effect of CaBP upon C a 2 + transients and NMDAR-mediated cell death. In these cell lines 81 diverse mechanisms were employed to transiently increase [Ca 2 + ] i . CaBP was able to influence the pattern of [Ca2+]j whether the C a 2 + was from the extracellular space (NMDAR-mediated, 4-Br-A23187, Capacitative-Ca2+), intracellular stores (ATP-induced) or released globally by the caged C a 2 + compound NP-EGTA. Regardless of the mechanism, a Ca2+-buffering effect was evident in all experiments, with the most consistent effect being a prolongation of the recovery from the induced C a 2 + load (Table 2). This prolongation of recovery is a consistency which runs through other CaBP-transfection studies reported in the literature. Aside from directly buffering Ca 2 + , CaBP may also have an effect upon the mechanisms of Ca entry, for example, by modulating 2_j  Ca -dependent inactivation (NMDAR) or activation (^/Ryanodine receptors). Evidence was also obtained to support a protective function for CaBP, based upon its Ca2+-buffering capability. Specifically, it has been demonstrated that the presence of CaBP is capable of protecting cells from calcium-dependent delayed, most likely apoptotic, cell death. However, if the Ca2+-buffering capacity of CaBP is overwhelmed by excessive C a 2 + entry, or if the stimulus also results in a large influx of sodium and chloride, the presence of CaBP fails to protect against cell death. In the latter case, the rapid cell death is most likely by necrosis. The protection conferred by CaBP against delayed, Ca2+-mediated cell death, may be due to the preferential binding of C a 2 + to CaBP, resulting in a limitation of the available free C a 2 + that would otherwise be taken up by mitochondria, with the potential of triggering the release of mitochondrial apoptotic factors, leading to cell death. The ability of CaBP to protect only against delayed cell death may explain some of the contradictory results in the literature. Chapter 4 Effect of Recombinant CaBP IP3 Mediated Calcium Oscillations 83 Chapter Summary 1. The effects of artificial C a 2 + buffers and transfected CaBP on histamine-induced Ca2" oscillations in HeLa cultures were examined. 2. Artificial C a 2 + buffers modified the oscillatory pattern of intracellular C a 2 + in histamine-stimulated HeLa cells. When loaded with artificial buffers, fewer cells demonstrated a distinct "baseline" oscillatory pattern. 3. Individual C a 2 + spikes in cells containing CaBP have a "flattened" profile; peak 2+ 2+ [Ca ]i is lowered, the rate of increase in [Ca ]\ is slower and transients are prolonged. 4. A novel method for resolving C a 2 + waves was utilized to investigate the effect of artificial C a 2 + buffers and CaBP upon the propagation of IP3-induced C a 2 + waves. 94- * 5. Artificial Ca buffers and recombinant CaBP were found to retard the propagation 2_|_ velocity (by up to 60%) of Ca waves in HeLa cells. 84 Introduction The discussion of excitotoxicity in the previous chapter has emphasized that C a 2 + buffering proteins can influence the dynamics, specifically the peak [Ca 2 +]i and the rates of rise and decay, of intracellular transients and thereby possibly affect the outcome during excitotoxic influx of Ca 2 + . Moreover, Gwag et al (1999) have shown that two different modes of cell death can be induced, depending upon the amplitude of the rise in [Ca2 +]j. The role of C a 2 + in cell death is clearly one related to pathophysiological conditions, with the exception of the type of programmed cell death that occurs, for example, during development. Under normal physiological conditions various means of C a 2 + signaling play a vital role in neuron function, including development and synapse formation, neurotransmitter release (Katz and Miledi, 1967), the mechanism underlying learning and memory (Bliss and Collingridge,1993), and the regulation of protein synthesis (Ghosh et al., 1994). Of increasing interest in the literature is the exploration of how Ca 2 + , a ubiquitous second messenger, can specifically mediate such a wide variety of cellular processes. Emerging hypotheses suggest that through the regulation of spatial and temporal variations in intracellular Ca , different signal transduction pathways can be activated (reviewed by Berridge, 1997). Intracellular signals are not only encoded by amplitudes of [Ca 2 +]i increases, but also by patterns of C a 2 + influx or release, often in the form of oscillations that can be observed at a whole cell level, and C a 2 + waves at the sub-cellular level (Berridge, 1997). Depending upon their dynamics, such as the amplitude of a Ca2+-response or the frequency of oscillations, different intracellular pathways may be activated. For example, variations in amplitude and 85 duration of Ca signals in lymphocytes has been shown to mediate differential activation of transcription factors (Dolmetsch et al., 1997). It has also been demonstrated that variation in frequency of Ca oscillations can modulate gene expression (Dolmetsch et al., 1998, L i et al., 1998). Providing a specific example in developing neurons, Gu and Spitzer (1995) have demonstrated that C a 2 + oscillations can influence neurite outgrowth. Our understanding of the mechanisms leading to Ca2+-oscillations and the generation of C a 2 + waves has been greatly advanced by studies utilizing histamine stimulation of HeLa cells (Thorn, 1995; Bootman et al., 1997). These studies have resulted in a model in which the generation of C a 2 + oscillations and waves are dependent on the bell-shaped sensitivity curve of Ca on IP3R. Further studies have suggested that endogenous C a 2 + buffers may play a role in modulating the information transmitted by Ca2+-oscillations and C a 2 + waves (Clapham, 1995). In order to investigate this possibility, a HeLa cell line was generated which stably expresses CaBP (HeLa-CaBP), along with a control cell line transfected with the pCINeo vector alone (HeLa-pCINeo), to determine if the C a 2 + buffering properties of CaBP demonstrated in other cell types could affect the dynamics of C a 2 + oscillations and waves induced by the action of histamine. 86 Results Effects of Indicator Dyes on Ca2* Oscillations In initial experiments with HeLa cells it was determined that the loading 2_|_ o 4-concentration of the Ca indicator fluo-3 significantly influenced the Ca -oscillatory behaviour of HeLa cells. When loaded with 3.1 u.M fluo-3 for 60 min at room 94-temperature, very few cells displayed oscillations in which the level of [Ca ]j returned to near-baseline levels prior to the initiation of the subsequent C a 2 + spike (data not shown). It was determined empirically that reducing the loading concentration of fluo-3 resulted in more defined C a 2 + spikes. Therefore in all subsequent experiments cells were loaded with 0.75 u M fluo-3 for 60 min, both in studies of fields of oscillating cells and 94-in high magnification studies of Ca waves in individual cells. In a typical field of cells, it was observed that almost all cells displayed a change in [Ca 2 +]i in response to superfusion with l u M histamine. However, in agreement with Missiaen et al., (1993), we also observed a variety of different responses from cell to cell. Therefore, the type of C a 2 + oscillations were categorized as shown in Fig. 18, following the nomenclature of Missiaen et al. (1993). The majority of cells responded with slow (~45s duration) C a 2 + oscillations with each successive peak typically being of lower magnitude than its predecessor. In additional experiments using fura-2, we observed that the influence of this indicator dye upon the type of response of HeLa cells to histamine, was even more pronounced than that observed with fluo-3. This problem was alleviated (but not to the 87 Baseline Oscillator 74% Sinusoidal 16% Time (s) Time (s) Single Spike 6% Monophasic <1% Time (s) Time (s) No response or other: 4% Figure 18 Classification of Calcium Oscillatory Types in HeLa cells. Representative traces of individual HeLa cells loaded with fluo-3 and exposed to continuous perfusion of 1 u M histamine at room temperature. Indicated percentages are from a population of 196 individual cells. The horizontal bar indicates the 300s l u M histamine perfusion. 88 extent seen with the use of lower loading concentrations of fluo-3) by reducing the loading concentration of fura-2 to the minimum possible levels for detection on our imaging system. Typically the use of fura-2 resulted in a greated proportion of cells exhibiting a "sinusoidal" oscillatory pattern (see Fig. 18). In view of our observed differences between responses using either fluo-3 or fura-2, initial experiments examining the effect of artificial C a 2 + buffers and CaBP on C a 2 + oscillatory patterns were performed utilizing fluo-3. However, some experiments were also performed using cells minimally loaded with fura-2, particularly when we considered it important to attempt to quantify the observed changes in [Ca 2 +]i. Effect of Artificial Calcium Buffers on the Oscillatory Behavior of HeLa Cells HeLa cell cultures were loaded with fluo-3 alone or combinations of fluo-3 and cell-permeant artificial C a 2 + buffers, then superfused for 300s with 1 p M histamine. The intensity of fluorescence from individual cells was determined on an Attofluor imaging system (63x Objective) by placing a single "region of interest" over each cell in the field of view. When compared to HeLa cells loaded with fluo-3 alone, the effect of loading cells with 2.5 u M E G T A - A M was a small increase in the percentage of cells demonstrating a more sinusoidal response at the expense of a reduction in those having an oscillatory response (Fig. 19). Increasing the loading concentration of E G T A - A M to 89 c o ro C L O CL 0) o n— o • Baseline H Sinusoidal • Single • Monophasic g No Response No Half EGTA-AM EGTA-AM Treatment BAPTA-AM 2.5uM 10uM BAPTA-AM BAPTA-AM 1uM 10uM Figure 19 Effect of Calcium buffers on Oscillatory Behaviour of H e L A Cells. HeLa cells co-loaded with fluo-3 and the indicated buffer were exposed to 300s perfusions of 1 p M histamine at room temperature. The pattern of calcium oscillation was classified according to Figure 18. Bars indicate the proportion of cells displaying the indicated oscillatory pattern. Total cells counted for each experiment were as follows: No Treatment: N=5, 196 cells total; Half-BAPTA: N=3, 106 cells total; 2.5 u M EGTA: N=3, 65 cells total; 10 u M EGTA: N=3, 102 cells total; 1 u M BAPTA: N=3, 88 cells total; 10 u M B A P T A : N=3, 149 cells total. 90 10 u M had the effect of reducing the total number of baseline oscillators and increasing the percentage of cells which responded with a single C a 2 + spike. Cells loaded with 1 u M B A P T A - A M displayed a significantly different pattern of C a 2 + response to 1 u M histamine (Fig. 19), with the majority of cells responding with a single C a 2 + transient. A loading concentration of 10 u M B A P T A - A M completely abolished repetitive oscillations and most cells exhibited a "monophasic" C a 2 + response to histamine. In addition, 43% of cells loaded with this concentration of B A P T A failed to show any detectable increase in fluo-3 fluorescence. Effect Of CaBP on Calcium Oscillations Continuous superfusion of HeLa-CaBP cells with 1 uM histamine produced oscillatory patterns that were not significantly different from HeLa-pCINeo cells (Fig. 20). A n additional analysis of the frequency of C a 2 + oscillations was carried out by counting the number of peaks in [Ca2+]i over 400s and this also revealed no significant differences between HeLa-pCINeo (0.93±0.04 oscillations/min, N=35) and HeLa-CaBP cells (0.95±0.03 oscillations/min, N=66). Unfortunately, the wide variety of oscillatory patterns and changing frequencies over time precluded any definitive conclusions as to the effect of CaBP on oscillation frequency. 91 90 "5 60 Q. O Q- 50 a> O 40 "co o 30 O 20 10 Baseline Sinusoidal Single Spike Other Oscillatory Pattern Figure 20 Effect of recombinant CaBP on calcium oscillatory pattern in stable HeLa cell lines. HeLa cells stably transfected with CaBP or control vector were loaded with fluo-3 and the fluorescence response of individual cells recorded during a 300s perfusion with 1 uM histamine at room temperature. The pattern of calcium oscillation was classified according to figure 18. Bars indicate the proportion of cells (HeLa-pCINeo N=4; 143 cells total; HeLa-CaBP: N=5, 154 cells total) which displayed the indicated oscillatory pattern. Error bars indicate S E M . 92 2-F When individual Ca oscillations were examined, it was found that the mean time for a complete oscillation in HeLa-CaBP cells was significantly prolonged (45.3111.94 s, N=32), when compared to that in HeLa-pCINeo cells (39.54±1.6 s, N=33 p<0.05) (Fig. 21). This prolongation effect was confirmed through analysis of span times of normalized fluorescence values (25%-span: HeLa-pCINeo: 19.14±0.74 s, N=33; HeLa-CaBP: 23.81+1.53 s, N=32; 50%-span: HeLa-pCINeo: 11.81+0.64 s HeLa-CaBP: 14.98+1.13 s, p<0.05). In addition, high-speed imaging studies, carried out in conjunction with wave resolution experiments described in the following section, 2+ revealed that, when treated with lOOuM histamine, the rate of increase in [Ca ], (expressed as AF/F s"1) was significantly reduced from 1.362+ 0.164 AF/F s"1, N=7 in HeLa-pCINeo cells to 0.914+0.096 AF/F s"1, N=19 in HeLa-CaBP cells (Fig. 22). In order to assess any effect of CaBP upon the magnitude of [Ca ]\ responses, an analysis of HeLa cell responses was also carried out using cells minimally loaded with the ratiometric dye fura-2. In the majority of both HeLa-CaBP and HeLa-pCINeo cells, a "sinusoidal" pattern of C a 2 + oscillation was observed and the elevated levels of [Ca 2 + ] i persisted until the histamine was removed (Fig. 23A). In order to assess the role of external C a 2 + in the response of HeLa cells to histamine, these experiments were repeated with 200uM E G T A and no added C a 2 + in the perfusate. Under these conditions, [Ca 2 + ] i returned to baseline levels prior to the washout of the histamine (Fig. 23B). The peak [Ca2 +]i appeared to be unaffected: maximum ratios (Fig. 24A) were not significantly different from cells superfused with media containing 1.8mM C a 2 + and there was still a significant difference between HeLa-pCINeo and HeLa-CaBP cells 93 Total Time 25%Span 50%Span Figure 21 Influence of CaBP upon individual calcium transients in oscillating stable HeLa cells. Single calcium transients in the oscillatory pattern of HeLa-pCINeo and HeLa-CaBP cells exposed to l u M histamine at room temperature were analyzed. The time of the total individual oscillation was determined as was the span (time) at 25% and 50% of the maximum response. Error bars indicate SEM. (* p<0.05; ** p<0.01) 94 120 , 0 1 2 3 4 5 6 7 8 Time (s) Figure 22 The effect of CaBP upon rate of C a 2 + increase in stable HeLa cells treated with lOOuM histamine. Representative traces of two stably transfected HeLa cells imaged at 15 frames per second following addition of lOOuM histamine at RT. Fluorescence data was converted to % AF/F and the rate of increase was determined between 25% and 50% of the maximum fluorescence measured. The inset chart displays the pooled data of 7 HeLa-pCINeo and 33 HeLa-CaBP cells. Error bars indicate S E M . (*p<0.05) 94 95 o 00 CO co ra: B o C O C O C O C O .2 1.0 CO OH 100 200 300 400 Time (s) 500 600 600 Figure 23 Effect of CaBP on response of HeLa cells loaded with fura-2 and treated with l u M histamine. (A) Mean [Ca 2 +]i response of HeLa-pCINeo (N=32) and HeLa-CaBP (N=42) cells treated for 400s with l u M histamine RT. Bar indicates treatment period. Inset: Single cell (HeLa-pCINeo) oscillator example. (B) Mean [Ca 2 +]i responses of HeLa-pCINeo (N=34) and HeLa-CaBP (N=42) cells treated for 400s with l u M histamine in the absence of external C a 2 + (BSS + 200 u M EGTA; No added Ca Z T) 2+N 96 (HeLa-pCINeo: 2.6+0.11 N=32; HeLa-CaBP: 2.0±0.08 N=42; Ca2+-calibrated values: HeLa-pCINeo: 710 nM ; HeLa-CaBP: 472 nM). However, the average maximum 334/380 fluorescence ratio (Fig. 24A) was significantly less in CaBP-containing cells (HeLa-pCINeo: 2.6+0.13, N=32; HeLa-CaBP: 1.9±0.08, N=42; Ca2+-calibrated values: HeLa-pCINeo: 710 nM ; HeLa-CaBP: 437 n M ) . Figure 24B summarizes the effect of CaBP upon the relative total intracellular Ca in histamine treated HeLa cells as determined by the integrated areas under the graphs of fluorescence ratio vs time. The presence of CaBP resulted in a significant reduction in average integrated area (HeLa-pCINeo: 455.4±16.06, N=32; HeLa-CaBP: 361.8+13.36, N=42). In similar experiments where C a 2 + was removed from the bathing media, a significant reduction in integrated area (HeLa-pCINeo: 294+10.08, N=34; HeLa-CaBP: 235±8.20, N=42) was also observed (Fig. 24B). The sinusoidal oscillatory pattern exhibited by the majority of fura-2 loaded cells prohibited an analysis of span times and recovery to baseline. Calcium Wave Propagation in Stable HeLa Cell Lines The above experiments described the changes of C a 2 + observed in entire cells in response to histamine. In order to resolve properties of the C a 2 + waves in HeLa cells, a technique employing a standard epifluorescence Ca2 +-imaging apparatus was developed (other reports of C a 2 + wave velocity have typically employed confocal laser scanning 97 A 3.0 N o rm al C a Ic iu m Ze ro C a l c i u m Figure 24 Effect of recombinant CaBP upon C a 2 + influx in fura-2 loaded HeLa cells treated with histamine. (A) Comparison of peak fura-2 ratios of HeLa-pCINeo (N=34) and HeLa-CaBP (N=42) cell lines loaded with fura-2 and treated with 1 u M histamine in the presence of 1.8mM calcium or absence of external C a 2 + . (B) Integrated area under traces of fluorescence response of fura-2 loaded HeLa-pCINeo (N=34) and HeLa-CaBP (N=42) to l p M histamine in the presence or absence of external C a 2 + . Error bars indicate SEM. * =p<0.05. 98 microscopy). A very similar technique for measuring Ca2 +-wave velocity was subsequently described by Wang et al (1997). Following superfusion of 100 u M histamine, a region of interest, (1 pixel wide and up to 250 pixels or 75pm long) was placed over single HeLa cells and scanned at 15 frames per second under high magnification (lOOx objective). The fluorescence intensity of each pixel along this strip was quantified. Averaged fluorescence values from these scans were used to determine global changes in cell [Ca2 +]i (see Fig. 22), while an analysis of individual pixels was 9-t-used to resolve the "front" of Ca waves. The percentage change in AF/F values were calculated for each pixel along the scanned line and the time at which each point surpassed a 50% increase in AF/F was plotted against its position along the scanned line. From this plot the approximate origin of the C a 2 + wave could be determined along with the propagation velocity of the wave. We observed, as have other authors (Bootman et al.,1997), that C a 2 + waves with single initiation sites were more reproducibly generated with the application of 100 uM histamine. In the example shown in Fig. 25, a single initiation site is evident. Cells did not appear to have a common region where waves were initiated, as both sites near the nucleus and in the cell periphery were observed, resulting in wave propagation away from or towards the center of the cell respectively. Successive histamine responses in the same cell resulted in wave initiation from the same region (data not shown) and some cells displayed more than one initiation site (Fig. 26) or an indistinct Ca2 +-wave. These were excluded from the detailed analysis of wave velocity as described below. 99 A 13 0 2 0 4 0 6 0 8 0 Cell Length ( m i l ) Figure 25 Example determination of wave velocity for an individual HeLa cell treated with lOOuM histamine. (A). Digitized fluorescence image (inverted greyscale for clarity) of a single HeLa Cell loaded with fluo-3. The horizontal bar indicates the region scanned for changes in fluorescence. The horizontal dimension is preserved in the X-axis of the graph in B. (B) Plot of calcium wave propagation in a fluo-3 loaded HeLa cell. The plot indicates the time at which that particular point along the linescan reaches 50% AF/F. The line of best fit (solid line) was used to determine wave velocity. The dotted line represents the minimum time required for a 50% AF/F at the initiation site. 100 Figure 26 Resolution of two calcium waves, initiated at distinct sites in an individual HeLa cell treated with lOOuM histamine. (A). Pseudo-colour digitized fluorescence image of a single HeLa Cell loaded with Fluo-3. The region indicated in the micrograph (1) was scanned for changes in fluorescence. The pseudo-colour scale is indicated on the left. The horizontal dimension ofthe micrograph is preserved in B and C. (B) Pseudo-colour composite of linescans of a single HeLa cell treated with lOOuM histamine. %AF/F values were calculated from fluorescence images obtained from the linescan depicted in panel A. Pseudo-colouring was applied to each image as indicated by the bar left of the image, and images were stacked vertically. Arrows indicate the regions of calcium wave initiation. (C) Plot of 50% threshold of linescan. Individual points along the line represent the time at which the pixel at the position indicated on the x-axis surpassed the 50%AF/F threshold. Arrows indicate regions of wave initiation. 101 102 Effect of Artificial Calcium Buffers and CaBP upon Calcium Wave Propagation The velocity of C a 2 + waves were determined in HeLa cells loaded with C a 2 + buffers (see Methods section). In addition, the velocities of C a 2 + waves in HeLa-pCINeo controls were compared to those in HeLa-CaBP cells. Controls cells had an average velocity of 81.4±10.5 pm s"1 (N=10) whereas loading with 2.5uM E G T A - A M , l u M B A P T A - A M or lOuM B A P T A - A M significantly reduced the average velocity to 50.215.3 pm • s"1 (N=7), 38.2±6.1 pm • s"1 (N=5), and 22.0±4.6 pm s"1 (N=5), respectively (Fig. 27). When wave velocity determination was carried out in HeLa-CaBP cells, a reduction in velocity to 49.5+6.66 pm • s"1 (N=12), similar to the effect of 2.5 p M E G T A was observed. 103 100 Untransfected HeLa-pCINeo 2.5 uM EGTA-AM 1uM BAPTA-AM 10uM BAPTA-AM HeLa-CaBP Figure 27 Summary of resolved calcium wave velocities in HeLa cells. Bars indicate mean calculated velocity of calcium waves in transfected (HeLa-p3 and HeLa-CaBP) HeLa cells and untransfected HeLa cells loaded with the indicated calcium buffer. Indicated significant differences are in comparison to the appropriate control * p<0.05 ** p<0.01; *** pO.001. The mean value of bars indicated by common letter code(s) are not significantly different from each other. 104 Discussion It has been recognized for some time that second messenger signaling by C a 2 + is not solely enacted through simple increases or decreases in [Ca2 +]i. Although information can be encoded by the amplitude of the increase in [Ca 2 +]i , oscillatory frequency has also been shown to mediate numerous cellular processes (for a review see Berridge, 1997). It has therefore been proposed that proteins which bind Ca may modulate this signal and hence influence the transduction of information in cells (Clapham, 1995). The availability of HeLa cells stably transfected with CaBP, enabled us to examine the effects of CaBP upon histamine-induced Ca2 +-signalling in this cell line and to compare these effects with that of artificial Ca2+-buffers. Calcium Oscillations in HeLa Cells The oscillatory behaviour of fluo-3 loaded HeLa cells observed in these experiments were similar to those observed by other authors (Missiaen et al, 1993). The majority of fluo-3 loaded cells demonstrated repetitive C a 2 + transients in which C a 2 + levels returned to baseline prior to the onset of the next spike. This oscillatory behaviour was dependent upon refilling of internal stores by external Ca (Fig. 23B). Ca oscillations in response to histamine are a result of the bi-phasic sensitivity curve ofthe IP 3R. Bezprozvanny et al. (1991) demonstrated that the IP3R open-probability increases 105 with increasing [Ca2+]j to a maximum at -250 nM [Ca2+]j. Increasing [Ca 2 +]i above this level results in a progressively lower open-probability. With this Ca2+-sensitivity in mind, the initial impetus for these experiments was to determine whether or not C a 2 + buffers and/or CaBP might have an effect upon the frequency of C a 2 + oscillations in HeLa cells. However our initial examination of oscillatory behaviour of populations of wild-type HeLa cells revealed that cell-to-cell heterogeneity of oscillatory frequency, and changing frequencies in individual cells, might preclude any definitive examination of frequency modulation in this cell type. Nonetheless, modulation of C a 2 + oscillatory behaviour was detected by an examination of the relative proportion of the various oscillatory patterns. Effects of Indicator Dyes Our observation that the oscillatory behaviour is significantly affected by the loading duration and concentration of fluo-3 suggests that the Ca2+-buffering effect of this indicator dye may interfere with the natural physiological response of HeLa cells to histamine. Indeed, this is not entirely unexpected since the properties of fluo-3 (see Table 1) are similar to its parent compound, BAPTA. In the particular case of fluo-3 however, an alternative explanation is possible since it has been shown that fluo-3 can act as a competitive antagonist of IP3 receptors (Richardson and Taylor, 1993). In our experiments, reducing the loading concentration of fluo-3 resulted in more consistent oscillatory responses. This observation further emphasizes the importance of minimal 106 loading with Ca indicators and the need for caution in interpretation of experiments utilizing these compounds. Ca2+ Diffusion, Fixed and Mobile Ca2+ Buffers 2+ 2+ Given that generation of Ca waves is dependent upon Ca released from the ER, and that the diffusion of this C a 2 + is responsible for the propagation of C a 2 + waves, it becomes important to understand how mobile buffers, such as B A T A , E G T A and CaBP, influence C a 2 + diffusion within living cells. It has been shown in many cell types, including gonadotrophs, G H 3 cells, and neurons, that 99% of the Ca which enters the cell is rapidly bound by endogenous, immobile C a 2 + buffers (Tse et al., 1994; Lledo et al., 1992; Fleet et al., 1998; Fierro and Llano, 1996; Palecek et a l , 1999). The remaining 1% is free ionic C a 2 + which can interact with mobile C a 2 + indicators and/or C a 2 + buffers, which in turn may modulate Ca oscillations and waves. Ca2+ Oscillations and Artificial Ca2+ Buffers in HeLa Cells To characterize the effect of artificial C a 2 + buffers on C a 2 + oscillations, HeLa cells minimally loaded with fluo-3 were also loaded with either E G T A or B A P T A . In experiments where HeLa cells were loaded with low (2.5 uM) concentrations of E G T A -107 24-A M , "baseline" Ca oscillations gave way to more sinusoidal oscillatory patterns, which in turn, were replaced by single C a 2 + transient that returned to baseline when the loading concentration was increased to 10 pM. In comparison, B A P T A abolished histamine-induced C a 2 + oscillations at all tested loading concentrations. At loading concentrations of 1 p M , HeLa cells predominately responded to histamine with a single C a 2 + spike, whereas cells loaded with 10 p M B A P T A responded with a delayed, slow, monophasic 2+ rise in [Ca ] i followed by a slow return to basal levels upon wash-out of histamine or failed to respond. • * 2+ Histamine-induced Ca oscillations in HeLa cells are a product of the bi-phasic sensitivity of the IP 3R to C a 2 + (Bezprozvanny et al., 1991). At low levels (< 250nM), 94- 94-increasing intracellular Ca acts synergistically with IP 3 to induce Ca release from the 2"P ER. This positive feedback system is responsible for the rising phase in [Ca ]\ 2"P oscillations. At high intracellular concentrations, Ca inhibits the action of IP 3 on its receptor, resulting in a return to near basal levels of [Ca 2 +]i as C a 2 + is cleared by Ca 2 + -2+ 2"P ATPases or by other intracellular Ca homeostatic mechanisms. Ca waves are initiated from so-called "hot-spots" in the cell, which are likely clusters of IP 3 receptors (Bootman et al.,1997). The effect of this clustering is the overlapping of microdomains of released Ca 2 + , resulting in an increased local [Ca 2 +]i and an increased open-probability of adjacent receptors. This Ca2+-mediated recruitment of adjacent receptors is thought to 2_|_ 24- • be the underlying mechanism of Ca oscillation initiation and Ca wave propagation. In attempting to predict the effect of a C a 2 + buffer upon IP3-mediated C a 2 + 94-release both the speed of binding and the level at which [Ca ]\ would be buffered (i.e. the K d of the buffer) must be taken into consideration. A buffer that binds C a 2 + 108 significantly slower than the IP3R itself is unlikely to have a significant effect. A n appropriate analogy would be the fact that, with respect to neurotransmitter release in the squid giant axon synapse, the fast Ca2+-buffer B A P T A significantly reduces release 2+ whereas the slow Ca -buffer, EGTA has little effect (Adler et al., 1991). With respect to the Kd, a buffer with a low Kd (~ 250 nM or less) may maintain [Ca 2 +]i, in the range where it promotes IP3-mediated Ca2+-release whereas one with a high Kd (> 500 nM) may maintain [Ca2 +]j in the range where it inhibits IP3-mediated Ca2+-release. Based upon the known properties of B A P T A (Kd=0.50 uM; Pethig et al., 1989, k o n= 6x108 M" 1 1 2+ s" ; Tsien, 1980, see Table 1) and the ranges of [Ca required for activation and inhibition of IP3-mediated Ca2+-release it is likely that B A P T A may initially inhibit IP3R activation by competing with adjacent receptors in "hot spots" for released Ca 2 + (or prevent channel activation altogether if the buffer concentration is high enough). Following activation however, B A P T A may enhance additional C a 2 + release and prolong elevation of [Ca2 +]i by buffering [Ca2+]j near the concentration where C a 2 + acts synergistically with IP 3 to open the IP3R (-250 nM; Bezprozvanny et al, 1991). This hypothesized action of B A P T A is supported by the results presented in this chapter in that HeLa cells loaded with B A P T A either fail to respond to histamine, or respond with a slow-onset, prolonged, monophasic rise in [Ca2 +]i which slowly returns to basal levels upon washout of histamine. It must also be considered that the observed effect of B A P T A could be mediated by B A P T A directly inhibiting the IP3 channel (as described above for fluo-3). It should be noted also that, as with the case of fluo-3-AM and fura-2-AM, the final intracellular concentration of the AM-derivatives of artificial Ca buffers is 109 unknown. Due to the mechanism of loading - A M derivatives, it is possible for the cells to accumulate the buffer to levels well above that of the loading solution. For example, it has been estimated that intracellular concentration of - A M derivatives of B A P T A analogs may reach 20 to 40 times the loading concentration (Wang and Thompson, 1995). The dramatic difference in effects of EGTA and B A P T A are interesting in that the two buffers have a similar Kd (see Table 1, Introduction). However, there is a 300-fold difference in the compounds on-rates (K o n ) (BAPTA: 600x l0 6 M _ 1 s_1 EGTA: 2x l0 6 M " 1 s"1). Therefore, these experiments suggest that the on-rate is more critical in the ability of C a 2 + buffers to interfere with the mechanisms underlying C a 2 + oscillations. It 2+ is possible that within the clusters of IP3 receptors ("hot-spots") where Ca waves are 2+ initiated, the close proximity of IP3 receptors may preclude the action of Ca buffers with slow on-rates. A similar correlation between K o n and localized Ca2+-buffering has been reported previously by Adler et al., (1991). In this study, B A P T A was found to be more effective at blocking Ca -mediated neurotransmitter release in the squid giant synapse than EGTA, and this effect was attributed to its faster Ca2 +-binding kinetics. 110 Modulation of [Ca J, Oscillations by CaBP Stable expression of CaBP did not appear to have an effect upon the frequency or general pattern of histamine-induced [Ca2+]j oscillations in HeLa cells. Similar to the case for EGTA, this result is perhaps due to CaBP's binding kinetics, which would not allow it to bind C a 2 + rapidly enough between IP3 receptors in initiation sites to affect initiation of Ca oscillations. Once again, by analogy with the study from Adler et al (1991), there is no evidence that the presence of CaBP in neuronal terminals interferes 2_j_ with neurotransmitter release, suggesting that it is likely to act as a relatively slow Ca -buffer and is therefore unable to influence Ca2+-dependent events in which the source of C a 2 + entry or release is very close to the subsequent target of the action of Ca 2 + . Although we were unable to show an effect of CaBP upon initiation of C a 2 + oscillations, an effect of CaBP was observed upon global changes in [Ca ]j when individual C a 2 + spikes were examined. In HeLa-CaBP cells the rate of increase in [Ca2+]i was slower and the duration of the spikes was prolonged when compared to those observed in the absence of CaBP. Given that the pCINeo-transfected HeLa cells are an appropriate control, we can conclude that any observed differences can be unequivocally ascribed to the effects of CaBP. In order to quantify the increase in [Ca2+]i associated with histamine induced C a 2 + release, additional experiments were performed utilizing fura-2. These experiments were limited by the effect of the indicator dye on oscillatory patterns, but nonetheless revealed that the peak C a 2 + response was significantly lower in HeLa-CaBP cells when I l l compared to HeLa-pCINeo cells. Overall the effect of CaBP on HeLa cell [Ca 2 +]j oscillation is similar to the effect observed in H E K 293 cells; a dampening of the [Ca 2 +]i response with transients that are lower in magnitude but longer in duration. It is possible that the apparent effect of CaBP might be mediated through a direct interaction with the IP3R, similar to that reported for fluo-3. However, this seems unlikely in light of the similarities of the observed effect upon individual C a 2 + 9-4-oscillations to those [Ca ]\ transients described in the previous chapter, mediated by 9+ caged Ca or structurally unrelated channels. CaBP in these experiments had an 9+ apparent buffering effect even when the source of the Ca was not the IP3R. Therefore unlike B A P T A , CaBP does not appear to be able to modulate the pattern of C a 2 + oscillations owing to its slower binding kinetics, however CaBP clearly modulates the histamine response in HeLa cells by globally buffering Ca . Calcium Waves It has been previously reported that individual C a 2 + oscillations are comprised of C a 2 + waves which propagate throughout the cell (Bootman and Berridge, 1996). Activation of phospholipase C following application of histamine results in the production of intracellular IP3 which, in turn, stimulates the release of C a 2 + from the ER. Any C a 2 + released from one IP3R can act synergistically with IP3 on adjacent receptors, resulting in propagation of the C a 2 + wave (Bootman, 1996). Typically these waves have been examined employing confocal laser-scanning imaging, and it has been determined 112 that the propagation velocity is in the order of 10 pm s"1 (Bootman et al., 1997). We have developed a method for resolving C a 2 + waves employing a standard epiflourescence Ca imaging apparatus in order to determine the effect of CaBP upon wave propagation velocity. Histamine-induced Ca waves were found to be similar to those observed with confocal microscopes in that they had a defined origin, propagated throughout the cell, originated from central or peripheral sites, and were initiated occasionally at more than one site. However, compared to previously reported values the actual velocity of wave propagation was in the order of 80 pm s"1. This is ~8 times faster than the average velocity found by Bootman et al (1997). However the authors of this paper describe the measured wave velocity as a result of a "partially desensitized response": the wave velocity was measured in a cell after having been repeatedly treated with histamine. In our experiments with individual HeLa cells, we also observed that wave velocity decreased with repeated applications of histamine (data not shown). Mechanism of the Attenuation of Ca Wave Velocity We have shown that EGTA, B A P T A and CaBP all significantly retarded the propagation of C a 2 + waves in HeLa cells stimulated with histamine. Propagation of IPs-mediated Ca2+-waves is reliant upon diffusion of C a 2 + from the activated IP3R to adjacent receptors. The C a 2 + -buffering action of EGTA, B A P T A and CaBP appears to inhibit this process, likely by competing with adjacent receptors for Ca 2 + , inhibiting the 113 synergistic activation with IP3. Simple competition for C a 2 + however does not take into account the fact that these buffers are mobile in the cytoplasm. Feher (1983) demonstrated using a 3-chamber diffusion model that CaBP can "facilitate" the diffusion of Ca 2 + . In the model of C a 2 + wave propagation, facilitated diffusion would 2_|_ effectively lower the local Ca concentration surrounding the activated IP3R limiting the contribution of released Ca to synergistic activation of adjacent receptors. Quantitative modeling of C a 2 + influx through Ca2+-channels has suggested that the presence of a mobile C a 2 + buffer will limit the effect of C a 2 + on adjacent receptors (Simon and Llinas, 1985; Roberts, 1994). Therefore the term "facilitated diffusion" is 2"P 2"T" somewhat of a misnomer. Diffusion of Ca is facilitated by mobile Ca buffers, in that such buffers will facilitate distribution of C a 2 + throughout the cytoplasm. However, the coefficient of diffusion for CaBP bound to C a 2 + (D=20pm2 s"1) is an order of magnitude 2_|_ 2 1 slower than the coefficient of diffusion for free Ca in cytoplasm (D=200pm s" ) (Roberts, 1994). Therefore Roberts (1994) suggests the term "buffered diffusion" as an alternative. The net result of this buffered diffusion in the immediate vicinity of an open C a 2 + channel is a reduction in the size of the plume of free C a 2 + emanating from the channel and hence a limiting of the effect of this released Ca on adjacent receptors. Evidence supporting such an effect of CaBP has been reported in theoretical papers (Roberts, 1994) and experimentally in neurons using electrophysiological techniques (Chard etal, 1995). As with the effect of buffers upon histamine-evoked Ca2+-oscillations, it appears that the k o n value is critical in determining the degree of modulation; the faster on-rate buffer B A P T A has a much greater effect upon C a 2 + wave velocity than the slower on-114 rate buffers E G T A or CaBP. The k o n for CaBP has been estimated at 0.77x108 M " 1 s"1 (I. Mody, personal communication) which is intermediate between B A P T A (6.0x108 M " 1 s"1) and E G T A (2xl0 6 M" 1 s"1) (Roberts , 1993). This correlation of k o n to the efficacy of modulatory effect on C a 2 + dependent processes has also been reported in posttetanic potentiation in neurons (Chard et al., 1995). In the present study, the absolute intracellular concentrations of E G T A or B A P T A are unknown, and would likely have an effect on wave propagation as illustrated by the effects of the two loading concentrations for B A P T A - A M upon Ca wave velocity. However, the concentration of CaBP in HeLa-CaBP is similar to that in a number of neurons (see Chapter 2 and Fig. 5) although less than that in cerebellar Purkinje cells. It is likely that increasing the concentration of CaBP would have a similar effect as increasing concentrations of EGTA or BAPTA. Summary / Conclusions In summary, it has been demonstrated that CaBP has a significant effect upon the velocity of IP3 mediated C a 2 + waves in HeLa cells, presumably through its ability to buffer Ca 2 + . In support of this, similar effects were observed in cells loaded with artificial C a 2 + buffers, although direct inhibitory effects upon IP3R (in the case of BAPTA) may have contributed. The propagation of Ca2+-waves requires the diffusion of C a 2 + away from the cluster of IP3 receptors in the initiation site to other IP3 receptors that are likely 115 separated by larger distances. As suggested by others (Adler et al., 1991; Chard et al., 1995), the scope of the effects of Ca buffers such as CaBP may be dictated by their binding kinetics. In this case the slower on-rate (relative to fast Ca2+-buffers such as BAPTA) is sufficient to bind and slow down the rate of diffusion of C a 2 + to IP3 receptors involved in the propagation of a Ca -wave. In a similar manner, the relatively slow buffering of C a 2 + would account for the effect of CaBP on the global C a 2 + -responses, namely the reduction in each of the rate of rise, peak value and rate of fall of the individual peaks of an oscillatory Ca -response. The fact that B A P T A was much more effective than CaBP at altering the pattern of an oscillatory C a 2 + response, even to the degree of completely preventing them, is consistent with this being an effect largely restricted to the site(s) of initiation. The close proximity of the IP3 receptors at the initiation sites dictates that C a 2 + released from one IP3 receptor need diffuse only a very small distance to promote the further release of C a 2 + from adjacent IP3 receptors. As such, only a Ca2+-buffer with a sufficient capacity and fast on-rate would be capable of influencing this essential interaction. This effect is entirely analogous to the ability of B A P T A (but not EGTA) to inhibit neurotransmitter release in the squid giant synapse where the calcium channels are spatially located very close to the release sites for the vesicles containing the neurotransmitter (Adler et al., 1991). It is possible that cell-to-cell heterogeneity, along with changing oscillation frequencies over time may have masked any subtle effect of CaBP on oscillatory frequency, and that higher levels of CaBP may be more effective. However, it is more likely that the levels of CaBP naturally expressed in cells, including neurons, are insufficient to influence very rapid Ca2+-dependent effects such as neurotransmitter 116 release or the initiation of Ca2+-waves. However, given sufficient time, CaBP can be an effective Ca2+-buffer capable of limiting the global Ca2+-response of a cell and of influencing events that require significant diffusion of C a 2 + from the source o f C a 2 + entry or release. Examples would be the ability of CaBP to slow down the velocity of a 7+ 74-Ca -wave (as described above), and to limit mitochondrial Ca -uptake or the activation of other processes leading to Ca2+-induced cell death. We therefore propose that CaBP may not only be a buffer which protects against possibly harmful increases in [Ca ]\ but may also affect signal transduction events by modulating C a 2 + transients. This has wide ranging significance in all cell types that exhibit C a 2 + waves and oscillations (for a review see Berridge et al., 1998). A n example 2_|_ in neurons, reported by Gu and Spitzer (1995), demonstrated that the frequency of Ca waves can regulate neurite extension and differentiation in spinal neurons. Modulation of this process by CaBP has not as yet been examined, but it is conceivable that modulation of Ca2+-oscillations could influence neurite outgrowth or the expression of developmental genes that do not manifest as structural changes. From this point of view it is of interest that it has been reported that some neurons only transiently express CaBP during early stages of development (Friauf, 1993). Chapter 5 General Discussion & Conclusions 118 CaBP and Ca Transport Calbindin-D28K (CaBP) was isolated from chick intestinal epithelium in 1966, where its expression is dependent upon the active metabolite of vitamin D, 1,25-dihydroxy-cholecalciferol (Wasserman and Taylor, 1966). It was subsequently found to be expressed in many vertebrate tissues, including the nervous system, where its distribution has been mapped extensively (Wasserman and Fullmer, 1982; Jande et al., 1981; Celio, 1990). The putative function of CaBP in the intestinal epithelium was 9+ hypothesized to be a dual role related to transcellular Ca transport; facilitating diffusion of C a 2 + across the cytoplasm, while buffering C a 2 + and thus preventing large increases in [Ca ]\ in the transporting cells. In mammals, CaBP is retained in the nervous system but is replaced by a smaller, genetically unrelated, protein calbindin-D9K, in C a 2 + transporting cells including those in the gut, kidney and bone (Christakos et al., 1989). CaBP as a Trigger Protein Although CaBP was discovered in intestinal tissues, phylogenetic evidence suggests that its expression most likely evolved first in the nervous system (Parmentier, 1990). In the mammalian nervous system, where expression of CaBP is not dependent upon vitamin D, the function of CaBP remains controversial. Evidence from structural 119 and molecular studies reveal that CaBP is highly conserved across species, demonstrating a 98% sequence identity among mammals and 79% identity between mammalian and avian forms of the protein. Interestingly, this strong sequence conservation is evident in regions of the protein not involved in binding C a 2 + (non-EF-hand regions). This observation led Parmentier (1990) to suggest that, in addition to 2_)_ buffering Ca , the function of CaBP may involve a protein-protein interaction. This has been supported by Berrgard et al (2000), who utilized spectroscopic analysis of CaBP to examine structural changes following addition of C a 2 + (also see Fig. 4, Chapter 2). They found that upon binding Ca 2 + , or in response to an increase in H + , CaBP undergoes a conformational change which exposes hydrophobic regions of the protein. These conformational changes are similar to, though of a lesser magnitude than those exhibited by "trigger" proteins such as calmodulin. Interestingly, in the absence of Ca , calmodulin has very little by way of exposed hydrophobic surfaces, whereas CaBP does. Based therefore on the ability of Ca2 +-binding to result in the exposure of hydrophobic 2_|_ surfaces, CaBP falls neither into the classical trigger sub-group of Ca -binding proteins, nor into the classical buffer sub-group. More recently, it has been suggested that the presence of hydrophobic surfaces even in the absence of Ca2 +-binding, may account for the Ca2+-independent binding of CaBP to caspase-3 reported by Bellido et al (2000). They reported that transfection of osteoblasts with CaBP decreases tumor necrosis factor-a induced apoptosis by inhibiting caspase-3 activity. 120 CaBP as a C a z + Buffer A large body of circumstantial evidence supports a Ca -buffering role independent of any cellular effects mediated through possible CaBP-protein interaction. The objective of the present study was therefore to examine the Ca2+-buffering properties of CaBP in cellular systems and to determine the effects of this Ca2+-buffering upon cell survival and C a 2 + signaling. Our experiments, taken together, suggest that CaBP can act as a C a 2 + buffer and that this Ca2+-buffering property alone may have direct effects on Ca2+-mediated events, both physiological and pathophysiological. Studies which have compared whole cell [Ca2 +]i with fluorescent probes, to the net flux of C a 2 + through channels measured as whole cell Ca2+-currents, have revealed that Ca 2 +-influx is very effectively buffered by endogenous C a 2 + buffers present in all cells. In gonadotrophs, for example, it was found that only 1% of the total number of C a 2 + ions which enter a cell remained as the free ion (Tse et al., 1994). Similar studies employing melanotrophs (Thomas et al., 1990), GH3 cells (Lledo et al., 1992) and neurons (Fleet et al., 1998; Fierro and Llano, 1996; Palecek et al., 1999) also revealed this large proportion of C a 2 + bound to endogenous buffers. The identity of these endogenous C a 2 + buffers is unknown, however studies have shown that they must be immobile and have a low affinity for C a 2 + (Kd ~ lOOpM in chromaffin cells) since they are not saturated at l p M [Ca 2 +]i (Xu et al., 1997; Neher and Augustine, 1992; Zhou and Neher, 1993). Candidates would include any immobilized, negatively charged cytosolic proteins, perhaps anchored to, or components of, the cytoskeleton. 121 In light of the studies described above, whatever Ca2+-buffering role CaBP may play, the pool of Ca that CaBP is capable of influencing is a very small percentage of the total net C a 2 + flux into the cell. Nonetheless, Lledo et al (1992), using CaBP-transfected G H 3 cells, demonstrated that indeed CaBP does have an additional and significant C a 2 + buffering effect on this remaining 1% of Ca 2 + . In our studies, utilizing stable cell lines expressing CaBP, we have unequivocally demonstrated that CaBP can buffer this pool of Ca 2 + . This buffering was evident in all experiments, manifested by a reduction in rate of [Ca 2 +]i rise, a reduction in peak [Ca 2 +], and/or a prolongation of the recovery to baseline [Ca Moreover, utilizing the various methods of inducing Ca transients, we have provided evidence that this buffering effect is independent of the source of the Ca entry or release. Furthermore, our experiments demonstrate this source-independent C a 2 + buffering can have consequences upon physiological (Ca 2 + waves and oscillations) and pathophysiological (NMDA mediated excitotoxicity) events. From a theoretical point of view, a Ca2+-buffer would be expected to act to lower peak [Ca2 +]i and reduce the rate of rise and decay of [Ca 2 +]i. In a variety of experimental models, we have demonstrated that transfected CaBP can produce some or all of these effects. Most notably, we observed that in H E K treated with ATP or 4-Br-A23187, transfected CaBP produced all of the effects ascribed to a C a 2 + buffer. Therefore, due to the use of simple cellular systems in which the only variable altered between control and experimental groups was the presence or absence of CaBP, this study provides unequivocal evidence in favour of a Ca2+-buffering role for CaBP. Additionally, our studies demonstrating C a 2 + wave attenuation provide evidence that CaBP is acting as a mobile C a 2 + buffer. As described above, the vast majority of 122 Ca entering the cell is bound by immobile endogenous buffers. A fixed buffer would retain a high concentration of C a 2 + locally while a mobile C a 2 + buffer would be expected to have the effect of facilitating "buffered diffusion" of C a 2 + (Roberts, 1993). This would 2_|_ limit the local rise of Ca around activated IP3 receptors and therefore the activation of adjacent receptors (Roberts, 1993; Hall, et al 1997; Wang and Thompson, 1995) and 9+ attenuate the rate of Ca wave propagation. 2+ Possible Consequences of Ca Buffering by CaBP Intracellular C a 2 + is a signaling molecule for a wide variety of biological 2_|_ processes. However, it has been shown that a prolonged elevation of [Ca ]\ can be harmful to cells and Ca2+-induced cell death has been implicated in numerous pathologies (reviewed by Doble, 1999). Many studies have also examined the potential correlation between the expression of CaBP and cell survival in many neuropathologies (reviewed by Heizmann and Braun, 1992). There is evidence of neuronal sparing of cells expressing CaBP in Parkinson's disease (Yamada et al., 1990; German et al., 1992), and MPTP-induced Parkinson's (German et al.,1992; Iacopino et al., 1992; Lavoie and Parent, 1991), while others have shown altered expression of CaBP in Alzheimers disease (Lally et al., 1997; Ichimiya et al., 1988), epilepsy (Sloviter et al., 1991; Mody et al., 1987; Magloczky et al., 1997), and Down syndrome (Kobayashi et a l , 1990). However, there are many inconsistencies in the argument for a neuroprotective role for 123 CaBP based upon anatomical data, likely due to the correlative nature of many of these studies. In addition to these correlative observations between CaBP and neuropathologies, a number of experimental approaches have been made in order to determine whether or not either artificial C a 2 + buffers, or naturally occurring molecules such as CaBP that may act as C a 2 + buffers, are capable of preventing C a 2 + mediated cell death. It has been hypothesized that CaBP may mediate this resistance through its ability to buffer increases in intracellular calcium (Baimbridge and Parkes, 1981; Baimbridge et al., 1982; Jande et al., 1981; Mody et al., 1987). Following this rationale, several studies have examined the effect of artificial calcium buffers in preventing Ca2+-mediated cell death. The potential utility of the rapid intracellular calcium buffer B A P T A as a (neuro)protective agent remains controversial, as recent studies have reported protective (Tymianski et al., 1993b) or deleterious (Adbel-Hamid and Baimbridge, 1997) effects of B A P T A in the face of C a 2 + challenges. Similar contradictory reports of the protective properties of endogenous CaBP have been published using various cell models (Mattson et al., 1991; Mockel and Fisher, 1994; Abdel-Hamid et al., In Preparation). However, recent studies, including those presented in Chapter 3, employing molecular biological techniques to introduce CaBP into cells which do not normally express CaBP, have provided evidence that transfected CaBP enhances survival following influx of toxic levels of C a 2 + (McMahon et al., 1998; Meier et al., 1998). In our studies, it would appear that CaBP is protective in our transfected H E K 293 model of Ca2+-mediated cell death. Specifically, CaBP appears to prevent delayed, presumably apoptotic cell death cell death 24 hours following excitotoxicity treatments. 124 These experiments are in agreement with the results reported by Gwag et al (1999), in that there is a Ca2 +-toxicity threshold, above which cells die by necrosis, whereas more moderate [Ca ]•, challenges induce apoptotic cell death. In our experiments, CaBP was unable to prevent cell death which occurred within 6 hours following excitotoxic treatments, presumably due to entry of C a 2 + in excess of the total buffering capacity of CaBP. However, in cells that were exposed to more moderate levels of [Ca 2 +]j, our study suggests that CaBP was able to successfully buffer levels of [Ca 2 + ] i and prevent cell death that is likely to be apoptotic in nature. CaBP, Mitochondria and Cell Death In the past decade it has been established that mitochondria, in addition to their essential role as generators of ATP, are also capable of sequestering and therefore contributing to the buffering of C a 2 + transients (Thayer and Miller, 1990, Werth and Thayer, 1994; Sidky and Baimbridge, 1997). However, it has also been reported that mitochondria constitute a potential "cellular poison cupboard", which, in response to physiological or pathophysiological signals (including Ca2+-uptake) will induce apoptotic or necrotic cell death (Earnshaw, 1999; Miller, 1998; Nicholls and Bud, 1998; 2000). In support of this function, it has been recently demonstrated that inhibition of mitochondrial C a 2 + uptake can prevent excitotoxic cell death (Stout et al., 1998). This mechanism raises the possibility that CaBP may act to prevent cell death by inhibiting uptake of C a 2 + into mitochondria. This hypothesis was first proposed by Rasmussin and 125 Gustin (1978), who, by comparing the relative Ca2+-affinities of CaBP (Kd -0.5 pM) and mitochondria (Kd -1.0 pM), suggested that the preferential binding of C a 2 + to CaBP may prevent Ca2+-uptake into mitochondria in Ca2+-transporting cells. This is one of the potential mechanisms to explain the results of the C a 2 + excitotoxicity studies described in Chapter 3, where CaBP appeared to prevent delayed cell death. By preventing C a 2 + uptake into mitochondria, CaBP may serve to protect the cell against death i f C a 2 + uptake into mitochondria is absolutely required for death to occur, as has been shown by Stout etal (1998). CaBP as Modulator of Intracellular Signaling Another potential role for CaBP suggested by the studies presented in this thesis is that of a modulator of intracellular C a 2 + signals. In transfected HeLa cells, CaBP significantly attenuated the velocity of histamine-induced C a 2 + waves. As well, we observed a Ca2+-buffering effect upon C a 2 + oscillations, with peak [Ca 2 +]i and the rates of increase and recovery of [Ca2 +]i all being reduced. We did not observe a direct effect upon the frequency of C a 2 + oscillations, likely due to the binding kinetics of CaBP which would not allow it to influence rapid C a 2 + events between the closely spaced IP3R at initiation sites. However this is consistent with studies that have suggested that buffers with slow on-rates may not be able to influence local, rapid Ca2+-mediated events but can effect global changes or events that involve greater distances between the source of C a 2 + entry and the effector molecules (Adler et a l , 1991; Chard et al., 1995). 126 As discussed in Chapter 4, CaBP may, in addition, be capable of modulating the frequency of C a 2 + oscillations, given a high enough intracellular concentration of the protein. CaBP-mediated modulation of C a 2 + oscillations is of particular interest in the light of recent studies that have demonstrated variations in amplitude (Dolmetch et a l , 1997) and/or frequency (Dolmetch et al., 1998, L i et al., 1998) of calcium oscillations can modulate gene expression. Another report by Gu and Spitzer (1995), demonstrated that modulation of calcium oscillations can affect neurite outgrowth. Taken together with our findings, this raises the possibility CaBP could affect gene expression and hence development through modulation of calcium transients or oscillations. Initial investigation of CaBP knockout mice have not revealed any structural abnormalities (Klapstein et al., 1998), however the possibility remains that CaBP could influence gene expression during development which is not manifested by structural changes. A n interesting example of a specific temporal and spatial pattern of CaBP expression during development is the transient appearance of CaBP in the superior olivary complex of developing rats (Friauf,1993). This expression could also be interpreted as a safeguard against cytotoxic levels of C a 2 + during active Ca2 +-signalling throughout development. 127 Future Studies Studies in Transfected HEK Cells Further studies on H E K 293 stable cell lines should focus upon the exact mechanism of cell death. The type of cell death (necrotic/apoptotic) in response to activation of transfected NR1/NR2A receptors should be examined. The influence of CaBP upon mitochondrial activity in transfected H E K 293 cells could be examined in two ways. Firstly, mitochondrial C a 2 + could be examined directly using the C a 2 + -indicator dye Rhod-2 that, in its reduced form, is preferentially taken up by the mitochondria. If CaBP does indeed delay mitochondrial Ca uptake, then we would expect that mitochondrial C a 2 + would not increase in parallel with increases in cytoplasmic Ca 2 + , measured with cytoplasmic-specific dyes. Secondly, the Ca 2 + -mediated disruption of the mitochondrial membrane potential that is hypothesized to be a step in apoptotic and necrotic cell death should be examined. Rhod-123 is a mitochondrial membrane potential sensitive dye which is significantly quenched inside mitochondria. Upon disruption of the inner membrane potential, the dye is released into the cytoplasm where its fluorescence intensity increases. If CaBP is capable of binding C a 2 + and preventing mitochondrial C a 2 + uptake, we would expect that the transfection of CaBP would prevent or delay the depolarization of the mitochondrial membrane potential, thereby preventing or delaying cell death. 128 Excitotoxicity studies of transfected H E K cells to date have employed simple cell counts to assess cell death. These studies should be followed up with experiments utilizing the live cell indicator 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl tetrazolium bromide (MTT), that is metabolized to a dark blue formazan product by live cells which can be quantified by absorption spectroscopy. Furthermore, NR1/NR2A transfected H E K cultures subjected to excitotoxic stimuli should be examined for specific signs of apoptotic cell death. This could be carried out by staining with 4,6-diamindino-2-phenylindole (DAPI) to determine the degree of peripheral nuclear condensation and Western Blotting for cytochrome C released from the mitochondria. Another possibility is the examination of caspase-3 activation, which is also associated with apoptotic cell death. This might be particularly interesting in light of the recent observations of Bellido et al (2000) that CaBP may exert its anti-apoptotic action through inhibition of caspase-3 activation. Studies in Primary Hippocampal Cultures Concurrent with the studies described in this report, we have recently generated a replication deficient adenovirus that will be employed to express CaBP in primary hippocampal cultures. Additional viruses are being generated which will co-express green fluorescent protein (GFP) and CaBP. This will allow for assessment of exogenous CaBP expression in live cells. The efficacy of CaBP-expressing herpes simplex viruses in preventing glutamate-mediated cell death have been recently described (Meier et al., 129 1998). These finding should be confirmed and extended to studies of the effect of CaBP upon mitochondria function, employing the mitochondrial-specific dyes described above. In Vivo Studies The hypothesized protective role of CaBP could also be tested in in vivo models of neuronal cell death. Stereotacic injections of CaBP expressing adenovirus could be employed to assess the protective effects of CaBP in ischemia models in rats and in the MPTP mouse model of Parkinsons Disease. These experiment could also be performed with the newly-available CaBP-knockout mice to eliminate any effects of endogenous CaBP. 130 Conclusions and General Summary 1. To overcome the potential confounding phenotypic effects of cells which naturally do or do not express CaBP, human CaBP was cloned from a cDNA library, inserted into mammalian expression vectors, and used to generate stable cell lines (with appropriate control cell lines). Therefore any differences between control and experimental groups in the experiments described can be unequivocally ascribed to the effect of CaBP. 2. We have obtained evidence that directly supports a Ca buffering role for CaBP. This C a 2 + buffering was evident in all experiments, despite different routes of C a 2 + entry or release. 3. In a model of excitotoxicity employing transfected N M D A receptors, the presence of transfected CaBP has been directly correlated to protection from delayed, presumably apoptotic cell death. This effect is possibly due to prevention of C a 2 + uptake by mitochondria. 2"f" 4. We have explored the possibility that CaBP may influence other modes of Ca signaling, particularly through the generation of C a 2 + oscillations and C a 2 + waves in stable-transfected HeLa cells. While we were unable to measure any direct effect upon the frequency of C a 2 + oscillations, the profile of the component C a 2 + spikes was flattened and CaBP significantly attenuated the velocity of C a 2 + waves. 131 5. Our results are consistent with a Ca2+-buffering action of CaBP that, compared to B A P T A , is relatively slow. Thus, events that are spatially coupled to the source o f C a 2 + entry or release are influenced by B A P T A but not CaBP. Examples would be neurotransmitter release and the feedback effects of C a 2 + on IP3 receptors. 6. When there is significant diffusion of C a 2 + between the source of entry or release and the target, (e.g. mitochondria or IP 3 receptors activated during an intracellular C a 2 + wave) CaBP may have a significant influence. An example would be the ability of CaBP to reduce the velocity of a Ca wave. One of the intriguing facets in the study of intracellular signaling is the fact that a single ubiquitous messenger such as calcium mediate a myriad of diverse cellular processes, including cell death. Four possible mechanisms have been proposed : amplitude modulation, frequency modulation, control of route of calcium entry (or release) and differential expression of downstream signaling molecules (reviewed by 2+ Berridge, 1997). In this study it has been demonstrated that CaBP can influence Ca 2+ mediated cellular processes through the spatial and temporal buffering of cytosolic Ca . The fourth possibility (not addressed experimentally in this work) may explain the differential susceptibility of certain cell types to cytotoxic calcium. Absence of expression of downstream signaling molecules in the cell death pathway may effectively protect cells regardless of CaBP expression. Similarly, the ability of CaBP to modulate C a 2 + signals may depend upon the expression levels and sensitivity ofthe target 132 mechanisms. In summary, the context of expression of CaBP may prove to be a very important determinant in the protective or modulatory capabilities of CaBP. 133 Abbreviations A M Acetoxymethyl ester A P V 2-Amino-5-phosphonovalerate ATP Adenosine triphosphate B A P T A 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid BBSS Balanced Buffered Salt Solution BSS Buffered Salt Solution C a 2 + Ionized calcium [Ca 2 +]i Intracellular ionized calcium concentration C A B P Calbindin-D28K cDNA Complementary Deoxyribonucleic Acid C R A C Calcium Release Activated Channel D M E M Dulbecco's modified Eagle's medium DMSO Dimethyl suifoxide D N A Deoxyribonucleic acid dNTP Deoxynucleoside triphosphate E A A Excitatory amino acids E G T A Ethyleneglycol-tetraacetic acid ER Endoplasmic Reticulum FBS: Fetal bovine serum HBS Hepes Buffered Saline H E K 293 Human Embryonic Kidney (cells) IPs Inositol 1,4,5-triphosphate IP3R Inositol 1,4,5-triphosphate-Receptor K d Dissociation constant LTP Long Term Potentiation Mito Mitochondrion mRNA Messenger ribonucleic Acid N M D A N-Methyl D-Aspartate N M D A R N-Methyl D-Aspartate Receptor NP-EGTA o-nitrophenyl E G T A PCR Polymerase Chain Reaction PBS Phosphate Buffered saline P M Plasma Membrane P M C A Plasma Membrane Ca 2 +-ATPase PTP Mitochondrial Permeability Transition Pore ROCC Receptor-operated C a 2 + channel ROI: Region of interest RyR Ryanodine Receptor SEM: Standard error of the mean SERCA Sarco/Endoplasmic Reticulum Ca 2 +-ATPase v o c e Voltage-operated C a 2 + channel 134 References Abdel-Hamid K . M . , Baimbridge K . G . The effects of artificial buffers on C a 2 + responses and glutamate-mediated excitotoxicity in cultured hippocampal neurons. Neuroscience. 81:673-687,1997. Abdel-Hamid K . M . , Rintoul, G.L. and Baimbridge K . G . The effects of Calbindin-D28K on C a 2 + responses and glutamate-mediated excitotoxicity in cultured hippocampal neurons. In Preparation - Unpublished Results. Adler E M . Augustine GJ. Duffy SN. Charlton MP. Alien intracellular calcium chelators attenuate neurotransmitter release at the squid giant synapse. Journal of Neuroscience. 11:1496-507, 1991. Airaksinen MS. Eilers J. Garaschuk O. Thoenen H. Konnerth A . Meyer M . Ataxia and altered dendritic Ca2+ signaling in mice carrying a targeted null mutation of the calbindin D28k gene. Proceedings of the National Academy of Sciences of the United States of America. 94:1488-93, 1997 . AnegawaNJ. Lynch DR. Verdoorn TA. Pritchett DB. Transfection of N-methyl-D-aspartate receptors in a nonneuronal cell line leads to cell death. Journal of Neurochemistry. 64:2004-12, 1995. Ankarcrona M . Dypbukt JM. Bonfoco E. Zhivotovsky B. Orrenius S. Lipton SA. Nicotera P. Glutamate-induced neuronal death: a succession of necrosis or apoptosis depending on mitochondrial function. Neuron. 15:961-73, 1995. Armstrong C M . Hille B. Voltage-gated ion channels and electrical excitability Neuron. 20:371-80, 1998. 135 Ausubel, Frederick M . Current protocols in molecular biology. Published: New York : Published by Greene Pub. Associates and Wiley-Interscience : J. Wiley, cl987-Baimbridge, K.G. In "The dentate gyrus and its role in seizures". Eds. C E . Ribak, C M . Gall and I. Mody. Elsevier Science. pp211-220,1992. Baimbridge K.G. , Celio M.R., Rogers, J.H. Calcium-binding proteins in the nervous system. Trends in Neurosciences 15: 303-308, 1992. Baimbridge K G . Miller JJ. Immunohistochemical localization of calcium-binding protein in the cerebellum, hippocampal formation and olfactory bulb of the rat. Brain Research. 245:223-9, 1982. Baimbridge K G . Miller JJ. Hippocampal calcium-binding protein during commissural kindling-induced epileptogenesis: progressive decline and effects of anticonvulsants. Brain Research. 324:85-90, 1984. Baimbridge K .G . , Miller J.J., Parkes, C O . Calcium-binding protein distribution in the rat brain. Brain Research 239: 519-525, 1982. Baimbridge K G . Mody I. Miller JJ. Reduction of rat hippocampal calcium-binding protein following commissural, amygdala, septal, perforant path, and olfactory bulb kindling. Epilepsia. 26:460-5, 1985. Baimbridge K G , Parkes CO. Vitamin D dependent calcium-binding protein in the chick brain. Cell Calcium. 2:65-67,1981. Baimbridge K G . Peet MJ . McLennan H. Church J. Bursting response to current-evoked depolarization in rat CA1 pyramidal neurons is correlated with lucifer yellow dye coupling but not with the presence of calbindin-D28k. Synapse. 7:269-77, 1991. 136 Bellido T, Huening M , Raval-Pandya M , Manolagas SC, Christakos S. Calbindin-D28K is expressed in osteoblastic cells and suppresses their apoptosis by inhibiting caspase-3 activity. Journal of Biological Chemistry, [electronic publication ahead of print] Jun 1 2000. Berggard T, Silow M , Thulin E, Linse S. Ca 2 + - and H+-Dependent Conformational Changes of Calbindin D(28k). Biochemistry. 39:6864-6873, 2000. Berridge MJ . Capacitative Ca2+ entry. Biochemical Journal. 312 :1-11, 1995. Berridge MJ. The A M and F M of calcium signalling. Nature. 386:759-60, 1997. Berridge MJ. Neuronal calcium signaling. Neuron. 21:13-26, 1998. Bezprozvanny I. Watras J. Ehrlich BE. Bell-shaped calcium-response curves of Ins(l,4,5)P3- and calcium-gated channels from endoplasmic reticulum of cerebellum. Nature. 351:751-4, 1991. Bischof G. Serwold TF. Machen TE. Does nitric oxide regulate capacitative Ca influx in H E K 293 cells?. Cell Calcium. 21:135-42, 1997. Bliss TV. Collingridge GL. A synaptic model of memory: long-term potentiation in the hippocampus. Nature. 361:31-9, 1993. Boeckman FA. Aizenman E. Pharmacological properties of acquired excitotoxicity in Chinese hamster ovary cells transfected with N-methyl-D-aspartate receptor subunits. Journal of Pharmacology & Experimental Therapeutics. 279:515-23, 1996. 137 Bonfoco E. Krainc D. Ankarcrona M . Nicotera P. Lipton SA. Apoptosis and necrosis: two distinct events induced, respectively, by mild and intense insults with N-methyl-D-aspartate or nitric oxide/superoxide in cortical cell cultures. Proceedings of the National Academy of Sciences of the United States of America. 92:7162-6, 1995. Bootman M D . Hormone-evoked subcellular Ca2+ signals in HeLa cells. Cell Calcium. 20:97-104, 1996. Bootman M D . Berridge MJ. Subcellular Ca2+ signals underlying waves and graded responses in HeLa cells. Current Biology. 6:855-65, 1996. Bootman M . Niggli E. Berridge M . Lipp P. Imaging the hierarchical Ca2+ signalling system in HeLa cells. Journal of Physiology. 499:307-14, 1997. Bootman M D . Young K W . Young JM. Moreton RJ3. Berridge MJ. Extracellular calcium concentration controls the frequency of intracellular calcium spiking independently of inositol 1,4,5-trisphosphate production in HeLa cells. Biochemical Journal. 314:347-54, 1996. Bredderman PJ. Wasserman RH. Chemical composition, affinity for calcium, and some related properties of the vitamin D dependent calcium-binding protein. Biochemistry. 13:1687-94, 1974. Buchan A M . Baimbridge K G . Distribution and co-localization of calbindin D28k with VIP and neuropeptide Y but not somatostatin, galanin and substance P in the enteric nervous system of the rat. Peptides. 9:333-8, 1988. Burke RE. Baimbridge K G . Relative loss of the striatal striosome compartment, defined by calbindin-D28k immunostaining, following developmental hypoxic-ischemic injury. Neuroscience. 56:305-315, 1993. 138 Celio MR. Calbindin D-28k and .paralbumin in the rat nervous system. Neuroscience. 35:375-475, 1990. Celio MR. Norman A W . Nucleus basalis Meynert neurons contain the vitamin D-induced calcium-binding protein (Calbindin-D 28k). Anatomy & Embryology. 173:143-8, 1985. Chard P.S., Bleakman D., Christakos S., Fullmer C.S., Miller, R.J. Calcium buffering properties of calbindin-D28k and parvalbumin in rat sensory neurones. Journal of Physiology 472: 341-57, 1993. Chard PS. Jordan J. Marcuccilli CJ. Miller RJ. Leiden JM. Roos RP. Ghadge GD. Regulation of excitatory transmission at hippocampal synapses by calbindin D28k. Proceedings of the National Academy of Sciences of the United States of America. 92:5144-8, 1995. Cheung WT., Richards DE., Rogers JH. Calcium binding by chick calretinin and rat calbindin D28k synthesised in bacteria. European Journal of Biochemistry 215: 401-10, 1993. Choi, D.W. Ionic dependence of glutamate neurotoxicity. Journal of Neuroscience 7:369-379, 1987. Choi DW. Calcium-mediated neurotoxicity: relationship to specific channel types and role in ischemic damage. Trends in Neurosciences. 11:465-9, 1988. Choi DW. Excitotoxic cell death. Journal of Neurobiology. 23:1261-76, 1992. Choi DW. Calcium and excitotoxic neuronal injury. Annals of the New York Academy of Sciences. 747:162-71, 1994. 139 Choi DW. Calcium: still center-stage in hypoxic-ischemic neuronal death. Trends in Neurosciences. 18:58-60, 1995. Christakos S. Gabrielides C. Rhoten WB. Vitamin D-dependent calcium binding proteins: chemistry, distribution, functional considerations, and molecular biology. Endocrine Reviews. 10:3-26, 1989. Cik M . Chazot PL. Stephenson FA. Optimal expression of cloned N M D A R 1 / N M D A R 2 A heteromeric glutamate receptors: a biochemical characterization. Biochemical Journal. 296:877-83, 1993. Clapham DE. Calcium signaling. Cell. 80:259-68, 1995. Crompton M . Costi A . Hayat L. Evidence for the presence of a reversible Ca2+-dependent pore activated by oxidative stress in heart mitochondria. Biochemical Journal. 245:915-8, 1987. Davenport AP. Augood SJ. Lawson DE. Emson PC. The use of quantitative immunocytochemistry (QICC) to measure calbindin D28k-like immunoreactivity in the rat brain. Cellular & Molecular Biology. 36:1-11, 1990. Diop A G . Lesort M . Esclaire F. Dumas M . Hugon J. Calbindin D28K-containing neurons, and not HSP70-expressing neurons, are more resistant to HIV-1 envelope (gpl20) toxicity in cortical cell cultures. Journal of Neuroscience Research. 42:252-258, 1995. Doble A . The role of excitotoxicity in neurodegenerative disease: implications for therapy. Pharmacology & Therapeutics. 81:163-221, 1999. 140 Dolmetsch RE. Lewis RS. Goodnow CC. Healy JI. Differential activation of transcription factors induced by Ca2+ response amplitude and duration. Nature. 386:855-8, 1997. Dolmetsch RE. X u K. Lewis RS. Calcium oscillations increase the efficiency and specificity of gene expression. Nature. 392:933-6, 1998-Doutrelant O. Poulain P. Carette B. Comparative distribution of calbindin and Met-enkephalin immunoreactivities in the guinea-pig lateral septum, with reference to electrophysiologically characterized neurons in the mediolateral part. Brain Research. 615:335-41, 1993. Dowd DR. MacDonald PN. Komm BS. Haussler MR. Miesfeld RL.. Stable expression of the calbindin-D28K complementary D N A interferes with the apoptotic pathway in lymphocytes. Molecular Endocrinology. 6:1843-8, 1992. Dubinsky JM. Effects of calcium chelators on intracellular calcium and excitotoxicity. Neuroscience Letters. 150:129-32, 1993. Dubinsky JM. Rothman SM. Intracellular calcium concentrations during "chemical hypoxia" and excitotoxic neuronal injury. Journal of Neuroscience. 11:2545-51, 1991. Duchen MR. Contributions of mitochondria to animal physiology: from homeostatic sensor to calcium signalling and cell death. Journal of Physiology. 516:1-17, 1999. Earnshaw WC. Apoptosis. A cellular poison cupboard. Nature. 397:387, 389, 1999. Eimerl S. Schramm M . The quantity of calcium that appears to induce neuronal death. Journal of Neurochemistry. 62:1223-6, 1994. 141 Ellis-Davies G.C.R., Kaplan, J.H. Nitrophenyl-EGTA, a photolabile chelator that selectively binds Ca2+ with high affinity and releases it rapidly upon photolysis. Proceedings of the National Academy of Sciences of the United States of America. 93: 187-191, 1994. Feher JJ. Facilitated calcium diffusion by intestinal calcium-binding protein. American Journal of Physiology. 244:C303-7, 1983. Feher J.J., Fullmer C.S., Wasserman, R.H. Role of facilitated diffusion of calcium by calbindin in intestinal calcium absorption. American Journal of Physiology 262: C517-26, 1992. Fierro L . Llano I. High endogenous calcium buffering in Purkinje cells from rat cerebellar slices. Journal of Physiology.496:617-25, 1996. Finch EA. Turner TJ. Goldin SM. Calcium as a coagonist of inositol 1,4,5-trisphosphate-induced calcium release. Science. 252:443-6, 1991. Fleet A . Ellis-Davies G. Bolsover S. Calcium buffering capacity of neuronal cell cytosol measured by flash photolysis of calcium buffer NP-EGTA. Biochemical & Biophysical Research Communications. 250:786-90, 1998. Freund TF. Buzsaki G. Leon A. Baimbridge K G . Somogyi P. Relationship of neuronal vulnerability and calcium binding protein immunoreactivity in ischemia. Expt. Brain Research. 83:55-66, 1990. Friauf E. Transient appearance of calbindin-D28k-positive neurons in the superior olivary complex of developing rats. Journal of Comparative Neurology. 334:59-74, 1993. 142 Gerfen CR. Baimbridge K G . Miller JJ. The neostriatal mosaic: compartmental distribution of calcium-binding protein and parvalbumin in the basal ganglia of the rat and monkey. Proceedings of the National Academy of Sciences of the United States of America. 82:8780-4, 1985. Gerfen CR. Baimbridge K G . Thibault J. The neostriatal mosaic: III. Biochemical and developmental dissociation of patch-matrix mesostriatal systems. Journal of Neuroscience. 7:3935-44, 1987. German DC. Manaye KF . Sonsalla PK. Brooks B A . Midbrain dopaminergic cell loss in Parkinson's disease and MPTP-induced parkinsonism: sparing of calbindin-D28k-containing cells. Annals of the New York Academy of Sciences. 648:42-62, 1992. Ghosh A. Ginty DD. Bading H . Greenberg M E . Calcium regulation of gene expression in neuronal cells. Journal of Neurobiology. 25:294-303, 1994. Goodman JH. Wasterlain CG. Massarweh WF. Dean E. Sollas A L . Sloviter RS. Calbindin-D28k immunoreactivity and selective vulnerability to ischemia in the dentate gyrus of the developing rat. Brain Research. 606:309-314, 1993. Gotoh H . Takahashi A. Mechanical stimuli induce intracellular calcium response in a subpopulation of cultured rat sensory neurons. Neuroscience. 92:1323-9, 1999. Grant ER. Bacskai BJ. Pleasure DE. Pritchett DB. Gallagher MJ. Kendrick SJ. Kricka LJ . Lynch DR. N-methyl-D-aspartate receptors expressed in a nonneuronal cell line mediate subunit-specific increases in free intracellular calcium. Journal of Biological Chemistry. 272:647-56, 1997. Green DR.. Reed JC. Mitochondria and apoptosis. Science. 281:1309-1312, 1998. 143 Gross M D . Nelsestuen GL. Kumar R. Observations on the binding of lanthanides and calcium to vitamin D-dependent chick intestinal calcium-binding protein. Implications regarding calcium-binding protein function. Journal of Biological Chemistry. 262:6539-45, 1987. Grynkiewicz G. Poenie M . Tsien R Y . A new generation of Ca2+ indicators with greatly improved fluorescence properties. Journal of Biological Chemistry. 260:3440-50, 1985. Gu X . Spitzer NC. Distinct aspects of neuronal differentiation encoded by frequency of spontaneous Ca2+ transients. Nature. 375:784-7, 1995. Gwag BJ. Canzoniero L M . Sensi SL. Demaro JA. Koh JY. Goldberg MP. Jacquin M . Choi DW. Ca2+ ionophores can induce either apoptosis or necrosis in cultured cortical neurons. Neuroscience. 90:1339-48, 1999. Hall JD. Betarbet S. Jaramillo F. Endogenous buffers limit the spread of free calcium in hair cells. Biophysical Journal. 73:1243-52, 1997. Harafuji H . Ogawa Y . Re-examination of the apparent binding constant of ethylene glycol bis(beta-aminoethyl ether)-N,N,N',N'-tetraacetic acid with calcium around neutral pH. Journal of Biochemistry. 87:1305-12, 1980. Hartley D M . Kurth M G . Bjerkness L. Weiss JH. Choi DW. Glutamate receptor-induced 45Ca2+ accumulation in cortical cell culture correlates with subsequent neuronal degeneration. Journal of Neuroscience. 13:1993-2000, 1993. Heizmann CW. Braun K. Changes in Ca2 +-binding proteins in human neurodegenerative disorders. Trends in Neurosciences. 15:259-64, 1992. 144 Hersham K. M . , Flemming W. W., and Taylor D. A . A quantitative method for assessing protein abundance using enhanced chemiluminescence. Biotechniques 15: 790-795, 1993. Ho B K . Alexianu M E . Colom L V . Mohamed A H . Serrano F Appel SH. Expression of calbindin-D28K in motoneuron hybrid cells after retroviral infection with calbindin-D28K cDNA prevents amyotrophic lateral sclerosis IgG-mediated cytotoxicity. Proceedings of the National Academy of Sciences of the United States of America. 93:6796-6801, 1996. Hollmann M . Heinemann S. Cloned glutamate receptors. Annual Review of Neuroscience. 17:31-108, 1994. Hunter DR. Haworth RA. The Ca2+-induced membrane transition in mitochondria. III. Transitional Ca2+ release. Archives of Biochemistry & Biophysics. 195:468-77, 1979. Hyrc K. Handran SD. Rothman SM. Goldberg MP. Ionized intracellular calcium concentration predicts excitotoxic neuronal death: observations with low-affinity fluorescent calcium indicators. Journal of Neuroscience. 17:6669-77, 1997. Iacopino A M . Christakos S. Specific reduction of calcium-binding protein (28-kilodalton calbindin-D) gene expression in aging and neurodegenerative diseases. Proceedings of the National Academy of Sciences of the United States of America. 87:4078-82, 1990. Iacopino A. , Christakos S., German D., Sonsalla P.K., Altar C A . Calbindin-D28K-containing neurons in animal models of neurodegeneration: possible protection from excitotoxicity. Molecular Brain Resesearch. 13: 251-261, 1992. 145 Ichimiya Y. Emson PC. Mountjoy CQ. Lawson DE. Heizmann CW. Loss of calbindin-28K immunoreactive neurones from the cortex in Alzheimer-type dementia. Brain Research. 475:156-9, 1988. lino M . Endo M . Ca -dependent immediate feedback control of inositol 1,4,5-triphosphate-induced Ca2+ release. Nature. 360:76-8, 1992. Jande S.S., Maler L. , Lawson, D.E. Immunohistochemical mapping of vitamin D-dependent calcium-binding protein in brain. Nature. 294: 765-767, 1981. Johnson JW. Ascher P. Glycine potentiates the N M D A response in cultured mouse brain neurons. Nature. 325:529-31, 1987. Katz B. Miledi R. Ionic requirements of synaptic transmitter release. Nature. 215:651, 1967. Kawaguchi Y . Katsumaru H. Kosaka T. Heizmann CW. Hama K. Fast spiking cells in rat hippocampus (CA1 region) contain the calcium-binding protein parvalbumin. Brain Research. 416:369-74, 1987. Kawaguchi Y . Kubota Y . Correlation of physiological subgroupings of nonpyramidal cells with parvalbumin- and calbindinD28k-immunoreactive neurons in layer V of rat frontal cortex. Journal of Neurophysiology. 70:387-96, 1993. Klapstein GJ. Vietla S. Lieberman DN. Gray PA. Airaksinen MS. Thoenen H . Meyer M . Mody I. Calbindin-D28k fails to protect hippocampal neurons against ischemia in spite of its cytoplasmic calcium buffering properties: evidence from calbindin-D28k knockout mice. Neuroscience. 85:361-73, 1998. 146 Kobayashi K. Emson PC. Mountjoy CQ. Thornton SN. Lawson DE. Mann D M . Cerebral cortical calbindin D28K and parvalbumin neurones in Down's syndrome. Neuroscience Letters. 113:17-22, 1990. Kohr G. Lambert CE. Mody I. Calbindin-D28K (CaBP) levels and calcium currents in acutely dissociated epileptic neurons. Experimental Brain Research. 85:543-51, 1991. Lally G. Faull RL . Waldvogel HJ. Ferrari S. Emson PC. Calcium homeostasis in ageing: studies on the calcium binding protein calbindin D28K. Journal of Neural Transmission (Budapest). 104:1107-12, 1997. Lavoie B. Parent A. Dopaminergic neurons expressing calbindin in normal and parkinsonian monkeys. Neuroreport. 2:601-4, 1991. Legendre R., Rosendund C , Westbrook, G.L. Inactivation of N M D A channels in cultured hippocampal neurons by intracellular calcium. Journal of Neuroscience. 13: 674-684. 1993. Lemasters JJ. Nieminen A L . Qian T. Trost L C . Elmore SP. Nishimura Y. Crowe R A . Cascio WE. Bradham CA. Brenner DA. Herman B. The mitochondrial permeability transition in cell death: a common mechanism in necrosis, apoptosis and autophagy. Biochimica et Biophysica Acta. 1366:177-96, 1998. L i Z. Decavel C. Hatton GI. Calbindin-D28k: role in determining intrinsically generated firing patterns in rat supraoptic neurones. Journal of Physiology. 488:601-8, 1995. L i W. Llopis J. Whitney M . Zlokarnik G. Tsien R Y . Cell-permeant caged InsP3 ester shows that Ca2+ spike frequency can optimize gene expression. Nature. 392:936-41, 1998. 147 Lledo P M . Somasundaram B. Morton AJ . Emson PC. Mason WT. Stable transfection of calbindin-D28k into the GH3 cell line alters calcium currents and intracellular calcium homeostasis. Neuron. 9:943-54, 1992. Lu Y M . Yin HZ. Chiang J. Weiss JH. Ca2+-permeable AMPA/kainate and N M D A channels: high rate of Ca influx underlies potent induction of injury. Journal of Neuroscience. 16:5457-65, 1996. MacDermott A B . Mayer M L . Westbrook GL. Smith SJ. Barker JL. NMDA-receptor activation increases cytoplasmic calcium concentration in cultured spinal cord neurones. Nature. 321:519-22, 1986. Magloczky Z. Halasz P. Vajda J. Czirjak S. Freund TF. Loss of Calbindin-D28K immunoreactivity from dentate granule cells in human temporal lobe epilepsy. Neuroscience. 76:377-85, 1997. Malenka RC. Nicoll RA. Long-term potentiation—a decade of progress?. Science. 285:1870-4, 1999. 9+ Manev H . Favaron M . Guidotti A . Costa E. Delayed increase of Ca influx elicited by glutamate: role in neuronal death. Molecular Neurobiology. 36:106-112, 1989. Mattson M.P., Rychlik B., Chu C , Christakos, S. Evidence for calcium-reducing and excito-protective roles for the calcium-binding protein calbindin-D28k in cultured hippocampal neurons. Neuron. 6: 41-51, 1991. McMahon A. Wong BS. Iacopino A M . Ng M C . Chi S. German DC. Calbindin-D28k buffers intracellular Ca2+ and promotes resistance to degeneration in PC 12 cells. Brain Research. Molecular Brain Research. 54:56-63, 1998. McPherson PS. Campbell KP. The ryanodine receptor/Ca2+ release channel. Journal of Biological Chemistry. 268:13765-8, 1993. 148 Meier TJ. Ho D Y . Park TS. Sapolsky R M . Gene transfer of calbindin D28k cDNA via herpes simplex virus amplicon vector decreases cytoplasmic calcium ion response and enhances neuronal survival following glutamatergic challenge but not following cyanide. Journal of Neurochemistry. 71:1013-23, 1998. Meier TJ. Ho D Y . Sapolski R M . Increased expression of calbindin D-28k via herpes simplex virus amplicon vector decreases calcium ion mobilization and enhances neuronal survival after hyperglycemic challenge. Journal of Neurochemistry. 69:1039-1047, 1997. Michaels RL . Rothman SM. Glutamate neurotoxicity in vitro; agonist pharmacology and intracellular calcium concentration. Journal of Neuroscience. 10:283-292, 1990. Miller RJ. Mitochondria - the Kraken wakes:. Trends in Neurosciences. 21:95-7, 1998. Missiaen L. Parys JB. Bootman M D . De Smedt H. Droogmans G. Casteels R. Normal Ca2+ signalling in glutathione-depleted and dithiothreitol-treated HeLa cells. Pflugers Archiv - European Journal of Physiology. 423:480-4, 1993. Mockel V . Fischer G. Vulnerability to excitotoxic stimuli of cultured rat hippocampal neurons containing the calcium-binding proteins calretinin and calbindin D28K. Brain Research. 648:109-120, 1994. Mody, I. Personal Communication. Unpublished results of experiments employing recombinant protein generated by this author as described in materials and methods. Mody I. Baimbridge K G . Miller JJ. Distribution of Calbindin-D28K 1 (CaBP) in the cerebral cortex and hippocampus of the epileptic (El) mouse. Epilepsy Research. 1:46-52, 1987. 149 Monyer H. Sprengel R. Schoepfer R. Herb A. Higuchi M . Lomeli H. Burnashev N . Sakmann B. Seeburg PH. Heteromeric N M D A receptors: molecular and functional distinction of subtypes. Science. 256:1217-21, 1992. Moriyoshi K. Masu M . Ishii T. Shigemoto R. Mizuno N . Nakanishi S. Molecular cloning and characterization of the rat N M D A receptor. Nature. 354:31-7, 1991. Mudrick L A . Baimbridge K G . Long-term structural changes in the rat hippocampal formation following cerebral ischemia. Brain Research. 493:179-84, 1989. Neher E. Augustine GJ. Calcium gradients and buffers in bovine chromaffin cells. Journal of Physiology. 450:273-301, 1992. Nicholls DG. Budd SL. Neuronal excitotoxicity: the role of mitochondria. Biofactors. 8:287-99, 1998. Nicholls DG.. Budd SL. Mitochondria and neuronal survival. Physiological Reviews. 80:315-60, 2000. Nowak L. Bregestovski P. Ascher P. Herbet A. Prochiantz A . Magnesium gates glutamate-activated channels in mouse central neurones. Nature. 307:462-5, 1984. Olney JW. Ho OL. Rhee V . Cytotoxic effects of acidic and sulphur containing amino acids on the infant mouse central nervous system. Experimental Brain Research. 14:61-76, 1971. Olney J. Price W. Samson MT. Labruyere J. The role of specific ions in glutamate excitotoxicity. Neuroscience Letters. 65:65-71, 1986. Palecek J. Lips M B . Keller B U . Calcium dynamics and buffering in motoneurones of the mouse spinal cord. Journal of Physiology. 520 Pt 2:485-502, 1999. 150 Parmentier M . Lawson DE. Vassart G. Human 27-kDa calbindin complementary D N A sequence. Evolutionary and functional implications. European Journal of Biochemistry. 170:207-15, 1987. Parmentier M . Structure of the human cDNAs and genes coding for calbindin D28K and calretinin. Advances in Experimental Medicine & Biology. 269:27-34, 1990. Pasteels JL. Pochet R. Surardt L. Hubeau C. Chirnoaga M . Parmentier M . Lawson DE. Ultrastructural localization of brain 'vitamin D-dependent' calcium binding proteins. Brain Research. 384:294-303, 1986. Peng TL Greenamyre JT. Privileged access to mitochondria of calcium influx through N-methyl-D-aspartate receptors. Molecular Pharmacology. 53:974-80, 1998. Peterson GL. Determination of total protein. Methods in Enzymology. 91:95-119, 1983. Pethig R. Kuhn M . Payne R. Adler E. Chen TH. Jaffe LF. On the dissociation constants of BAPTA-type calcium buffers. Cell Calcium. 10:491-8, 1989. Petralia RS. Yokotani N . Wenthold RJ. Light and electron microscope distribution of the N M D A receptor subunit N M D A R 1 in the rat nervous system using a selective anti-peptide antibody. Journal of Neuroscience. 14:667-96, 1994. Phillips RG. Meier TJ. Giuli L C . McLaughlin JR. Ho D Y . Sapolsky R M . Calbindin D28K gene transfer via herpes simplex virus amplicon vector decreases hippocampal damage in vivo following neurotoxic insults. Journal of Neurochemistry. 73:1200-5, 1999. Price C.J., Rintoul G.L., Baimbridge K.G. , Raymond, L . A . Inhibition of calcium-dependent NMDAreceptor current rundown by calbindin-D28k. Journal of Neurochemistry. 72: 634-642, 1999. 151 Putney JW Jr. McKay RR. Capacitative Ca2+ entry channels. Bioessays. 21:38-46, 1999. Randall RD. Thayer SA. Glutamate-induced C a 2 + transient triggers delayed C a 2 + overload and neurotoxicity in rat hippocampal neurons. Journal of Neuroscience. 12:1882-95, 1992. Rasmussen H . Gustin M C . Some aspects of the hormonal control of cellular calcium metabolism. Annals of the New York Academy of Sciences. 307:391-401, 1978. Raymond L A . Moshaver A. Tingley WG. Huganir RL. Glutamate receptor ion channel properties predict vulnerability to cytotoxicity in a transfected nonneuronal cell line. Molecular & Cellular Neurosciences. 7:102-15, 1996. Richardson A . Taylor CW. Effects of C a 2 + chelators on purified inositol 1,4,5-trisphosphate (InsP3) receptors and InsP3-stimulated C a 2 + mobilization Journal of Biological Chemistry. 268:11528-1153, 1993. Roberts W M . Spatial calcium buffering in saccular hair cells. Nature. 363:74-6, 1993. Roberts W M . Localization of calcium signals by a mobile calcium buffer in frog saccular hair cells. Journal of Neuroscience. 14:3246-62, 1994. Rogers JH. Calretinin: a gene for a novel calcium-binding protein expressed principally in neurons. Journal of Cell Biology. 105:1343-53, 1987. Rothman SM. The neurotoxicity of excitatory amino acids is produced by passive chloride influx. Journal of Neuroscience 5:1483-1489, 1985. Rothman S M . Thurston JH. Hauhart RE. Delayed neurotoxicity of excitatory amino acids in vitro. Neuroscience. 22:471-480, 1987. 152 Roy J. Minotti S. Dong L. Figlewicz DA. Durham HD. Glutamate potentiates the toxicity of mutant Cu/Zn-superoxide dismutase in motor neurons by postsynaptic calcium-dependent mechanisms. Journal of Neuroscience. 18:9673-9684, 1998. Sambrook, Joseph. Molecular cloning : a laboratory manual / J. Sambrook, E.F. Fritsch, T. Maniatis. Cold Spring Harbor, N . Y . : Cold Spring Harbor Laboratory, 1989. Sanghera M K . Manaye KF . Liang CL. Lacopino A M . Bannon MJ. German DC. Low dopamine transporter mRNA levels in midbrain regions containing calbindin. Neuroreport. 5:1641-4, 1994. Schanne FA. Kane A B . Young EE. Farber JL. Calcium dependence of toxic cell death: a final common pathway. Science. 206:700-2, 1979. Scharfman HE. Schwartzkroin PA. Protection of dentate hilar cells from prolonged stimulation by intracellular calcium chelation. Science. 246:257-60, 1989. Schinder A F . Olson EC. Spitzer NC. Montal M . Mitochondrial dysfunction is a primary event in glutamate neurotoxicity. Journal of Neuroscience. 16:6125-6133, 1996. Seto-Ohshima A . Emson PC. Lawson E. Mountjoy CQ. Carrasco L H . Loss of matrix calcium-binding protein-containing neurons in Huntington's disease. Lancet. 1:1252-5, 1988. Sidky A.O. , Baimbridge K . G . Calcium homeostatic mechanisms operating in culture post-natal rat hippocampal neurones following flash photolysis of NP-EGTA. Journal of Physiology (Lond). 504: 579-590, 1997. Simon S M . Llinas RR. Compartmentalization of the submembrane calcium activity during calcium influx and its significance in transmitter release. Biophysical Journal. 48:485-98, 1985. 153 Sloviter RS. Calcium-binding protein (calbindin-D28k) and parvalbumin immunocytochemistry: localization in the rat hippocampus with specific reference to the selective vulnerability of hippocampal neurons to seizure activity. Journal of Comparative Neurology. 280:183-96, 1989. Sloviter RS. Sollas A L . Barbaro N M . Laxer K D . Calcium-binding protein (calbindin-D28K) and parvalbumin immunocytochemistry in the normal and epileptic human hippocampus. Journal of Comparative Neurology. 308:381-96,1991. Stout A K . Raphael H M . Kanterewicz BI. Klann E. Reynolds IJ. Glutamate-induced neuron death requires mitochondrial calcium uptake. Nature Neuroscience. 1:366-73, 1998. Stout A K . Reynolds IJ. High-affinity calcium indicators underestimate increases in intracellular calcium concentrations associated with excitotoxic glutamate stimulations. Neuroscience. 89:91-100, 1999. Thastrup O. Cullen PJ. Drobak B K . Hanley MR. Dawson AP. Thapsigargin, a tumor promoter, discharges intracellular Ca2+ stores by specific inhibition of the endoplasmic reticulum Ca 2 +-ATPase. Proceedings of the National Academy of Sciences of the United States of America. 87:2466-70, 1990. Thayer SA. Miller RJ. Regulation of the intracellular free calcium concentration in single rat dorsal root ganglion neurones in vitro. Journal of Physiology. 425:85-115, 1990. Thomas P. Surprenant A . Aimers W. Cytosolic Ca 2 + , exocytosis, and endocytosis in single melanotrophs of the rat pituitary. Neuron. 5:723-33, 1990. Thorn P. C a 2 + influx during agonist and Ins(2,4,5)P3-evoked Ca2+ oscillations in HeLa epithelial cells. Journal of Physiology. 48:275-81, 1995. 154 Tong G., Jahr C.E. Regulation of glycine-insensitive desensitization of the N M D A receptor in outside-out patches. Journal of Neurophysiology. 72:754-61, 1994. Tse A. Tse FW. Hille B. Calcium homeostasis in identified rat gonadotrophs. Journal of Physiology. 477:511-25, 1994. Tsien R Y . New calcium indicators and buffers with high selectivity against magnesium and protons: design, synthesis, and properties of prototype structures. Biochemistry. 19:2396-404, 1980. Tymianski M . Charlton MP. Carlen PL. Tator CH. Source specificity of early calcium neurotoxicity in cultured embryonic spinal neurons. Journal of Neuroscience. 13:2085-104, 1993a. Tymianski M . Wallace M C . Spigelman I. Uno M . Carlen PL. Tator C H . Charlton MP. Cell-permeant Ca2+ chelators reduce early excitotoxic and ischemic neuronal injury in vitro and in vivo. Neuron. 11:221-35, 1993b. Varghese S, Lee S, Huang Y C , Christakos S. Analysis of rat vitamin D-dependent calbindin-D28k gene expression. Journal of Biological Chemistry. 263:9776-84, 1988. Vergun O. Keelan J. Khodorov B.I. Duchen M.R. Glutamate-induced mitochondrial depolarisation and perturbation of calcium homeostasis in cultured rat hippocampal neurones. Journal of Physiology. 519 Pt 2:451-66, 1999. Wang SS. Thompson SH. Local positive feedback by calcium in the propagation of intracellular calcium waves. Biophysical Journal. 69:1683-97, 1995. Wang Z. Tymianski M . Jones OT. Nedergaard M . Impact of cytoplasmic calcium buffering on the spatial and temporal characteristics of intercellular calcium signals in astrocytes. Journal of Neuroscience. 17:7359-71, 1997. 155 Wasserman RH, Fullmer CS. Vitamin D-induced calcium binding proteins. In: Calcium and cell function, Vol 2. (Cheung WY, Ed.), pp 175-216. 1982. Wassermann RH, Taylor A N . Vitamin D3-induced calcium binding protein in chick intestinal mucosa. Science 152:791-793 1966. Watanabe M . Inoue Y. Sakimura K. Mishina M . Distinct distributions of five N-methyl-D-aspartate receptor channel subunit mRNAs in the forebrain. Journal of Comparative Neurology. 338:377-90, 1993. Werth JL. Thayer SA. Mitochondria buffer physiological calcium loads in cultured rat dorsal root ganglion neurons. Journal of Neuroscience. 14:348-56, 1994. Williams RJ. Calcium and cell steady states. Advances in Experimental Medicine & Biology. 269:7-16, 1990. Wu K D . Lee WS. Wey J. Bungard D. Lytton J. Localization and quantification of endoplasmic reticulum Ca 2 +-ATPase isoform transcripts. American Journal of Physiology. 269:C775-84, 1995. X u T. Naraghi M . Kang H . Neher E. Kinetic studies of C a 2 + binding and C a 2 + clearance in the cytosol of adrenal chromaffin cells. Biophysical Journal. 73:532-45, 1997. Yamada T. McGeer PL. Baimbridge K G . McGeer EG. Relative sparing in Parkinson's disease of substantia nigra dopamine neurons containing calbindin-D28K. Brain Research. 526:303-7, 1990. Zhou Z. Neher E. Mobile and immobile calcium buffers in bovine adrenal chromaffin cells. Journal of Physiology. 469:245-73, 1993. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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


Related Items