Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Pre- and postsynaptic actions of pentobarbital on corticothalamic transmission Ran, Israeli 2005

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

Item Metadata

Download

Media
831-ubc_2005-105585.pdf [ 12.78MB ]
Metadata
JSON: 831-1.0092402.json
JSON-LD: 831-1.0092402-ld.json
RDF/XML (Pretty): 831-1.0092402-rdf.xml
RDF/JSON: 831-1.0092402-rdf.json
Turtle: 831-1.0092402-turtle.txt
N-Triples: 831-1.0092402-rdf-ntriples.txt
Original Record: 831-1.0092402-source.json
Full Text
831-1.0092402-fulltext.txt
Citation
831-1.0092402.ris

Full Text

P R E - A N D P O S T S Y N A P T I C A C T I O N S OF P E N T O B A R B I T A L O N C O R T I C O T H A L A M I C T R A N S M I S S I O N by Israeli Ran B . S c , Tel A v i v University, 1995 M.Sc . , University of Calgary, 1998 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T F O R T H E D E G R E E O F D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E STUDIES Neuroscience T H E U N I V E R S I T Y OF BRITISH C O L U M B I A July 2005 © Israeli Ran, 2005 Abstract This thesis examined the pre- and postsynaptic actions of an anesthetic barbiturate, pentobarbital, on neurons of the corticothalamocortical system in vitro. The in vivo system mediates conscious and sleep states. The thesis focuses on pentobarbital actions that induce network oscillations, and modify responses of single thalamocortical neurons to corticothalamic stimulus trains. The thesis addressed the following: (1) does pentobarbital induce oscillations in thalamic slices? (2) what receptors contribute to oscillations? (3) how does pentobarbital interact with modulators of excitability? (4) what are pentobarbital effects on post- and presynaptic parameters of glutamatergic transmission during short-term depression (STD)? (5) how do the effects o f pentobarbital on S T D compare with selective action potential blockade? (6) given the well-known actions of pentobarbital on metabolism, do its effects on S T D mimic glucose deprivation? Pentobarbital at a subanesthetic concentration induced 1-15 H z oscillations, requiring glutamatergic excitation, but not elevated temperature or low extracellular [ M g 2 + ] . Glycine receptors mediated oscillations in ventrobasal nuclei, disconnected from nucleus reticularis thalami (nRT). y-aminobutyrate ( G A B A ) receptors mediated oscillations in isolated nRT. B y acting on N-methyl-D-aspartate ( N M D A ) receptors, spermine modulated membrane rectification, firing threshold, and decay of excitatory postsynaptic potentials (EPSPs). These interactions occurred at the polyamine site on N M D A receptors. Pentobarbital enhanced S T D of excitatory postsynaptic currents (EPSCs) by decreasing quantal size. These use-dependent effects persisted during blockade of desensitization and saturation of glutamate receptors and hence, likely were presynaptic. Pentobarbital decreased apparent quantal size and amplitude in the post-stimulus train, evoked miniature EPSCs (minEPSCs) but not ongoing, pre-train minEPSCs, reaffirming a presynaptic action. Pentobarbital eliminated E P S C facilitation early in a train, due to high extracellular [K + ] ( [K + ] e ) . Partial blockade of action potentials by tetrodotoxin reduced the apparent quantal size and evoked minEPSC size, without effect on pre-stimulation minEPSC. Like pentobarbital, glucose deprivation reduced quantal size and rundown o f quantal contents. Glucose deprivation abolished S T D and intra-train, post-gap jump in E P S C amplitude. In summary, this thesis describes several new types of synaptic modulation by pentobarbital that complement known postsynaptic mechanisms of anesthesia. The analysis techniques provide a new approach for examining the pre- and postsynaptic drug effects on transmission in the brain. Table of Contents Abstract — i i Table o f Contents iv List o f Tables v i i i List o f Figures x Abbreviations x i i i Acknowledgements xv Chapter 1. Introduction 1 1.1. Scope of thesis 1 1.2. Background 2 1.2.1. E E G activity and brain oscillations 2 1.2.2. Receptor mediation of sleep-like oscillations in the C T C system 3 1.2.3. Non-synaptic mechanisms of sleep-like oscillations 4 1.2.4. Synaptic composition of the C T C system 5 1.2.5. G A B A receptors modulate the frequency of sleep-like oscillations 6 1.2.6. Glycine receptors contribute to thalamic oscillogenesis 7 1.2.7. Barbiturate anesthetics induce sleep-like oscillations 7 1.2.8. Polyamine modulation of barbiturate action 8 1.2.9. Polyamine enhancement of corticothalamic transmission: Relevance to thalamic oscillogenesis 8 1.2.10. Pre- and postsynaptic mechanisms of short-term depression (STD) 11 1.2.10.1. Presynaptic mechanisms in S T D 11 1.2.10.1.1. Depletion of quantal store 12 1.2.10.1.2. Reduction in transmitter content • 15 1.2.10.1.3. Modification of the presynaptic action potential 15 1.2.10.1.4. Inactivation of C a 2 + channels • 15 1.2.10.1.5. Interference with endocytosis 16 1.2.10.2. Postsynaptic mechanisms of S T D 16 1.2.10.2.1. Receptor desensitization 16 1.2.10.2.2. Receptor saturation 18 1.2.11. Pre- and postsynaptic effects of pentobarbital 19 1.2.11.1. Postsynaptic effects 19 1.2.11.1.1. Effects on receptor systems 19 1.2.11.1.2. Effects on non-receptor systems 20 1.2.11.2. Presynaptic effects of barbiturates 20 1.2.11.2.1. Effects on ion channels 21 - iv -1.2.11.2.1.1. N a + channels 21 1.2.11.2.1.2. Voltage-gated C a 2 + channels 21 1.2.11.2.1.3. K + channels 21 1.2.11.2.2. Effects on the release machinery 23 1.2.11.2.3. Effects on energy metabolism 23 1.2.11.2.4. Effects on transmitter release • 24 1.2.11.2.4.1. Evoked release 24 1.2.11.2.4.2. Spontaneous release 25 1.2.12. Theory of depletion model and fluctuation analysis 26 1.3. Rationale 29 1.4. Objectives and research approach 31 1.5. Major questions 34 Chapter 2. Methods 36 2.1. Slice preparation 36 2.2. Drug application 38 2.3. Extracellular recording 38 2.4. Whole-cell recording 39 2.5. Detection of signals 40 2.5.1. Direct method 40 2.5.2. Deconvolution method 40 2.5.3. First and second derivative method 41 2.6. Repetitive stimulation 43 2.7. Induction of plastic modifications of corticothalamic synaptic responses 43 2.8. Analysis of extracellular recordings 44 2.9. Fluctuation analysis of corticothalamic synaptic responses 44 2.10. Statistical comparisons 51 Chapter 3. Results 52 - V -Part I. Pentobarbital oscillations in vitro in ventrobasal thalamus 52 3.1. Extracellular effects of pentobarbital 52 3.1.1. Pentobarbital application and internal capsule stimulation 52 3.1.2. Effects of reduced extracellular M g 2 + • 52 3.1.3. Effects of raised temperature 55 3.1.4. Application of a high dose of pentobarbital 55 3.1.5. Effects of synaptic receptor blockade 58 3.1.6. Pentobarbital-induced oscillations in separated thalamic nuclei 61 3.1.7. Discussion 66 Part II. Modulation of N M D A receptors in corticothalamic transmission 68 3.2.1. Effects of spermine • 68 3.2.1.1. Tonic firing 68 3.2.1.2. Passive membrane properties 69 3.2.1.3. Action potential threshold 71 3.2.1.4. Membrane rectification 75 3.2.1.5. Low threshold C a 2 + spike (LTS) firing 80 3.2.1.6. Excitatory and inhibitory postsynaptic potentials 84 3.2.2. Pentobarbital effects on corticothalamic EPSPs 90 3.2.3. Interactions of Z n 2 + with spermine and pentobarbital 91 3.2.4. Antagonism of polyamine site 93 3.2.5. Discussion 93 Part III. Effects of pentobarbital on short-term depression 99 3.3.1. Behaviour of EPSCs in trains during short-term depression (STD) 99 3.3.1.1. Passive membrane properties 99 3.3.1.2. Frequency - dependent fade (STD) of corticothalamic E P S C s 100 3.3.2. Effects of alterations in extracellular C a 2 + concentration ([Ca 2 + ] e ) 107 3.3.2.1. L o w [ C a 2 + ] e perfusion 107 3.3.2.2. Elevated [ C a ^ e perfusion 108 3.3.3. Receptor desensitization and saturation • 113 3.3.3.1. Effects of blockade of receptor desensitization 113 3.3.3.2. Combined blockade of receptor desensitization and saturation 113 3.3.4. Effects of pentobarbital on S T D 117 3.3.4.1. E P S C behaviour in trains 117 3.3.4.2. S T D in raised Ca 2 +concentration 122 - vi -< 3.3.4.3. S T D in reduced Ca 2 +concentration 127 3.3.4.4. Combined cyclothiazide and kynurenate blockade • 128 3.3.5. Effects of altered extracellular K + concentration ( [K + ] e ) 128 3.3.5.1. High [ K + ] e perfusion 132 3.3.5.2. L o w [ K + ] e perfusion 133 3.3.6. Effects of tetrodotoxin 139 3.3.7. Effects of glucose deprivation on STD 148 3.3.8. Discussion. 154 4. General discussion 163 Bibliography 182 - vii -List of Tables 1.1. Summary of synaptic and non-synaptic actions of pentobarbital 22 2.1. Comparison of peak detection methods 42 3.1. Effects of spermine on EPSP variables 86 3.2.1 A . Summary of parameters of corticothalamic STD at different frequencies 102 3.2. I B . Derived parameters of STD at different frequencies 106 3.2.2A. Summary of effects of altered [ C a 2 + ] e on parameters of S T D I l l 3.2.2B. Effects of altered [ C a 2 + ] e on derived parameters of S T D 112 3.2.3 A . Summary of effects of C T Z and K Y N on parameters of S T D 115 3.2.3B. Effects of C T Z and K Y N on derived parameters of S T D 116 3.2.4A. Summary of pentobarbital effects on parameters of S T D 119 3.2.4B. Effect of pentobarbital on derived parameters of S T D 121 3.2.5A. Pentobarbital effects on parameters of S T D in raised [ C a 2 + ] e 123 3.2.5B. Pentobarbital effects on derived parameters of STD in raised [ C a 2 + ] e 124 3.2.6A. Pentobarbital effects on parameters of STD in low [ C a 2 + ] e 125 3.2.6B. Pentobarbital effects on derived parameters of STD in low [ C a 2 + ] e 126 3.2.7A. Pentobarbital effects on parameters of S T D during co-applied C T Z and K Y N 130 3.2.7B. Pentobarbital effects on derived parameters of S T D during co-applied C T Z and K Y N 131 3.2.8A. Summary of effects of high [ K + ] e , pentobarbital on parameters of S T D 135 3.2.8B. Derived parameters of STD for high [ K + ] e , pentobarbital 136 3.2.9A. Summary o f effects of low [ K + ] e , pentobarbital on parameters of S T D 137 - viii -3.2.9B. Derived parameters of S T D in low [ K + ] e , pentobarbital 138 3.2.1 OA. Summary of T T X effects on parameters of STD 143 3.2.10B. T T X effects on derived parameters of S T D 144 3.2.11 A . T T X effects on parameters of S T D during co-applied C T Z + K Y N 146 3.2.1 I B . T T X effects on derived parameters of S T D during co-applied C T Z and K Y N 147 3.2.12 A . Summary of parameters of S T D at different glucose concentrations • 151 3.2.12B. Derived parameters of S T D at different glucose concentrations • 153 4.1. Receptor involvement in pentobarbital-induced oscillations 167 List of Figures 1.1. Diagram of the corticothalamocortical circuit 5 1.2. Depression of end-plate potentials (EPPs) during tetanic stimulation 13 1.3. Intra- and intersite variability of transmitter release 14 1.4. Mechanisms of short-term depression and their presumed site of action 17 2.1. Direct method of peak detection 40 2.2. Peak detection obtained by using deconvolution method 41 2.3. First and second derivative method for peak detection 41 3.1. Pentobarbital induces extracellular oscillations in ventrobasal nuclei during electrical stimulation of internal capsule at 0.05 H z 53 3.2. Pentobarbital-induced oscillations in low M g 2 + medium 54 3.3. Effects of raised temperature on pentobarbital oscillations 56 3.4. Time dependence of effects of increasing concentrations on pentobarbital oscillations 57 3.5. Antagonists of G A B A , and glycine receptors modulate frequency of pentobarbital oscillations 60 3.6. Photomicrograph of sagittal slice shows complete separation (asterisk) o f V B nuclei from nRT 63 3.7. Pentobarbital oscillations in electrically stimulated V B nuclei, after surgical separation from nRT 64 3.8. Pentobarbital oscillations in electrically stimulated nRT, before and after its surgical separation from V B nuclei 65 3.9. Spermine enhanced tonic firing in a concentration-dependent manner in M G B neurons 70 3.10. Spermine (100 u M ) decreased the action potential threshold 72 3.11. Spermine increased tonic firing by interacting with N M D A receptors 74 3.12. Effects of spermine (100 u M , 3 min) on membrane rectification 77 2+ 3.13. Alterations in extra- and intracellular Ca influence spermine effects on depolarizing current - voltage (V -1) relationships in M G B neurons 79 2+ 3.14. Effects o f spermine (100 u M , 3 min) on the low threshold Ca spike firing 83 3.15. Spermine (100 u M , 3 min) prolonged late component of corticothalamic EPSPs mediated by N M D A receptors 85 3.16. Spermine (100 uM) prolonged the EPSPs by interacting with the polyamine-sensitive site on N M D A receptor 89 3.17. Pentobarbital effects on NMDA-mediated corticothalamic EPSPs 92 3.18. Pentobarbital reversal of spermine EPSP prolongation involves interactions at the polyamine site on N M D A receptor 94 3.19. Frequency-dependence of corticothalamic STD 101 3.20. Validation of the corrected variance-mean method during corticothalamic STD-104 3.21. Pre- and post-stimulation miniature EPSCs vary in size 105 3.22. Persistence of STD in media - containing low C a 2 + 109 3.23. C a 2 + modification of corticothalamic STD 110 3.24. Effects of blockade of receptor desensitization and saturation on S T D 114 3.25. Dose-dependence of pentobarbital enhancement of corticothalamic S T D 118 3.26. Quantal alterations mediate pentobarbital effects on STD 120 3.27. Pentobarbital enhancement of STD during combined blockade of receptor desensitization and saturation 129 3.28. Effects o f altered K + concentration on pentobarbital enhancement of S T D 134 3.29. Tetrodotoxin enhanced S T D by reducing quantal size 141 3.30. T T X decreased the size of evoked miniatures EPSCs without affecting spontaneous miniature E P S C size 142 3.31. T T X effects on S T D during blockade of receptor desensitization and saturation .145 3.32. Effects of glucose deprivation on STD 150 3.33. Evoked and spontaneous miniature EPSCs during glucose deprivation 152 4.1. Possible synaptic targets for pentobarbital actions during corticothalamic STD. . . . 181 - X l l -Abbreviations A C S F Artificial cerebrospinal fluid A M P A a-amino-3-hydroxy-5methyl-4-isoxazoleproprionate A N O V A Analysis of variance A P V 2-amino-5-phosphono-valerate A T P Adenosine-5'-triphosphate c A M P Cycl ic 3' 5' -adenosine-monophosphate C S F Cerebrospinal fluid C N Q X 6-cyano-7-nitroquinoxaline C N S Central nervous system EC50 Concentration of drug that produces a half-maximal effect E E G Electroencephalogram E G T A Ethylene glycol-bis-(P-aminoethyl ether) N,N,N rN'-tetraacetic acid E P S P Excitatory postsynaptic potential E P S C Excitatory postsynaptic current G A B A Y-aminobutyric acid GluR Glutamate receptor h hour H E P E S N-[2-hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid] H z Hertz (s"1) IC50 Concentration of a drug that produces a half-maximal inhibition I h Hyperpolarization activated inward current Iicir Inwardly rectifying K + current - xiii -Iieak Voltage-independent leak current I N a P Persistent N a + current I x L o w threshold C a 2 + current IPSP Inhibitory postsynaptic potential IPSC Inhibitory postsynaptic current L T S Low threshold C a 2 + spike M G B Medial geniculate body min Minute N M D A N-methyl-D-aspartate nRT Nucleus reticularis thalami p H Hydrogen concentration; pH-log[H + ] p K a Dissociation constant; pH-log[base]/[cation] Ri Input resistance R E M Rapid eye movement S E M Standard error about the mean T Time constant; time required to reach (1-1/e) of a steady state value T E A Tetraethylammonium T T X Tetrodotoxin V m Membrane potential V B Ventrobasal complex of the thalamus V P L Ventral posterior lateral thalamic nucleus - xiv -Acknowledgements I would like to express my deepest gratitude to my supervisor Dr. Ernie Pui l who not only provided me sheer intellectual and moral support but also encouraged me to pursue my own original ideas. I am extremely grateful to Dr. David Mathers for the fruitful discussions that contributed to the work. I am extremely and equally grateful to Dr. David Quastel who provided excellent guidance in analysis and interpretation of the results. Without his help this work would not have been possible. I wish to thank Dr. T i m Murphy for excellent comments and suggestions. The present work was supported by the Canadian Institutes for Health Research, Mathematics in Information Technology and Complex Systems, and the Jean Templeton Hugi l l Foundation. I thank M s . Viktoriya Dobrovinska for preparation of materials and solutions and M r . Christian Caritey for excellent technical support. I thank Douglas Brown for his assistance with photography of brain slices. M y gratitude for invaluable intellectual and emotional support goes to my family and friends. Xiang Wan and Amer Ghavanini each deserve a special thank you for sharing a set-up and discussing scientific issues. The assistance of Mitrut Isbasescu was extremely helpful in programming and running various software. I'd also like to thank the new XV recruits o f our laboratory Sarah McCarthy and Stephanie Lee for asking interesting science-related questions. Last but not least, I'd like to thank my beloved girlfriend Andrea Krawzcyk whose moral support was needed at many exhausting moments. - xvi -Chapter 1. Introduction I Ran Chapter 1 INTRODUCTION 1.1. Scope of thesis The thesis describes in vitro experiments that delineate the modulation of corticothalamic transmission by pre- and postsynaptic actions of the barbiturate, pentobarbital. Introduced in the first half o f the 20 t h century, barbiturates have received extensive clinical use, based on their pharmacological and pharmacokinetic properties. Drugs in this class are sedative-hypnotics, anti-epileptics, and general anesthetics. Occasionally used in humans, pentobarbital is still the most widely used general anesthetic in experimental animals. Pentobarbital produces a wide range of in vivo effects. Unl ike other barbiturates, pentobarbital does not have anti-epileptic properties at subanesthetic doses. Like other barbiturates in anesthetic doses, pentobarbital is capable o f terminating convulsions. The anesthetic effects of pentobarbital are due to a depression of neuronal excitability in the central nervous system (CNS). This reduced responsiveness occurs in all C N S regions, contributing to the overall loss of awareness of environment during induction of pentobarbital anesthesia. It is generally accepted that barbiturate-induced depression involves postsynaptic interactions of neurons in the cortico-thalamocortical (CTC) system. A s with other drugs, there is limited evidence for presynaptic actions of pentobarbital at central synapses. This gap in knowledge stems from the inability to obtain reliable electrical recording at axon terminals, which regulate neurotransmitter release. This thesis explores the method Chapter 1. Introduction I Ran - 2 -of fluctuation analysis of synaptic responses to repetitive stimulation. The methodology described here facilitates the separation of pre- from postsynaptic drug effects. It provides a new approach for assessing actions of drugs, such as pentobarbital. The overall hypothesis of this thesis is that pentobarbital has actions at pre- and postsynaptic sites on neurons; reduction of excitatory transmitter release compromises transmission in the C T C system. If these effects occur at anesthetic doses in vivo, they could contribute to a loss o f consciousness. 1.2. Background 1.2.1. EEG activity and brain oscillations In the conscious brain, sleep and wake states correspond to varying degrees of synchronized oscillations in the electroencephalogram (EEG), correlative with rhythmic electrical activity of networks of neurons in the C T C system. Voltage oscillations o f neurons in the C T C network produce this synchrony (Steriade, 2003). During the early stages of sleep, there is prevalent spindling E E G activity, which reflects 6-14 H z oscillations of cortical neurons. A s sleep evolves into deeper stages, this spindling behaviour transforms to the slower delta activity in the 1 - 4 H z range. Lesions in the thalamus result in a disruption of the characteristic rhythmic E E G pattern in cortical neurons (Villablanca and Salinas-Zeballos, 1972). Thalamic neurons can generate and maintain oscillations that are independent of cortical inputs (Villablanca and Marcus, 1972). In summary, the thalamus is an essential component of the network for generating oscillations during natural sleep. Chapter 1. Introduction I Ran - 3 -1.2.2. Receptor mediation of sleep-like oscillations in the CTC system The synaptic interactions that mediate thalamic oscillations are complex (Jones, 2002). The type of receptors involved in generation of oscillations may alter their frequency range. For example, antagonism of ionotropic receptors for a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid ( A M P A ) enhances the light stage of non-rapid eye movement ( N R E M ) slow wave sleep resulting in 6-14 H z oscillations. On the other hand, N-methyl-D-aspartate ( N M D A ) receptor antagonism increases the deep stages while reducing the lighter stage of N R E M sleep, favoring oscillations in the 1-4 H z range (Juhasz et al., 1990). Activation of thalamic metabotropic glutamate receptors results in 1-4 H z membrane oscillations (Emri et al., 2003). Thus, the involvement of multiple excitatory receptors in thalamic oscillogenesis provides a mean for regulating the transition between the light and deep stages of sleep. Reduction in synaptic transmission through ionotropic receptors for the inhibitory transmitter, y-aminobutyric acid ( G A B A ) results in sustained thalamic oscillations. Blockade of type A G A B A ( G A B A A ) receptors with the potent antagonist, bicuculline, produces delta (1-4 Hz) oscillations, characteristic of deep sleep (von Krosigk et al., 1993). Penicillin, a much weaker G A B A A receptor antagonist, is also effective in inducing sleep oscillations in vivo (Vital-Durand et al., 1972). In both cases, the oscillations result, presumably, from a reduced inhibitory receptor-mediated transmission, which heightens neuronal excitability and primes the C T C system for rhythmic activity. Chapter 1. Introduction I Ran - 4 -1.2.3. Non-synaptic mechanisms of sleep-like oscillations Non-synaptic interactions provide an additional mechanism to explain thalamic oscillations. These mechanisms include intrinsic ionic conductances and gap junctions. According to this view, oscillations initiate with a depolarization o f thalamic neurons which brings their resting membrane potential above threshold and generates regenerative C a 2 + spikes. Upon inactivation of the C a 2 + channels, activation of voltage sensitive K + channels hyperpolarizes the membrane potential, which is then repolarized by the hyperpolarization-activated cation conductance (IR). The contribution of ionic conductances has been demonstrated in vitro and was obtained by the use o f patch clamp recordings in thalamic neurons (Leresche et al., 1991, McCormick and Pape, 1990). In these studies, oscillations occurred in isolated neurons that express T-type and L-type 2+ C a channels and were independent of synaptic inputs (Alonso and Llinas, 1989; Leresche et al., 1991). Pharmacological blockade of these channels abolished the ability to generate rhythmic patterns. More recent studies (Hughes et al., 2002) have implicated gap junctions as mediators o f thalamic synchrony. This appears to be the case in a small subset of ventrobasal thalamic neurons, but may not be significant for thalamic synchronization. Hence, intrinsic conductances are an essential requirement for generation of synchronized activity in thalamic neurons in addition to synaptic components of the C T C system. Chapter 1. Introduction I Ran - 5 -1.2.4. Synaptic composition of the CTC system The C T C system consists of excitatory glutamatergic and inhibitory G A B A e r g i c pathways, as well as 'presumed' glycinergic pathways (Ran et al., 2004). Thalamocortical relay neurons, in the dorsal thalamic nuclei, receive extensive connections from the Intemeuron 9lu? j; Thalamocortical neuron / Dorsal thalamus Figure 1.1. Diagram of the corticothalamocortical circuit. Thalamocortical relay neurons in the dorsal thalamus receive excitatory glutamatergic inputs from pyramidal neurons in the neocortex. In the neocortex, pyramidal neurons receive reciprocal excitatory glutamatergic inputs from thalamocortical relay neurons. G A B A e r g i c neurons in the reticular thalamic nucleus receive glutamatergic inputs from pyramidal neurons in the neocortex and from thalamocortical relay neurons. Reticular neurons send inhibitory projections to thalamocortical relay neurons in the dorsal thalamus. G A B A e r g i c reticular neurons contain local inhibitory collateral and dendro-dendritic synapses. Chapter 1. Introduction I Ran - 6 -cortex via glutamatergic synapses which constitute the corticothalamic pathway (Figure 1.1). The projection of thalamocortical neurons back to the cortex is a reciprocal pathway, mediated by glutamate (Jones, 2002). There is also a unidirectional excitatory glutamatergic pathway which originates in the cortex and terminates in the reticular thalamic nucleus. In addition, neurons of the dorsal thalamus excite inhibitory G A B A e r g i c reticular thalamic neurons by a glutamatergic pathway (Ohara, 1988). Thus, corticothalamic transmission triggers a loop of excitatory and inhibitory inputs in the C T C system. Corticothalamic activation occurs at ionotropic and metabotropic glutamate receptors (Jones, 2002). Ionotropic receptors include AMPA/kainate and N M D A receptors that mediate fast and slow excitatory postsynaptic responses, respectively. Slow excitation is also possible by activation o f metabotropic glutamate (mGlu) receptors. Overall, a wide spectrum of excitatory receptors that are subject to modulation, together with inhibitory receptors, govern the pattern of C T C rhythmic activity. 1.2,5. GAB A receptors modulate the frequency of sleep-like oscillations Inhibitory transmission, mediated by ionotropic G A B A A receptors and metabotropic G A B A B receptors, originates from G A B A neurons of the nucleus reticularis thalami (nRT) which terminate on dorsal thalamic neurons (reviewed by Llinas et al., 2005). There are inhibitory interneurons intrinsic to the dorsal thalamus that are estimated to make up 1-2 % of the total neuronal population. Activation of G A B A A and G A B A B receptors contributes to the transition between the delta and spindle rhythms (Bal et al., Chapter 1. Introduction I Ran - 7 -1995a,b). G A B A A receptors have been implicated in the generation of spindle rhythms (von Krosigk et al., 1993). G A B A B receptors are thought to exist extra-synaptically and mediate the slower delta (1-4 Hz) oscillations (Bal et al., 1995a). The different durations and locations of G A B A A and G A B A B receptor mediated responses provide frequency modulation. 1.2.6. Glycine receptors contribute to thalamic oscillogenesis A few studies have implicated glycine receptors in thalamic inhibition in addition to G A B A receptor mediated inhibition (Tebecis, 1974; Ghavanini et al., 2005). Glycine receptors have been shown to mediate theta-like oscillations (6-15 Hz) induced by pentobarbital in isolated thalamic ventrobasal slices (Ran et al., 2004). This novel pathway is quite intriguing, in light of the fact that glycine receptors mediate oscillations independent o f G A B A transmission in spinal neurons. Glycine receptor antagonism induces oscillations in other systems, such as in spinal cord neurons (Bracci et al., 1996). Thus, the rhythmic activity of the C T C network may be shaped by at least two types of inhibitory ionotropic receptors. 1.2.7. Barbiturate anesthetics induce sleep-like oscillations Barbiturates induce cortical as well as thalamic oscillations, in vivo, similar to those during natural sleep. This ability is attributable to changes in membrane properties o f neurons in the C T C system. The oscillations occur at frequencies in the delta (1-4 Hz) and theta (7-14 Hz) ranges, similar to deep and/or light natural sleep. The most commonly used barbiturates with this property include pentobarbital and thiopental. Less Chapter 1. Introduction I Ran - 8 -commonly, volatile anesthetics induce in vivo oscillations (Keifer et al., 1994). One of the mechanism by which barbiturates induce oscillations may depend on the excitable state of thalamic neurons, involving modulation of membrane properties and receptor interactions at corticothalamic synapses. 1.2.8. Poly amine modulation of barbiturate action The anesthetic potency of barbiturates depends on actions on corticothalamic transmission through N-methyl-D-aspartate ( N M D A ) receptors, subject to polyamine modulation. For example, polyamines decrease the rate at which N M D A receptors desensitize (Lerma, 1991) whereas pentobarbital has the opposite effect (Charlesworth et al., 1995). However, the effects of polyamines at N M D A receptors include both enhancement and inhibition (Benveniste and Mayer, 1993). Spermine, among other polyamines, enhances the property of barbiturates to induce general anesthesia. This enhancement may result from dual inhibitory actions of spermine and pentobarbital at the M g 2 + site on N M D A receptors (Daniell, 1992). However, pentobarbital depression o f corticothalamic transmission may prevent the prolongation of N M D A responses caused by spermine. 1.2.9. Polyamine enhancement of corticothalamic transmission: Relevance to thalamic oscillogenesis Polyamines are endogenous biogenic amines that interact with synaptic and nonsynaptic targets and may modulate corticothalamic transmission or enhance thalamic excitability. The C T C circuit may be primed for oscillatory activity as a result of polyamine-mediated Chapter 1. Introduction I Ran - 9 -prolongation of EPSPs. Similar to other polyamines, spermine acts at both sides o f neuronal membrane. These interactions influence ion channels and transmitter-gated channels (reviewed by Will iams, 1997). Intracellular spermine enhances membrane rectification by blocking inward rectifier K + (KFR) channels (Schuber, 1989) and confers an inwardly rectifying property on receptor-gated channels activated by acetylcholine (Haghighi and Cooper, 2000) and A M P A (Koh et al., 1995). Extracellularly, spermine interacts with receptors for N M D A resulting in the prolongation of postsynaptic responses (reviewed by Rock and Macdonald, 1995). Low concentrations of spermine enhance N M D A - e v o k e d currents whereas high concentrations produce a voltage-dependent block o f these currents in hippocampal neurons (Benveniste and Mayer, 1993). The effects of spermine and its ability to influence barbiturate actions are unknown in the C T C system. Spermine and other polyamines are present at micromolar extracellular concentrations in the brain, including the thalamus (Harman and Shaw, 1981). This may indicate a possible role as a modulator of membrane excitability. Recent binding studies have challenged the validity of the previous measurements and estimate the extracellular concentration of spermine at < 1 u M (Dot et al., 2000). Neurons and glia release spermine during electrical stimulation, depolarization by high external [ K + ] , and activation of NMDA-receptors (Harman and Shaw, 1981; Fage et al., 1992). Uptake o f spermine maintains low extracellular concentrations, presumably resulting from a constitutive release of spermine (Dot et al., 2000). Spermine uptake is regulated by various transporters in the dendrites, cell bodies, and nerve terminals, as well as by a high Chapter 1. Introduction I Ran - 10 -affinity transporter expressed in glia (Laube and Veh, 1997). Spermine, at concentrations > 50 u M , enhances long-term potentiation (Pussinen et al., 1998; Toth et al., 2000) and neuroprotection (Trout et al., 1993; Mui r and Lees, 1995; Ferchmin et al., 2000). Hence, alterations in extracellular spermine concentration can provide a unique mechanism for modulation, prolonging or shortening the duration of excitatory responses to corticothalamic stimulation. A deficiency of spermine may have important consequences in dysfunctional states, whereas excessively high extracellular concentrations may predispose neurons in the C T C network to hyperexcitable states. For example, pharmacological inhibition o f polyamine synthesis decreases polyamine concentrations in the cochlea (Schweitzer et al., 1986). This deficiency induces a temporary hearing loss in humans and rats (reviewed by McCann and Pegg, 1992), possibly due to decreased spermine-regulation o f NMDA-receptor-mediated activities in cochlear neurons (Petralia et al., 2000). On the other hand, exceptionally high concentrations of spermine may exist in several neurological disorders, including neurodegenerative diseases (Yatin et al., 2001), stroke (Lukkarinen et al., 1997), global and focal ischemia (Baskaya et al., 1997; Dogan et al., 1999) and kindling epilepsy in an animal model (Hayashi et al., 1992; Herberg et al., 1992; De Sarro et al., 1993; Halonen et al., 1993). Hence, the alterations o f excitable states imposed by endogenous modulators may determine the outcome of barbiturate actions on thalamic neurons. However, barbiturates may modify corticothalamic transmission independent of the excitable state of thalamic neurons during short-term alterations of synaptic strength or, short-term plasticity. Chapter 1. Introduction I Ran -11-1.2.10. Pre- and postsynaptic mechanisms of short-term depression (STD) Synaptic plasticity has long been the focus of many neuroscientists, in view o f its potential roles in signal processing, learning, and memory (Fortune and Rose, 2000; Zucker and Regehr, 2002). Plastic changes in synaptic efficacy include both enhancement and reduction of neurotransmission, depending upon experimental conditions and stimulation paradigms. Enhancement can be observed both on short- and long-time scales, termed facilitation and long-term potentiation (LTP), respectively, and reduction also can be short- (seconds to minutes) or long-lasting (minutes to hours), termed short- and long-term depression (STD and L T D ) , respectively (von Gersdorff and Borst, 2002; Voronin, 1994). Both facilitation and depression are demonstrable on short time scales in synaptic responses evoked by pairs of evoked stimuli, generally referred to as paired-pulse facilitation or paired-pulse depression. These plastic alterations also occur during and after trains of stimuli at certain frequencies of stimulation. Although paired-pulse studies have shed light on short time scale modifications of synaptic strength, the dynamics and progression of these phenomena are rather limited. Studies that use intermediate (5-20 pulses) to long (>20 pulses) trains of stimulation pulses have provided a more comprehensive view of short-term depression and a fuller expression o f these processes (reviewed by von Gersdorff and Borst, 2002). 1.2.10.1. Presynaptic mechanisms of STD S T D reflects reductions in the number of quanta released per stimulus (Elmqvist and Quastel, 1965a). This could be because of depletion of a releasable presynaptic store o f quanta, consequent to release, as proposed by Li ley and North (1953). However, there are Chapter 1. Introduction I Ran -12-other possibilities: inactivation of presynaptic C a 2 + channels, changes in presynaptic action potential configuration, or decreased rate of endocytosis (see below). A reduction in the amount of transmitter per quantum would also be classified as presynaptic (Elmqvist and Quastel, 1965b). It should be noted that it is often assumed that there is a 1:1 relationship between quantum and vesicle (del Castillo and Katz, 1954). This relationship has been confirmed by increases in membrane capacitance measured during vesicular fusion (Aimers and Neher, 1987). However, the relationship is in fact controversial (Matthews, 1996; Vautrin and Barker, 2003). For the present purposes, a quantum is defined as an elementary pulse of a packet of transmitter which generates a brief synchronized postsynaptic current. Presynaptic mechanisms contribute to, or dominate, STD at low to moderate stimulus frequencies (< 10Hz ; Elmqvist and Quastel 1965a,b; Charlton et al., 1982; Emptage et al., 2001). However, at higher stimulation frequencies, postsynaptic mechanisms such as receptor desensitization or saturation, may also contribute to S T D (Figure 1.4). 1.2.10.1.1. Depletion of quantal store A simple model that assumes the release of a constant fraction of 'available quanta' with each action potential, at a constant rate of 'refill', might explain S T D on the basis o f store depletion (Vere-Jones, 1966). These assumptions are too simplistic for neuromuscular (Elmqvist and Quastel, 1965a) or hippocampal synapses (Rosemund and Stevens, 1996) and the calyx o f Held (Wu and Borst, 1999; Schneggenburger et al., 1999). For example, Chapter 1. Introduction I Ran - 13-the rate of refill o f transmitter packets is evidently accelerated at high frequencies of stimulation (Elmqvist and Quastel, 1965a; Wang and Kaczmarek, 1998), resulting in a faster recovery from depletion of the readily releasable pool of transmitter packets. The adjustment of the rate of refill o f transmitter packet may provide a regulatory mechanism that diminishes S T D in response to high frequency inputs. Figure 1.2: Depression of end-plate potentials (EPPs) during tetanic stimulation. Extracellularly recorded EPPs decrease and reach a plateau after 8 stimuli. Plateau reflects a steady-state between depletion of transmitter packets and refill o f quanta and o f transmitter per quanta. Stimulation frequency was 180 H z (From Liley and North 1953). The release probability, presumably the fraction of 'available quanta' that are released per stimulus, varies in different preparations. Release probabilities have a heterogeneous distribution at different release sites (Wu and Borst, 1999; Sakaba and Neher, 2001). The heterogeneity of release probabilities could depend on the position of transmitter packets with respect to C a 2 + channels (positional heterogeneity) and/or their sensitivity to C a 2 + (biochemical heterogeneity). These observations led to the following modifications in the depletion model (Liley and North, 1953; Elmqvist and Quastel, 1965a, Mi led i and Chapter 1. Introduction I Ran -14-Thies, 1967): 1) a C a - and activity-dependent enhancement of refill (Elmqvist and Quastel, 1965a; Zimmermann and Whittaker, 1977); 2) the existence of multiple groups of transmitter packets with varying probabilities of release (Auger and Marty, 1997); and, 3) a decrease in the number of participating release sites (Weis et al., 1999). A n increase in presynaptic intracellular C a 2 + concentration ([Ca 2 +];) accelerates the uptake of transmitter into packets or increases the number of releasable packets to the same extent in the absence or presence of an exogenous C a 2 + buffer (Wu and Borst, 1999). This implies that the actions of internal C a 2 + are indirect and may involve second messengers. Indeed, interactions of [Ca 2 +]j with calmodulin activate protein kinases which accelerate vesicular refill and reduce store depletion (Sakaba and Neher, 2001). The existence of groups of packets with varying release probabilities does not alter the rate of store depletion at any one site. Instead, there is a heterogeneous distribution of groups of transmitter packets in different release sites (Figure 1.2; Auger and Marty, 1997). Intrasite quantal variance Intersite quantal variance Figure 1.3: Intra- and intersite variability of transmitter release. Chapter 1. Introduction I Ran -15-1.2.10.1.2. Reduction in transmitter content A second possible mechanism for S T D is reduction in amount o f transmitter per quantum (Elmqvist et al., 1965b; Naves and Van Der Kloot, 2001; reviewed by Vautrin and Barker, 2003). Blocking acetylcholine synthesis reduces the transmitter content early (Van der Kloot and Molgo, 1994) or only late (Elmqvist et al., 1965b) after prolonged neuromuscular stimulation. A t C N S synapses, there are both supporting (Ishikawa et al., 2002) and contradicting (Sakaba and Neher, 2001) reports about the correlation between reduction in transmitter content and depression. Studies on rates of transmitter refill, recycling and endocytosis would clarify these discrepancies. 1.2.10.1.3. Modification of the presynaptic action potential A third presynaptic mechanism of S T D is modification of presynaptic action potential configuration (Brigant and Mallart, 1983; Smith, 1983). Also , in cultured hippocampal neurons, N a + channel inactivation produces failure in nerve conduction at presynaptic branch points, enhancing depression (Debanne et al., 1997; Brody and Yue, 2000; He et al., 2002). However, these, changes occur in conjunction with changes in C a 2 + currents. A t the calyx of Held, action potentials are reduced in amplitude and increased in duration much to the same degree as C a 2 + currents, leading to STD (Borst and Sakmann, 1999). Hence, this mechanism may co-exist with Ca2 +-dependent aspects of S T D . 1.2.10.1.4. Inactivation of Ca2+ channels A fourth mechanism of STD involves enhanced inactivation of C a 2 + channels at synaptic terminals. A t calyx of Held neurons, the inactivation of release sites contributes to S T D Chapter 1. Introduction I Ran -16-during prolonged high frequency stimulation (Forsythe et al., 1998). The inactivation o f 9 + C a channels is subject to modulation and coupling to various G-proteins by glutamate-(mGluRs), G A B A B - , adenosine-, and noradrenaline-receptors (Barnes-Davis and Forsythe, 1995; Isaacson, 1998; Kajiwara, 1997; Takahashi et al., 1996; W u et al., 1998). In calyx of Held neurons, modulation by mGluRs seems to contribute to up to 10 % o f 9 + depression due to inhibition of Ca currents (von Gersdorff et al., 1997). In summary, inactivation of C a 2 + channels might impair the release process enhancing S T D at corticothalamic synapses. 1.2.10.1.5. Interference with endocytosis The fifth mechanism, and last to be discussed here, is the regulation of endocytosis, discovered in shibire mutant flies. These mutants suffer from severe S T D and paralysis due to defective endocytosis at neuromuscular synapses (Poodry and Edgar, 1979). Recent investigations show an enhanced STD in response to genetic interference with endocytosis (Delgado et al., 2000; Luthi et al., 2001). Blockade o f the action of dynamin, a regulator o f vesicle endocytosis, markedly enhances STD and prolongs the recovery time at calyx of Held synapses (Takahashi et al., 2000). 1.2.10.2. Postsynaptic mechanisms of STD 1.2.10.2.1. Receptor desensitization A t high stimulation frequencies, postsynaptic receptor desensitization and saturation contribute to STD. Receptor desensitization of A M P A receptors was observed during paired-pulse depression at retinogeniculate synapses (Chen et al., 2002; Kielland and Chapter 1. Introduction I Ran 17-Inhibitory autoreceptors: - Metabotropic - Adenosine - Noradrenaline -GABA Reduction of postsynaptic sensitivity: - Desensitization - Saturation - Lowered excitability Altered Ca"- secretion coupling: - Depletion of releasable vesicles - Inactivation of release sites - Change in sensitivity to Ca" - Modulation of vesicle recruitment Figure 1.4: Mechanisms of short-term depression and their presumed site of action. Different mechanisms of synaptic depression, acting at distinct points in the synaptic vesicle cycle have been proposed and can be directly studied in synapses with large synaptic terminals (Adapted from von Gersdorff and Borst, 2002). Chapter 1. Introduction I Ran - 18 -Heggelund, 2002). However, the long recovery time (>4 s) of postsynaptic responses after S T D induced by 5 or 10 H z stimulus trains implies that receptor desensitization does not contribute to S T D at all stimulation frequencies. The faster time scale of recovery from desensitization of A M P A and N M D A receptors means that these receptors do not likely contribute to S T D at low frequencies of stimulation. Since A M P A and N M D A receptors desensitize on different time scales, a postsynaptic mechanism also would depend on their recovery times. However, the recovery time from S T D is identical in the responses evoked on activation of both receptors. This means that receptor desensitization makes minimal contribution to STD at low frequencies of stimulation. 1.2.10.2.2. Receptor saturation A saturation of postsynaptic A M P A receptors might also contribute to STD. Although a single quantum does not saturate A M P A receptors (Ishikawa et al. 2002), intensive stimulation of glutamate release leads to significant receptor saturation at calyceal synapses (Wu and Borst, 1999). Non-stationary fluctuation analysis methods reveal a contribution of receptor saturation to STD (Scheuss et al., 2002). Like receptor desensitization, saturation is not significant at low stimulation frequencies and is minimal at the onset of repetitive stimulation when release of transmitter is maximal (Matveev and Wang, 2000). Hence, receptor saturation does not likely contribute to depression at low stimulation frequencies. Chapter 1. Introduction I Ran - 19 -Pentobarbital effects on a wide variety of synaptic and non-synaptic targets might provide a window of corticothalamic transmission for examining the mechanisms mentioned above. 1.2.11. Pre- and postsynaptic effects of pentobarbital 1.2.11.1. Postsynaptic effects 1.2.11.1.1. Effects on receptor systems The most studied postsynaptic receptor target for pentobarbital action is the receptor for G A B A of subtype A ( G A B A A R ; Macdonald and Olsen, 1994). B y prolonging its decay, pentobarbital enhances the GABA-mediated CI" current, in a wide variety of brain preparations, including ventrobasal thalamic neurons (Wan and Pui l , 2002; Table 1.1). Pentobarbital can also directly activate CI" currents through G A B A A R (Mathers, 1987). In hippocampal neurons, pentobarbital has use-dependent postsynaptic actions o f promoting A M P A receptor desensitization (Jackson et al., 2003), which would contribute to corticothalamic STD. Other studies have demonstrated an action of pentobarbital to promote the desensitization of the GluR2 subtype of A M P A receptors (Taverna et al., 1994). The actions of pentobarbital at this receptor subtype, also expressed in thalamic neurons (Spreafico et al., 1994), are very sensitive to the actions of cyclothiazide (CTZ) , a blocker of A M P A receptor desensitization (Jackson et al., 2003). Chapter 1. Introduction I Ran - 20 -Pentobarbital has depressant actions on N M D A receptor channels (Charlesworth et al., 1995). These actions include a reduction in the probability of channel opening, a shortening of mean open time, and a decrease of burst length. 1.2.11.1.2. Effects on non-receptor systems The postsynaptic effects of pentobarbital on K + currents have been studied in cerebellar and hippocampal neurons (Carlen et al., 1985), and extensively studied in ventrobasal thalamic neurons (Wan et al., 2003; Table 1.1). These effects include: 1) increasing the input conductance by activating a leak current, 2) activating a voltage-dependent K + conductance, and 3) decreasing the hyperpolarization-activated N a + / K + inward current (Ih). Pentobarbital reduces Ca currents by increasing channel inactivation in dissociated spinal cord (Werz and Macdonald, 1985) and hippocampal neurons (ffrench-Mullen et al., 1993). These observations may explain pentobarbital effects in decreasing low-threshold spike firing in thalamic neurons (cf. Wan and Puil , 2002). 1.2.11.2. Presynaptic effects of barbiturates Sherrington (1906) initially suggested that anesthetics reduce synaptic transmission rather than decreasing nerve conduction. Larrabee and Posternak (1952) showed that general anesthetics depress transmission through sympathetic ganglia, without effects on nerve conduction. Most investigations have dealt with the postsynaptic effects (Franks and Lieb, 1994). The contribution of presynaptic components to barbiturate anesthetic Chapter 1. Introduction I Ran - 21 -properties has been neglected due to the difficulty in recording from axon terminals and distinguishing between their pre- and postsynaptic effects. Only two reports have indicated a reduction of transmitter release without changes in action potential configuration in C N S neurons (Mathews and Quilliam, 1964; Weakly, 1969). 1.2.11.2.1. Effects on ion channels 1.2.11.2.1.1. Na+ channels A t anesthetic doses, barbiturates broaden action potentials at the frog neuromuscular junction (Thompson and Turkanis, 1973). The broadening of the action potential might result from a hyperpolarizing shift in activation (Wartenberg et al., 1999) or a use-dependent block of the N a + channel (Rehberg et al., 1995). In principle, a broader action potential should promote transmitter release, by prolonging the depolarization of the terminal and Ca influx. 1.2.11.2.1.2. Voltage gated Ca channels C a 2 + imaging studies show pentobarbital suppression of C a 2 + entry into terminal branches of hippocampal neurons (Baudoux et a l , 2003). These observations suggest that 9-1-pentobarbital enhancement of Ca channel inactivation (ffrench-Mullen et al., 1993) may promote S T D by reducing Ca entry into the nerve terminals. 1.2.11.2.1.3. channels K + channels are abundant at presynaptic terminals and are highly involved in regulating transmitter release (reviewed by Dodson and Forsythe, 2004). However, there is little Chapter 1. Introduction I Ran Table 1.1: Summary of synaptic and non-synaptic actions of pentobarbital Site Effect ED 5 o or IC50 Neuron Reference -22-Receptor A M P A use-dependent inhibition 20 | i M Hippocampal Jackson et al., 2003 (culture) Decreased EPSP Amplitude N M D A reduced mean channel open time G A B A A prolonged current decay time 50 u M Thalamocortical Wan et al., 2003 (slice) 250 u M Olfactory Charlesworth et al., 1995 (culture) 53 u M Thalamocortical Wan et a l , 2003 (slice) increased mean channel open time Glycine prolonged current decay time 100 u M Thalamocortical Wan et al., 2003 (slice) 30 u M Spinal dorsal horn L u and X u , 2002 (culture) Ion channel Voltage-dependent Conduction block 3 m M N a + Voltage-dependent decreased I n , IKIR 8 pJVI K + Voltage-independent Increased Iieak 8 u M K + Low-threshold no effect > 1 0 0 u M Ca' 2+ Lobster (slice) Blaustein, 1968 on h Thalamocortical Wan et al., 2003 (slice) Thalamocortical Wan et al., 2003 (slice) Thalamocortical Wan et al., 2003 (slice) Voltage-dependent C a 2+ Enhanced 3 | j M Hippocampal inactivation (culture) ffrench-Mullen et al., 1993 Chapter 1. Introduction I Ran - 23 -direct evidence for anesthetics acting on K + channels at nerve terminals. The function and properties o f K + channels vary widely at presynaptic locations. These include: 1) dampening of the presynaptic action potential by low-voltage-activated K + channels; 2) faster repolarization (termination of the action potential) by presynaptic high-voltage-activated K + channels; and, 3) activity-dependent modulation of transmitter release by interplay of K + - and Na+-current activation. The structure of presynaptic voltage-gated K + channels which mediate inactivating transient and non-inactivating delayed rectifier currents (IA and IKDR) shows some similarity to K + channels located in the soma (Dodson and Forsythe 2004), suggesting possible interactions with pentobarbital. 1.2.11.2.2. Effects on the release machinery There are no reports on pentobarbital interactions with the soluble N-ethylmaleimide-sensitive factor attachment protein receptors ( S N A R E ) proteins, which mediate exocytotic transmitter release at active zones (cf. Duman and Forte 2003). However, imaging techniques demonstrated a pentobarbital effect of suppression of transient increases in intracellular [Ca 2 + ] at axon terminals associated with a decrease in the amplitude of spontaneous EPSPs (Baudoux et al., 2003). Since S N A R E proteins are 9-1-sensitive to intracellular Ca alterations, one can assume an indirect effect that could impair or increase release. 1.2.11.2.3. Effects on energy metabolism Pentobarbital impairs respiratory metabolism as well as glutamate synthesis, presumably at presynaptic terminals. Early studies have shown that pentobarbital reduces cellular Chapter 1. Introduction I Ran - 24 -respiratory processes (Quastel and Wheatley, 1932; Aldridge and Parker, 1960; Chance and Hollunger, 1963), including neuronal glycolysis (Crane et al. 1978). The majority of neuronal mitochondria are located at nerve terminals where A T P consumption is required for synthesis and release of transmitters such as glutamate (Schwartz et al., 1979; Hertz and Zielke 2004). Hence, pentobarbital suppression of metabolism produces a state of energy deprivation, contingent on activity (Hertz et. al., 1986), which could eventually impair transmitter release. 1.2.11.2.4. Effects on transmitter release 1.2.11.2.4.1. Evoked release The presynaptic actions of pentobarbital were initially studied in spinal motoneurons (Weakly, 1969), and involve a reduction in evoked transmitter release. Subanesthetic concentrations of pentobarbital decreased the number of released quanta, without changing the input resistance and firing threshold (quantal content; Figure 1.1). Hence, the study demonstrated for this site a presynaptic effect of pentobarbital with a negligible postsynaptic effect. In later studies, pentobarbital was found to enhance evoked transmitter release at the neuromuscular junction by increasing the number of quanta released by nerve stimulation (Thomson and Turkanis, 1973; Seyama and Narahashi, 1975; Weakly and Proctor, 1977). The contrasting enhancement and reduction of transmission were dose-dependent and attributable to the combined pre- and postsynaptic actions of pentobarbital (Proctor and Weakly, 1976). The enhanced quantal release was assessed from the ratio of evoked to Chapter 1. Introduction I Ran - 25 -spontaneous end-plate potentials (EPPs), whereas postsynaptic effects were identified as decreased miniature amplitudes. Although initially the increased quanta was presumed to result from the increased duration of a broader presynaptic action potential (Thompson and Turkanis, 1973), subsequent studies demonstrated no such effect (Weakly and Proctor, 1977). In summary, barbiturates have dose-dependent effects attributed to dual pre and postsynaptic actions that include both increase and decrease of quantal release and reduction of quantal size (Figure 1.2). 1.2.11.2.4.2. Spontaneous release Studies on spontaneous release provided greater detail about the complex nature o f barbiturate effects. Barbiturate application increased the frequency and decreased the amplitude o f spontaneous acetylcholine release, independent of extracellular [Ca 2 + ] (Pincus and Insler, 1978). These observations were quite similar to the [Ca 2 + ] -independent enhancement of miniature EPP frequency by ethanol (Quastel et al., 1971). These results implied that an effect of barbiturates on C a 2 + uptake determined the magnitude of the changes in quantal content (Rahamimoff et al., 1976). In summary, barbiturates have opposing effects at the neuromuscular junction - a presynaptic action that increases miniature EPP frequency and quantal content and a postsynaptic action that increases membrane conductance, reducing amplitude of spontaneous and evoked EPPs. Due to lack of reliable methods of assessment, the effects of barbiturates on transmitter release have not received study at central synapses. A s investigated in this thesis, the use of fluctuation analysis can provide an accurate measure of barbiturate modulation of quantal parameters during short-term synaptic plasticity. Chapter 1. Introduction I Ran - 26 -1.2.12. Theory of depletion model and fluctuation analysis Statistical estimation of synaptic plasticity dates back to the work of Del Castillo and Katz (1954) who analyzed amplitude fluctuations of spontaneous and evoked responses at the neuromuscular junction. The resemblance of incremental amplitude fluctuation to the mean amplitude o f spontaneous miniature synaptic events formed the basis o f a 'quantal hypothesis of transmitter release' (Del Castillo and Katz, 1954). According to this hypothesis, 3 parameters describe transmitter release at a given synapse: (1) the average amplitude of the postsynaptic response (Q); (2) total number of independent release sites at the synapse (N); and, (3) the average probability of release across all sites (p). Changes in p and Q constitute respectively the pre- and postsynaptic strength of synaptic connectivity and either one or the other must be altered whenever there is plastic modification in synaptic transmission. Alterations in these parameters reflect a drug's mechanism of action, e.g., a drug that acts exclusively at postsynaptic receptors must modify Q without an effect on p or N . Since the classical approach to fluctuation analysis (del Castillo and Katz, 1954; Boyd and Martin, 1955) of the synaptic response depends upon quantal content being low enough that there are many 'unit' responses, alternative statistical methods for analyzing quantal release, collectively called variance-mean analysis are often employed (Elmqvist and Quastel, 1965a; Vere-Jones, 1966; Clements, 2003). The variance-mean analysis uses the relation between the mean and the variance of iterated synaptic responses. The binomial model of transmission is valid i f the variance has a parabolic relation to the mean of the response (Clements, 2003); the variance/mean ratio gives a measure o f Q Chapter 1. Introduction I Ran - 27 -o (Elmqvist and Quastel, 1965a,b; Clamann et a l , 1991; Silver et al., 1998; Scheuss and Neher, 2001). A major limitation of the variance mean approach is that it is restricted to steady-state responses under stationary conditions of release, i.e., all release sites are assumed to be independent and have the same release probability. When applied to long trains of synaptic responses, the variance-mean analysis is useful for the study o f short-and long-term modification of synaptic plasticity (Elmqvist and Quastel, 1965a). The present study utilized a modification of the variance-mean analysis (see later in this section). The use of the classical quantal analysis presents some difficulties in interpreting changes in p, Q, and N . This method necessitates the use of clearly detectable mean response amplitudes, plotted in an amplitude histogram. In many preparations and various C N S neurons, however, the quanta are not easy to detect and do not form clear peaks in the histogram. Such difficulty may result from sampling error or low signal to noise ratio. Additional factors that interfere with detection of quantization include high quantal content and variability in quantal size. In some preparations, quantal size is highly variable in the range of 44-90%, as observed in distributions of miniature excitatory postsynaptic currents (mEPSCs; Frerking and Wilson, 1996). The heterogeneity in the probability of transmitter release is in the range of 22-71% in spinocerebellar tract neurons (Walmsley et al., 1988) and > 50% in hippocampal neurons (Murthy et al., 1997). The variability in quantal size and release probability necessitate modifications of the binomial model, such as the compound binomial, multinomial and compound Chapter 1. Introduction I Ran - 28 -multinomial models (Brown et al., 1976; Redman, 1990; Quastel, 1997; Silver et al., 1998). Despite limitations, the variance-mean method provides an independent approach for comparing the experimental means and variances with the model-based fits. A n advantage is that changes in synaptic parameters due to deviations from the simple assumptions of the binomial model are reflected in the slope of the linearized variance-mean plot (Silver et al. 1998). The variance-mean method also is useful for analysis of initial synaptic responses in long trains at different frequencies, and at various probabilities of release due to systematic variations in the external [Ca 2 + ] . In cases where quantal content varies within the train, however, the variance-mean method, per se, cannot follow gradations of the synaptic parameters (p and Q) within the train. Another difficulty in the classical quantal hypothesis is the assumption that there is constant number of participating release sites, defined as stationary. During repetitive stimulation, however, this assumption is not valid as the number of participating sites is continuously altered and hence is non-stationary. Vere-Jones (1966) and Quastel (1997) pointed out an inherent variation of N that occurs during release. Such variations can be estimated by using covariance analysis methods derived by Vere-Jones (1966), Quastel (1997), and Scheuss and Neher (2001) from the mathematical model o f Vere-Jones (1966). These methods essentially correct the variance/mean ratio for the effect o f p and provide, in principle, estimates of Q at successive responses in iterated trains. The methods depend upon the validity of the binomial model, which is indicated by the Chapter 1. Introduction I Ran - 29 -existence of negative correlations between responses to successive stimuli (Elmqvist and Quastel, 1965a; Scheuss and Neher, 2001). Assuming the maximal release o f one packet o f transmitter per stimulus per release site and little 'refill', a negative correlation between successive responses occurs because no release by one stimulus precludes release at the next (Vere-Jones, 1966). The correlation between successive stimuli might be influenced by presynaptic modulators of release (V iz i and Somogyi, 1989). A sudden increase in the rate o f refill should show up as a positive interstimulus correlation (Matveev and Wang, 2000). B y using the binomial depletion model discussed above, one can assess the effects of barbiturates on quantal parameters of transmission in the thalamus during short-term alterations in synaptic plasticity such as STD. 1.3. Rationale W h y study S T D in the thalamus? STD contributes to the generation o f oscillations, an essential behaviour of thalamic neurons (Steriade 1999; Castro-Alamancos and Calcagnotto, 1999). For example, the effects of STD on thalamic firing behaviour depend on the composition, desensitization, and saturation of postsynaptic receptors in thalamic neurons (Chen et al., 2002; Bartlett and Smith, 2002). There are no reports on the presynaptic mechanisms that mediate short-term plasticity at corticothalamic synapses. Not surprisingly, most neuroscientists have assumed that thalamic plasticity is entirely postsynaptic (Castro-Alamancos and Calcagnotto, 1999). This view has received Chapter 1. Introduction I Ran - 30 -challenge by a demonstration of a presynaptic form of L T P at corticothalamic synapses (Castro-Alamancos and Calcagnotto, 1999). A frequency- and C a 2 + - dependence and decreased paired-pulse facilitation characterized this form of L T P . However, the previous study lacked a continuous measure of the plastic alterations, which would emerge during longer stimulation train, en route to L T P . W h y study the effects of barbiturates on STD? A barbiturate-induced depletion o f transmitter packets at nerve terminals might promote S T D during corticothalamic transmission. The best support for such a mechanism is based on the enhanced quantal release observed in spinal motoneurons (Weakly, 1969). The increase in released quanta depletes the store of transmitter packets, which could exaggerate STD. The interactions of barbiturates at the neuromuscular junction and on spinal motoneurons provide some rationale for studying the pre- and postsynaptic aspects of corticothalamic transmission. Thalamocortical neurons in vivo are indeed sensitive because barbiturates disrupt their characteristic firing behaviour (Contreras and Steriade, 1996). Current theory maintains that these effects are postsynaptic and are due to enhanced G A B A e r g i c inputs. In some studies (Castro-Alamancos and Calcagnotto, 1999), thalamocortical neurons display postsynaptic mechanisms of plasticity. Recent studies have shown that N M D A receptors may enhance, whereas G A B A receptors may reduce the synaptic strength of corticothalamic responses. The plastic effects of S T D would contribute to synaptic connectivity of the thalamus during development (Bartlett and Smith, 2002). Both N M D A and G A B A receptors are well-documented postsynaptic targets for Chapter 1. Introduction I Ran -31-anesthetics but there is little or no information about potential presynaptic mechanisms that may explain the observed barbiturate-depression of synaptic responses (Richards, 1971; Sawada and Yamamoto, 1985). 1.4 Objectives and research approach One o f the objectives of the present study was to assess the changes in quantal content and size, number of release sites, and rate of vesicular refill during repetitive stimulation, by using a corrected version of the variance-mean method. The correction of the variances was obtained using the covariances between successive synaptic stimuli (Vere-Jones, 1966; Quastel, 1997; Scheuss and Neher, 2001). The present study examined how barbiturate anesthetics alter these corrected parameters. The hypothesis was that any effect of pentobarbital on corticothalamic transmission, manifested in a change of S T D during repetitive stimulation, must be reflected in changes of quantal content and/or quantal amplitude (Q). The study examined the anesthetic effects on synaptic transmission during repetitive stimulation of corticothalamic axons, in order to allow an expression of short-term depression. The binomial model was used to estimate changes in synaptic parameters such as quantal size and content, number of release sites, and rate of refill o f transmitter packets. These include STD, observed as decreases in amplitude ('rundown') of the initial synaptic responses and increases in amplitude of synaptic responses observed after a recovery from intra-train gap between stimuli. The findings enabled an assessment of the Chapter 1. Introduction I Ran -32-interactions of endogenously released glutamate with A M P A receptors while differentiating pre- from postsynaptic mechanisms of STD. B y pharmacological inhibition of A M P A receptor desensitization, it was possible to distinguish a postsynaptic contribution to S T D during repetitive stimulation. The validity of the covariance-corrected variance/mean method of determining quantal amplitude was established in control experiments which showed consistency with the binomial model - in particular, negative correlations between responses to successive stimuli that conformed with predictions of the model. These investigations represent a determination of anesthetic effects on both pre- and postsynaptic aspects o f excitatory synaptic transmission, for the first time in neurons of the C N S . The anesthetic interactions with corticothalamic transmission presumably pertain to the mechanism of barbiturate anesthesia. The investigations validated a method for estimating the pre- and postsynaptic contributions to synaptic plasticity. Hence, the present study obtained new knowledge about anesthetic interactions with the mechanisms of plasticity, perhaps relevant to drug-induced amnesia as well as unconsciousness. The effects of pentobarbital on axonal conduction (Blaustein, 1968) and shunting o f thalamocortical neuron firing (Wan and Pui l , 2002), suggested a presynaptic blockade o f action potential. Thus, it was worthwhile to compare the effects of N a + channel blockade with tetrodotoxin to those of pentobarbital. Chapter 1. Introduction I Ran - 3 3 -The high energy demand of repetitive stimulation and the effects of pentobarbital on metabolism (Quastel and Wheatley, 1932; Crane et al. 1978) provided rationale to examine whether conditions of energy shortage would promote STD. For this reason, S T D was examined during conditions of energy shortage imposed by glucose deprivations. The known depressant actions of barbiturates provided rationale to examine whether pentobarbital could reduce the effects of drugs that heighten excitability. The present studies investigated the effects of pentobarbital on thalamic hyperexcitability due to extracellular spermine. First, it was necessary to examine whether extracellular spermine would heighten excitability by actions on membrane electrical properties, synaptic activation, and firing modes of thalamic neurons. Only a few investigations have addressed this issue in C N S neurons, finding that millimolar concentrations o f spermine depress population excitatory postsynaptic potentials (EPSPs) mediated by N M D A and non-NMDA-receptors, as well as inhibitory postsynaptic potentials (IPSPs) in hippocampal C A 1 neurons (DiScenna et al., 1994; Eterovic et al., 1997). Secondly, I examined the interactions of pentobarbital with spermine for the presumed opposing actions o f the drugs on corticothalamic transmission, which modulated thalamic excitability. The in vivo effects suggested that pentobarbital might promote oscillogenesis in slice conditions. The present investigations also addressed pentobarbital effects on oscillatory behaviour in the corticothalamocortical network of neurons. In vitro oscillations were Chapter 1. Introduction I Ran - 34 -induced by corticothalamic stimulation in combination with pentobarbital application at sub- and anesthetic concentrations. Participating receptors were identified by pharmacological blockade as well as surgical separation of V B nuclei from the nRT. The frequency distribution of pentobarbital oscillations was determined by using spectrocorrelograms, obtained for continuous extracellular multi-unit recordings. These investigations facilitated the identification of a pro-oscillatory action of a subanesthetic concentration of pentobarbital on the C T C network. 1.5. Major questions Alterations in corticothalamic transmission may have a crucial role in oscillogenesis and modulatory mechanisms of anesthesia. This thesis w i l l focus on actions o f pentobarbital on thalamocortical excitability, including the modification of pre- and postsynaptic aspects of synaptic transmission. The studies addressed the following questions: 1. Is pentobarbital capable of inducing thalamic oscillations under slice conditions? 2. What receptors contribute to pentobarbital oscillations in vitro? 3. Does spermine enhance neuron excitability and does pentobarbital modify spermine action? 4. Are the assumptions made by the binomial depletion model valid during S T D at corticothalamic synapses? Does the analysis reveal changes in quantal parameters that are consistent with the binomial depletion model? Chapter 1. Introduction I Ran - 35 -5. Does pentobarbital alter STD? How do plastic alterations relate to changes in pre- and postsynaptic parameters? 6. During STD, in what ways are the actions of pentobarbital similar to selective N a + channel blockade? Does pentobarbital affect energy metabolism? Does glucose deprivation mimic pentobarbital actions? Chapter 2. Methods I Ran -36-Chapter 2 M E T H O D S 2.1. Slice preparation The Animal Care Committee at The University of British Columbia approved the procedures for these experiments. Experiments were performed on young adult Sprague-Dawley rats or gerbils (age 12-15 days) since they lack extensive myelination and are therefore ideal for proper formation of patch-clamp seals. Animals were decapitated while under deep isoflurane anesthesia. The cerebral hemispheres were quickly removed (~1 min) from the cranial vault and immersed for 1-2 min in ice-cold (0-2 °C) sucrose solution. The sucrose solution contained (in m M ) : sucrose, 248; NaHCC>3, 26; glucose, 10; KC1, 2.5; C a C l 2 , 2; M g C l 2 , 2; and N a H 2 P 0 4 , 1.25. The brain was quickly transferred to artificial cerebrospinal fluid (ACSF) , which had the same composition except for 124 m M N a C l instead of sucrose. In repetitive stimulation experiments, glucose concentration was increased from 10 to 25 m M and the following drugs added (in m M ) : myo-inositol 3, Na-pyruvate 2, and ascorbic acid 0.4. This altered A C S F composition (320 mOsm) enhanced to the vitality of repetitively stimulated neurons. The A C S F , on saturation with 95% O2 and 5% C O 2 , was adjusted to a p H of 7.3-7.4. The brain was trimmed into a cube (-0.5 cm ) containing the cortex and thalamus. A Vibroslicer (Campden Instruments, London, England) was used to cut 250 (im slices, for whole-cell recording, or 500 |om slices, for extracellular recordings. The slices were cut in a hybrid coronal plane that formed a 45° angle with a sagittal plane. The slices used for extracellular recordings were somewhat thinner than similar slices described by Tancredi et al. (2000). Chapter 2. Methods I Ran - 37 -The increased thickness (500 | jm compared to 250 Jim) facilitated the induction o f pentobarbital oscillations in vitro. The use of a hybrid cut slice maximized the corticothalamic fiber content and enhanced the ability to evoke excitatory postsynaptic potentials (EPSPs) or currents (EPSCs). In a series of extracellular recording experiments, a razor blade was used to surgically separate the ventrobasal ( V B ) nuclei from nRT. The slices were electrically stimulated by using a bipolar tungsten electrode (tip diameter -100 itm), placed in the slice at 0.2-0.3 mm from the recording electrode. For experiments performed in gerbils, horizontal slices were cut at 250 tun thickness that contained medial geniculate ( M G B ) and inferior collicular nuclei. In these slices, EPSPs were evoked by stimulating at a position mediodorsal to the M G B and near corticothalamic axons. Stimulation at this position resulted only in EPSPs. The stimuli consisted of single pulses of approximately 30 V in amplitude (range, 10-100 V ) and 100-200 |is in duration. The stimulation rate was 0.5 Hz . Using these stimulus parameters, it was possible to evoke inhibitory postsynaptic potentials (IPSPs) when the electrode was placed in the brachium, midway between the inferior colliculus and M G B . In experiments performed in rat thalamus, slices included portions of the V B thalamus, internal capsule, and nucleus reticularis thalami (nRT). The postsynaptic potentials and currents were averaged and fitted with an a- function (pClamp 8 software), yielding the rise and decay time constants. Finally, the slices were incubated for 2-3 h in A C S F at room temperature (22-25 °C), until required for recording, which was carried out at 21-25°C. Chapter 2. Methods I Ran - 38 -2.2. Drug application The slices were perfused on a nylon mesh with oxygenated A C S F and drugs. The drugs were prepared in distilled water, firstly as stock solutions at -1000 times the required concentration and then frozen. Just before the experiment, the stock solutions were thawed and diluted in A C S F for application. Pentobarbital, Mg 2 +-adenosine 5'-triphosphate salt (MgATP) , the C a 2 + chelators, ethylene glycol-bis-(p-aminoethyl-ether)-A W A ^ - t e t r a a c e t i c acid ( E G T A ) or l,2-bis(2-aminophenoxy)ethane-/A^V r/V'^V'-tetraacetate ( B A P T A ) , Na+-guanosine 5'-triphosphate (NaGTP), H E P E S , QX-314, C s C l , D-2-amino-5-phosphono-valerate ( A P V ) , cyclothiazide (CTZ), kynurenate ( K Y N ) , and the inorganic chloride salts were obtained from Sigma (St. Louis, M O ) . The drugs were diluted in A C S F and the p H adjusted in the range of 7.3-7.4. Extracellular solutions were delivered in two ways: 1) bath applications performed using a roller-type pump at a rate o f 2 ml/min through a submersion-type of chamber with a volume o f ~ 0.3 ml , and 2) local application by a glass pipette (-100-200 (am tip diameter) connected by polyethylene tubes to the various drug reservoirs. The local application approach allowed for a rapid switch (within < 5 s) between the various drugs (Quastel et al., 1971). 2.3. Extracellular recording Extracellular multiunit activity was recorded in lateral and medial portions of ventral posterior thalamus of submerged slices. The glass electrodes had tip diameters o f -1 | i m and resistances of 2-5 M Q when filled with 4 M N a C l . Aided by light microscopy (Zeiss Axioskop, Jena Germany), the electrode was positioned at 25-50 jam under the slice Chapter 2. Methods I Ran -39-surface and recorded multiple unit potentials with an isolation amplifier (World Precision Instruments, Sarasota, F L , U S A ) . The signals were bandpass filtered with series, low and high-pass 8 pole Bessel filters (cutoffs, 1 and 16 Hz). The signals were digitized at 5 kHz , and stored and analyzed (in part) with Axoclamp 8.2 software (Axon Instruments, Foster City, C A ) . 2.4. Whole-cell recording The electrical recordings were made in the current- and voltage-clamp modes of an Axoclamp 2 A amplifier (Axon Instruments, Foster City, C A ) . A pClamp 8.2 software (Axon Instruments) was used on a Pentium computer for data acquisition, storage and analysis. The voltage values were corrected for a measured junction potential of -11 m V . For voltage-clamp recordings, the recording electrode were coated with Sylgard and the volume of the bath solution lowered in order to minimize electrode and bath capacitance, respectively. In current-clamp experiments, the intracellular patch solution contained (in m M ) : K-gluconate, 140; /V-2-hydroxyethylpiperazine-/V-2-ethanesulfonate (HEPES) , 10; KC1, 5; N a C l , 4; adenosine 5-triphosphate (disodium salt), 3; guanosine 5-triphosphate (trisodium salt), 0.3; E G T A , 10; and C a C l 2 , 1. This combination of E G T A and C a 2 + yielded a final [Ca ] of 10 n M (calculated with M a x Chelator software). In experiments with B A P T A , E G T A was substituted with an equimolar concentration of B A P T A , which yielded a final [Ca 2 + ] of 1 n M . In voltage-clamp experiments, the patch solution contained (in m M ) : Cs-gluconate 125; TEA-C1 , 20; the lidocaine derivative QX-314, 3; H E P E S , 10; Na-phosphocreatine, 5; M g - A T P , 4; G T P , 0.4; and E G T A , 10, p H 7.3, 295-Chapter 2. Methods I Ran -40-300 mOsm. The intracellular presence of these drugs minimized the contribution o f postsynaptic N a + , K + , and C a 2 + channels while achieving an improved space clamp (Cahalan and Aimers 1979; Konishi 1990; Budde et al. 1994). 2.5. Detection of signals 2.5.1. Direct method A n even number of points (typically 4-10) were averaged around the point of largest value (Figure 2.1). This approach reduced the error due to noise at the local maximum point. The amplitude of the initial response was subtracted from baseline. Overlapping responses were obtained by subtraction of the single exponential fit o f the late component from the preceding response. . Highest point, highest noise Figure 2.1: Direct method of peak detection 2.5.2. Deconvolution method Another method for obtaining peak amplitude was deconvolution. The recorded signal represents the convolution of the time course of release of multiple quanta by the time course of a single quantum (Figure 2.2). Deconvolving the signal by the long x o f a quantum w i l l yield a value independent of superposition on the tail o f a previous signal. Chapter 2 . Methods I Ran -41-If a = e~l/T where x is the time constant of decay of individual quanta, then the deconvolution of a signal y; to produce: y j - yj-ayi_i/(l-a) excludes the components of signal amplitude due to the 'tail' o f a previous signal. The resulting peaks are, however, much noisier than the original signal. Recorded signal Deconvolved signal time Figure 2.2: Peak detection obtained by using deconvolution method. 2.5.3. First and second derivative method Peak were also detected by looking for zero crossings of the first and second derivatives (Figure 2.3). In this method, the zero crossing of the first derivative occurs at the location of the peak. The second derivative crosses zero at the point of maximal rise of the signal. The two points were used to obtain a single exponential fit that contained the peak Figure 2.3: First and second derivative method for peak detection. The first and second derivatives were used to detect the location of the peak (1) and the maximal rate of rise (2), respectively. The single exponential fit contained the peak estimate (1), above the zero crossing of the first derivative (Chen and Regehr, 1999). Chapter 2. Methods I Ran -42-Table 2.1: Comparison of peak detection methods Method Parameter Direct Deconvolution 1 s t and 2 n d derivative Si (nA) 1.25 ± 0 . 3 5 1.25 ± 0 . 2 4 1.23 ± 0 . 2 7 Sio (nA) 0.56 ± 0 . 1 1 0.53 ± 0.08 0.55 ± 0.09 S i / S 2 0.58 ± 0.08 0.56 ± 0.06 0.57 ± 0.09 S12/S10 1.35 ± 0 . 1 2 1.31 ± 0 . 1 3 1.34 ± 0 . 1 0 Var(Si) 0.023 ± 0.007 0.021 ± 0.005 0.024 ± 0.009 Cov(Si ,S 2 ) -0.010 ± 0 . 0 0 3 -0.012 ± 0 . 0 0 4 - 0.009 ± 0.005 Values are Mean ± S E M ; Si is E P S C amplitude where i corresponds to E P S C number Var - variance, Cov - covariance (averages from 5 neurons) Table 2.1 compares E P S C parameters obtained by the 3 peak detection methods. There were no significant differences between the 3 methods (P < 0.05, A N O V A test). Values obtained by the deconvolution method were normalized to the 1 s t peak value and rescaled by the mean amplitude obtained by the direct method. The direct and 1 s t and 2 n d derivative methods did not require any scaling. Since a choice of the peak detection method did not produce major errors in parameter values, the direct method was used throughout the study, for the data presented in the R E S U L T S section. Chapter 2. Methods I Ran - 43 -2.6. Repetitive stimulation Excitatory post-synaptic potentials (EPSPs) or currents (EPSCs) were evoked by stimulating a small portion of corticothalamic projections to V B thalamic neurons. For stimulation, a bipolar tungsten electrode was placed in the internal capsule (IC). The IC was first stimulated with a strong (100V, 50-400 LIS) stimulus, the amplitude of which was reduced as the electrode approached the surface of the fibers. Once the electrode was just above the IC fibers the amplitude of the stimulus was reduced to a value that was twice the minimal required to evoke EPSCs (or EPSPs). This indicated a contribution o f a small number of stimulated fibers. Under these conditions of stimulation, trains o f 20 stimuli were evoked at various frequencies (2.5-20 Hz) and statistical analysis performed using a binomial model. 2.7. Induction of plastic modifications of corticothalamic synaptic responses In early experiments, short-term depression was induced in a long train o f repetitive stimulation (50 pulses) at 2.5, 5, 10, and 20 Hz . Trains were applied with a 20 s inter-train interval to allow complete post-stimulation recovery. In most experiments, trains were 20 pulses; in preliminary experiments, the time course of post-stimulation recovery was assessed by applying trains with alternating inter-train intervals o f 1 and 10 s or more. Trains at the various frequencies were applied in a sequence that included all possible combinations. This approach avoided biasing results by possible 'memory' in the response from the previous frequency. For example, for frequencies 2.5, 5, 10 and 20 H z the sequence cycle was as follows: 10, 10, 20, 20, 5, 5, 2.5, 2.5, 10, 2.5, 10, 20, 5, 20, 2.5, 5. These stimulation protocols avoided time dependent correlations at a particular Chapter 2. Methods I Ran -44-frequency. B y omitting the 11 t h stimulus in trains of 20 stimuli, it was possible to assess the jump in response that reflects more time for 'refill' in the doubled gap between stimuli. 2.8. Analysis of extracellular recordings For multi-unit analysis, signal-to-noise ratio (SNR) was used for comparing the relative power density of voltage fluctuations (Gabbiani and Koch, 1998). Fast Fourier Transforms were applied to filtered voltage traces, binned at 20 ms over 1-15 H z range. Dominant frequencies were visualized with spectrocorrelograms ( M A T L A B 6.1). Autocorrelation functions ( M A T L A B 6.1) of consecutive data sweeps (10 ms bin width) were used to assess stationarity. 2.9. Fluctuation analysis of corticothalamic synaptic responses Amplitudes o f EPSCs at various locations in the train were summarized in tables showing their descriptive parameters. Namely, the E P S C amplitudes of the 1 s t (Si), 2 n d (S2), and averaged responses 15 t h - 20 t h (Plateau(Si5_2o)). The descriptive tables also contained E P S C amplitude ratios early (S2/S1) in the train, around the intra-train omitted stimulus (S12/S10), and between the plateau and the 1 s t response (Plateau/Si). The following equations were used to estimate the variance and covariance. For 2 successive EPSCs of amplitude S; and Sj+i that are repeated r times: Chapter 2. Methods I Ran -45-Var(S) = y (S -S )2 ll var is calculated using sequentia ' r-\~l ''r , r + 1 pairs of repeats r-x cov is calculated using cov(5,., SM) = — — £ fe, - St<r+l \Si+ir - Si+Ur+l )/2 sequential pairs of repeats of r r = 1 successive responses Theory of corrected variance/mean as a measure of quantal amplitude Assuming a binomial behavior (Vere-Jones, 1966; Quastel, 1997): The release probability p is the product of the output probability p 0 , determined by the readily available pool, and the eligibility probability p A , which depends on the rate o f refill and the stimulation frequency. Hence, p = P 0 P A where p 0 is the probability for an output from an 'available' site and p A is the probability for a site to be available. _ m _ m Q Also , p - — - ~~ ~ m - quantal content; N - number of release sites N NQ Var is expected to be: Var = m • Q2 • (1 - p) • (1 + CV2B) Between site variations Var = m • Q2 • (1 - p) • (1 - p + CV2) Within site variations where C V - coefficient of variation The mean is always simply m-Q. With predominantly between-site variation in Q (Auger and Marty, 2000), the variance to mean ratio gives: ^ = e o + c ^ ) . ( 1 - ^ ) = e . ( i + c r 1 j ) . ( 1 - = ) mean N * NQ Chapter 2. Methods I Ran -46-Consider now covariances. A t any one site i f there is no refill, i f there is a response at any one stimulus then there is none at the next, and vice versa cov(Sj, Sj) = < Sj Sj > - < S j>< Sj > cov(Sj, Sj) = 0 - < S; >< Sj > while Var (Sj) = < Sj >Q ( 1 + C V Q 2 ) - < Sj > 2 For N sites, S's, Var 's , and cov's are multiplied by N , hence Var (SO = < Si >Q(1+CV Q 2 ) - < S; > 2 /N cov (Si, Sj) = - < Sj >< Sj >/N That is, cov (Si, Sj) multiplied by < S; >/< Sj > is the same as the term in Var (Sj) that involves N . Thus, one can define an apparent quantal size, Q', that is the same as i f quantal release were Poisson distributed, for each (average) member o f the train (Scheuss andNeher, 2001) "apparent quantal size" = Q' = Q ( 1 + C V Q 2 ) = Var (Sj) /< Sj > - cov (S i ; Sj)/ < Sj > The same result is obtained i f variation of quantal size is within sites. If refill (a) is appreciable, < Si-Sj > becomes positive and covariances become less negative with increased separation of j from i (Vere-Jones, 1966; Quastel, 1997). A corollary is that correction of variance/mean ratio to give Q' is incomplete, and the best one can do is use j = i + 1 for choosing which covariance to use. Chapter 2. Methods I Ran -47-If, and only if, a covariance is non-zero, one can in principle estimate N from any two responses: < S >< S > N = - }-— valid only for cov(Sj, S,) < 0 in the absence o f refill cov / O O \ cov(S„Sj) c o v ( S „ S , ) However, 1 / Ncov ~ ~ — is a derived number that should correspond to the N calculated from the decline in calculated quantal contents of responses in trains. It is notable also that i f there is nonstationarity of Q between trains, e.g. i f local changes in conditions change 'shunting' between genesis and recording of signals, one obtains Var(Si) =< S, > 0(1 +CV2)(1 + f ) - + S,.2(f ) N (no refilD cov(Si,Sj) = < S ' > ^ S j > + < St x 5 . > (f) where y 2 is the between-train variance/mean2 which can be large relative to 1/N although much less than 1 (Quastel, personal communication). The net result is that between-train variation can lead to a diminished, nulled, or positive covariance, obscuring the negative covariance due to the binomial nature of the system. However, corrected variance-mean, Q', comes out the same as in the absence of between-train non-stationarity. The same result is obtained with between-train non-stationarity of N . Estimation of quantal content of signals The number of quanta, also referred to as the quantal content (m), was obtained separately from the initial five responses where EPSCs decline, and the subsequent Chapter 2. Methods I Ran - 48 -plateau. From the 6 t h to 10 t h stimulus, the quantal contents were obtained from the ratio of the size response to the corresponding quantal size: 'quantal content' = m, = <Si>/Q'j where i > 10 The quantal contents of the 1 s t response were obtained by dividing the response size by an average of the quantal size at the first stimulus of various frequencies (2.5, 5, 10, and 20 Hz) . The quantal contents of response 2-5 were obtained by dividing the response size by the quantal size averaged between the 2 n d and 5 t h response. In the absence of refill, theoretical covariances are, for any pair of outputs, in terms of S's, cov(Sj,Sj+i) = - <Sj>-<Sj+i>/N (see above), but with refill this becomes (Vere-Jones, 1966; Quastel, 1997): cov (st ,Sj)= '—^—J— + /(refill = a) That is, the negative covariance becomes smaller the higher cc, the probability of refill between stimuli. Evidently, this probability is greater the more stimuli are separated in time, causing cov(Si,Sj)/(<S;><Sj>) to be most negative when j = i + 1, and progressively less negative with j = i + 2, i + 3, etc. The presynaptic 'releasable store' and 'refill' According to the depletion model, signals fall in a train (STD) because each release o f quanta from a pre-existing store leaves the next stimulus with fewer quanta to draw upon. If release probability is constant, this model gives an exponential decline of signals to a near steady state at which outputs balance refill of the store. Because the data of Elmqvist and Quastel (1965a) showed consistently higher S2's than expected on this basis, they Chapter 2. Methods I Ran - 49 -therefore used a plot of <S> versus accumulated sum of previous <S>'s to obtain by extrapolation a number representing a total 'apparent presynaptic store' presumably equivalent to Q N , 'refill' being apparently small early in the train. This method was also used by Scheuss and Neher (2001). However, given negligible refill, one may also reason as follows: the first stimulus releases mi quanta leaving N - m\, the second m 2 quanta, leaving N - m\ - m 2 , etc. Soon, few are left i f fractional release (P 0) is more than ~ 0.3. Then the release is P 0 mult ipl ied by (N - mi - m2 - m^ ...etc.). The store (N) is therefore mx +m2 + m3/Po, or ra, + m 2 + m 3 + ra4 / P o , or m, + m 2 + w 3 + m 4 + ra5 / P0 etc. Therefore, given an estimate of P 0 , one has various estimates o f N , which become overestimates, because there is some refill, the further one goes along the train. A fair compromise between underestimation of N (at low P 0 ) and overestimation (because o f refill) is at m, + m2 + m3 + m4 IP0, or of Q N , using S's instead of w's, with P G estimated as 5, /(51, + S2 + S 3 + 5 4 ) o r S2 l(S2 + S3+ S4) whichever is larger (Quastel, personal communication), with the same assumption that Q's are constant. The result is less overestimation of Q N , when refill (a) is not negligible, than with the method of Elmqvist and Quastel (1965). A still better estimate of the 'releasable store' is Q N ' = QN/(1 + a) and in theory a is obtainable (Vere-Jones, 1966) since at equilibrium, Seq=QaP0-N/(a+P0-aP0) while SX=Q-P0N giving P0 = Sx l{Q • N) Chapter 2. Methods I Ran -50-Therefore, defining x = Seq I Si = a I (a + Po - a • PQ), rearranging gives a = x-Pgl(\-x + xP0) However, this estimate is highly sensitive to error in the estimate of Q N , and therefore o f P 0 and i f a is not small in the early part of STD wi l l result in an underestimate of P 0 and a that cannot be corrected without information as to how P 0 and a change during the train. In the tables summarizing the data, Q N was used simply as a descriptive measure, equal to Sx + S2 + S3 + S41p , with p being Sx + S2 + 53 + S4)or S2 /(S2 + S 3 + 54) whichever is larger. The jump after an omitted stimulus In the plateau phase, outputs are the same. The number of quanta present at stimulus j is mj and release is my = p • nj where p is fractional release. This leaves a store of nj - p • n, and refill is a- (N - nj • (1 - p)). Since the next store is the same, n j + 1 = n j = n j - (1 - p) + a- ( N - n j - (1 -p)) and rearranging, nj - nj- (1 - p) + a- nj- (1 - p) = a-N nj • (p + a - a-p) = a-N nj = a-N/(p + a - a-p) This expression, derived by Vere-Jones (1966), was used above to obtain a. Now, i f stimulus j+1 is omitted, there is further filling, namely, a- (N - nj). Chapter 2. Methods I Ran -51-After the gap, the store is n j+2 = rij + a- (N - rij) = rij • (1 - a ) + a -N = rij • (1 - a ) + rij-(p + a - a-p) = n j - ( l + p - a - p ) On the assumption that p does not change after the gap, m j + 2 /mj = n'j+2/n'j = 1 + p- (1 - a ) 2.10. Statistical comparisons A Student's Mest was used for comparing two groups and an analysis of variance ( A N O V A ) test for comparing more than 2 groups. In some cases, a Wilcoxon signed rank test was used for matched pairs of groups. P < 0.05 was considered significant. Chapter 3. Results I Ran -52-Chapter 3 R E S U L T S Parts of the results in the foregoing section have been published (Ran et al., 2004). Part I. Pentobarbital oscillations in vitro in ventrobasal thalamus 3.1. Extracellular effects of pentobarbital 3.1.1. Pentobarbital application and internal capsule stimulation Pentobarbital (PB) application (20 or 200 | i M ) did not produce oscillations in 6 out o f 6 V B slices. Electrical stimulation of the internal capsule also did not produce oscillations in 12 out of 12 slices. Spontaneous oscillations were not observed during these conditions. However, P B at 20 u M in combination with stimulation at 0.05 H z induced oscillations at 0.3-1 s intervals (Figure 3.1A) in all slices (n = 69). The oscillations typically lasted < 20 s from the stimulus onset and had a frequency distribution that remained approximately stationary (Figure 3.IB), as confirmed by autocorrelation analysis. Figure I C shows that P B increased the S N R near 3 and 8 Hz , with smaller increases near 11 and 13 Hz . Since electrical stimulation was essential, P B application was a necessary, but not a sufficient condition for evoking oscillations. For convenience, these oscillatory responses wi l l be referred to as "PB-induced oscillations". 3.1.2. Effects of reduced extracellular Mg2+ (fMg2+Je) Application o f P B increased the oscillations in slices made hyperexcitable in low [ M g 2 + ] e media (cf. Tancredi et al. 2000; Jacobsen et al. 2001). Perfusion with low [ M g 2 + ] e (0.65 m M ) with, or without combined P B application did not result in oscillations Chapter 3. Results I Ran •53-A Control PB 20 uM H ^ j v ^ ^ - j l H ^ - - - — I n d * — | t | Wash 150 uV 1 s B 15 -10 o c o BT 5 Control . PB (20 uM) ( Wash r 10 20 Time (min) 5 10 Frequency (Hz) 15 Figure 3.1. Pentobarbital induces extracellular oscillations in ventrobasal nuclei during electrical stimulation of internal capsule at 0.05 Hz . A ) Pentobarbital (PB) reversibly evoked oscillatory discharge at 0.3-1 s intervals. B) Spectrocorrelogram of activity in slice (A) illustrates that P B (horizontal bar) induced oscillations in the 1-13 H z range. The greyscale gradations correspond to the magnitude (black, highest; white, zero) of oscillations as a function of frequency. C) Signal-to-noise (SNR) is shown as a function of frequency, averaged from n = 6 slices in control (open circles) and n = 6 slices bathed in 20 j i M P B (closed circles). Chapter 3. Results I Ran - 5 4 -Low [Mg2T B PB 20 uM JWMfa H'M = T' M. i J '• • i 'il n Wash 150 1 s low [Mg2*] PB (20 uM) Wash 10 20 Time (min) -i r 5 10 Frequency (Hz) 15 Figure 3.2. Pentobarbital-induced oscillations in low M g 2 + medium. A ) Experiment similar to Figure 3.1 was conducted in a slice bathed in 0.65 m M [Mg 2 + ] (low [ M g z + ] e ) ,2+n Oscillations appeared in low [ M g 2 + ] e during electrical stimulation. B) P B application reversibly extended the frequency range of firing from 5-10 Hz , to 1-15 Hz . C) S N R versus frequency plots were averaged from 6 slices in low [ M g 2 + ] e medium, before (open circles) and after 20 u M P B (closed circles). P < 0.001 in C. Chapter 3. Results I Ran - 55 -in 6 out 6 slices (Figure 3.2). Perfusion with low [ M g 2 + ] e (0.65 m M ) and electrical stimulation resulted in oscillations at 5 to 9 H z in 14 out of 14 slices (cf. Figure 3.1). Under these conditions, P B application increased oscillatory activity in 10 out of 10 additional slices. Figure 3.2B shows that P B application reversibly intensified the oscillations and extended their frequency range from 5-9 Hz , to 1-15 H z . Application o f P B increased the S N R predominantly near 8 Hz , and to lesser extent near 3, 11, and 13 H z (Figure 3.2C). PB-induced oscillations did not apparently depend on the hyperexcitability due to low extracellular [ M g 2 + ] e . 3.1.3. Effects of raised temperature. It was necessary to determine i f the frequencies of PB-induced oscillations were temperature dependent, as shown in vivo (Andersen and Andersen, 1968). Spontaneous oscillations were not observed in slices at 34 °C. Figure 3.3 illustrates a 10 °C increase in temperature did not produce significant alterations in the S N R over the 1-15 H z range (n = 6). Since the higher temperature did not significantly affect PB-induced oscillations, the study was conducted using slices maintained near 24 °C, promoting slice viability. 3.1.4. Application of high dose pentobarbital A 10-fold increase in P B concentration induced oscillations which subsequently ceased, despite continuing P B application with internal capsule stimulation (n = 6). Figure 3.4 shows this biphasic effect for P B at 200 j i M . The frequency range (1-15 Hz) at 3 min was similar to that observed with P B at 20 | i M (Figure 3.4B). The frequency range Chapter 3. Results I Ran -56-PB 20 uM, 24 ° C 3 4 ° C 150 uV 1 s B 5 10 Frequency (Hz) Figure 3.3. Effects of raised temperature on pentobarbital oscillations. A n increase in temperature from 24 °C to 34 °C did not alter the discharge frequency in a slice (A) or the mean S N R (B) during oscillations induced by P B (20 u\M). Electrical stimulation (0.05 Hz) was applied throughout, as in Figure 3.1. n = 6, P > 0.05 in B . Chapter 3. Results I Ran - 5 7 -Figure 3.4. Time dependence of effects of increasing concentrations on pentobarbital oscillations. A ) A t 200 (i.M, P B application evoked oscillations in a slice (24 °C) at 3 min which disappeared within 9 min of the application. B) S N R (n = 6) showed an increase at 3 min (closed circles) which disappeared within 9 min of the application (open circles). Electrical stimulation (0.05 Hz) applied throughout, as in Figure 3.1. P < 0.01 in B. Chapter 3. Results I Ran - 58 -narrowed to 1-4 H z before disappearance of the oscillations at ~9 min. The biphasic effect of P B at 200 u M occurred in 6 out of 6 additional slices under low [ M g ] conditions and electrical stimulation (not shown). Hence, subsequent studies focused on 9+ the stable oscillations induced by P B at 20 | i M under normal [Mg ] conditions. 3.7.5. Effects of synaptic receptor blockade A possible involvement of glutamate receptors in PB-induced oscillations was examined, as found for the oscillations induced by electrical stimulation and low [ M g 2 + ] conditions (Tancredi et al., 2000; Jacobsen et al., 2001). Application of kynurenate (1 m M ) , an ionotropic glutamate receptor blocker, reversibly abolished PB-induced oscillations in 5 out of 5 slices (not shown). Hence, glutamatergic transmission was essential for the P B -induced oscillations. Experiments using applications of G A B A antagonists showed that GABA-receptors l ikely participate in PB-induced oscillations. During electrical stimulation, bicuculline methiodide (50 | j M ) reversibly induced oscillations at 1-4 H z (n = 3, not shown). In comparison with P B , co-application of bicuculline with P B resulted in reduced oscillogenesis at 5-15 H z in 6 out of 6 slices, without apparent changes in the 1-4 H z oscillations (Figure 3.5A). However, recent studies have shown that bicuculline methiodide may have effects in addition to G A B A A receptor blockade (Debarbieux et al., 1998; Seutin et al., 1997) that could account for the depression of oscillations. Chapter 3. Results I Ran - 59 -The question of receptor specificity was further examined by applying another G A B A A antagonist, gabazine (Uchida et al., 1 9 9 6 ; Seutin et al., 1 9 9 7 ) . Application o f gabazine (20 L I M ) reversibly induced oscillations at 1-4 H z in 6 out of 6 slices, similar to bicuculline (not shown). In comparison with P B , co-application of gabazine (20 j i M ) with P B resulted in reduced oscillogenesis at 5 - 1 5 H z in 6 out of 6 slices, without apparent changes in oscillations at 1-4 H z (Figure 3 . 5 B ) . Therefore, it seemed likely that gabazine- and bicuculline-sensitive G A B A A receptor interactions participated in P B -induced oscillations at frequencies above 4 Hz . Application of the G A B A B receptor antagonist, C G P 35348 (100 nM) during electrical stimulation, reversibly induced oscillations at 5-15 H z in 3 slices (not shown). Co-application of C G P 35348 with P B resulted in reduced oscillogenesis at 1-4 H z and 11-15 Hz , sparing the 5 to 10 H z range in 6 out of 6 slices (Figure 3.5C). Thus, G A B A B antagonism induced oscillations centred near 8 Hz , and during P B application, resulted in suppressed oscillations at lower and higher frequencies. Combined antagonism by bicuculline and C G P 35348 did not evoke oscillations during electrical stimulation in 5 slices (not shown). Unexpectedly, oscillations in the 5-10 H z range persisted during co-application of G A B A receptor antagonists with P B in 6 out o f 6 slices (Figure 3.5D). Hence, G A B A A and G A B A B mediated inhibition was apparently not an absolute requirement for oscillations during P B application. Chapter 3. Results I Ran - 6 0 -B C Frequency (Hz) Frequency (Hz) Figure 3.5 Antagonists of G A B A , and glycine receptors modulate frequency of pentobarbital oscillations. A - F (n = 6 in each panel) show mean S N R as a function o f frequency during application of 20 u\M P B . A ) Bicuculline (BIC, 50 | i M ) decreased the S N R of the oscillations at middle, and especially at high frequencies (P < 0.01). B ) Gabazine ( G B Z , 20 | j M ) also decreased the S N R of the oscillations in the middle and high frequencies, similar to bicuculline in A (P < 0.01). C) C G P 35348 (CGP, 100 u M ) decreased the S N R of the oscillations at low and high frequencies (P < 0.01). D) Combined application of BIC and C G P 35348 did not eliminate P B enhancement of the S N R in the middle frequency range (P < 0.01). E) Strychnine (STR; 1 ixM) decreased the S N R of the oscillations at low and high frequencies, similar to C G P 35348 application (P < 0.01). F) S T R (1 nM).combined with BIC and C G P 35348 abolished P B enhancement of S N R (P < 0.01). In A - F , the S N R for control P B oscillations is shown as a dotted reference plot. Horizontal line, S N R = 1. Chapter 3. Results 1 Ran - 61 -In view o f the incomplete blockade of all oscillation frequencies by G A B A receptor antagonism, it was necessary to apply picrotoxinin, which blocks at a different site on the G A B A A receptor than bicuculline or gabazine, as well as inhibiting homomeric glycine receptors (Yoon et al., 1998). Picrotoxinin (50 (iM) during electrical stimulation did not evoke oscillations in 6 out of 6 slices. Application of picrotoxinin (50 u M ) (n = 4), alone, or co-applied with C G P 35348 (n = 2), eliminated PB-induced oscillations (not shown). These results are consistent with picrotoxinin blockade o f G A B A A and glycine receptors. The next experiment determined the effects of strychnine, a glycine-receptor antagonist, on PB-induced oscillations. B y itself, strychnine (1 u.M) induced oscillations, but not during G A B A receptor antagonism. These effects were not studied further in these investigations. During P B application, strychnine (1 L I M ) reversibly decreased the oscillations at low and high frequencies, sparing those at 5-10 H z in 5 out o f 5 slices (Figure 3.5E). This effect was similar to that of C G P 35348. During G A B A receptor antagonism, co-application of strychnine with P B resulted only in rudimentary oscillations in 5 out of 5 slices (Figure 3.5F). Hence, both G A B A and glycine receptors apparently modulated PB-induced oscillations. 3.1.6. Pentobarbital-induced oscillations in separated thalamic nuclei. The possibility that G A B A e r g i c connections from nRT were necessary for PB-induced oscillations was examined by studying the effects of P B in V B nuclei and nRT, isolated from each other (Figure 3.6). In V B nuclei surgically isolated from nRT (Figure 3.7A), P B application evoked oscillations at > 5 H z during electrical stimulation at a V B Chapter 3. Results I Ran -62-site that was 1-2.5 mm from the recording electrode (Figure 3.6). G A B A receptor blockade by co-applied bicuculline and C G P 35348 had no significant effect on the oscillations (not shown, n = 5). Hence, receptors for G A B A did not appear to have a modulatory role in V B nuclei, isolated from nRT. In view o f the persisting oscillations during G A B A receptor blockade, the effects o f strychnine in V B nuclei were determined, after isolation from nRT. Strychnine (1 p M ) reversibly eliminated PB-induced oscillations (Figure 3.7; n = 6). This implied that P B actions on glycine receptor systems were essential for the oscillations in V B nuclei, deprived of G A B A e r g i c inputs. Since disconnection from nRT altered the frequency of oscillations in V B nuclei, the effects o f P B in nRT were determined, before and after its disconnection from V B nuclei. Before disconnection, P B application (20 p M ) induced oscillations at 1-12 H z in nRT during electrical stimulation of the internal capsule in 5 out of 5 slices (1-10 H z in Figure 3.8). Surgical disconnection from V B nuclei in these 5 slices did not significantly affect the ability of P B application to induce nRT oscillations in a similar frequency range (Figure 3.8). Hence, P B can induce oscillations in nRT, at a site distinct from V B nuclei. G A B A receptor blockade by co-applied bicuculline and C G P 35348 abolished the oscillations induced by P B in nRT (not shown). This occurred with (n = 6), or without Chapter 3. Results I Ran -63-Figure 3.6. Photomicrograph of sagittal slice shows complete separation (asterisk) o f V B nuclei from nRT. Cx , cortex; Hipp, hippocampus; IC, internal capsule; nRT, nucleus reticularis thalami; Po, posterior thalamic nucleus. Vertical arrows: rostral, R; caudal, C . Chapter 3. Results I Ran -64-Isolated VB (no nRT) PB PB+STR 5 10 Frequency (Hz) 1 5 Figure 3.7. Pentobarbital oscillations in electrically stimulated V B nuclei, after surgical separation from nRT. Top: In a slice (top and middle), P B (20 p M ) evoked 6-15 H z oscillations in V B nuclei, after separation from nRT. Middle: Co-application with S T R (1 ixM) abolished these oscillations. Bottom: Plot of S N R as a function of frequency quantifies the effects of STR in 11 slices (P < 0.01, Mest). Chapter 3. Results I Ran -65-Recording in nRT 0 10 20 30 0 10 20 30 Time(min) Time(min) Figure 3.8. Pentobarbital oscillations in electrically stimulated nRT, before and after its surgical separation from V B nuclei. A ) P B (20 p M ) induced oscillations at 1-10 H z in nRT during stimulation of the internal capsule. B) in a another slice, P B (20 p M ) induced oscillations at 1-9 H z in nRT, electrically stimulated at 0.05 H z after surgical separation from dorsal thalamic nuclei. Chapter 3. Results I Ran - 66 -(n = 6) connections to V B nuclei. Strychnine application had no significant effect on the oscillations induced by P B in nRT, with (n = 5) or without (n = 5) connections to V B nuclei (not shown). Hence, P B can induce purely G A B A e r g i c oscillations in nRT, separated from V B nuclei. 3.1.7. Discussion Pentobarbital oscillations required ionotropic glutamate excitation, but not elevation of temperature from 24° to 34° or low [Mg 2 + ] conditions. Although they can occur spontaneously under different conditions (Jacobsen et al., 2001), oscillations were never observed without electrical stimulation, in the present study. The oscillations had a broader frequency range than seen with low extracellular [ M g 2 + ] . Hence, pentobarbital l ikely acted on M g independent sites. Receptors for G A B A modulated pentobarbital-induced oscillations. G A B A A antagonism with bicuculline methiodide or gabazine eliminated oscillations at 10-15 Hz , but not in the lower frequency range. Bicuculline and gabazine had equivalent effects, with and without co-applied pentobarbital. The modulation of pentobarbital-induced oscillations was not likely due to unselective actions of bicuculline methiodide (cf. Debarbieux et al., 1998). Gabazine blocks G A B A A receptors (Uchida et al., 1996) without producing the effects of bicuculline methiodide on intrinsic membrane currents in C N S neurons (Seutin et al., 1997). The above results imply that G A B A A receptors modulated the oscillations in the high frequency range. Chapter 3. Results I Ran - 67 -Picrotoxinin abolished oscillations induced by pentobarbital, in marked contrast to the other G A B A A antagonists. Picrotoxinin itself did not induce oscillations, in agreement with previous reports (Jacobsen et al., 2001). In addition to blocking G A B A A receptors, picrotoxinin inhibits homomeric glycine receptors (Yoon et al., 1998). Thus, a picrotoxinin-blockade of both G A B A A and glycine receptors apparently mimicked the effects of co-applied strychnine and bicuculline. Apparently, G A B A B receptors modulated the oscillations in both low and high frequency ranges. The oscillations persisted in the 5-10 H z range during C G P 35348 (Jacobsen et al., 2001 and Ziakopoulos et al., 2000) or strychnine applications. Unexpectedly, G A B A receptor blockade did not abolish the oscillations, rather they persisted in a 5-10 H z range. This reflected an altered network function because pentobarbital-induced oscillations in the same frequency range in ventrobasal nuclei deprived of reticular G A B A e r g i c inputs. After this surgical isolation, or during G A B A receptor blockade, strychnine application eliminated the oscillations. In nRT disconnected from ventrobasal nuclei, strychnine did not affect the oscillations mediated by glutamate and G A B A . In summary, pentobarbital induced oscillations in isolated networks of the ventrobasal and reticular nuclei, mediated by glutamate receptors and modulated by overlapping interactions at G A B A A , G A B A B , and glycine receptors. Chapter 3. Results I Ran - 68 -Parts of the results in the following section have been published (Ran et al., 2003). Part II. Pentobarbital modulation of N M D A receptors in corticothalamic transmission The present section of the thesis addressed the issue of how modulation of N M D A receptors affects the excitability of thalamic neurons. N M D A receptors are located at distal synapses on thalamic neurons and receive extensive cortical inputs. Abnormal modulation of N M D A receptors may result in thalamic hyperexcitability, which contributes to some forms of epilepsy. A depressant action of pentobarbital may reduce the effects caused by such abnormal modulation. The following experiments examined how pentobarbital affects the modulation of thalamic excitability by spermine, an endogenous polyamine with extracellular modulatory effects on N M D A receptors. 3.2.1 Effects of spermine 3.2.1.1. Tonic firing Spermine application reversibly increased the number of action potentials in all neurons depolarized from rest by current pulse injection. Spermine (100 |JM) applied for 3 - 6 min induced tonic firing of action potentials on top of subthreshold responses. When action potentials were present in the control, spermine application increased the rate of firing (Figure 3.9A). Long recovery times of 35 - 45 min characterized spermine's effects on thalamic firing modes after 6 min applications. In the neuron of Figure 3.9A, substantial recovery occurred at ~ 32 min after discontinuing the spermine application. Chapter 3. Results I Ran -69-The spermine-induced increase in the firing frequency was concentration-dependent over the range of 5 0 and 5 0 0 J J M (n = 19 , Figure 3 . 9 B ) . In addition, application o f 1 | i M spermine did not affect firing (n = 2); however, at 1 m M , there was a marked increase in the firing rate (n = 2), without any apparent recovery (data not shown). Spermine, applied at an ED50 o f 1 0 0 L L M (cf. Figure 3 . 9 B ) , reversibly increased the number o f action potentials per pulse by an average of ~ 8 0 % in nine neurons (control, 1.8 ± 0.3 action potentials/pulse and spermine application, 3.3 ± 0 .4 action potentials/pulse, paired t-test, P < 0 . 0 1 ) . 3.2.1.2. Passive membrane properties The increased firing due to spermine did not likely result from changes in the passive membrane properties which did not greatly change during 3 to 6 min spermine applications (cf. subthreshold responses in Figures. 3.9A and 3.11 A ) . The average resting potentials were -67 ± 4 m V during the control period, and -66 ± 5 m V during applications of spermine at 50 - 500 \iM (n = 19). Spermine application did not significantly change the mean membrane time constant (x m = 64 ± 6 ms in control, and 76 ± 6 ms during spermine application, paired Mest) and mean input resistance (Rf= 772 ± 3 8 M Q in control, and 756 ± 6 1 M Q during spermine application, paired Mest), computed from the responses to hyperpolarizing current pulse injections in 19 neurons held at -65 m V . The lack o f effect on passive membrane properties may result from spermine actions on distal dendrites. Hence, the spermine-induced effects on the passive properties could not account for the changes in firing threshold. Chapter 3. Results 1 Ran - 7 0 -Figure 3.9. Spermine enhanced tonic firing in a concentration-dependent manner in M G B neurons. A ) Spermine application (100 L I M , 3 min) enhanced action potential firing evoked by current pulses (25 and 50 p A , 500 ms, 1.5><threshold). Holding potential, - 65 m V . Vertical upper bar, 30 m V and lower bar, 60 pA; horizontal bar, 150 ms. B ) Increase in number of action potentials per pulse was concentration dependent. The control firing was 1.8 ± 0.3 action potentials/pulse (n =9). The ED50 was - 1 0 0 L I M for spermine-enhanced firing which approached saturation at 200 j i M . Chapter 3. Results I Ran - 7 1 -3.2.1.3. Action potential threshold Spermine (100 u M ) decreased the latency to tonic firing by decreasing the threshold (Figure 3.10). Spermine decreased the firing threshold from -51.0 ± 0.6 m V to -57.1 ± 2.2 m V (Figure 3.10B). Significant changes in action potential amplitude did not accompany the decreased threshold. Figure 3.10B summarizes the effects of spermine on firing threshold for six neurons. The possibility that N M D A receptors mediated the effects on the firing threshold and tonic firing rate was examined by determining the interactions of spermine and the competitive antagonist, A P V (50 piM). A s shown in Figure 3.10B, spermine reduced the threshold voltage for an action potential evoked with a 500 ms current pulse, by an average of 6.2 ± 1.1 m V . On recovery from spermine (washout, Figure 3.10B), application of A P V alone, or in combination with spermine, did not significantly change the firing threshold (during A P V , - 52.1 ± 1.9 m V and A P V + spermine, -53.0 ± 1.7 m V ; n = 6) or changes in membrane properties that could account for the blockade o f the spermine-induced reduction in firing threshold. This signified that A P V acted on N M D A receptors to completely block spermine action. Since these neurons had received a spermine application prior to A P V , A P V was also applied to five neurons that had not previously received a drug application in order to assess the possibility of constitutive release o f glutamate in the slice. Here, A P V produced an increase in threshold, which remained largely unaltered by a subsequent, combined application with spermine (Figure 3.10C). A l l neurons showed substantial recovery at 15 min after discontinuing the Chapter 3. Results I Ran -72-A J Control Spermine Washout APV +APV Washout + APV Figure 3.10. Spermine (100 p M ) decreased the action potential threshold. (A) Spermine application produced a leftward shift in action potential latency (current pulse duration, 500 ms). Arrows point to threshold in control and spermine. (B) N M D A receptors mediated spermine effects on action potential threshold. The control threshold was -50.9 ± 0.6 m V , which spermine reduced to -57.1 ± 2.2 m V (n = 6). Partial recovery was observed after 15 min (-52.4 ± 0.6). Blockade of N M D A receptors by A P V (50 p M ) reduced the threshold by < 1 m V . A reduction in threshold was not observed during co-application of A P V and spermine (-0.9 ± 0.6 m V , n = 6). (C) A P V (50 p M ) increased firing threshold in five neurons from -52.3 ± 0.7 to -48.7 ± 0.5 m V . Subsequent spermine application did not alter the increased threshold (-49.1 ± 0.6 m V ) . Two-way A N O V A ; * in (B) indicates P < 0.01 and in (C) P < 0.05. Vertical bar in (A), 15 m V ; horizontal bar, 50 ms. Chapter 3. Results I Ran -73-application. These data implicated an N M D A receptor mechanism in the spermine-induced decrease in the threshold. N M D A receptors also mediated the spermine-induced increase in firing rate. In the neuron o f Figure 3.11 A , spermine application (100 ( i M , 3 min) reversibly increased the number of action potentials during a 500 ms current pulse injection from one action potential in the initial control, to three action potentials. In the neurons that had not previously received a spermine application, an increase in the current pulse amplitude during action potential blockade due to A P V application, produced a return o f the action potential (Figure 3.1 IB) . The A P V blockade of spermine-induced increase in firing also was overcome by an increase in the current amplitude. The APV-induced blockades o f action potentials and spermine enhancement of firing were not attributable to an increased input conductance and were completely reversible. The graph of Figure 3.1 I C summarizes the data that implicated N M D A receptor mediation. It was necessary to examine the possibility that n o n - N M D A receptors for glutamate contributed to the increased firing during spermine application. These studies determined the interactions of spermine with an A M P A receptor antagonist, C N Q X . Application o f C N Q X (30 f iM) for 6 min did not result in significant changes in evoked action potential firing, configuration, or membrane electrical properties. Subsequently, combined C N Q X and spermine application did not greatly alter the reduction in threshold and ~ 200% increase in firing rate evoked by current pulses (amplitude ~1.5 x threshold), as observed with prior spermine application in all five neurons ( C N Q X , 1.4 ± 0.3 action potentials per Chapter 3. Results 1 Ran - 7 4 -Control Spermine Washout APV APV+ Washout Spermine , Figure 3.11. Spermine increased tonic firing by interacting with N M D A receptors. (A) Spermine application (100 L I M , 3 min) reversibly induced firing. After a 15 min washout from spermine, A P V (50 | i M , 6 min), an N M D A receptor antagonist, blocked the evoked action potential. Firing was not observed during co-application of A P V and spermine (3 min). Washout shows recovery at 10 min after discontinuing the co-application. Lower traces show hyperpolarizing tests for input resistance. (B) Application of A P V (50 u M , 6 min) abolished firing induced by just-threshold current pulse (40 pA) . A subsequent 3 min co-application of spermine and A P V did not alter this suppression (lower superimposed traces in middle panel). A two-fold increase in current amplitude overcame the blockade during A P V application, alone, or during co-application with spermine (upper superimposed traces in middle panel). Recovery was observed after 10 min washout. (C) Summary of spermine effects on firing in six neurons. A N O V A ; *P < 0.01 - significantly different from control, **P < 0.05 - significantly different from spermine. Vertical bar, 30 m V ; horizontal bar, 200 ms. Chapter 3. Results I Ran -75-pulse, and CNQX+spermine, 4.2 ± 0.4 action potentials/pulse; data not shown). Hence, the increase in tonic firing rate due to spermine application did not likely involve A M P A receptor interactions. 3.2.1.4. Membrane rectification The following experiments examined the possibility that the spermine-induced increase in tonic firing involved voltage-dependent membrane properties. For example, thalamocortical neurons exhibit larger voltage responses to depolarizing, compared to hyperpolarizing current pulses (Tennigkeit et al., 1996; Parri and Crunelli, 1998; cf. Figure 3.12B). Spermine (100 p M ) application for 3 min increased inward rectification in a range between the rest and firing threshold, but did not appreciably change the responses at hyperpolarized potentials, down to - 100 m V (n =19; Figure 3.12A). Quantification of the increase in rectification on depolarization was difficult because spermine application shortened the latency to firing (cf. arrows in Figures. 3.1 OA and 3.12A). Application of A P V (50 p M , 6 min) completely blocked the rectification in the upper right quadrant of the voltage - current (V - 1) relationship. A subsequent co-application with spermine (100 p M ) did not greatly change this curve. The graph of Figure 3.12 A (right) summarizes these findings for six neurons. There was little or no involvement of A M P A receptors in the spermine-induced (100 p M , 3 min) enhancement of rectification produced by depolarizing current pulses. C N Q X (30 p M , 6 min) did not alter spermine's effects on the rectification in five neurons. The average voltage response during co-application of C N Q X and spermine (15.9 ± 0.6 m V ) Chapter 3. Results I Ran - 76 -was significantly different from control (11.5 + 0.5 mV) or C N Q X application (11.3 ± 0.4 m V ; A N O V A , P < 0.05). Hence, the spermine-induced increase in the depolarizing responses involved N M D A receptors, but likely not A M P A receptors. The next investigations examined whether spermine increased the rectification in the upper right quadrant of the V - I relationship by interacting with a persistent N a + conductance. The rectification observed on depolarization from — 7 0 m V to threshold involved a persistent N a + conductance, sensitive to T T X blockade (Tennigkeit et al., 1996; Parri and Crunelli, 1998). Blockade of voltage-dependent N a + channels with T T X (0.6 p M , 6 min) decreased the slope of the V -1 relationship, more in the depolarizing quadrant than in the hyperpolarizing quadrant. The blockade with T T X nullified the ability o f spermine (100 p M , 3 min) to increase rectification on depolarization (n = 6 ; Figure 3.12B). The results imply that spermine increased rectification in the upper right quadrant of the V -1 relationship by increasing a TTX-sensitive, voltage-dependent N a + conductance. The spermine-induced enhancement of rectification on depolarization of the neuron also 2+ may depend on extra or intracellular Ca , as in neocortical neurons (Cri l l , 1996). Hence, the spermine-induced enhancement of voltage responses to depolarizing current injections were measured during intracellular application of B A P T A (10 m M ) and 2+ extracellular perfusion with nominally Ca - free A C S F . In the neuron of Figure 3.13A, 2+ perfusion of C a - free A C S F did not greatly alter these depolarizing responses. In four neurons, a 50 p A current pulse evoked average responses of 9.8 ± 0.6 m V in control Chapter 3. Results I Ran -77-Figure 3.12. Effects of spermine (100 | J M , 3 min) on membrane rectification. (A) Voltage -current (V - I) relationship of a neuron shows that spermine increased depolarizing response which was abolished during combined application (3 min) with A P V . A P V (50 L I M , 6 min), alone, reduced rectification in upper right quadrant. V -1 curve after 15 min washout shows substantial recovery. Graph at right summarizes effects o f spermine, A P V , and their co-application on rectification. The response on depolarization increased from 12.5 ± 0.5 m V (control) to 17.5 ± 0.8 m V during spermine application, and after 15 min washout, recovered to 12.9 ± 0.7 m V . Subsequent A P V application reduced the response on depolarization from 12.9 ± 0.7 m V (first washout) to 9.2 ± 0.4 m V ( A P V ) . Co-applied A P V and spermine did not greatly alter rectification (8.7 ± 0.3 m V ) . Recovery from A P V and spermine occurred after 15 min (12.7 ± 0.5 m V ) . Holding potential, -70 m V . (B) V - I relationship for a neuron shows that T T X application (0.6 u M , 6 min) decreased rectification on depolarization over a ~10 m V range. Co-applied with T T X , spermine did not alter rectification in upper right quadrant. A 20 p A pulse was sufficient to observe rectification on depolarization, whereas a - 50 p A pulse produced little or no rectification on hyperpolarization. Graph at right summarizes effects of T T X and spermine on rectification. Rectification on depolarization decreased from 13.1 ± 0 . 8 m V (control) to 10.2 ± 0 . 8 m V during T T X application. A subsequent co-application with spermine did not greatly alter depolarizing responses (10.8 ± 0.7 m V ) . Inserts in upper left quadrants of (A) and (B) show superimposed responses (7 m V ) to depolarizing and hyperpolarizing current pulses (duration 500 ms) of 60 and - 60 p A , during control (C), spermine (S), and at 3 min of co-application of T T X and spermine (TTX+S). Bar graph values are mean ± S . E . M . A N O V A test, n = 6; * and **P < 0.05. Chapter 3. Results 1 Ran - 78 -2+ A C S F and 10.0 ± 0.5 m V in 0 m M [Ca ]. The response increased to 15.2 ± 1 m V during 2+ spermine application (in 2 m M Ca A C S F ; A N O V A , P < 0.01) which did not 2+ significantly change during combined application of spermine and 0 m M [Ca ] (10.2. ± 0.3 m V , n = 4 , P > 0.05, A N O V A ) . 2+ In contrast, the intracellular application of B A P T A , a more rapid C a chelator than E G T A , eliminated the spermine-induced enhancement of voltage responses, observed on depolarization (Figure 3.13B). In neurons recorded with BAPTA-containing pipettes, spermine application did not alter the responses to depolarizing currents (average of 12.4 ± 1 m V in control and 12.4 ± 0.7 during spermine ; n = 4). This implied that spermine 2+ 2 + induced either Ca entry into the neuron or a Ca -dependent conductance which enhanced the subthreshold depolarizing responses and promoted rectification in the upper right quadrant of the V-1 relationship. The spermine-induced increase in rectification on depolarization would reduce firing threshold. This provided rationale to determine i f the spermine-induced reduction o f the 2+ action potential threshold also depended on extra-and intracellular [Ca ]. The omission 2+ 2+ of Ca from the A C S F , which normally contained 2 m M C a , decreased the threshold from -53.7 ± 2.9 m V to -60.6 ± 3.1 m V (n = 3, A N O V A , P < 0.05). On application o f 2+ spermine in A C S F that was nominally Ca - free, there was no change in the threshold (control, -60.6 ± 3 . 1 m V , and spermine, -60.2 ± 3 . 2 m V ; n = 3). Recovery o f the 2+ threshold from the effects of the Ca - free solution occurred at 10 min after returning to Chapter 3. Results I Ran - 7 9 -Internal EGTA Spermine Control 0 Ca2*, Spermine.Washout J + 0 Ca2* Current step amplitude (pA) B Internal BAPTA Spermine Current pulse amplitude (pA) 2+ Figure. 3.13. Alterations in extra- and intracellular Ca influence spermine effects on depolarizing current - voltage (V - I) relationships in M G B neurons. (A) Voltage responses to current pulses (80 p A , 500 ms in upper traces) and V -1 diagram show that 2+ removal of extracellular Ca from A C S F abolished the increase in voltage responses 2+ induced by spermine during internal application of E G T A (10 m M ) . Perfusion o f C a -2+ free media for 6 min (0 Ca ), alone, and with spermine (100 | i M , 3 min) did not alter voltage response. After a 10 min washout, and perfusion with control media containing 2 2+ m M C a (10 min), spermine application (100 | i M , 3 min) increased the voltage response. V -1 diagram in same neuron shows that spermine did not change the slope of the voltage 2+ responses during Ca - free perfusion. When applied during extracellular perfusion with 2+ normal [Ca ], spermine increased the voltage responses to current pulses that were > 25 p A . (B) Voltage responses to current pulses (80 p A , 500 ms in upper traces) and V -1 diagram show that spermine did not increase the voltage responses in a neuron recorded during internal B A P T A (10 m M ) . Vertical bar, 10 m V ; horizontal bar, 100 ms. Chapter 3. Results I Ran -80-2+ normal C a perfusion. A subsequent application of spermine for 3 min in normal 2+ solution (2 m M C a ) reversibly reduced the firing threshold to -60.0 ± 2.6 m V (n = 3, A N O V A , P < 0.05). These effects, observed when the pipette solution contained 10 m M E G T A , were largely reversible (recovery, -55.3 ± 2.3 mV) . The spermine-induced 2+ reduction in action potential threshold was re-examined using the fast C a chelator, B A P T A (10 m M ) , applied internally. A 3 min application of spermine did not significantly change the threshold in four neurons recorded with BAPTA-conta in ing pipettes (control, -49.3 ±2.1 m V , and spermine, -49.3 ± 2.3 m V ) . These experiments 2+ demonstrated that the effects of spermine on action potential threshold depended on C a entry. 3.2.1.5. Low threshold Ca2+ spike (LTS) firing 2+ Application of Ca - free A C S F abolished the transient, low threshold spike (LTS), evoked at the offset of hyperpolarizing current pulses or on step depolarization in neurons 2+ held at hyperpolarized potentials. This blockade confirmed the Ca mediation of the L T S (Tennigkeit et al., 1996). Spermine application increased action potential firing on top of a L T S in only 10 out of 19 neurons, in contrast to the increased tonic firing rate on spermine application, observed in all neurons. A s shown in Figure 3.14A, spermine induced an action potential on the rebound depolarizing response at the termination o f hyperpolarizing current pulses. In 5 out of the 10 neurons hyperpolarized with D C to -80 m V , spermine application induced one or two action potentials on a subthreshold rebound response to hyperpolarizing current pulses. Spermine application increased the amplitude Chapter 3. Results I Ran -81-of the rebound L T S that did not reach action potential threshold in the remaining 5 neurons. The effects of spermine were reversible, requiring 20 to 40 min for recovery. Blockade of voltage-dependent N a + channels with T T X did not significantly alter the ability of spermine to enhance the L T S in 6 neurons (cf. Figure 3.14A and B) . During T T X blockade, the spermine enhancement of the L T S depended on the holding potential. In these experiments, the L T S was evoked by injecting hyperpolarizing currents o f different amplitude into neurons held at different membrane potentials (Figure 3.14C). Co-application of spermine (100 uM) and T T X induced an L T S in neurons at potentials that caused marked inactivation of the L T S . A t potentials where an L T S was present, a spermine application increased its amplitude and rate of rise (dV/dt). There was a greater increase in the dV/dt of the L T S when the neuron was held at -55 m V than at -85 m V (Figure 3.14C). After spermine application, the L T S evoked in a neuron held at - 55 m V had an average dV/dt of 3.1 ± 0 . 2 mV/ms, compared to 1.5 ± 0.3 mV/ms in the control during T T X application. The average rate of decay was - 1.7 ± 0.3 mV/ms (n = 6) with fast (26 ± 5 ms, n = 5) and slow (146 ± 12 ms, n =5) components (trace 2, Figure 3.14B). Figure 3.14C summarizes the effects of spermine on the dV/dt of the L T S , showing a maximal effect at a holding potential ( V n ) = - 55 m V and a minimal increase at Vh = -85 m V in 6 neurons (paired Mest, P < 0.01). Since the hyperpolarization-activated current influences the rate of rise o f the L T S , the next experiments examined whether spermine affected the voltage sag, mediated by this current (Tennigkeit et al., 1996). The voltage sag was not prominent in the majority o f Chapter 3. Results I Ran - 82 -neurons. Spermine application (100 p M ) produced no change in the voltage sag induced by a hyperpolarizing current pulse in three neurons. Hence, the increase in the rate o f rise o f the L T S did not likely involve interactions with this hyperpolarization-activated conductance. Due to the previous observations, it was necessary to establish i f the spermine potentiation of the L T S involved interactions at N M D A receptors (n = 8). In six o f these neurons, A P V application decreased the amplitude of the subthreshold responses to current pulses (cf. Figure 3.14D, lower traces). A s shown for the neuron o f Figure 3.14D, A P V application also decreased an L T S burst to a single action potential and L T S rate of rise, increasing the latency to the first action potential on top of the L T S . Despite A P V antagonism of N M D A receptors, spermine application transformed the subthreshold response into an L T S , as well as decreased the latency to an action potential on top o f the L T S (cf. A P V and spermine traces in Figure 3.14D). In eight out of eight neurons, A P V application (50 p M , 6 min) reduced the average rate of rise of the L T S from 1.6 ± 0.3 m V / ms in naive controls to 1.2 ± 0.2 mV/ms. A subsequent co-application of A P V and spermine caused a significant increase in the rate of rise of the L T S to 2.2 ± 0 . 1 mV/ms ( A N O V A , P < 0.05). In light of the previous observations, it was necessary to determine i f the spermine potentiation of the L T S involved interactions with A M P A receptors. During blockade o f A M P A receptors with C N Q X (30 p M , 6 min), spermine significantly increased the dV/dt of the L T S (control, 1.8 ± 0.1 mV/ms and spermine, 3.6 ± 0.2 mV/ms; n = 5, A N O V A , P Chapter 3. Results I Ran -83 -A Spermine Washout Control APV + APV (in APV) J 1 I I I I _f Figure 3.14. Effects of spermine (100 p M , 3 min) on the low threshold Ca spike (LTS) firing. (A) Superimposed voltage responses (control, spermine, and recovery) show that spermine induced an L T S on termination of a hyperpolarizing current pulse ( - 40 p A , 500 ms). (B) Spermine increased the rate of rise and amplitude of the L T S at the end of a hyperpolarizing current pulse ( - 80 pA) just before (1), during (2), and after (3) spermine application during T T X blockade (0.6 p M ) of voltage-dependent N a + conductances. Holding potential, - 55 m V . (C) Bar graph summarizes spermine effects on dV/dt of the L T S at the end of hyperpolarizing responses and during T T X blockade. Spermine increased dV/dt in neurons held at - 85, - 75, - 65, and, maximally, at - 55 m V (n = 6, paired t-test, *P < 0.01, **P < 0.005). (D) A P V (50 p M ) did not block the effects of spermine on the L T S , as shown by sub-and suprathreshold responses to current pulses (60, 120 pA) during application of A P V , alone, and co-application with spermine. Application of A P V reduced the subthreshold response, L T S rate of rise, and number o f action potentials. Co-application (3 min) of A P V and spermine transformed a subthreshold response to an L T S , increased L T S rate of rise, and shortened the latency to the action potential. Recovery (in A P V ) was observed after a 10 min washout. Vertical bar, 15 m V in (A) and (B) ; 30 m V in (D). Horizontal bar, 150 ms. Insert in (B) (right): Vertical bar, 3 m V ; horizontal bar, 30 ms. Chapter 3. Results I Ran -84-< 0.05). Hence, the effects of spermine on the L T S did not likely involve N M D A or A M P A receptor interactions. 3.2.1.6. Excitatory and inhibitory postsynaptic potentials Spermine application (100 | i M ) to 18 neurons resulted in bursts of action potentials on EPSPs evoked by electrical stimulation of corticothalamic projections (Figure 3.15A). Spermine had little or no effects on the rate of rise of the EPSP, but always prolonged the decay phase. The EPSP amplitude increased slightly ( 3 - 5 m V ) during spermine application to five neurons, but this was not a consistent finding in the 18 neurons. The spermine-induced action potentials on the EPSPs were reversible in all neurons. Complete recovery was observed in 13 of 18 neurons at 35 min after terminating the application. Spermine prolonged the EPSP decay time constant ( idecay) , as estimated with an a -function fit o f the EPSPs (Figure 3.15B). This promoted the occurrence of action potentials on top of the EPSPs (Figure 3.15A). The ED 5 o for the spermine-induced increase in T u e C ay o f EPSPs was - 1 0 0 pJVI which was approximately the same for the spermine-induced increase in firing (cf. Figure 3.9B). Recovery to the control value occurred after 30 min (148 ± 1 5 ms). Figure 3.15B summarizes these results for 15 neurons. Table 3.1 summarizes the effects of single or cumulative applications of spermine on the 90 - 10% decay time in 18 neurons. Spermine application (100 u M ) did not significantly affect the amplitude or time course of depolarizing potentials evoked by Chapter 3. Results I Ran -85-Figure 3.15. Spermine (100 u M , 3 min) prolonged late component of corticothalamic EPSPs mediated by N M D A receptors. (A) Spermine increased EPSP amplitude and duration, resulting in three action potentials. (B) Spermine delayed the late component (2) o f the EPSPs. The bar graph summarizes the spermine-induced prolongation of EPSP decay time constant (Xtay), expressed as % of the control. Control x ^ w a s 142 ± 8.5 ms (n = 15, paired t-test, *P < 0.01). (C) Spermine did not affect EPSPs during N M D A receptor blockade by 50 uJVl A P V or significantly change remaining EPSP components. Bar graph summarizes the reduction in EPSP Xtay by A P V and lack o f spermine effect during A P V blockade, expressed as % of control. Control Tdecay W3.S 143.6 ± 14 ms (n = 8, A N O V A , *P < 0.01). Vertical bar, 5 m V ; horizontal bar, 200 ms. Chapter 3. Results 1 Ran Table 3.1: Effects of spermine on EPSP variables Amplitude Rise Decay Half-width n (mV) (ms) (ms) (ms) Control 6.3 ± 1 . 8 37 ± 6 477 ± 11 238 ± 1 1 19 Spermine 9.1 ± 2 . 3 2 0 ± 11 710 ± 2 3 * 297 ± 85 19 A P V 6.4 ± 2.1 1 2 ± 1.1 147 ± 9 . 5 59 ± 4.2 9 A P V + 6.1 ± 1 . 9 12 ± 1 . 6 153 ± 12 63 ± 2.4 9 Spermine C N Q X 6.2 ± 0.8 90 ± 5 534 ± 1 8 247 ± 9.5 9 C N Q X + 6.3 ± 0 . 9 95 ± 4 ** 925 ± 24 433 ± 20** 9 Spermine Arcaine 4.5 ± 1 . 4 36 ± 7 419 ± 11 204 ± 40 5 Arcaine + 4.6 ± 0 . 6 42 ± 1 1 394 ± 15 205 ± 38 3 Spermine Glycine 6.9 ± 1 . 1 24 ± 5 190 ± 7 245 ± 41 3 Glycine + 9.0 ± 2.8 28 ± 9 400 ± 2 1 * 437 ± 4 5 3 Spermine Values are m e a n ± SE.*P< 0.05 , ** P< 0.01, Mest Chapter 3. Results I Ran - 87 -stimulation of the brachium colliculi inferioris (n = 6 ; data not shown). These potentials, 100 to 200 ms in duration, were likely IPSPs mediated by G A B A A receptors because they reversed at potentials near Eci (-55 m V ) and were sensitive to blockade by bicuculline (50 LiM, n = 6). Spermine prolonged the decay of the late EPSP component mediated by an N M D A - t y p e receptor. The application of A P V (50 nJVl), blocked the late component, resulting in shorter rise (10 - 90%) and decay (90 - 10%) times of the EPSPs (Table 3.2). During N M D A receptor blockade, EPSP mean idecay was 100 ± 14 ms, compared to 143 ± 15 ms in control (Figure 3.15C; n = 8, A N O V A , P < 0.01). This mean value did not change significantly during combined spermine and A P V application (Figure 3.15B). Interactions with A M P A receptors, which might prolong the EPSPs, were examined by applying spermine in combination with C N Q X (30 | i M , 6 min). Spermine prolonged the EPSP 90 - 10% decay time during C N Q X blockade to the same extent as in the absence of A M P A receptor blockade (Table 3.1). In 3 additional neurons, spermine was applied before the co-application with C N Q X . The co-application resulted in a significant prolongation of the EPSP to the same extent as in the absence of C N Q X blockade of A M P A receptors (Table 3.1). This confirmed that spermine affected only the N M D A -mediated component. Co-application of C N Q X (30 |JM) and A P V (50 L I M ) then abolished the early and late components of the EPSP which remained absent despite a subsequent spermine application (n = 4). These results suggest that spermine increased the duration of the EPSP decay phase by interacting with N M D A receptors. Chapter 3. Results I Ran - 88 -The possibility was considered that spermine prolonged the EPSPs by acting on an extracellular polyamine-sensitive site of the N M D A receptor (cf. Benveniste and Mayer, 1993). This required an investigation of the interactions of spermine and arcaine, an antagonist that acts at the polyamine-sensitive site on the N M D A receptor (Reynolds, 1990). In these studies, arcaine (40 p M ) , spermine (100 p M ) , arcaine and spermine, were sequentially applied each for 3 min (n = 3). Arcaine, alone, did not greatly alter the configuration of the EPSP (Figure 3.16A) or produce changes in the EPSP amplitude, 90 - 10% decay time, and half-width (Table 3.1). After a 15 min washout from arcaine application, spermine significantly prolonged the EPSP Xdecay to 180 ± 32 ms from 101 ± 16 ms in the control (Figure 3.16A). A subsequent co-application of spermine and arcaine abolished the actions of spermine, resulting in Tdecay o f 118 ± 16 ms (Figure 3.16A). The graph of Figure 3.16A summarizes the spermine-induced increases in EPSP Tdecay and arcaine blockade of spermine effects. A possibility was tested that spermine increased the NMDA-mediated component o f the EPSP by potentiating the actions of glycine on the N M D A receptor. In the presence o f a saturating concentration of glycine (40 p M ) , spermine still prolonged the EPSP by ~ 49% (Figure 3.16B). In 3 neurons, spermine increased EPSP Tdecay from 255 ± 44 ms to 379 ± 53 ms ( A N O V A , P < 0.05). In summary, spermine actions on the EPSP likely involved an extracellular polyamine-sensitive site, and not a glycine-sensitive site o f the N M D A receptor. Chapter 3. Results I Ran -89-Spermine Control Spermine Washout Control Arcaine Spermine + Arcaine (Glycine) + Glycine (Glycine) Control Spermine Spermine Control Spermine Washout + Arcaine Figure 3.16. Spermine (100 U . M ) prolonged the EPSPs by interacting with the polyamine-sensitive site on N M D A receptor. (A) Arcaine (40 | i M ) , a blocker at the polyamine-sensitive site on N M D A receptors, almost eliminated the spermine-induced prolongation o f the EPSP , expressed as % control Xdecay which was 100.4 ± 16 ms (n = 3, A N O V A , *P < 0.05). Note that there was a 10 min washout after arcaine, prior to spermine application. (B) Spermine (100 u M ) prolonged the EPSP Xdecay during co-application with glycine (10 JIM). Control Xdecay (glycine present) was 255 ± 44 ms (n = 3, paired Mest, *P < 0.05). Bar graphs summarize the effects of spermine on EPSP Xdecay during arcaine and glycine applications. Holding potential, - 60 m V . Vertical bar, 5 m V ; horizontal bar, 100 ms. Chapter 3. Results 1 Ran -90-2+ 2+ A contribution of extracellular Ca or M g to the spermine-induced enhancement o f EPSPs was assessed in the next experiments. Spermine application did not alter the am-2+ plitude or duration of the EPSPs during a 6 min perfusion of C a - free A C S F in three neurons (data not shown). Hence, spermine effects on NMDA-mediated EPSPs l ikely 2+ depended on Ca entry. 2+ In 2 neurons, the omission of M g from A C S F perfusion resulted in subthreshold oscillations of the membrane potential and spontaneous firing of action potentials. These observations were consistent with previous studies on thalamocortical neurons (Jacobsen 2+ et al., 2001) which prevented critical assessment of a co-agonist role o f M g at the polyamine-sensitive site on the N R 2 B receptor subunit (Kew and Kemp, 1998). 3.2.2. Pentobarbital effects on corticothalamic EPSPs The following experiments were performed in thalamic neurons of the ventroposteriolateral thalamic nucleus of the rat (see Methods). In non-thalamic neurons, pentobarbital, at anesthetic doses, inhibits NMDA-media ted currents (Charlesworth et al., 1995). This provided some rationale to test whether pentobarbital could depress evoked corticothalamic responses mediated by N M D A receptors in V B neurons. Pentobarbital at 200 p M , but not at 50 p M shortened the duration of the NMDA-mediated EPSPs (Figure 3.17A). This result was consistent with pentobarbital's shortening of N M D A receptor mean open time observed in hippocampal neurons (Charlesworth et al., 1995). Chapter 3. Results I Ran -91-The depressant effect of pentobarbital suggested that it would reduce EPSP prolongation caused by spermine. A n application of spermine, at 100 p M , resulted in ~ 45% prolongation of EPSPs (Figure 3.17B). This observation was consistent with E P S P prolongation in medial geniculate neurons of gerbils (cf. Figure 3.15). During spermine application, a subsequent co-application of pentobarbital at 50 p M did not produce a significant change in the duration or amplitude of NMDA-mediated EPSPs. However, a subsequent increase in the pentobarbital dose to 200 p M produced a reversal of the spermine-mediated prolongation of the EPSP (Figure 3.17). Pentobarbital reversal o f the prolongation of EPSPs caused by spermine implicated an action at specific modulatory sites. 3.2.3. Interactions ofZn2+ with spermine and pentobarbital The first possibility examined was that pentobarbital reversal of spermine prolongation of EPSPs involved interactions at the Z n 2 + binding site on N M D A receptors. This was done by applying Z n 2 + , a negative modulator of N M D A receptors at a site distinct from that o f polyamines (Forsythe et al., 1988). Application of Z n 2 + (20 p M , 1 min) resulted in a 32 % reduction in EPSP decay time constant (Figure 3.17C). Z n 2 + application also reduced EPSP mean amplitude, however, this was not significant (6.1 ± 1 m V in controls compared to 4.9 ± 0.9 m V after Z n 2 + , n = 5, P > 0.05). During Z n 2 + application, co-applied spermine (100 p M , 1 min) prolonged the EPSP decay by 58 % (137 ± 12 ms). A subsequent co-application of pentobarbital reversed the spermine-mediated prolongation of EPSP to control levels (with Z n 2 + present, Figure 3.17), similar to the effects in the Chapter 3. Results 1 Ran - 9 2 -Figure 3.17. Pentobarbital effects on NMDA-mediated corticothalamic EPSPs. (A) Pentobarbital at 200 f i M , but not at 50 p M shortened the duration of the N M D A -mediated EPSPs. The bar graph summarizes the effects on EPSP decay time constant (Tdoay), expressed as % of the control. Control T ^ y was 118 ± 13 ms (n = 6, A N O V A test, *P < 0.05- significantly different from control). (B) Pentobarbital at 200 p M , but not at 50 p M reversed the spermine-mediated (100 p M , 1 min) prolongation of the N M D A -mediated EPSPs. The bar graph summarizes the pentobarbital-reversal o f spermine prolongation of EPSP decay time constant (tdecay), expressed as % of the control. Control Tdecay was 123 ± 11 ms (n = 5, ANOVA-tes t , *P < 0.05 - significantly different from spermine 100 p M + pentobarbital 0 pM)) . (C) Z n 2 + (20 p M , 1 min) reduced the amplitude and shortened the duration of NMDA-mediated EPSPs. Graph summarizes the effects on Tdecay as in A (n = 5, student Mest, *P < 0.05- significantly different from control). (D) The presence of Z n 2 + did not alter spermine's ability to prolong the N M D A -mediated EPSP. A subsequent co-application with pentobarbital still reversed the EPSP prolongation caused by spermine. Graph summarizes the effects on prolongation and its reversal by pentobarbital. Tdecay as in A (n = 5, ANOVA-tes t , *P < 0.05- significantly different from control or spermine + pentobarbital). EPSPs were evoked by internal capsule stimulation (50 V , lOOps). Holding potential, - 60 m V . C N Q X (30 p M ) was applied throughout the experiment. EPSP traces are averages of 5 samples each. Chapter 3. Results I Ran - 93 -absence of Z n 2 + . These observations suggested that pentobarbital acted at sites distinct from those of Z n , possibly the polyamine site. 3.2.4. Antagonism of polyamine site The lack of effect of Z n 2 + on pentobarbital reversal of spermine prolongation of EPSPs implicated interactions at the polyamine site. To test this hypothesis, spermine and pentobarbital were co-applied during blockade of the polyamine site with arcaine. B y itself, arcaine (40 L I M , 1 min) decreased the duration of EPSPs to 63 % o f the control (Figure 3.18); an effect which indicated that endogenously-released spermine contributed to EPSP decay. During arcaine application, a subsequent co-application of spermine did not alter the amplitude or duration of EPSPs (Figure 3.18). A subsequent co-application with pentobarbital (200 m M , 3 min) did not change EPSP duration. The results suggested that the reversal of spermine prolongation of EPSPs involved an allosteric modulatory action of pentobarbital at the polyamine site on N M D A receptors. 3.2.5. Discussion Spermine application increased the decay time constant of corticothalamic EPSPs mediated by N M D A receptors. This finding is consistent with the increased amplitude of N M D A - e v o k e d currents during spermine application to cultured hippocampal and spinal neurons (Lerma, 1992; Benveniste and Mayer, 1993). In the present studies, the actions of spermine were selective and required extracellular C a 2 + because they were abolished in three neurons by brief Ca 2 +-free perfusion. Spermine also did not significantly alter the Chapter 3. Results I Ran -94-A Figure 3.18. Pentobarbital reversal of spermine EPSP prolongation involves interactions at the polyamine site on N M D A receptor. A ) Application of arcaine (40 | i M , 1 min) shortened the duration of NMDA-mediated EPSPs without affecting their amplitude. The bar graph summarizes s the effects on Xdecay in five neuorns. B) With arcaine present, spermine (100 u\M, 3 min), when applied alone, or in combination with pentobarbital (200 | i M , 3min) did not cause any prolongation of the EPSP. Holding potential, - 60 m V . The bar graph summarizes this lack of effect (n = 5, P > 0.05 with or without spermine). EPSP traces in A and B are averages of 5 samples each. Chapter 3. Results I Ran -95-early EPSP component mediated by A M P A receptors, or appreciably affect IPSPs mediated by G A B A receptors that were sensitive to bicuculline antagonism. Application of A P V completely blocked the spermine-induced increase in the EPSP decay time constant. This implicated N M D A receptors in spermine actions to increase excitation. The effects of spermine on M G B neurons involved a polyamine-sensitive site on the N R 2 B subtype of N M D A receptors, as demonstrated with arcaine and glycine applications. Arcaine, itself, did not have significant effects on the passive and active membrane properties but reversed the spermine-induced decrease of the EPSP decay. Previous studies have shown that arcaine blocks spermine actions by inverse agonism, antagonism, and open-channel blockade of the polyamine-sensitive site on N M D A receptors (Reynolds, 1990; Pritchard et al., 1994). The actions of spermine at this site decreased the EPSP decay, despite saturating concentrations o f glycine. These observations are consistent with the glycine-independent potentiation of N M D A currents by spermine at the N R 2 B receptor subunit in cultured hippocampal neurons (Benveniste and Mayer, 1993). In thalamocortical neurons, the persistence during high glycine concentrations and arcaine reversal imply that spermine acted independently o f the glycine site at a specific polyamine-sensitive site on the N R 2 B receptor subunit o f the N M D A receptor. Chapter 3. Results (Ran -96-The N R 2 B subunit may modulate the decay time constant of the N M D A receptor-mediated EPSP during the development in M G B neurons. A t the end of the second postnatal week, thalamocortical neurons express an abundance of the N R 2 B polyamine-sensitive receptor subtype in the M G B and lateral geniculate body ( L G B ) o f the rat (Chen and Regehr, 2000). The duration of EPSPs mediated by N M D A receptors in L G B neurons of the rat is similar at P14 to that in gerbil M G B neurons. The decay time constant in L G B neurons is longer at P14 in rats than at earlier (P7-P13) or later (PI 6 -P28) stages of development (Chen and Regehr, 2000; cf. also rat M G B at P21-P42, Bartlett and Smith, 1999). Hence, spermine modulation of the N R 2 B subunit may cause the longer EPSP duration in M G B neurons at PI4. Spermine enhanced excitability by increasing inward rectification on depolarization, without greatly affecting the passive properties of M G B neurons. It is not known i f the passive and active membrane properties of M G B neurons mature by P14 in gerbil, as in the rat (Tennigkeit et al., 1998). Thalamocortical neurons of the adult guinea pig and P 7 -P28 rat inwardly rectify because the activation of persistent N a + conductance on depolarization results in an amplification of the voltage response (Jahnsen and Llinas, 1984; Tennigkeit et al., 1998; Parri and Crunelli, 1998). In the present studies, blockade of the TTX-sensitive rectification or N M D A receptors eliminated the spermine-induced enhancement of rectification on depolarization. These findings imply that spermine interactions with N M D A receptors led to activation of a persistent N a + conductance in M G B neurons. Chapter 3. Results I Ran - 97 -9+ 9+ A n elevation in intracellular Ca concentration ([Ca ];) due to N M D A receptor activation (Jahr, 1992) may have enhanced rectification on depolarization. In the present study, there was no evidence for a spermine-induced increase in voltage responses on depolarization during Ca 2 +-free perfusion or rapid chelation of C a 2 + with intracellular B A P T A . It seems likely that an elevation of [Ca 2 +]j initiated by spermine actions at N M D A receptors activated intracellular messengers and increased this rectifying behavior. In neocortical neurons, transmitter activation of dendritic N M D A receptors increases C a 2 + entry (Schwindt and C r i l l , 1995) that may increase channel phosphorylation (Siekevitz, 1991) and a persistent N a + conductance (Schwindt et al., 1992). Hence, the spermine-induced enhancement of TTX-sensitive rectification on depolarization may result from NMDA-mediated C a 2 + entry in M G B neurons. The effects of spermine on membrane rectification and firing threshold may involve the recruitment of a Ca 2 +-dependent second messenger, subsequent to N M D A receptor activation. Activation of N M D A receptors enhances C a 2 + entry, resulting in a C a 2 + gradient in the dendrites (Connor et al., 1988) and activation of a protein kinase C ( P K C ) pathway. A rise in intracellular [Ca 2 + ] also may activate calmodulin kinase II which enhances N a + currents (Carlier et al., 2000). P K C activation increases membrane excitability by shifting the activation curve for the persistent N a + current along the voltage axis to more hyperpolarized potentials (Astman et al., 1998; Franceschetti et al., 2000). This voltage range is consistent with the range of spermine enhancement of voltage rectification in our experiments. Chapter 3. Results I Ran -98-The increased rectification on depolarization may have reduced the threshold for an action potential in M G B neurons (cf. neocortical neurons, Stafstrom et al., 1982). Antagonism of N M D A receptors, perfusion with Ca 2 +-free A C S F or rapid chelation o f C a with B A P T A , eliminated the reduction in threshold and increased tonic firing due to spermine application. Hence, the modulation of N M D A receptor-mediated C a 2 + entry likely increased membrane rectification on depolarization and reduced firing threshold. This mechanism explains the ability of spermine to increase postsynaptic excitability and tonic firing in M G B neurons. Spermine facilitated L T S firing by a mechanism that did not involve interactions with N M D A receptors. Spermine increased the rate of rise and amplitude o f the L T S , despite A P V blockade of N M D A receptors. This was evident on depolarization to action potential threshold where there is a smaller gradient for C a 2 + as well as greater inactivation of T-type C a 2 + channels (Hernandez-Cruz and Pape, 1989). Spermine enhanced the L T S during blockade of voltage-dependent N a + channels by T T X . Hence, a change in some parameter of the T-type C a 2 + current, e.g. voltage dependence of the inactivation-activation relationship, may have increased the L T S . Pentobarbital modulated NMDA-media ted corticothalamic EPSPs producing more transient responses. A t an anaesthetic concentration, pentobarbital shortened the duration o f NMDA-media ted corticothalamic EPSPs. Pentobarbital reversed spermine Chapter 3. Results I Ran -99-prolongation o f corticothalamic EPSPs by a mechanism that was independent o f Z n interactions. The lack of pentobarbital effects during arcaine blockade implicated an action at the polyamine site on N M D A receptors. These effects are consistent with pentobarbital shortening of burst durations of NMDA-mediated single channel currents (Charlesworth et al., 1995). Pentobarbital actions on NMDA-mediated corticothalamic transmission may contribute to its anti-epileptic effects. The depressant effects of pentobarbital on NMDA-mediated synaptic responses provided a rationale to examine its short-term effects on pre- and postsynaptic parameters o f non-N M D A mediated transmission presented in the next section. Part III. Effects of pentobarbital on short-term depression 3.3.1. Behaviour of EPSCs in trains during short-term depression 3.3.1.1 Passive membrane properties For proper assessment of short-term depression (STD), QX-314 and Cs-gluconate were applied intracellularly to block, respectively, N a + and K +-channels and reduce postsynaptic currents. Wi th this pipette solution, the input resistance (Rj) increased by ~ 81 % (380 ± 25 MQ., n = 10, P < 0.05) compared to values obtained using solutions containing K +-gluconate and no QX-314 (210 ± 15 MQ, n = 9). During combined C s + and QX-314 blockade, pentobarbital (200 p M ) did not alter the R i throughout 3 - 5 min of application (365 ± 34 M Q , n = 10; cf. Wan et al. 2004). Hence, intracellular blockade Chapter 3. Results I Ran -100 -of N a + and K + channels reduced their postsynaptic contributions to stimulus-evoked depression (Cahalan and Aimers 1979; Konishi 1990; Budde et al. 1994). The experiments were performed in neurons voltage clamped at -80 m V in order to minimize postsynaptic contributions of voltage-dependent C a 2 + conductances (Hernandez-Cruz and Pape, 1989). This allowed the study of the frequency-dependent aspect of corticothalamic S T D while minimizing postsynaptic temporal summation. 3.3.1.2. Frequency - dependent fade (STD) of corticothalamic EPSCs Repetitive stimulation in the 2.5 - 20 H z range produced S T D of EPSCs (Figure 3.19). Wi th increasing stimulation frequencies, the train of EPSCs decreased in amplitude to a plateau (Si 5. 2o) that was 49 to 21 % of the 1 s t E P S C amplitude (Si ; see Plateau/Si ratio in Table 3.2.1 A ) . The relation between the plateau and stimulation frequency is illustrated in their product value (plateau x Hz) which increased significantly at stimulation frequencies > 5 H z (Table 3.2.1 A ) . The apparent Q N decreased at stimulation frequencies > 10 Hz , indicating substantial refill at lower frequencies (Table 3.2.1 A ) . The ratio of the amplitude of the 12 t h E P S C , subsequent to the omitted 11 t h stimulus to the 10 t h E P S C amplitude (S12/S10), increased at stimulation frequencies > 5 H z (Table 3.2.1 A ) , indicating high values of fractional release in the plateau. The frequency-dependent characteristic o f S T D was similar to other observations at corticothalamic synapses (L i et al. 2003; Reichova and Sherman, 2004). Chapter 3. Results I Ran - 101 -A 2.5 Hz 5 Hz ({{fmfrrrrcrrm— 10 Hz 20 Hz B 1.00 Ui% 0.7S | | 0 C ° ' Z 0.25 ooo-^ x I f 1 nA 10 15 »> sllmulus # !0 15 B> stimulus # Figure 3.19. Frequency-dependence of corticothalamic STD. A ) Traces o f trains of E P S C s during STD. Increasing the frequency of stimulation from 2.5 to 20 H z enhanced STD. Expanded traces below show the 1 s t to 5 t h EPSCs (middle) and the 10 t h to 14 t h E P S C s around the omitted 11 t h stimulus (bottom). B) Normalized E P S C amplitudes at four stimulation frequencies. A t 10 Hz , EPSCs reached a plateau of 40 % of the 1 s t response whereas at 20 H z E P S C amplitudes reached a plateau of 25 % o f the 1 s t response. A t 20 H z stimulation, the mean amplitude of the 11 t h response just subsequent to the missing 10 t h stimulus nearly doubled, consistent with the depletion model. Traces in A are averages of 6 repeats from 1 neuron. Values in B are averages o f 6 neurons. S E M indicates between neuron variations. Neurons were held at - 80 m V . Chapter 3. Results I Ran Table 3.2.1 A : Summary of parameters of corticothalamic S T D at different frequencies Frequency (Hz) 2.5 5 10 20 Parameter S , ( n A ) 1.19 ± 0 . 3 5 1.21 ± 0 . 3 8 1.18 ± 0 . 2 9 1.17 ± 0 . 4 1 S 2 ( n A ) 0.67 ± 0.23 0.70 ± 0.25 0.74 ± 0 . 1 9 0.56 ± 0.32 S2/S1 0.56 ± 0 . 1 9 0.57 ± 0.2 0.63 ± 0 . 1 6 0.47 ± 0.27 S12/S10 1.02 ± 0 . 0 8 1.06 ± 0 . 0 9 1.31 ±0 .10* 1.86 ± 0 . 1 1 * Plateau (Si 5-20) (nA) 0.58 ± 0 . 1 6 0.52 ± 0 . 1 9 0.41 ± 0 . 1 7 0.25 ± 0 . 1 5 Plateau/Si 0.49 ± 0 . 1 3 0.43 ± 0 . 1 5 0.35 ± 0 . 1 3 0.21 ± 0 . 1 2 * Plateau x H z (nA/s) 1.4 ± 0 . 4 2.6 ± 0 . 5 4.1 ± 0 . 6 * 5.0 ± 0 . 6 * Apparent Q N (nA) 4.25 ± 0 . 1 1 4.22 ± 0 . 1 0 4.09 ± 0 . 1 3 3.26 ± 0 . 1 4 * Values are mean ± S E M between neurons; n = 6, * (relative to 2.5 Hz) P < 0.01, A N O V A test Chapter 3. Results I Ran -103 -S T D was found to be associated with negative covariances between pairs of the first 5 EPSCs , consistent with the binomial/depletion model (Figure 3.20). Theoretically, in the absence of refill, the covariance divided by the product of the mean amplitudes o f the 1 s t and 2 n d EPSCs (<Si>-<Sj>) equals the negative of the reciprocal of the number of release sites (-1/N). The plot of Figure 3.20A shows the negative of the covariance, expressed as -cov(Si,Sj)/ <Si>-<Sj> between the 1 s t E P S C and the 2 n d to 5 t h EPSCs . A t each particular frequency, the covariance term decreased as the distance between the pairs of stimuli increased (Figure 3.20). This attenuation of the covariance term is expected from refill which reduces the negativity of the covariance term. A s shown in Figure 3.20, an increase in the stimulation frequency worked in the opposite direction, as expected from a lower a between stimuli. The covariance data were entirely consistent with the binomial/depletion model (see Methods). This validated the use o f the covariance term to correct the variance to mean ratios and to estimate the apparent quantal sizes during S T D , at each stimulus in the train. The results indicated an inconsistency with the simple model. Namely, S T D was also characterized by a frequency-dependent decrease in the apparent quantal size along the E P S C train (Figure 3.20 B , Table 3.2.IB). After 2 stimuli, the apparent quantal size declined to a plateau value lower than the initial. The equivalence with variance/mean ratios indicated no error in the correction, which is intrinsically small when signals are much less than the initial (Figure 3.20A). The apparent quantal size values, in the 26-35 p A range, were not significantly different from the mean amplitudes of evoked miniature Chapter 3. Results I Ran - 104-B i ° °^  2.5 Hz 10 .15 20 Stimulus U 5 Hz 10 Hz 20 Hz 2 0.02 5" 0.04 1 0.03 1 0.02 I 0.01 aoo IS 20 Ifi. 20 SHi-Hii H t f H H 12 _ 50 S •e 40 8 •g » 10 15 20 : stimulus ft • 10 15 20 stimulus # 10 15 20 stimulus # Figure 3.20. Validation of the corrected variance-mean method during corticothalamic STD. A ) Negative covariances within the first five EPSCs during train-evoked S T D . Attenuation o f the covariance term calculated for pairs of the 1 s t E P S C relative to the 2 n d to 5 t h EPSCs . The attenuation increased with decreasing stimulus frequencies in the 2.5-20 H z range, likely due to refill o f depleted packets. B) Quantal size estimates during S T D . Note the frequency-dependent reduction in quantal size early in the train and the lack o f change in quantal size after an intra-train gap at the 11 t h stimulus. C) Alterations in quantal content during S T D at 2.5 - 20 Hz . Note the similarity in reduction of quanta to the rundown in E P S C amplitude (Figure 3.19) and the post intra-train gap increase in quanta reflecting a presynaptic provenance. Data were obtained from same neurons as in Figure 3.19. S E M indicates between neuron variations. The relative jump after the gap theoretically equals POutputx(l-o0. Chapter 3. Results I Ran - 1 0 5 -A Pre-stimulation minEPSCs 1 s 50 pA Post-stimulation minEPSCs B 2.5 Hz (a) Before trains 1-.0, _ 0 . 8 f . | 0.6 9 OA ?o:2 0.0 46 minis in 30s •1,1.8*2.5 f>A 0 10 20 30 40 Size (pA) (b) After trains e i 37 mins in 6 s 0 10 20 30 40 Size(pA) (b)-(a) a 3 I2: I o 31.613:0 pA — ± 3 , 0 l i f l l 20 30 40 Size (pA) 5Hz 43 minis in 30s r ^ - * - 11.4 ± 2.1 pA a. 0 10 20 30 40 Size (pA) 44 minis in 6 s' :1\ 0 10 20 30, 40 ,Size{pA) ? 3 I 2 I o -1 3 2 , 3 ± 3 : 3 p A » ^ 0 ' 1 * 20 30. 40 Size (pA) 1.0 „0.B> | 0.6. I 0 ' | 0 . 2 . 0.0 10 Hz 48 minis in 30s • 11 1 ±2.4pA .0 10 20 30 40 Size(pA) 0 10 20 30 40 Size (pA) 6; 4 I * | 2 : 32.55 2 . 7 p A - « ^ 0 ^ior=M^""30*''~*4o" Size(pA) 1.0, ,0.8] i ; BA I 0.4 ! !"o,2-0.0 20 Hz 53 minis in 30s — 10.9 ±3.1 pA 0 10 20: 30 40 Size (pA) 0 10 20 30 40 Size (pA) 30.9 ± 2.4 p A : - » / ! 0 '-1a 20 30 • 1 - Size(pA) Figure 3.21. Pre- and poststimulation miniature EPSCs in a neuron vary in size. A ) Sample records of miniature EPSCs (minEPSCs), 5 s in duration, before and after a 10 H z stimulus train. B) (a) Amplitude histograms of spontaneous minEPSCs counted 5 s before the stimulus train (6 repeats, 30 s in total), (b) Amplitude histograms of evoked minEPSCs counted 1 s after the stimulus train (6 repeats, 6 s total). Total minEPSC count (minis) is indicated above the histograms. Evoked minEPSC sizes were obtained by subtraction o f spontaneous from evoked minEPSC histograms ((b) - (a)). Values next to black arrows are the mean size ± S E M . The apparent quantal sizes (Q') were (in pA) : 34.6 at 2.5 Hz , 33.1 at 5 Hz , 32.9 at 10 Hz , and 34.1 at 20 Hz . Chapter 3. Results I Ran 106-Table 3.2.IB: Derived parameters of STD at different frequencies Frequency (Hz) 2.5 10 20 Parameter cov(Si,S 2 ) (nA 2 ) Q'(Si) (pA) Q'(S 2) (pA) QXS15-S20) (PA) Q T S i s ^ a i Q'(Si) Var/Mean (S15-S20) (PA) Evoked minEPSC size (PA) Pre-stimulation m i n E P S C size (PA) mi m 15-20 mu/mio 0.016 ± 0 . 0 0 9 -0.018 ± 0 . 0 1 0 -0.021 ± 0 . 0 1 3 -0.024 ± 0 . 0 1 2 36.1 ± 9 . 1 35.5 ± 8 . 7 31.3 ± 6.1 30.7 ± 7.4 30.5 ± 3 . 7 11.4 ± 3.1 53 ± 4 34 ± 3 37.3 ± 7 . 3 35.3 ± 6 . 2 36.5 ± 8.2 34.9 ± 5 . 5 34.5 ± 6 . 4 36.1 ± 5 . 9 29.0 ± 7 . 3 26.1 ± 5 . 7 25.5 ± 5 . 4 0.85 ± 0 . 0 7 0.78 ± 0 . 1 0 0.74 ± 0.08 0.68 ± 0.06 31.2 ± 6 . 6 27.2 ± 4 . 9 26.4 ± 5 . 6 31.4 ± 4 . 1 32.0 ± 5 . 3 29.1 ± 6 . 1 11.2 ± 2.9 52 ± 3 31 ± 4 11.1 ± 3 . 4 10.5 ± 2 . 5 52 ± 8 27 ± 3 53 ± 3 21 ± 4 * 1.08 ± 0 . 1 1 1.13 ± 0 . 1 8 1.25 ± 0 . 2 0 1.96 ± 0 . 1 5 Values are mean ± S E M between neurons; n = 6 ; * (relative to 2.5 Hz) P < 0.05; data from same neurons as Table 3.2.1 A . Chapter 3. Results I Ran -107 -EPSCs (minEPSC) observed 1 s after the stimulus train (Figure 3.21; Table 3.2.1B). The frequency o f these minEPSCs increased with stimulation frequency (Figure 3.21). During STD, the quantal content (m) also decreased to a plateau that depended on the stimulation frequency. At 20 H z stimulation, m reached a plateau value (m 15.20) o f - 40 % of the initial m (m\, Figure 3.20C). The m value increased subsequent to the gap at the omitted 11 t h stimulus (mn/mo', Figure 3.20C, Table 3.2.IB), consistent with the increased amplitude of the 12 t h E P S C amplitude (20 H z in Figure 3.19, Table 3.2.IB). This observation was indicative of a refill process that restores the apparent number o f releasable quanta. In summary, an intra-train reduction in quantal content mediated much of the frequency-dependent component of STD. 3.3.2. Effects of alterations in extracellular Ca2+ concentration ([Ca2+]e) 3.3.2.1 Low [Ca2+Je perfusion Since C a 2 + has been proposed to be a key factor for release probability and therefore a mediator of STD, the following experiments examined i f S T D persisted under conditions of low release probability. Washing extracellular C a 2 + with E D T A resulted in an overall reduction in the amplitude of the E P S C train (Figure 3.22) and a rundown to a plateau of 28 % o f the initial E P S C amplitude (Figure 3.22; Table 3.2.2A). A subsequent application of a 0.2 m M C a 2 + media with no E D T A increased the amplitude of the E P S C train, confirming an effective reduction of extracellular C a 2 + concentration ([Ca 2 + ] e ) by the previous E D T A solution. Contrary to the expectation, there was a significant drop in the apparent Q N in low [ C a 2 + ] e with or without E D T A , rather than a fall in fractional release Chapter 3. Results I Ran - 108 -(Table 3.2.2A). In the neuron shown in Figure 3.22, an application of D M S O (1%), subsequent to 0.2 m M [Ca ] e , produced higher E P S C amplitudes and faster rundown by increasing the quantal contents at the beginning of the train. Hence, a substantial decrease 2_|_ in [Ca ] e did not result in a loss of STD. The reduction in Q N was, apparently, due to a decrease in apparent quantal size throughout the train in low [Ca 2 + ] e , with or without E D T A (Figure 3.22; Table 3.2.2B). On application of very low [ C a 2 + ] e (0.1 m M C a 2 + in E D T A ) , the apparent quantal size was decreased already at the 1 s t response and in the remainder of the train. The subsequent switch to 0.2 m M C a 2 + (without E D T A ) increased the quantal content at the beginning of the train without having significant effect on quantal size (Table 3.2.2B). These effects of low C a 2 + on the quantal parameters suggested the participation o f high probability release sites consistent with a compound binomial model (Brown et al. 1976; Walmsley et al., 1988). However, the high probability release sites were associated with relatively small quantal responses, perhaps at sites on dendrites relatively protected in some way from reduction in local [Ca 2 + ] . 3.3.2.2. Elevated [Ca ]e perfusion A n increase in [ C a 2 + ] e enhanced STD. Raising [ C a 2 + ] e from 2 to 8 m M resulted in increased amplitudes of the initial EPSCs (Figure 3.23). The EPSCs plateau increased by -50 %, whereas, the ratio of the plateau to the 1 s t E P S C (plateau/Si) decreased (Table 3.2.2A). The ratio of EPSCs around the intra-train gap (S i 2 /S i 0 ) increased from 1.31 ± 0.08 to 1.56 ± 0.07 (n = 5, P < 0.05, Mest). Consistent with increases in fractional Chapter 3. Results I Ran -109 -Control EDTA mm i r fW 0.2 mM Ca'' f/ffCfrfrrfrmrrr 1% DMSO (0.2 mM Ca") mrrr frr 0,21 nAL 0.2s / f t « 1 nA 0.5s Figure 3.22. Persistence of STD in media containing low [Ca ] e . Top traces: E P S C trains from a neuron during control, after a 1 min application of 1.1 m M E D T A , after a 2 min wash in 0.2 m M Ca + , and subsequent to a 1 min application of 1% (vol/vol) D M S O . Lower traces, a x3 magnification of the initial five EPSCs at the beginning of the train are for E D T A and 0.2 m M C a 2 + . Bottom Left: E P S C amplitudes in low C a 2 + media. Bottom middle: quantal size estimates. Bottom right: quantal content estimates. Data in bottom plots are averages of 5 neurons. Error bars in 1 s t response indicate between-neuron variations. Chapter 3. Results I Ran - 110-0.1 mM [Ca2*] 0.5 nA 50 ms 3n 2 O a. 1 UJ Control HiCa2+ Lo Ca2+ u 55 CL UJ . 1.0H 1 g to 0 . 5 ^ 0 . 0 • 2 mM fCa*j • C.lmMjCa 1 ] ***•# •+*•*-§-•-«.-*-• 5 10 15 stimulus # 20 Figure 3.23. [ C a 2 + ] e modification of STD. Top: Traces showing rundown of the 1 s t - 5 th EPSCs in 0.1 m M C a 2 + (light grey), 2 m M C a z + (dark grey), and 8 m M C a 2 + (black). Bottom Left: Scatter plot of 1 s t E P S C amplitude in control (2 m M ) , high C a 2 + (Hi C a 2 + ; 8 2  m M ) , and low C a 2 + (Lo C a z + ; 0.1 mM) . Horizontal lines indicate mean. Bottom right: 2  Normalized E P S C amplitudes in 0.1 m M C a 2 + (squares), 2 m M C a z + (circles), and 8 m M ,2  C a 2 + (triangles). Note increased amplitude after the intra-train gap in the control and high [Ca 2 + ] and lack of increase in amplitude in low C a 2 + . Data are from 10 neurons. Holdings potential was -80 m V . Values are expressed as Mean ± S E M . Chapter 3. Results 1 Ran - Ill -Table 3.2.2A: Summary of effects of altered [Ca ] e on parameters of S T D Control (2 m M C a 2 + ) High C a 2 + ( 8 m M ) Control C a 2 + / E D T A ( 2 m M C a 2 + ) (0.1 m M C a 2 + ) L o w C a 2 + (0.2mM) Parameter S i ( n A ) 1.42 ± 0 . 4 1 2.53 ± 0 . 3 4 1.32 ± 0 . 5 3 0.47 ± 0.40 0.92 ± 0.42 S 2 (nA) 0.74 ± 0 . 1 9 1.42 ± 0 . 3 5 0.68 ± 0.27 0.34 ± 0 . 3 1 0.63 ± 0 . 1 9 S2/S1 0.52 ± 0 . 1 2 0.56 ± 0 . 1 3 0.51 ± 0 . 2 0 0.72 ± 0 . 1 6 0.68 ± 0 . 1 8 S12/S10 1.31 ± 0 . 0 8 1.56 ± 0 . 0 7 * 1.28 ± 0 . 1 1 1.03 ± 0 . 0 6 1.13 ± 0 . 0 5 Plateau (Si 5-20) (nA) 0.62 ± 0.09 0.91 ± 0 . 1 1 * 0.59 ± 0 . 1 3 0.13 ±0.10** 0.24 ± 0.08** Plateau/Si 0.48 ± 0.05 0.34 ± 0.04* 0.44 ± 0.07 0.28 ± 0.04** 0.26 ± 0.06** Apparent Q N (nA) 3.93 ± 0 . 1 1 7.2 ± 0.25 4.13 ± 0 . 1 7 1.07 ±0 .21** 1.55 ±0.19** Values are mean ± S E M between neurons; n = 5 in high and low Ca groups, * P < 0.05, Mest .** P< 0.05, A N O V A test. The controls were different neurons for high and low [Ca 2 + ] e . Chapter 3. Results I Ran 112-9+ Table 3.2.2B: Effects of altered [Ca ] e on derived parameters of S T D Control High Ca 2+ Control C a 2 + / E D T A L o w C a 2 + ( 2 m M C a 2 + ) (8 m M ) (2mM C a 2 + ) ( 0 . 1 m M C a 2 + ) (0.2mM) Parameter cov(S,,S2) -0.018 ±0.010 -0.026 ±0.008 -0.021 ±0.013 -0.010 ±0.009 -0.015 ±0.010 Q'(S0(pA) 33.4 ±7 .4 31.2 ±8 .6 35.4 ±8 .9 12. 6 ±8.6* 13.5 ±7.5* Q'(S 2)(pA) 32.4 ±6.5 29.5 ± 7.8 33.6 ±7 .5 11. 9 ±6.5* 13.1 ±4.2* Q'(S,5-S2 0) 28.3 ± 6.9 (PA) O Y S ^ - S ^ 0.85 ±0.17 Q'(S,) Var/mean 29.4 ± 4.9 (S15-S20) (PA) Evoked 30.8 ± 4.2 minEPSC size (pA) Pre-stimulation minEPSC size 10.6 ±2 .9 (pA) mi 47 ± 9 W15.20 26 ± 5 mnlmw 1.21 ±0.17 26.9 ±9 .1 213 ±1.1 12.3 ±7 .0 14.2 ±8 .1 0.87 ±0.11 0.78 ±0.21 0.97 ±0.11 1.05 ±0.18 28.6 ±6 .6 27.9 ±8 .5 32.7 ±7 .1 33.1 ±5 .7 13.2 ±4.1 12.4 ±3 .7 78 ± 11* 33 ± 8 37 ± 13 19 ± 10 1.52 ±0.09* 1.15 ±0.11 13.6 ±9 .2 14.6 ±7 .8 14.5 ±6 .3 15.1 ±6 .0 13.9 ±2 .7 12. 7 ±2 .2 36 ± 19 22 ± 7 68 ± 15 1 7 ± 6 1.02 ±0.09 1.13 ±0.10 Values are mean ± S E M between neurons; n = 5 in high and low [ C a 2 + ] e groups; data from same neurons as Table 3.2.2A; * P < 0.05, Mest .** P< 0.05, A N O V A test. Chapter 3. Results I Ran -113-release, the changes in STD parameters occurred in parallel to an increase in quantal content at the beginning of the train and around the intra-train gap (m^/mio) with no effect on quantal size (Table 3.2.2B). Hence, raising [ C a 2 + ] e promoted S T D by increasing the quantal content, and increasing fractional release. The apparent increase in Q N with 2*1" high [ C a ^ e was also reported for neuromuscular junction (Elmqvist and Quastel, 1965a) and was consistent with a compound binomial model (Quastel, 1997) in which fractional release varied between sites. 3.3.3. Receptor desensitization and saturation 3.3.3.1. Effects of blockade of receptor desensitization The reduction in quantal size observed during S T D (Figure 3.20B; Table 3.2.IB) suggested that a substantial component arises from receptor desensitization (Scheuss et al., 2002). Application of a blocker of A M P A receptor desensitization, cyclothiazide ( C T Z ; 50 u M ) had little effect on the development of S T D (Figure 3.24; Table 3.2.3A), but estimates of quantal size decreased less than in controls reaching a higher plateau (Figure 3.24; Table 3.2.3B). These data demonstrated a moderate contribution of receptor desensitization to the decrease in the apparent quantal size during STD. 3.3.3.2. Combined blockade of receptor desensitization and saturation To test whether receptor saturation, in addition to desensitization, contributed to the early drop in quantal size during STD, C T Z (50 u M ) was co-applied with kynurenate ( K Y N ; 50 u M ) ; an antagonist that rapidly dissociates from glutamate receptors. The result was abolition of the drop in quantal size early in the train (Figure 3.24C; Table 3.2.3B). A t the Chapter 3. Results I Ran - 1 1 4 -A Control C T Z ' i l ' | * | l l l J / i l l i i 0.5 ms 0.5 nA i !/ i/ y A /' ' 0.2 ms 0.5 nA B o Control • Cyclothiazide(50 uM) I cp. £• H 0 04 0.03 1 25 10 15 20 10 15 20 0 10 15 20 o Control Cyclothiazide (50 uM) * Kynurenate (50 uM) 1 VrTTTT! Figure 3.24. Effects of blockade of receptor desensitization and saturation on S T D . A ) . Traces of EPSCs (averages of 5 repeats) before and after application of 50 p M cyclothiazide (CTZ; 2 min). Traces on right show time expansions o f the l s t - 5 t h E P S C . B) Left: Mean E P S C amplitudes (6 neurons) before and during application of C T Z (50 p M ) . Note that the reduced depression resulted in a higher plateau in C T Z compared to control. Middle: quantal size estimates during C T Z application decreased to lesser extent compared to control reaching a higher plateau. Right: Quantal content estimates during C T Z application showed a decreased use-dependent reduction and a higher plateau. Note increase in content subsequent to the intra-train gap. C) Combined blockade of receptor desensitization and saturation abolished use-dependent alterations in quantal size but not S T D . Left: Mean E P S C amplitudes (6 neurons) before and during combined application o f C T Z (50 p M ) and K Y N (50 pM) . Note the reduced amplitude of the 1 s t E P S C which reached a plateau by the 5 t h response. Middle: quantal size estimates did not change during combined C T Z and K Y N application. Right: Quantal content estimates during combined C T Z and K Y N showed an increased use-dependent reduction and a lower plateau compared to control. Note increase in content subsequent to the intra-train gap. Error bars in 1 s t response show between neuron variations. (P < 0.001 in amplitude data; P < 0.05 in quantal size data; P < 0.01 in quantal content data, Mests). Chapter 3. Results I Ran - 115 -Table 3.2.3A: Summary of effects of C T Z and K Y N on parameters o f S T D Control C T Z (50 p M ) Control C T Z + K Y N (50 p M ) (50 p M ) Parameter S i ( n A ) 1.29 ± 0 . 2 3 1.43 ± 0 . 2 6 1.21 ± 0 . 1 7 0.81 ± 0 . 1 3 * S 2 (nA) 1.04 ± 0 . 2 1 1.27 ± 0 . 2 2 0.97 ± 0 . 1 4 0.62 ± 0 . 1 4 * s 2/s, 0.81 ± 0 . 1 3 0.73 ± 0 . 1 6 0.85 ± 0 . 1 0 0.88 ± 0 . 1 7 S12/S10 1.31 ± 0 . 1 5 1.23 ± 0 . 1 1 1.28 ± 0 . 1 9 1.21 ± 0 . 1 0 Plateau (Si 5-20) (nA) 0.59 ± 0 . 2 1 0.78 ± 0 . 1 9 0.55 ± 0 . 1 6 0.38 ± 0.24 Plateau/Si 0.45 ± 0 . 1 0 0.54 ± 0.09 0.45 ± 0 . 1 4 0.46 ± 0 . 1 7 Apparent Q N (nA) 3.93 ± 0.24 4.62 ± 0 . 3 1 * 3.99 ± 0 . 1 8 2.63 ± 0 . 4 1 * Values are mean ± S E M between neurons; n = 6, * P < 0.05, Mest. Data are from 10 H z trains. Chapter 3. Results I Ran -116 Table 3.2.3B: Effects of C T Z and K Y N on derived parameters of S T D Control C T Z Control C T Z + K Y N (50 p M ) (50 p M ) (50 p M ) Parameter cov(Si,S 2 ) -0.027 ± 0 . 0 0 8 -0.022 ± 0 . 0 1 1 -0.024 ± 0 . 0 1 3 -0.011 ± 0 . 0 0 8 Q' (Si ) (pA) 35.2 ± 2 . 5 38.3 ± 3.0 37.5 ± 5.7 25.3 ± 4.5* Q ' (S 2 ) (pA) 29.4 ± 6 . 4 37.2 ± 2.4 32.6 ± 3 . 9 26.1 ± 2 . 1 * Q'(Si 5-S 2o) 26.1 ± 5 . 7 30.3 ± 3 . 1 27.3 ± 2.8 24.8 ± 3 . 1 (pA) OYSIS-STT,) 0.74 ± 0 . 1 5 0.79 ± 0 . 1 9 0.71 ± 0 . 1 3 0.98 ± 0 . 1 0 * Q'(Si) Var/mean 27.2 ± 4 . 9 32.2 ± 4 . 1 29.4 ± 3 . 7 25.2 ± 5 . 1 (Sl5-S2o) (PA) Evoked minEPSC 32.0 ± 5 . 3 34.5 ± 3 . 9 31.0 ± 2 . 6 27.6 ± 4 . 6 size (pA) Pre-stimulation 11.1 ± 3 . 4 10.5 ± 2 . 2 12.3 ± 1 . 9 8.9 ± 2 . 4 m i n E P S C size (PA) m, 52 ± 8 53 ± 12 34 ± 14 37 ± 1 6 OTI5.20 23 ± 5 35 ± 7 24 ± 4 15 ± 5 wi 2 /mio 1.32 ± 0 . 2 1 1.16 ± 0 . 1 9 1.21 ± 0 . 1 8 1.28 ± 0 . 2 1 Values are mean ± S E M between neurons; n = 6 ; data from same neurons as Table 3.2.3A. * P < 0.05, Mest. Chapter 3. Results 1 Ran - 1 1 7 -beginning of the train, the quantal size decreased significantly compared to the control and remained unchanged throughout the train (Figure 3.24C; Table 3.2.3B). Despite a significant reduction in the amplitudes of the 1 s t and 2 n d EPSCs , the characteristic E P S C rundown during S T D (Plateau/Si) was unaffected by co-application of C T Z and K Y N (Table 3.2.3 A ) . Given the lack of effects on the quantal content, it was concluded that the effects o f co-applied K Y N and C T Z on STD were predominantly postsynaptic. The contribution of receptor desensitization and saturation to S T D seems paradoxical since postsynaptic receptors cannot desensitize or saturate unless release has already occurred at a given site, whereas the negative covariances indicated that sites activated by the 2 n d and 3 r d stimuli are those that were not involved previously. This paradox could be explained by an overflow of transmitter from neighbouring release sites to receptor sites which were not previously activated (Telgkamp et al., 2004). 3.3.4. Effects of pentobarbital on STD 3.3.4.1. EPSC behaviour in trains Pentobarbital enhanced S T D in a dose-dependent manner in the 2.5-20 H z stimulation range (Figure 3.25). The maximal enhancement of S T D was at a concentration of 200 pM which lowered the plateau producing a greater rundown of EPSCs (Plateau/Si; Table 3.2.4A). A t this concentration, pentobarbital increased the ratio of E P S C amplitudes around the intra-train gap (S12/S10; Table 3.2.4A). A parallel effect on the quantal content ratio around the intra-train gap (Table 3.2.4B), suggested an increase in P 0 . Pentobarbital also produced an apparent reduction of the product Q N (Table 3.2.4A). A correction o f Chapter 3. Results I Ran -118-1.0-0.5-0:0-1.0-0.5-0.0-2 .5 H z O Control • PB 50 uM • PB 200 uM 5 10 15 20 stimulus # 5 H z O Control • PB 50 uM • PB 200 uM 10 15 stimulus # 20 o w ft £ "8.3 N O. to £ E «> 1 o z 0.5 0.0 o w ft-g N Q. IS E E «»• 1.0 0.5 0.0 10 H z O Control • PB 50 \iM PB 200 uM 5 10 15 20 stimulus # 2 0 H z O Control • PB 50 uM • PB 200 uM 5 10 15 stimulus # 20 100-50-100! application time PB 50 uM PB200 uM 10 15 time (min) on 75-to Ul o. 50-<D X5 25-o control • P B 2 pM A P B 20 pM * PB 50pM • PB 100 pM • PB200uM 5 10 15 Stimulus frequency (Hz) 20 Figure 3.25. Dose-dependence of pentobarbital enhancement of STD. A) Pentobarbital, applied at 50 and 200 u M , enhanced S T D producing a greater and faster rundown o f E P S C s with increasing stimulation frequencies. Pentobarbital did not abolish the post-gap increase in the 12 t h E P S C amplitude. B) A decrease in the amplitude of the first E P S C amplitude was only observed after 4 min of pentobarbital application. The data were obtained from samples within less than 4 min of drug application. C) Dose dependence of S T D at 2.5, 5, 10, and 20 Hz . Pentobarbital, at 100-200 u M , enhanced S T D at 10 - 20 H z more than at lower stimulation frequencies (P < 0.05, A N O V A ) . Data represent averages obtained from 6 neurons. Chapter 3. Results I Ran Table 3.2.4A: Summary of pentobarbital effects on parameters of S T D Pentobarbital concentrations (uM) Control 50 200 Parameter Si (nA) 1.21 ± 0 . 3 5 1.15 ± 0 . 2 9 1.11 ± 0 . 3 6 S 2 (nA) 0.74 ± 0 . 1 9 0.66 ± 0 . 1 1 0.52 ± 0 . 1 3 S12/S10 1.20 ± 0 . 1 6 1.21 ± 0 . 2 1 1.67 ± 0 . 1 9 * S2/S1 0.63 ± 0.07 0.58 ± 0.08 0.47 ± 0.05 Plateau (S15-S20) (nA) 0.51 ± 0 . 1 0 0.34 ± 0.09 0.23 ± 0.07* Plateau/Si 0.42 ± 0 . 1 2 0.30 ± 0.08 0.21 ± 0 . 0 7 * Apparent Q N (nA) 4.31 ± 0 . 1 3 3.52 ± 0 . 2 1 * 2.70 ± 0 . 3 2 * Values are mean ± S E M between neurons; n = 6; * P < 0.05, A N O V A . Data are from 10 H z trains. Chapter 3. Results I Ran - 120-0.04 % 0.03 • Pentobarbital ( 5 0 | i M ) Pentobarbital ( 2 0 0 pM) T y - T - r f B (a) Before trains (b) After trains (b) - (a) iTo.4 Control 5 8 minis in 3 0 s •m- 10.1 ± 3 .7 pA 10 20 30 40 Size (pA) 6 4 minis in 6 s 0 10 20 30 40 Size (pA) i 3-3 1 . 8 ± 4 . 3 pA ~, 'o=Tib\~ 20 30 Size (pA) Pentobarbital (200 uM) 1.0-,0.8-j 0.6 foM 5 9 minis in 3 0 s 1 0 . 6 ± 3 . 2 pA 5-_ 4 1 10 20 30 Size (pA) 48 minis in 6 s 0 10 20 30 40 Size (pA) p-*i, — 11.1 ± 3 . 6 pA 0 10 20 30 40 Size (pA) Figure 3.26. Quantal alterations mediated pentobarbital effects on S T D . A ) Left, apparent quantal sizes during corticothalamic STD in response to 10 H z stimulation. During application of 200 u M pentobarbital, quantal sizes became significantly smaller starting at the 3 r d response until the end of the train (P < 0.01, A N O V A test). Right: Effects of pentobarbital on quantal contents during corticothalamic STD. A low dose of pentobarbital (50 (iM) decreased whereas a high dose increased the quantal content throughout the E P S C train. Note the significant increase in quantal content after the omitted 11 t h stimulus. During application of 50 or 200 uJVI pentobarbital, quantal content were significantly different than control starting at the 2 n d response until the end o f the train (P < 0.01, A N O V A test). Neurons held at -80 m V . Data obtained from 6 neurons. Error bars in 1 s t response show between neuron variations. Jump after gap o f quantal content implied high P 0 . B) Pentobarbital reduced the amplitude of evoked minEPSCs without affecting pre-stimulation minEPSC amplitude, (a) Histograms of spontaneous minEPSCs obtained 5 s prior to the onset of stimulation (6 repeats, 30 s in total), (b) Histograms of minEPSCs obtained 1 s after the end of the stimulus train (6 repeats, 6 s in total). Evoked minEPSC sizes were obtained after subtraction of pre- from poststimulation evoked minEPSCs (b - a). Total minEPSC counts ('minis') are indicated above histograms. Values next to black arrows are mean size ± S E M . Data are from 1 neuron. Chapter 3. Results 1 Ran 121 Table 3.2.4B: Effect of pentobarbital on derived parameters of S T D Pentobarbital concentrations (pM) Control 50 200 Parameter cov(S,,S 2 ) Q ' (Si ) (pA) Q'(S 2) (pA) Q'(Si5-s 2 o) (PA) Q'(Si) Var/mean (S15-S20) (pA) Evoked minEPSC size (PA) Pre-stimulation m i n E P S C size (pA) m\ m 15-20 mu/m\o 0.022 ± 0 . 0 1 1 35.2 ± 2 . 5 29.4 ± 6.4 25.1 ± 5 . 7 0.71 ± 0 . 1 3 27.2 ± 4.9 32.0 ± 5 . 3 11.1 ± 3 . 4 34 ± 8 20 ± 7 1.18 ± 0 . 2 4 0.025 ± 0 . 0 0 8 -0.031 ± 0 . 0 1 3 36.3 ± 3 . 0 29.6 ± 5 . 0 31.3 ± 8 . 6 19.8 ± 4 . 2 22.3 ± 4 . 2 20.4 ± 8 . 1 12.1 ± 3 . 1 31 ± 5 17 ± 6 21.6 ± 7 . 4 9.7 ± 5 . 1 * 0.55 ± 0 . 1 6 0.37 ± 0 . 1 1 13.5 ± 5 . 2 11.3 ± 7.1 10.8 ± 2 . 9 41 ± 7 23 ± 7 1.21 ± 0 . 1 9 1.85 ± 0 . 1 7 Values are mean ± S E M between neurons; n = 6 ; data from neurons of Table 3.2.4 A ; * P < 0.05, A N O V A . Chapter 3. Results I Ran -122 -Q N for a ((QN/l+a)) showed a similar effect of pentobarbital ( a values were in control 0.17 ± 0.03 and in 200 p M pentobarbital 0.09 ± 0.02, P < 0.05, Mest; the corrected Q N values were 3.68 ± 0.17 n A in control and 2.45 ± 0.26 n A in 200 p M pentobarbital, P < 0.05, Mest). These effects were use-dependent and, hence, did not affect the 1 s t E P S C in the train during application periods of < 4 minutes (Figure 3.25 B) . Wi th pentobarbital, a major component of STD appeared to be a use-dependent reduction in quantal size (Figure 3.26; Table 3.2.4B). The ratio of the plateau to the 1 s t apparent quantal size (Q'(S 15-20)/ Q'(Si)) decreased from 71 % in the control to 37% during pentobarbital application. The same effect was seen in reduction in the size o f evoked m i n E P S C without changes in the spontaneous pre-stimulation minEPSCs (Figure 3.26; Table 3.2.4B). The modulation of quantal parameters contingent on stimulation, and lack of effects on pre-stimulation minEPSCs, suggested that pentobarbital produced smaller size quanta either by a presynaptic action of selecting sites with small quanta, or a postsynaptic action confined to activated synaptic sites. 3.3.4.2. STD in raised Ca2+concentration The next set o f experiments examined i f pentobarbital effects on S T D could be modulated during conditions of high release probability. A s previously, raising [ C a 2 + ] e from 2 to 8 m M produced a greater rundown of EPSCs . Under these conditions, pentobarbital (200 p M ) produced an even greater rundown, reducing the plateau to the 1 s t E P S C ratio (plateau/Si) to 0.19 ± 0.08 (P < 0.05, A N O V A ; Table 3.2.5A). A t this concentration of pentobarbital, the amplitude ratio of the EPSCs around the intra-train Chapter 3. Results I Ran 94-Table 3.2.5A: Pentobarbital effects on parameters of STD in raised [Ca ] e Control High C a Pentobarbital Pentobarbital 50 p M 200 p M ( 2 m M C a 2 + ) (8 m M ) (8 m M C a 2 + ) (8 m M C a 2 + ) Parameter S i ( n A ) 1.11 ± 0 . 4 2 1.66 ± 0 . 3 5 1.59 ± 0 . 3 9 1.54 ± 0 . 4 1 S 2 (nA) 0.69 ± 0 . 1 9 0.94 ± 0.23 0.71 ± 0 . 1 5 0.55 ± 0 . 2 1 S2/S1 0.63 ± 0 . 1 1 0.57 ± 0 . 1 4 0.45 ± 0 . 1 0 0.36 ± 0 . 1 6 S12/S10 1.20 ± 0 . 1 0 1.35 ± 0 . 0 8 1.38 ± 0 . 1 7 1.91 ±0.21*** Plateau (Si 5-20) (nA) 0.49 ± 0 . 1 0 0.58 ± 0 . 1 3 0.45 ± 0 . 1 5 0.27 ± 0.09** Plateau/ Si 0.44 ± 0 . 1 0 0.36 ± 0 . 1 1 0.28 ± 0.09 0.19 ± 0 . 0 8 * Apparent Q N (nA) 3.90 ± 0 . 1 3 5.35 ±0 .19* 3.96 ±0 .21** *** 3.20 ± 0 . 1 7 Values are mean ± S E M between neurons; n = 6; * (relative to control) ** (relative to high C a 2 + ) ***(relative to control and C a 2 + ) P < 0.05, A N O V A . Data are from 10 H z trains. Chapter 3. Results I Ran - 124 -Table 3.2.5B: Pentobarbital effects on derived values of S T D in raised [Ca ] e Control (2 m M C a 2 + ) High Ca2+ (8 m M ) Pentobarbital 50 u M (8 m M C a 2 + ) Pentobarbital 200 u M (8 m M C a 2 + ) cov(Si,S 2 ) 0.017 ± 0 . 0 1 2 - 0.023 ± 0.009 -0.026 ± 0 . 0 1 3 -0.031 ± 0 . 0 1 7 Q' (Si ) (pA) 34.9 ± 8 . 1 35.4 ± 9 . 7 32.7 ± 8 . 1 30.5 ± 7.7 Q'(S 2) (pA) 33.8 ± 9 . 2 32.6 ± 8.6 29.2 ± 7.3 24.7 ± 9 . 1 Q'(Si 5-S 2o) (pA) 29.4 ± 7.6 28.9 ± 6 . 9 25.6 ± 8 . 2 18.5 ± 10.3 OYS„-S,n) Q'(Si) 0.84 ± 0 . 1 4 0.81 ± 0 . 1 0 0.78 ± 0 . 1 2 0.60 ± 0.08* Var/mean ( S 1 5 - S 2 0 ) (PA) 30.8 ± 6 . 7 29.7 ± 8 . 5 27.6 ± 9 . 1 22.1 ± 7 . 9 Evoked minEPSC size (pA) 33.1 ± 5 . 3 31.8 ± 9 . 4 28.3 ± 6.5 25.2 ± 8.4 Pre-stimulation 12.3 ± 2.9 minEPSC size (pA) 11.5 ± 4 . 3 10.6 ± 3 . 1 9.6 ± 3 . 7 m\ 31 ± 9 46 ± 11 48 ± 13 50 ± 15 m 15-20 16 ± 5 21 ± 8 1 7 ± 9 14 ± 13 mnlmxo 1.17 ± 0.13 1.36 ± 0 . 1 7 1.40 ± 0 . 1 2 1.95 ± 0 . 1 6 * Values are mean ± S E M between neurons; n = 5 ; data from neurons of Table 3.2.5A; * P < 0.05, A N O V A test. Chapter 3. Results I Ran - 125 -2_|_ Table 3.2.6A: Pentobarbital effects on parameters of S T D in low [Ca ] e Control (2 m M C a 2 + ) Low C a 2 + (0.1 m M ) Pentobarbital 50 p M (0.1 m M C a 2 + ) Pentobarbital 200 p M (0.1 m M C a 2 + ) Parameter S i ( n A ) 1.15 ± 0 . 4 2 0.41 ± 0.37 0.39 ± 0.22* 0.35 ± 0.27* S 2 ( n A ) 0.72 ± 0 . 3 1 0.29 ± 0.20 0.24 ± 0 . 1 9 0.19 ± 0 . 1 5 * S2/S1 0.63 ± 0 . 1 4 0.73 ± 0 . 1 1 0.64 ± 0 . 1 2 0.50 ± 0 . 1 5 S12/S10 1.20 ± 0 . 1 1 1.01 ± 0 . 0 5 1.21 ± 0 . 0 8 1.67 ± 0 . 0 9 * Plateau (Si 5-20) (nA) 0.50 ± 0 . 1 5 0.23 ± 0 . 1 1 * 0.19 ± 0 . 0 9 * 0.12 ± 0 . 1 0 * Plateau/Si 0.43 ± 0.09 0.56 ± 0.06 0.48 ± 0 . 1 0 0.34 ± 0.08** Apparent Q N (nA) 4 . 1 0 ± 0 . 1 5 1.75 ±0 .24* 1.43 ±0 .19* 1.17 ± 0.15** Values are mean ± S E M between neurons; n = 5, * (relative to control) ** (relative to low C a 2 + ) P < 0.05, A N O V A . Data are from 10 H z trains. Chapter 3. Results I Ran -126 • Table 3.2.6B: Pentobarbital effects on derived parameters of S T D in low [Ca ] e Control L o w C a 2 + Pentobarbital Pentobarbital 50 u M 200 u M ( 2 m M C a 2 + ) (0.1 m M ) ( 0 . 1 m M C a 2 + ) ( 0 . 1 m M C a 2 + ) Parameter cov(S,,S 2 ) -0.023 ± 0 . 0 1 1 -0.014 ± 0 . 0 0 8 -0.017 ± 0 . 0 0 9 -0.021 ± 0 . 0 1 3 Q' (Si ) (pA) 32.3 ± 9 . 1 14.2 ± 6 . 8 * 13.8 ± 8 . 8 * 12.7 ± 7.9* Q ' (S 2 ) (pA) 30.3 ± 8 . 1 13.7 ± 9 . 2 * 12.9 ± 7 . 5 * 12.1 ± 6 . 9 * Q'(Si5-S2o) 28.1 ± 7 . 7 14.4 ± 6 . 9 11.8 ± 8 . 3 * 9.4 ± 7.0* (PA) ** OYS^-STO) 0.87 ± 0 . 1 7 1.01 ± 0 . 1 3 0.85 ± 0 . 1 4 0.74 ± 0 . 1 1 Q'(Si) Var/mean 29.3 ± 9.0 14.5 ± 7 . 4 13.7 ± 8 . 2 11.5 ± 7 . 3 * (Si 5-S 2o) (PA) Evoked 35.2 ± 6 . 1 15.1 ± 8 . 8 13.4 ± 5 . 1 10.8 ± 4 . 7 * minEPSC size (pA) Pre-stimulation 11.7 ± 4.6 1 2 . 4 ± 5 . 8 11.4 ± 3.7 11.6 ± 4.1 minEPSC size (PA) mx 35 ± 10 28 ± 12 29 ± 9 27 ± 13 mi5-2o 17 ± 8 16 ± 9 16 ± 7 12 ± 6 mx2lmxQ 1.19 ± 0 . 0 9 0.99 ± 0 . 1 2 1.21 ± 0 . 1 4 1.60 ±0 .11** Values are mean ± S E M between neurons; n = 5 ; data from neurons of table 3.2.6A; * (relative to control) ** (relative to low C a 2 + ) P < 0.05, A N O V A . Chapter 3. Results 1 Ran -127 -gap (S12/S10) nearly doubled. The reduction in apparent Q N was similar to that observed 9+ in normal Ca (cf. Table 3.2.4A). Hence, pentobarbital effects on S T D were additive to the increased E P S C rundown caused by raised C a 2 + . Raising [Ca ] e did not greatly alter pentobarbital effects on quantal size and content (Table 3.2.5B). In raised [Ca 2 + ] e , pentobarbital decreased the ratio of the plateau to the 1 s t apparent quantal size and increased the ratio of the quantal contents around the intra-train gap, similar to the effects observed in 2 m M [ C a 2 + ] e (Table 3.2.5B, cf. Table 3.2.4B). However, reduction in Q' was much less than in 2 m M [Ca 2 + ] e . 3.3.4.3. STD in reduced Ca concentration 9+ Reducing [Ca ] e from 2 to 0.1 m M did not affect the pentobarbital enhancement o f S T D (Table 3.2.6A, B) . In low [Ca 2 + ] e , pentobarbital still produced a greater rundown then it did in normal [ C a 2 + ] e media. The effects of low [ C a 2 + ] e to reduce the apparent quantal size, variance-mean ratio, and evoked minEPSC size were farther modified by pentobarbital, which caused further rundown in Q', and an increased ratio o f the quantal contents around the intra-train gap. This unexpected latter effect implicates a presynaptic action on the high release pool of quantal packets which mediate S T D in low [ C a 2 + ] e media. In summary, pentobarbital enhancement of STD was resistant to reductions in [Ca 2 + ] e . Chapter 3. Results I Ran - 128 -3.3.4.4. Combined cyclothiazide and kynurenate blockade In order to unmask the presynaptic actions of pentobarbital, it was necessary to reduce the postsynaptic contributions to STD. In other neurons, pentobarbital has been reported to have postsynaptic actions of promoting A M P A receptor desensitization (Jackson et al., 2003), which would contribute to STD. Receptor saturation is also a possible contributor to S T D (Chen et al., 2002). For these reasons, pentobarbital actions on S T D were re-examined during pharmacological blockade of receptor desensitization and saturation. Pentobarbital enhancement of S T D was unaffected by a combined blockade o f receptor desensitization and saturation. During co-application of C T Z (50 p M ) with K Y N (50 p M ) , pentobarbital (200 p M ) still increased the rundown of EPSCs in a use-dependent manner (Figure 3.27; Table 3.2.7A). The apparent quantal size plateau (Q'(S 15-20)), ratio o f Q'(Si5-S2o)/ Q 'S i , variance-mean ratio, and evoked minEPSC size all decreased in response to pentobarbital (Table 3.2.7B). In addition, the quantal content plateau (m\s-2o) also increased significantly relative to control or co-applied C T Z + K Y N (Table 3.2.7B). The above data implicated an enhancement of S T D by a presynaptic action that reduced the quantal size. 3.3.5. Effects of altered extracellular concentration ([Jt]e) on STD Transmitter release is sensitive to [ K + ] e alterations, which modify the membrane potential in the nerve terminal (Hatt and Smith 1976; Saint et al. 1987). This provided a rationale to examine the effects of altering [ K + ] e on S T D and their modulation by pentobarbital. Chapter 3. Results I Ran - 129-O Control • Cyclothiazide (50 pM) + Kynurenate (50 p.M) • Cyclothiazide (50 pM) + Kynurenate (50 pM) +Pentobarbital (200 pM) o4 , 1 , , o.oo-l , , , , o-l 1 , 1 , 0 5 10 15 20 0 5 10 15 20 0 5 10 15 20 stimulus # stimulus # stimulus # Figure 3.27. Pentobarbital enhancement of S T D during combined blockade of receptor desensitization and saturation. Co-application of C T Z (50 uJVI) with K Y N (50 u M ) reduced the amplitude of EPSCs (left) and abolished the decrease in quantal size throughout the train (middle). The quantal content (right) had a time course similar to that o f E P S C amplitude (left) in the presence of co-applied C T Z and K Y N . Under these conditions, pentobarbital application enhanced E P S C depression (left) and reduced the quantal size (middle) throughout the duration of E P S C train. In parallel, pentobarbital increased the quantal content (right) during the train at a rate similar, and possibly due to the decrease in quantal size. Data are from 5 neurons. Holding potential was -80 m V . Error bars of 1 s t response show between neuron variations. Chapter 3. Results I Ran -130 -Table 3.2.7A: Pentobarbital effects on parameters of STD during co-applied C T Z and K Y N Control C T Z + K Y N (50 p M ) (50 p M ) Pentobarbital + C T Z + K Y N (200 p M ) (50 p M ) (50 p M ) Parameter S , (nA) 1.29 ± 0 . 1 7 0.89 ± 0 . 1 3 * 0.85 ± 0 . 1 4 * S 2 ( n A ) 1.04 ± 0 . 1 3 0.62 ±0 .14* 0.60 ± 0 . 1 8 * S2/S1 0.85 ± 0 . 1 0 0.88 ± 0 . 1 7 0.70 ± 0 . 1 1 S12/S10 1.28 ± 0 . 1 9 1.21 ± 0 . 1 0 1.15 ± 0.12 Plateau (S15-20) (nA) 0.61 ± 0 . 1 6 0.49 ± 0 . 1 4 0.34 ± 0 . 1 1 Plateau/Si 0.47 ± 0.08 0.55 ± 0.05 0.40 ± 0.06** Apparent Q N (nA) 3.99 ± 0 . 1 8 2.63 ± 0 . 4 1 3.42 ± 0.27 Values are mean ± S E M between neurons; n = 6;* (relative to control) ** (relative to C T Z + K Y N ) P < 0.05 A N O V A test. Data are from 10 H z trains. Chapter 3. Results I Ran -131 -Table 3.2.7B: Pentobarbital effects on derived parameters of S T D during co-applied C T Z and K Y N Control C T Z + K Y N Pentobarbital + C T Z + K Y N (50 uM) (50 uM) (200 u M ) (50 u M ) (50 u M ) Parameter cov(Si,S 2 ) -0.024 ± 0 . 0 1 3 -0.011 ± 0 . 0 0 8 -0.019 ± 0 . 0 1 1 Q' (Si ) (pA) 37.5 ± 5 . 7 25.3 ± 4.5 24.5 ± 5.3 Q' (S 2 ) (pA) 32.6 ± 3 . 9 26.1 ± 2 . 1 * 17.6 ± 1 . 9 * Q'(Si5-S 2 0) 27.3 ± 2 . 8 24.8 ± 3 . 1 4.1 ± 3 . 8 (PA) O ' fSiyS™) 0.71 ± 0 . 1 3 0.98 ± 0 . 1 0 0.17 ± 0 . 1 1 * Q'(Si) Var/mean 29.4 ± 3 . 7 25.2 ± 5 . 1 6.9 ± 4 . 4 * (Sl5-S2o) (PA) Evoked minEPSC 31.0 ± 2 . 6 27.6 ± 4 . 6 10.3 ± 3 . 8 * size (pA) Pre-stimulation 12.3 ± 1 . 9 8.9 ± 2.4 9.8 ± 2 . 1 minEPSC size (PA) mi 34 ± 14 37 ± 16 36 ± 1 0 /ni5_2o 24 ± 4 15 ± 5 78 ± 9 * mn/mio 1.21 ± 0 . 1 8 1.28 ± 0 . 2 1 1.09 ± 0 . 1 7 Values are mean ± S E M between neurons; n = 6 ; data from same neurons as Table 3.2.7A; * (relative to control) ** (relative to C T Z + K Y N ) P < 0.05, A N O V A . Chapter 3. Results I Ran - 132 -3.3.5.1. High [K^Je perfusion Raising [ K + ] e from 2.5 to 10 m M caused significant changes in the shape of S T D (Figure 3.28A; Table 3.2.8A). The amplitude of the 2 n d E P S C facilitated relative to the 1 s t E P S C (Figure 3.28A). The E P S C plateau increased whereas the rundown (Plateau/Si ratio) was raised compared to control [ K + ] e conditions (Table 3.2.8A). The apparent Q N nearly doubled in response to high [ K + ] e application (Table 3.2.8A), indicating an enhancement of refill or recruitment of sites previously with low fractional release, now to be included in the releasable pool (Quastel, 1997). Unexpectedly, high [ K + ] e application abolished the covariance between the 1 s t and 2 n d E P S C , yielding a small positive value (Table 3.2.8B). This might or might not be a statistical aberration. The apparent quantal size, variance-mean ratio and evoked minEPSC size were unaffected by high [ K + ] e (Table 3.2.8B). These data implicated a presynaptic action that increased the fractional release and pulse to pulse facilitation. In 10 m M [ K + ] e , pentobarbital had a marked effect on the shape of E P S C train. Pentobarbital effects included: 1) transforming the facilitation between the 1 s t and 2 n d E P S C into depression; 2) decreasing the plateau; and, 3) increasing the rundown of EPSCs , (Figure 3.28 A ; Table 3.2.8A). The apparent Q N also decreased, similar to the effects in normal [ K + ] e media (cf. Table 3.2.4A). A similar effect of pentobarbital was observed, after correcting Q N for a values (QN/(l+a); a values were 0.15 ± 0.04 in control, 0.16 ± 0.02 in high [ K + ] , and 0.12 ± 0.03 in pentobarbital/high [K + ] (P > 0.05, t-test). The corrected Q N values were (in nA) 5.11 ± 0.22 in control, 9.38 ± 0.41 in high Chapter 3. Results I Ran - 133 -[ K + ] e , and 5.04 ± 0.37 in pentobarbital and high [ K + ] e (P < 0.05, Mest). Pentobarbital also restored the negative covariance at the beginning of the train. The use-dependent effects of pentobarbital on the decline of the apparent quantal size early in the train (QYJS15-S2o)/Q'(Si) ratio) were similar to the effects in normal [ K + ] e (Table 3.2.8B). The total pentobarbital suppression of the increase in S2/S1 produced by 10 m M [ K + ] e indicated a presynaptic effect of pentobarbital distinct from what was seen in other experiments. The apparent increase in Q N by 10 m M K + reflects the raised S2, S3, S4 - presumably reflecting a combination of high a and rising P 0 . The unchanged Si in high [ K + ] e indicated that Q N (~6 nA) did not change at the beginning of the train. The data indicate fractional release of about 0.3 for Si and S2 under control conditions, with 10 m M [ K + ] e raising fractional release to about 0.9 or more at S2. Hence, hypothetically, fractional release could have increased greatly i f a after the first pulse remained low, or could have remained nearly unchanged, i f a were to become near 1, as suggested by the absence of negative covariance, or anywhere in between, depending upon how a is postulated to have changed. 3.3.5.2. Low [Kf]e perfusion Reducing [ K + ] e concentration did not affect S T D (Table 3.2.9 A , B) . However, the mean ratio of E P S C amplitudes and quantal contents around the intra-train gap decreased compared to normal [ K + ] e (Table 3.2.9A, B) . This observation suggested that the post-gap increases in E P S C size and quantal content were attenuated as a result o f the low [ K + ] e Chapter 3. Results I Ran -134-B 1 Control (2.5 mM K*) pWtt 0.1 mM K+ PB 200 uM (0.1 mM K) 1/nA 0.5s © C o n t r o l • Low K' • PB (Low K + ) ;._,_:.:.:::_^ Q.OO I t ... . .! : , • 0 i • - • T T " " t 1 •' 0 5 10 15 . 20 o 5 10: 15 20 0 5 10 15 20 stimulus # stimulus* .stimulus.*! Figure 3.28. Effects of altered [ K + ] e on pentobarbital enhancement o f S T D . A) Effects o f raised [ K + ] e . Top: E P S C traces during application of 2.5 m M [ K + ] e (left), 10 m M [ K + ] e (middle), and 200 u M pentobarbital in 10 m M [ K + ] e (right). Bottom: Mean E P S C amplitudes (left), quantal size estimates (middle), and quantal contents (right) from control, raised [ K + ] , and pentobarbital in raised [ K + ] e . Data are from 5 neurons. B) Effects of reduced [ K + ] e . Top: E P S C traces during application of 2.5 m M [ K + ] e (left), 0.1 m M [ K + ] (middle), and 200 u M pentobarbital in 0.1 m M [ K + ] e (right). Bottom: Mean E P S C amplitudes (left), quantal size estimates (middle), and quantal contents (right) from control, reduced [ K + ] e , and pentobarbital in reduced [ K + ] e . Traces in A and B are responses to single trains from one neuron each. Other data are averages from 5 neurons. Error bars in 1 s t response show between-neuron variations. Chapter 3. Results I Ran - 135-Table 3.2.8A: Summary of effects of high [ K + ] e , pentobarbital on parameters o f S T D Control High K + Pentobarbital 200 p M (2.5 m M K + ) (10 m M ) (in high K + ) Parameter S i ( n A ) 1.78 ± 0 . 2 5 1.75 ± 0 . 1 2 1.73 ± 0 . 1 1 S 2 (nA) 1.08 ± 0 . 1 5 1.92 ±0 .19* 1.24 ±0.14** S2/S1 0.60 ± 0 . 1 1 1.08 ±0 .17* 0.71 ±0.15** S12/S10 1.31 ± 0 . 1 0 1.05 ±0 .14* 1.45 ±0.17** Plateau (Si 5-20) (nA) 0.67 ± 0 . 1 5 0.93 ±0 .19* 0.52 ±0.16** Plateau/Si 0.38 ± 0 . 1 0 0.53 ± 0 . 1 9 * 0.31 ±0 .21** Apparent Q N (nA) 5.91 ± 0 . 2 5 10.83 ± 0 . 3 3 * 5.65 ± 0.47** Values are mean ± S E M between neurons; n = 5; * (relative to control) ** (relative to high K + ) P < 0.05 A N O V A test. Data are from 10 H z trains. Chapter 3. Results I Ran - 136 -Table 3.2.8B: Derived parameters of S T D in high [ K + ] e , pentobarbital Control High K + Pentobarbital 200 p M (2.5 m M K + ) (10 m M ) (in 10 m M K + ) Parameter cov(Si,S 2 ) -0.017 ± 0 . 0 0 7 0.003 ± 0.005* -0.011 ± 0 . 0 0 9 Q ' (S i ) (pA) 34.1 ± 7 . 4 33.8 ± 3 . 0 33.6 ± 2 . 5 Q*(S 2)(pA) 29.0 ± 6 . 9 28.7 ± 2 . 9 28.3 ± 2.6 Q ' (Si5-S 2 0) 28.3 ± 6 . 1 29.9 ± 1 0 . 8 18.0 ± 7 . 1 (pA) OYSis-SW) 0.83 ± 0 . 1 0 0.88 ± 0 . 1 7 0.53 ± 0.09* Q'(Si) Var/Mean 30.1 ± 4.4 32.3 ± 5.1 20.3 ± 6.2 (Si5-S2rj) (PA) Evoked minEPSC 32.0 ± 5 . 3 31.8 ± 6 . 6 24.3 ± 4 . 2 size (pA) Pre-stimulation 11.1 ± 3 . 4 12.3 ± 4.1 1 0 . 9 ± 3 . 7 minEPSC size (PA) m, 54 ± 8 52 ± 6 79 ± 1 0 ml5.2o 23 ± 3 32 ± 4 44 ± 6 mn/ml0 1.32 ± 0 . 0 7 1.09 ± 0 . 0 9 1.15 ± 0 . 1 1 Values are mean ± S E M between neurons; n = 5 ; data from same neurons as table 3.2.8A; * (relative to control) ** (relative to high K + ) P < 0.05 A N O V A test. Chapter 3. Results I Ran Table 3.2.9A: Summary of effects of low [ K + ] e , pentobarbital on parameters o f S T D Control (2.5 m M K + ) Low K + (0.1 m M ) Pentobarbital 200 p M (in 0.1 m M K + ) Parameter S i ( n A ) 1.65 ± 0 . 2 0 1.60 ± 0 . 1 7 1.55 ± 0 . 2 5 S 2 ( n A ) 1.10 ± 0.17 0.99 ± 0 . 1 0 1.42 ± 0 . 1 5 s 2/s, 0.66 ± 0 . 1 3 0.40 ± 0 . 1 8 0.91 ± 0 . 2 5 Si 2 /Sio 1.21 ± 0 . 0 7 1.01 ± 0 . 0 9 1.40 ± 0.11* Plateau (Sl5-2fj) (nA) 0.67 ± 0 . 1 8 0.60 ± 0 . 1 1 0.39 ± 0 . 1 7 Apparent Q N (nA) 5.90 ± 0 . 3 0 5.25 ± 0.23 6.25 ± 0.43* Values are mean ± S E M between neurons; n = 5; * (relative to low K + ) P < 0.05, A N O V A test. Data are from 10 H z trains. Chapter 3. Results I Ran - 138 Table 3.2.9B: Derived parameters of S T D in low [ K + ] e , pentobarbital Control Low K + Pentobarbital 200 p M (2.5 m M K + ) (0.1 m M ) (in 0.1 m M K + ) Parameter cov(Si,S 2 ) -0.019 ± 0 . 0 0 9 -0.015 ± 0 . 0 0 7 -0.009 ± 0 . 0 0 6 Q' (Si ) (pA) 33.4 ± 2 . 5 33.0 ± 11.3 32.3 ± 8.9 Q ' (S 2 ) (pA) 29.5 ± 2 . 0 28.7 ± 9 . 5 27.8 ± 1 2 . 1 Q'(Si5-S 2 0) 28.5 ± 5 . 7 28.0 ± 8 . 1 21.8 ± 6 . 9 (pA) OYSis-S?^ 0.85 ± 0 . 1 2 0.84 ± 0 . 1 9 0.67 ± 0 . 1 1 Q'(S.) Var/Mean 31.4 ± 6 . 1 30.5 ± 7.0 23.4 ± 8 . 2 (Si5-S2o) (pA) Evoked minEPSC 30.1 ± 4.1 29.7 ± 8.0 26.1 ± 5.1 size (PA) Pre-stimulation 9.7 ± 2 . 4 10.1 ± 3 . 2 9.9 ± 3 . 9 m i n E P S C size (pA) wi 54 ± 9 50 ± 5 51 ± 8 W15-20 23 ± 9 21 ± 6 17 ± 6 ml2/ml0 1.33 ± 0 . 2 3 1.04 ± 0 . 1 6 1.43 ± 0 . 1 4 * Values are mean ± S E M between neurons; data from same neurons as table 3.2.9A; n = 5, * (relative to low K + ) P < 0.05 A N O V A test. Chapter 3. Results I Ran - 139 -contrary to effect of nerve terminal hyperpolarization at the neuromuscular junction which would theoretically increase release (Hubbard et al., 1967). In low [ K + ] e , pentobarbital had few effects on S T D (Table 3.2.9A, B) . The sole significant effect was to increase the post-gap jump in E P S C amplitude (Table 3.2.9A). This effect is compatible with a raised fractional release (P 0), raised P 0-(l-oc), and/or lowered a, and occurred in conjunction with increases in the quantal content (Figure 28B, Table 3.2.9B). Hence, reducing [ K + ] e revealed release-promoting actions of pentobarbital, distinct from those observed in normal or raised [ K + ] e . 3.3.6. Effects of tetrodotoxin (TTX) The following investigations examined the effects of partial blockade o f voltage-gated N a + channels on STD. Application of 8 to 64 n M T T X caused significant changes in the configuration of S T D (Figure 3.29). A t 32 n M , T T X reduced the 1 s t and 2 n d E P S C amplitudes and increased the rundown of the E P S C train (Figure 3.29; Table 3.2.1 OA). A t this concentration, T T X also significantly reduced the apparent Q N . Unexpectedly, the ratio o f E P S C s around the intra-train gap increased to a value > 2. This observation was in contrast to the prediction of the binomial model that the amplitude of the E P S C following the intra-train gap could increase only to twice as much as the previous E P S C amplitude. This result suggested that some blockade of presynaptic action potentials reversed the increased gap between stimuli. A t 64 n M , T T X irreversibly abolished E P S C s (Figure 3.29), as expected i f presynaptic action potentials were completely blocked. Chapter 3. Results IRan - 1 4 0 -T T X effects on S T D also involved a use-independent reduction in quantal size in the entire E P S C train (Figure 3.29; Table 3.2.1 OB). The decrease in the apparent quantal size was significant already at the 1 s t and 2 n d responses (Q'(Si), Q'(S2))- The plateau o f the apparent quantal size (Q'(Si5-2o)) and its ratio to the 1 s t quantal size (Q'(Si5-S2o)/ Q'(Si)) also decreased significantly during T T X (Table 3.2.1 OB). These effects of T T X coincided with reductions in the variance-mean ratio and the evoked minEPSC size (Figure 3.30; Table 3.2.1 OB). The amplitude o f pre-stimulation minEPSCs was unaffected by T T X application, arguing against a postsynaptic action. Hence, T T X effects on S T D likely involved a presynaptic use-dependent action that produced smaller quanta, plus an effect to reduce quantal size that did not reverse in the 20 s between-train period. In order to unmask the net presynaptic actions of T T X , it was necessary to re-examine its effects in conditions that reduce postsynaptic contributions to STD. For this reason, the experiments were repeated with co-applied C T Z and K Y N . A s previously observed, co-application of C T Z (50 p M ) with K Y N (50 p M ) decreased the 2 n d and plateau E P S C amplitudes as well as the apparent Q N (cf. Table 3.2.3A). A subsequent application o f T T X (32 n M ) produced a lower E P S C plateau and apparent Q N values and a faster E P S C rundown, similar to the effects without co-applied C T Z + K Y N (cf. Table 3.2.10A). In these experiments, T T X affected quantal parameters nearly the same way as without co-applied C T Z and K Y N (cf. Table 3.2.10B). The apparent quantal size decreased throughout the train (Table 3.2.1 IB) . The plateau of apparent quantal size, variance-mean Chapter 3. Results 1 Ran -141 -B Control Mtyfmrr'yrcrmi "g 1.0 Q. E to a 0.5 TTX 8 nM TTX 32 nM o control • TTX 8nM • TTX 32 nM i+iHi Wr%i stimulus # TTX 64 nM 0.5 s 0.5 nA *4 ^ Figure 3.29. Tetrodotoxin enhanced STD by reducing quantal size. A ) Traces o f E P S C trains during application of T T X at 8, 32, and 64 n M . A t 8 - 32 n M T T X application enhanced S T D producing a greater and faster rundown of EPSCs . A t 32 n M , T T X decreased the amplitude of the EPSCs early in the train. Application o f T T X at 64 n M resulted in irreversible loss of EPSCs . B) Left: Mean E P S C amplitude from 6 neurons before and during application of 8 and 32 n M T T X . Middle: Apparent quantal size estimates during STD. A t 32 n M , T T X reduced the quantal size significantly (P < 0.01, t-test). The quantal size decreased and reached a plateau at response 4. Right: Quantal content during T T X enhancement of STD. Note the increase in quantal content following the intra-train gap. Data are from 5 neurons. Holding potential was - 80 m V . Error bars of first response show between-neuron variations. Chapter 3. Results I Ran - 1 4 2 -Control TTX 32 nM Before trains 66 minis In 30 «' 10.9 • 2.7 pA B 0 10 20 30 40 Size (pA) After trains - * 4 I 3' | 2 & i 69 minis in 6 s 0 10 20 30 40 Size (pA) B - A Mr f o . 59 minis in,3Q s ~f\\ 11.2 ± 2.3 pA ;0 10 20 30 40 Size (pA) 38 minis in 6 s 0 10 20 30 40 Size (pA) 32:9 ±2 .2 pA 0 ' 1.10, 20 30 40 Size(pA) 4 i . 3 12.4 ±3.4 pA -V 0 10 20 30 40 Size(pA) Figure 3.30. T T X decreased the size of evoked miniature EPSCs without affecting spontaneous miniature E P S C size. A ) Amplitude histograms of spontaneous minEPSCs 5 s before the stimulus train (6 repeats, 30 s in total). B) Amplitude histograms of evoked minEPSCs counted 1 s after the stimulus train (6 repeats, 6 s total). Total mini count is indicated above histograms. Evoked minEPSC sizes were obtained after subtraction of spontaneous from evoked minEPSC histogram (B - A ) . Values next to black arrows pointing at peaks are mean size ± S E M . Data are from 1 neuron. Chapter 3. Results I Ran Table 3.2.10A: Summary of T T X effects on parameters of S T D T T X concentrations (nM) Control 8 32 Parameter Si (nA) 1.19 ± 0 . 3 3 0.95 ± 0.36 0.49 ± 0.28* S 2 (nA) 0.87 ± 0 . 1 4 0.80 ± 0 . 3 4 0.40 ± 0 . 1 6 * S2/S1 0.73 ± 0.07 0.84 ± 0.09 0.81 ± 0 . 1 1 S12/S10 1.51 ± 0 . 1 1 1.40 ± 0 . 1 9 2.52 ± 0 . 4 1 * Plateau (Si 5-20) (nA) 0.39 ± 0.08 0.20 ± 0 . 1 2 0.12 ± 0 . 0 9 * Plateau/ Si 0.32 ± 0 . 1 1 0.21 ± 0.07 0.24 ± 0.08 Apparent Q N (nA) 4.40 ± 0 . 1 7 4.59 ± 0 . 3 1 1.82 ± 0 . 3 5 * Values are mean ± S E M between neurons; n = 5, * P < 0.05, A N O V A . Data are from 10 H z trains. Chapter 3. Results I Ran - 1 4 4 -Table 3.2.10B: T T X effects on derived parameters of S T D T T X concentrations (nM) Control 8 32 Parameter cov(S,,S 2 ) Q'(Si) (pA) Q'(S 2 ) (pA) Q'(Si5-S 2 0) (PA) Q X S J S ^ Q ) Q'(Si) Var/Mean (Si5-S2o) (PA) Evoked minEPSC size (pA) Pre-stimulation minEPSCsize (PA) m\ m 15 -20 mn/mio 0.026 ± 0 . 0 1 2 33.5 ± 2 . 5 27.8 ± 2.0 24.5 ± 3.6 0.73 ± 0 . 1 1 27.8 ± 4.7 32.1 ± 4 . 1 10.1 ± 2 . 0 35 ± 6 18 ± 3 1.52 ± 0 . 2 5 -0.019 ± 0 . 0 1 3 34.6 ± 6.3 30.0 ± 5 . 3 19.8 ± 4 . 3 0.57 ± 0.09 22.9 ± 5 . 1 22.3 ± 3.0 10.7 ± 1.6 27 ± 10 1 0 ± 7 1.42 ± 0 . 3 1 0.014 ± 0 . 0 0 9 17.8 ± 5 . 1 * 14.2 ± 3 . 6 * 7.9 ± 3 . 8 * 0.44 ± 0.07* 9.6 ± 2 . 8 * 12.9 ± 3 . 3 * 9.8 ± 1.9 28 ± 9 1 6 ± 4 2.52 ± 0.27* Values are mean ± S E M between neurons; data from same neurons as Table 3.2.1 OA; n = 5, * P < 0.05, A N O V A . Chapter 3. Results I Ran -145-O control • CTZ+KYN • CTZ+KYN+TTX T N-T f f T T * T 10 15 20 stimulus # "?7"T r r r r r r r r 8 75 40-5 10 15 20 stimulus # Figure 3.31. T T X effects on S T D during blockade of receptor desensitization and saturation. Co-application of C T Z (50 m M ) with K Y N reduced the amplitude of E P S C s without affecting the rundown (left). Under these conditions, the quantal size (middle) did not change throughout the train whereas the quantal content (right) followed the time course of the E P S C train. A subsequent application of T T X (32 nM) enhanced S T D reducing the plateau and increasing the rundown. T T X decreased the quantal size, already at the 2 n d response and increased the quantal content, from the 8 t h response until the end o f the train (right). Data are from 5 neurons. Holding potential was -80 m V . Error bars o f 1 s t response show between-neuron variations. Chapter 3. Results I Ran -146-Table 3.2.11 A : T T X effects on parameters of STD during co-applied C T Z + K Y N Control C T Z + K Y N (50 p M ) (50 p M ) T T X + C T Z + K Y N (32 nM) (50 p M ) (50 p M ) Parameter S i ( n A ) 1.29 ± 0 . 2 4 0.85 ± 0 . 1 8 0.83 ± 0 . 1 6 * S 2 ( n A ) 1.04 ± 0 . 1 3 0.62 ±0 .10* 0.51 ± 0 . 1 1 * S2/S1 0.80 ± 0 . 1 0 0.73 ± 0.07 0.61 ± 0;09 S12/S10 1.21 ± 0 . 1 3 1.15 ± 0 . 0 9 1.22 ± 0 . 0 8 Plateau (Si 5-20) (nA) 0.62 ± 0 . 1 0 0.38 ± 0 . 1 1 * 0.22 ± 0 . 1 2 * Plateau/Si 0.48 ± 0.09 0.44 ± 0.06 0.26 ± 0.05** Apparent Q N (nA) 5.04 ± 0.35 3.04 ± 0.25* 2.5 ±0 .31** Values are mean ± S E M between neurons; n = 5, * (relative to control) ** (relative to control and C T Z + K Y N ) P < 0.05, A N O V A . Data are from 10 H z trains. Chapter 3. Results 1 Ran - 147-Table 3.2.1 I B : T T X effects on derived parameters of STD during co-applied C T Z and K Y N Control C T Z + K Y N T T X + C T Z + K Y N (50 p M ) (50 p M ) (32 nM) (50 p M ) (50 p M Parameter cov(Si,S 2 ) -0.026 ± 0 . 0 1 0 -0.013 ± 0 . 0 0 7 -0.018 ± 0 . 0 1 3 Q ' ( S 0 ( p A ) 37.0 ± 6 . 1 24.7 ± 5 . 1 * 25.6 ± 4.3* Q ' (S 2 ) (pA) 32.6 ± 5 . 0 25.6 ± 3 . 7 19.1 ± 4 . 3 * Q'(Si5-S 2 0) 27.1 ± 3 . 7 24.8 ± 5 . 1 5.1 ±2.9** (PA) O Y S i s - S ^ 0.73 ± 0 . 1 9 1.03 ± 0 . 1 1 * 0.20 ±0.13** Q'(S.) Var/mean 29.4 ± 3 . 3 26.1 ± 2 . 8 7.2 ± 2.5 (Sl5-S 2n) (PA) ** Evoked minEPSC 32.0 ± 5 . 3 27.6 ± 4 . 2 9.4 ± 3 . 7 size (pA) Pre-stimulation 11.1 ± 3 . 4 10.9 ± 2 . 4 10.4 ± 3 . 1 minEPSC size (PA) mi 34 ± 1 1 33 ± 14 30 ± 9 ml5.20 23 ± 9 15 ± 8 50 ± 1 1 * * wi 2 /wio 1.23 ± 0 . 1 2 1.27 ± 0 . 1 4 1.43 ± 0 . 1 8 Values are mean ± S E M between neurons; data from same neurons as table 3.2.11 A ; n = 5, * (relative to control) ** (relative to control and C T Z + K Y N ) P < 0.05, A N O V A test. Chapter 3. Results I Ran - 148 -ratio, and evoked minEPSC size decreased as without C T Z and K Y N . Also , T T X co-applied with C T Z and K Y N did not affect the amplitude of pre-stimulation minEPSCs. Thus, receptor desensitization or saturation did not mediate T T X effects on quantal size. 3.3.7. Effects of glucose deprivation on STD Because of the extensive literature relating barbiturate actions to effects on metabolism (see Introduction), the next experiments explored how S T D might be altered during conditions o f limited energy supply, for which brief glucose deprivation was used. Overall, graded reductions in glucose concentrations caused a gradual nullification o f S T D . Stepping the glucose concentration to values lower than 25 m M caused a decrease in the amplitude of EPSCs both at the beginning of trains and subsequently (Figure 3.32; Table 3.2.12A). This was accompanied by an increase of the Plateau/Si ratio and a reduction in the apparent Q N . The post-gap jump was also reduced with decreasing glucose concentrations (Figure 3.32; Table 3.2.12A). Hence, low glucose produced an apparent reduction in fractional release. The gradual nullification of STD due to glucose deprivations was accompanied by a loss of the negative covariance terms early in the train (Figure 3.32 E). Figure 3.32E shows that for control conditions in 25 m M glucose, the covariance term, equal to - 1/N, decreased from cov(Si,S2) to cov(Si,S 5 ); The data suggest that either between-train nonstationarity becomes relatively high, and/or fractional release becomes so small that covariances become nearly undetectable. In principle, the variance is related to p by Chapter 3. Results I Ran -149 -Var(S) = <S>-Q' -(1 - p), which is scarcely affected by p when p is very small, with p = w / N . The covariance, -<Si>-<S2>/N = pQ'<S2> is correspondingly small and therefore readily obscured by between-train non-stationarity, e.g., random failure of some nerve terminals to participate in the response. A s glucose concentrations were stepped down to below 10 m M , the apparent quantal sizes, the variance-mean ratios as well as the size of evoked minEPSCs decreased significantly to values in the 9-12 p A range (Figure 3.32, 3.33; Table 3.2.12B). The ratio o f quantal contents around the intra-train gap also decreased, with lowering glucose concentrations. Glucose deprivation did not alter the size of pre-stimulation minEPSCs (Figure 3.33; Table 3.2.12B). The effects on apparent quantal size and evoked m i n E P S C size, and lack of effects on pre-simulation minEPSC size, implicate a presynaptic site (or sites), sensitive to glucose deprivation, particularly during periods o f intense synaptic activity, with little or no recovery in the 20 s between train period. The effects of nominally zero glucose conditions were examined on evoked and spontaneous EPSCs . Glucose omission from the perfusion media after 1 min resulted in no change in the holding current but an irreversible loss of both spontaneous and evoked EPSCs . In 3 neurons, large, irreversible increases in holding current and input conductance that occurred at 10 min signified a loss of cell viability. Despite 15-40 min periods o f observation in these neurons, re-establishing control perfusion after the 3 min omission did not result in any recovery of the EPSCs. Chapter 3. Results I Ran -150-Control (25 mM glucose) 10 mM glucose Wash (25 mM glucose) mmrmmr~ m 5 mM glucose iWuwrftttfrtr 2.5 mM glucose 1s o Contra) • 10 mM glucose * 5 mM glucose •. 2.5 mM glucose £25 mM giucosa) 1 nA B | •§ 1.0 Q. £ O 0.5 <J 0 04-trtrttrrtl TrritTTtt D 5 10 15 stimulus # "E 0.02 TO cr 0.01 0 00 S 0.03H T-»^> rT:Tirrrn mW-6 10 15 stimulus* 40 I 3 0 ' 3 20-to ffwH 0.03' K0 . 0.02 ^ CO 0:01 5 10 15 stimulus # i= 2 Figure 3.32. Effects of glucose deprivation on STD. A ) Current traces of corticothalamic EPSCs during S T D evoked during brief glucose deprivations. Top: Lowering glucose concentration from 25 to 10 m M (middle trace) decreased the amplitude of the entire E P S C train without changing the rundown. Middle: Lowering glucose concentration from 25 to 5 m M further decreased the amplitude of the entire E P S C train and also increased the rundown. Bottom: Reducing glucose from 25 to 2.5 m M abolished S T D (middle trace). B) Average EPSCs amplitudes (n = 6). C) Apparent quantal size estimates during glucose deprivation. D) Quantal content estimates from the neurons in B and C. E) Covariance estimates obtained by pairing the 1 s t E P S C with the 2 n d - 5 t h E P S C at different glucose concentrations. The linear regression line is proportional to the rate of vesicular replenishment. Chapter 3. Results I Ran Table 3.2.12 A : Summary of parameters of S T D at different glucose concentrations [glucose] (mM) 25 10 5 2.5 Parameter S i ( n A ) 1.20± 0.16 0.93 ± 0 . 1 3 0.63 ± 0 . 1 9 * 0.20 ± 0.08* S 2 ( n A ) 0.73 ± 0.07 0.42 ± 0.08 0.26 ± 0.06* 0.16 ± 0 . 0 8 * S2/S1 0.61 ± 0 . 1 1 0.45 ± 0.08 0.42 ± 0.06 0.79 ± 0.09 S12/S10 1.27 ± 0 . 0 8 1.29 ± 0 . 1 3 0.95 ± 0.09 1.04 ± 0 . 0 7 * Plateau (Si 5-20) (nA) 0.54 ± 0 . 1 6 0.40 ± 0 . 1 9 0.30 ± 0 . 1 1 0.21 ± 0 . 1 3 * Plateau/Si 0.45 ± 0 . 1 0 0.43 ± 0.09 0.47 ± 0.09 1.05 ±0.13** Apparent Q N 4.31 ± 0 . 3 3 2.69 ± 0 . 3 1 * 1.76 ± 0 . 2 9 * 1.09 ± 0 . 4 1 * (nA) Values are mean ± S E M between neurons; n = 5, * (relative to glucose 25 m M ) ** (relative to all other groups) P < 0.05, A N O V A test. Data are from 10 H z trains. Chapter 3. Results I Ran -152 -Glucose 25 mM Glucose 10 mM Glucose 5 mM Glucose 2.5 mM Before trains _0 . l I OJ 71 minis In 30 s . a * 10-9 i 4j) 0 10 20 30 40 Size (pA) B After trains i Ds 0 10 ' 20 30 40 Size(pA) B - A u — ' 10,914.0 8 6 \ 4 V 2 V nL... i 10 20 30 40 Size (pA) | 0.0 48 minis in 6s f Cl'1 10 20 30 40 Sizo (pA) mis'In 30 s • 10.9 .t 4.0 0 10 20 30 Stea (pA) 10 20 30 40 . , . , . . . . T . . ^ 4 l - . . L U - . J . , _ b 0'~{10J • 20: • 130 40 Size (pA) $  3 31.2 ±3.4 OiZTloj 20 30 40 Sue (pA)' 20.1 ±4.2 I 20 30 40 'Size (pA) •10 20 30 40 Sizs(pA) Figure 3.33. Evoked and spontaneous miniature EPSCs during glucose deprivation. A ) Amplitude histograms of spontaneous minEPSCs counted 5 s before the stimulus train (6 repeats, 30 s total) during glucose deprivation produced by stepping glucose concentration from 25 to 10,5, and to 2.5 m M . B) Amplitude histograms of evoked minEPSCs counted 1 s after the stimulus train (6 repeats, 6 s total) during the same conditions as in A . Total minEPSC count ('minis') is indicated above histograms. Evoked minEPSC sizes were obtained after subtraction of spontaneous from evoked m i n E P S C histogram (B - A ) . Values next to black arrows are mean size ± S E M . Data are from 1 neuron. Chapter 3. Results I Ran - 153 -Table 3.2.12B: Derived parameters of STD at different glucose concentrations [glucose] (mM) 25 10 5 2.5 Parameter cov(Si,S 2 ) -0.022 ± 0 . 0 1 3 - 0.009 ± 0.006 0.002 ± 0.003 4x l0" 5 (nA 2 ) Q ' (Si ) (pA) 35.9 ± 8 . 1 30.1 ± 9 . 1 18.2 ± 1 0 * 10.7 ± 4 . 8 * Q ' (S 2 ) (pA) 31.6 ± 7 . 9 29 ± 5 . 1 16.3 ± 3 . 8 9.2 ± 6 . 2 * Q'(Si5-S 2 0) 29.2 ± 3 . 8 25.6 ± 4 . 3 12.9 ± 2 . 1 * 8.9 ± 3 . 3 * (PA) OYSJJ-STJV) 0.81 ± 0 . 1 2 0.85 ± 0 . 1 5 0.70 ± 0 . 1 1 0.83 ± 0 . 1 4 Q'(Si) Var/Mean 28.3 ± 3 . 2 26.7 ± 2 . 5 14.6 ± 4 . 1 * 11.4 ± 2 . 1 * (Si5-S2o) (PA) ** Evoked minEPSC 30.5 ± 3 . 3 28.1 ± 3 . 4 18.6 ± 2 . 1 10.2 ± 2 . 8 size (pA) Pre-stimulation 11.8 ± 4 . 1 11.4 ± 2 . 8 10.8 ± 3 . 6 10.2 ± 2 . 9 minEPSC size (pA) mx 33 ± 4 26 ± 3 34 ± 7 19 ± 8 W15-20 19 ± 2 14 ± 3 23 ± 8 23 ± 7 mxilmxo 1.23 ± 0 . 0 7 1.26 ± 0 . 0 8 1.00 ± 0 . 0 5 * 0.98 ± 0.07* Values are mean ± S E M between neurons; data from same neurons as table 3.2.12A; n = 5, * (relative to glucose 25 and 10 m M ) ** (relative to all other groups) P < 0.05, A N O V A test. Chapter 3. Results 1 Ran -154 -3.8. Discussion The present studies have demonstrated that pentobarbital produced changes in quantal parameters during short-term depression of corticothalamic EPSCs . In control conditions, the quantal aspects of synaptic transmission was found to behave similarly to that at neuromuscular and calyx of Held synapses (Scheuss and Neher, 2001; Scheuss et al., 2002). In particular, there were significant negative covariances between responses to the first and second stimuli, as predicted by the binomial/depletion model. In addition, S T D , at different stimulation frequencies, was modulated in much the same way as at the neuromuscular junction (Elmqvist and Quastel, 1965a). The plateau phase of E P S C s subsequent to their rundown can be explained by refill eventually matching release characteristics, rather than a different population of release sites becoming involved. This proposal applies to the stimulation frequencies studied here but may not apply for higher frequencies where refill rates remain unaltered (Wesseling and Lo , 2002). This agreement with the binomial/depletion model provided a validation for using variance/mean ratios, corrected using covariances, to obtain estimates o f apparent quantal size Q' (theoretically Q ( l + C V Q 2 ) ) , at each stimulus. The results then showed that a substantial part of S T D arises from decline in Q', as at calyx of Held (Scheuss et al., 2002), another glutamatergic ( A M P A ) synapse, and unlike the neuromuscular junction (Elmqvist and Quastel, 1965a). The validity of the statistical estimates of Q' was also established by their agreement (late in trains) with the size of minEPSCs evoked by the stimulation (cf. Otsu et al., 2004). Chapter 3. Results I Ran -155 -When combined with a pharmacological approach, the statistical method provided estimates for changes in release parameters during corticothalamic S T D and their modulation by pentobarbital and other drugs or changes in the bathing medium. A combination o f postsynaptic receptor desensitization and saturation contributed to S T D by reducing the apparent quantal size. Presynaptically, depletion of readily-releasable quanta mediated the S T D that remained in the absence of receptor desensitization and saturation. These findings validate, for the first time, a depletion-based model at a relatively small central synapse and suggest a link between pentobarbital depressant effects and the metabolic requirement during intense processing tasks in the thalamus. 3.3.8.1. Behaviour of EPSCs in trains Repetitive stimulation produced a frequency-dependent rundown of E P S C s in thalamocortical neurons, illustrating a short-term form of depression, similar to other observations at corticothalamic synapses (L i et. al. 2003; Reichova and Sherman 2004). The detection of negative covariances between successive pairs o f EPSCs at the onset o f repetitive stimulation validated the application of the depletion model, similar to neuromuscular and calyx of Held synapses (cf. Elmqvist and Quastel, 1965a; Scheuss and Neher, 2001), to obtain estimates of apparent quantal size at each stage in the train. The negative covariances were necessary for corrections of the variance to mean ratios from the initial responses in a train of EPSCs , which provided an estimate of the apparent quantal size. This method also corrected for deviations in the variance to mean ratio attributable to failure of axonal conduction and/or action potential invasion into nerve terminals that might occur during prolonged train stimulation. Chapter 3. Results I Ran - 156 -However, the corrected variance-mean model estimated the apparent quantal size and not the "true" quantal size. In other words, this approximation, which is based on assumption o f binomial characteristics, relates to the true quantal size by additional contributions o f within- and between-site variabilities ((1 + C V B ) and (1 + C V w ), respectively; reviewed by Auger and Marty 2000). In other neurons, such variabilities have been estimated by recording at single synapses (Forti et al. 1997) or by comparing distributions o f minEPSCs originating at distant terminal branches (Bekkers and Clements 1999). If incorporated into the corrected variance-mean ratio, a future assessment o f within- and between site variability at corticothalamic synapses would approximate the true quantal size alterations during STD. Corticothalamic S T D occurred at a frequency range that allows thalamic relay neurons to filter high frequency cortical inputs (cf. Reinker et. al., 2004). A t other C N S synapses that receive high frequency inputs, STD occurs preferentially at frequencies > 100 H z , whereas, low frequency inputs induce primarily facilitation (von Gersdorff and Borst 2002). Thalamic relay neurons receive a variety of cortical inputs in the 2.5-20 H z range that drive or modulate their firing and firing patterns (Reichova and Sherman 2004). Corticothalamic S T D may function as a low-pass filter (Fortune and Rose 2000) and contribute to thalamic processing of somatosensory inputs particularly during conditions o f intense stimulation and various waking and sleep states. Chapter 3. Results I Ran - 157 -3.3.8.2. Effects of[Ca2+]e on STD Corticothalamic STD was influenced by but did not critically depend on extracellular [Ca 2 + ] . In high [Ca 2 +] e-containing media, S T D behaved as expected, producing a faster rundown of EPSCs and higher plateau (cf. Elmqvist and Quastel, 1965a; Scheuss et al., 2002). The heightened plateau of EPSCs was attributable to an increased quantal content throughout the E P S C train, indicating a presynaptic origin with high [ C a 2 + ] e increasing P 0 's . The greater E P S C size subsequent to an intra-train gap also increased in agreement with raised P 0 . These observations demonstrate a modulatory presynaptic effect of raised 2+ [Ca ] e on STD, closely resembling that seen at neuromuscular junction (Elmqvist and Quastel, 1965a). Surprisingly, reducing [Ca ] e did not abolish STD, disproving the assumption of a strict C a dependence. This persistent S T D was not characterized by changes in quantal content but rather was attributable to a decrease in quantal size. This apparent dependence of quantal sizes on [ C a 2 + ] e might implicate the existence of a heterogeneous population of release sites with varying transmitter contents. The concomitant persistence of S T D and reductions in quantal size suggest that reducing [ C a 2 + ] e caused a shift between populations of release sites. Alternatively, the smaller quanta could result from a partial release of transmitter content due to a reduced spread of the local C a 2 + signals. Such effects could provide means to regulate corticothalamic plasticity. Chapter 3. Results I Ran - 158 -3.3.8.3. Receptor desensitization and saturation during STD Receptor desensitization and saturation contributed to but did not exclusively mediate corticothalamic STD, consistent with observations in calyx of Held neurons (cf. Scheuss et al. 2002). Pharmacological blockade of desensitization and saturation slightly reduced but did not abolish the rundown of corticothalamic EPSCs during train stimulation. The effects demonstrated that STD is attributable to depletion of readily-releasable quanta, which is l ikely a major component. However, these results imply a paradox. The binomial/depletion model (and negative covariances) indicate that the signals early in the train after the first response are from sites where there was no previous release of a quantum of transmitter, especially at frequencies > 10Hz, where refill is relatively small. But those responses show lowered quantal size (Table 3.2.IB; Figure 3.20). A possible explanation is that desensitization and receptor saturation occur because of overflow of neurotransmitter from sites where quanta were released to neighbouring sites where quanta are subsequently released (Rossi and Hamann, 1998; Isaacson, 1999; DiGregorio et al., 2002; Telgkamp et al., 2004). This is in keeping with the observed morphology of corticothalamic synapses in the cat (Jones and Powell, 1969). 3.3.8.4. Pentobarbital effects on STD Pentobarbital enhanced S T D by a mechanism that involved a use-dependent decrease in quantal size, that was not sensitive to blockade of A M P A receptor desensitization and saturation. The reduction in the apparent quantal size, and evoked minEPSCs occurred Chapter 3. Results I Ran - 159 -without effects on ongoing minEPSC size observed before stimulation. There are several possibilities which can explain this result. First, the reduction in quantal size is presynaptic, there being a lowered amount of transmitter (glutamate) in each quantum. Second, it could be that with pentobarbital some sites no longer become stimulated presynaptically (e.g. because of action potential failure) and those that continue to be stimulated are associated with smaller quanta, because of filtering at distally located dendrites. Thirdly, it is possible that with stimulation there develops a preferential release of pre-existing quanta that have less than normal amounts of neurotransmitter. Lastly, pentobarbital might enhance or produce a kind of desensitization not seen normally and not blocked by a combination of cyclothiazide and kynurenate. Other studies have demonstrated a preferred action of pentobarbital to promote the desensitization o f the GluR2 subtype of A M P A receptors (Taverna et al. 1994). However, the actions of pentobarbital at this receptor subtype expressed in thalamic neurons (Spreafico et al. 1994), are sensitive to blockade by cyclothiazide (Jackson et al. 2003). 3.3.8.5. Effects of extracellular alterations on STD High [ K + ] e increased whereas low [ K + ] e reduced the number of released quanta without producing significant changes in quantal size, which indicated a predominantly presynaptic effect. Unexpectedly, raising [ K + ] e to 10 m M , while having no effect on the first E P S C s in trains, greatly increased the second and subsequent EPSCs (Figure 3.28; Table 3.2.8A, B) . In other words, facilitation from pulse 1 to pulse 2 was greatly increased. There was also a loss of the negative covariance between Si and S2, suggesting Chapter 3. Results I Ran - 160 -a large increase in the rate of replenishment (refill, a) of the readily releasable store (Kuromi and Kidokoro, 2004). Increasing [ K + ] e may have promoted the back propagation of an action potential which is normally limited by activation of Ca 2 +-gated K + channels subsequent to a local rise in presynaptic [Ca 2 + ] . These results imply that local increases in [ K + ] e subsequent to activity at nerve terminals could modify transmitter release, transforming a depressed into a facilitated synaptic response (cf. Poolos et al., 1987; Nishimura et al., 1993; Kamiya and Zucker, 1994). In 10 m M [ K + ] e , pentobarbital at 200 p M completely blocked the pulse-pulse facilitation produced by raised [ K + ] e . This result suggests that whatever the mechanism by which [ K + ] e has this effect (note it is opposite to presynaptic depolarization - see Hubbard et al., 1967) it should be sensitive to lower concentrations of pentobarbital which reduced fractional release without affecting quantal size (cf. Table 3.2.4B). A possible explanation for this is that pentobarbital activated a Ca 2 +-gated K + channel (Sailer et al., 2004) which blocked the propagation of an antidromic action potential in raised [ K + ] e . This possibility could be the focus of further studies. In low [ K + ] e , although STD was apparently unaffected, pentobarbital increased the amplitude of the E P S C after the intra-train gap, indicating an increase in P 0 , or (1 - a), which possibly reflected a selection of high probability release sites. In summary, altering [ K + ] e unmasked actions of pentobarbital on parameters of quantal release as well as apparent quantal size. Chapter 3. Results I Ran -161 -3.3.8.6. Effects of tetrodotoxin Because pentobarbital might at least partially act by modifying the presynaptic action potential (Blaustein, 1968), it was worthwhile to study the effect(s) of tetrodotoxin ( T T X ) , an agent considered to act only by blockade of voltage-gated N a + channels. A t a concentration (32 nM) half that producing block of EPSCs , T T X equally reduced the apparent quantal size and evoked minEPSC size, without affecting pre-stimulation minEPSC size (Table 3.2.10B; Figure 3.30). T T X effects presumably resulted from a reduced N a + entry into the nerve terminal rather than a decrease in depolarization per action potential, which would reduce effective m's, numbers of quanta released per stimulus. 3.3.8.7 Effects of lowered glucose In low ambient glucose (5 or 2.5 compared to 25 m M ; Table 3.2.12A, B) , there were two major effects, lowered quantal size and lowered rundown of quantal numbers (m's) in trains (abolition of S T D in 2.5 m M ) . Reductions in the jump after the omitted stimulus also were consistent with reduced P 0 , compared to controls. This reduction in P G could be related to the observation that interference with A T P hydrolysis impairs the transition o f packets from the resting/non-releasable to the release-competent pool (Heidelberger et al. 2002). However, this would amount to a reduction in a (refill o f releasable quanta) rather than reduction in P 0 . It is therefore likely that energy supply may be critical for both release, per se, and for maintenance of the readily releasable pool o f packets during periods of intense stimulation. Chapter 3. Results I Ran - 162 -A s with T T X , quantal sizes did not recover much between trains. Since a specific kind o f desensitization produced by low glucose is hardly likely, there are only two remaining possibilities for the lowered Q' (and evoked minEPSC size), namely, reduced amount o f transmitter per quantum and/or selection of synapses with small quanta (pre- or postsynaptically). O f these possibilities, the first, lowered neurotransmitter per quantum (or packet) seems by far the more likely, since the uptake of glutamate into synaptic packets critically depends on glucose (Ikemoto et. al., 2002). The present data (Table 3.2.12, A , B) suggest that glucose deprivation impairs the maintenance of transmitter content in the transmitter packets, which constitute the pool or store o f quanta from which outputs are evoked. General Discussion I Ran - 163 -4. General Discussion This thesis provides for the first time, evidence for dual pre- and postsynaptic actions o f the barbiturate, pentobarbital. The most notable actions were to promote oscillogenesis in a thalamic network and to cause short-term plastic modifications of cortical inputs to thalamic neurons. The details of these actions have received discussion at the end of each Results section. Therefore, I w i l l succinctly summarize the most pertinent findings and then, discuss their relevance to a context of anesthetic mechanisms. Summary of results Pentobarbital oscillogenesis. A subanesthetic concentration of pentobarbital induced thalamic oscillations in in vitro preparations. Sustained oscillations at 1-15 H z required electrical stimulation of the internal capsule, but not elevated temperature or low extracellular [ M g 2 + ] . Receptors for glutamate and glycine mediated oscillations in ventrobasal nuclei, disconnected from nRT. Receptors for glutamate and G A B A mediated oscillations in nRT, disconnected from ventrobasal nuclei. Pentobarbital oscillogenesis occurred in isolated networks of the ventrobasal and reticular nuclei, mediated by glutamate receptors with frequency modulation by G A B A A - , G A B A B - , and glycine-receptors. Spermine modulation. Extracellular spermine acted on the polyamine site of N M D A receptors, to increase membrane rectification on depolarization, to reduce firing threshold, and to slow the decay of corticothalamic EPSPs. The heightened excitability of thalamocortical neurons increased tonic firing evoked by depolarizing current pulses General Discussion I Ran -164 -and EPSP bursts of action potentials. Spermine increased the rates of rise and amplitudes of low threshold Ca spikes by an unknown mechanism, not mediated by N M D A receptors. B y increasing the efficacy of corticothalamic excitation, spermine actions take on importance in the transformation of somatosensory signals to tonic and burst discharge patterns during the juvenile stage of rat brain development. Pentobarbital-spermine interactions. Pentobarbital at an anesthetic concentration reversed spermine actions that prolonged corticothalamic EPSPs in thalamocortical neurons. This effect involved postsynaptic interactions at the polyamine site on N M D A receptors. Pentobarbital shortened the EPSP duration and reversed the prolongation caused by spermine. The opposing effects of pentobarbital and spermine on corticothalamic transmission provide a model for anesthetic modulation o f glutamate receptors in thalamic hyperexcitability. Pentobarbital effects on short-term depression (STD). Pentobarbital enhanced S T D of corticothalamic EPSCs by decreasing quantal size in a use-dependent manner. These actions were presumably presynaptic because they were insensitive to pharmacological blockade o f desensitization and saturation of A M P A receptors. Pentobarbital affected the statistically estimated quantal size (apparent quantal size) and the amplitude of evoked minEPSCs. Prior to electrical stimulation, pentobarbital did not affect the amplitude of ongoing minEPSCs, which reaffirmed a lack of postsynaptic action on spontaneous minEPSCs. The effects of pentobarbital to promote STD, may have resulted from General Discussion I Ran - 165 -preferred release of quanta with a small transmitter content, activation of a K + conductance-mediated shunt of the presynaptic action potential, or impairment of evoked release due to blockade of Ca channels (see below). Alterations in extracellular [K*] on short-term depression. High [ K + ] e increased, whereas low [ K + ] e reduced the number of released quanta. Significant changes in quantal size did not accompany these effects, indicating that pentobarbital acted predominantly at presynaptic sites. Pentobarbital completely blocked the high [K +]-induced facilitation early in a stimulus train. Low [ K + ] e by itself did not affect STD. In combination with low [ K + ] e , pentobarbital increased the amplitude of the E P S C after the intra-train gap, implicating an increased fractional release due to preferred release from high probability sites. Effects of tetrodotoxin (TTX) on short-term depression. Application o f T T X at a concentration that was half that producing blockade of evoked action potentials, reduced the apparent quantal size and evoked minEPSC size to the same extent. T T X did not affect pre-stimulation minEPSC amplitude, suggestive of a minimal postsynaptic action. The effects of T T X presumably resulted from a reduced N a + entry into the nerve terminal and not from a decreased depolarization by each action potential. This interpretation was in agreement with a lack of effect on numbers of quanta released per stimulus (effective m's). General Discussion I Ran - 166 -Glucose deprivation and short-term depression. Glucose deprivation had two major effects (Table 3.2.12A, B) - a reduction in quantal size and lesser rundown of quantal contents (m's). A ten-fold reduction in the extracellular glucose concentration abolished S T D and the post-gap jump in E P S C amplitude. These combined effects were consistent with reduced output probability of release, P 0 . The combined effects coupled with a reduced quantal size imply that glucose deprivation impaired the maintenance of transmitter content in the packets. It is plausible that the induced energy shortage modifies the pool o f quanta from which outputs are normally evoked because glucose deprivation did not affect spontaneous minEPSCs. Initiation of pentobarbital oscillations Induction o f oscillations by pentobarbital seems paradoxical for a drug usually viewed as a C N S depressant. Pentobarbital hyperpolarizes thalamocortical neurons, acting by multiple mechanisms over a subanesthetic concentration range (Wan and Pui l , 2002; Wan et al., 2003). Hyperpolarizing activation and deinactivation of intrinsic currents would promote oscillogenesis in a thalamic slice network (cf. McCormick and Pape, 1990), modulated by receptor-gated currents (cf. Steriade et al., 1997). In the present study, G A B A A , G A B A B , and glycine receptors modulated, but none was essential for pentobarbital-induced oscillations (cf. Table 4.1). Glycine receptors l ikely were critical for oscillations in a dorsal thalamic network, divested o f G A B A e r g i c inhibition. Glycine is not a recognized neurotransmitter in the thalamus, where G A B A General Discussion 1 Ran - 167 -Table 4.1: Receptor involvement in pentobarbital-induced oscillations Recording Site Activation of Receptors for Glutamate G A B A Glycine Ventrobasal nuclei Essential Modulate Modulate frequency frequency Ventrobasal nuclei, without nRT N o Essential apparent Essential role nRT Essential N o apparent role nRT, without ventrobasal nuclei Essential No apparent role - not tested General Discussion I Ran - 168 -mediates transmission by reticulothalamic and some interneuronal pathways (Steriade et al., 1997). The thalamus exhibits strychnine-binding sites, m R N A for glycine receptor subunits and expresses a subunit protein (Zarbin et al., 1981; Malosio et al., 1991; Naas et al., 1991). Conceivably, activation of glycine receptors may mediate synaptic inhibition in ventrobasal nuclei of rat (cf. Tebecis, 1974; Ghavanini et al., 2005). A subanesthetic concentration of pentobarbital induced oscillations in a corticothalamic network. These oscillations required glutamatergic excitation, but not G A B A e r g i c inhibition from nRT. Glycine receptors were essential for the oscillations in ventrobasal nuclei, isolated from nRT. G A B A receptors were essential for the oscillations in nRT, isolated from ventrobasal nuclei. Thus, pentobarbital can induce oscillogenesis in either ventrobasal nuclei or nRT, independent of their reciprocal synaptic connectivity. The pentobarbital-induced oscillations in slices may have some relevance to the genesis of spindling in the E E G during barbiturate administration (see Introduction). The stimulation may have mimicked the slow rhythm that neocortex imposes through glutamatergic pathways (Steriade et al., 1991, 1997; von Krosigk et al., 1993). The insensitivity to temperature was possibly due to compromised network behavior in the slice (cf. Andersen and Andersen, 1968). The sustained thalamocortical oscillations observed in the slice network may reflect pentobarbital actions that produce sedative-hypnotic effects in vivo (Nelson et al., 2002). Substances that mimic this oscillogenesis may have potential as sedative-hypnotic drugs. The stationary oscillations may represent a model o f sedation-hypnosis, amenable to pharmacological analysis. General Discussion I Ran - 169 -Pentobarbital l ikely acted on M g independent sites. In either normal or low external [ M g ], an anesthetic concentration (200 u M ; cf. Sato et al., 1995) first increased and then abolished oscillations. Jacobsen et al., (2001) observed that at lower concentrations (100 \\M) pentobarbital eliminated ventrobasal oscillations in low [ M g 2 + ] . Disappearance of oscillations may have resulted from a G A B A A receptor shunt of glutamate excitation (Wan et al., 2003 and Sawada and Yamamoto, 1985). The new and previous results are consistent with the marked C N S depression that occurs during deep anesthesia in vivo (Satoetal. , 1995). Spermine modulation of thalamic excitability The endogenous polyamine spermine applied in the 1-200 pJVl range enhanced the excitability of thalamocortical neurons in specific ways that were consistent with a neuromodulator role at P12-P15 stage of development. Spermine actions on N M D A receptors produced a heightened state of excitability which were viewed as prolonged EPSPs, and increased bursting and tonic firing of action potentials. To a large extent, these effects resulted from increased membrane rectification on depolarization and a reduced threshold for action potential genesis. Spermine also modulated the burst firing mode by increasing the rate of rise and amplitude of low threshold C a 2 + spikes (LTSs). This unusual effect did not involve interaction with glutamate receptors. The modulation of corticothalamic excitation and LTSs of M G B neurons may be critical in the transformation of auditory signals in gerbil thalamus at the P14 stage of development. General Discussion I Ran -170 -Spermine is widely distributed in rodent and human brain (Harman and Shaw, 1981; Morrison et al., 1995). A membrane transporter appears to maintain low extracellular concentrations of <1 f i M (Dot et al., 2000). These concentrations may increase on N M D A stimulation to > 50 f i M in striatal neurons of adult rat brain (Fage et al., 1992). The effects of spermine on N M D A receptors and low threshold C a 2 + spikes in juvenile M G B neurons (ED 5o= ~100 | i M ) are consistent with neuromodulatory actions at high micromolar concentrations (Williams, 1997). Given the role of N M D A receptors during development, such modulation by spermine is likely important for learning processes (Chidaetal . , 1992). The present results are relevant to the normal function of the central auditory system. The N M D A receptor-mediated effects of spermine would enhance the ability of thalamic neurons to detect simultaneous inputs, as in coincidence detection. For example, an overexpression of the spermine-sensitive N R 2 B subunit (Williams et al., 1994) prolongs EPSPs and shortens the time window between two coincident signals in hippocampal neurons (Tang et al., 1999). In thalamic neurons, the generation of synchronous activity may involve coincidence detection (Roy and Alloway, 2001) as well as amplitude selectivity in the M G B neurons (Kuwabara and Suga, 1993). The effects of spermine on the LTSs of M G B neurons may have relevance for conscious or sleep states and disorders of consciousness. The L T S is essential in the generation of General Discussion IRan -171-bursting and oscillatory activity in the auditory nuclei (Hu, 1995; Tennigkeit et al., 1996). B y increasing the rate of rise and amplitude of the L T S and slowing its decay, spermine modulation may increase an M G B neuron's responsiveness of neurons at hyperpolarized potentials (Hu et al., 1994) to inputs during these states (He and Hu, 2002). Modulation by spermine may have importance for bursting behavior during sleep states whereas excessive modulation may occur in absence epilepsy as in audiogenic seizures (Porta et al., 1981), sensitive to blockade by polyamine antagonists (Kotlinska and Liljequist, 1996). Pentobarbital reversal of spermine modulation Pentobarbital, at an anesthetic concentration, modulated corticothalamic transmission by postsynaptic actions at the polyamine site on N M D A receptors. Pentobarbital effects included shortening of the duration of EPSPs and reversal o f their prolongation by extracellular spermine. Consistent with pentobarbital shortening of burst durations and mean open times of NMDA-mediated single channel currents (Charlesworth et al., 1995), these actions on NMDA-mediated transmission may contribute to its depressant effects on oscillogenesis. Like other polyamines, spermine increases the potency of barbiturates to induce general anesthesia (Daniell, 1992). The basis for this enhancement is presently unclear because of the observed opposing actions of pentobarbital and spermine on the prolongation o f N M D A mediated EPSPs. Indeed, pentobarbital depression of transmission prevented the General Discussion I Ran - 172 -prolongation of N M D A responses caused by spermine. The effects and interactions of pentobarbital might have relevance for various forms of epilepsy. For example, N M D A receptors are activated in thalamocortical neurons during the development o f spike-and-wave discharges in a strain of genetic absence epilepsy rats (Koerner et al., 1996). Similarly, an injection of N M D A into subthalamic nuclei of rats generates audiogenic seizure behaviour (Faingold el al. 1989) and increases neuronal firing in M G B neurons (N'Gouemo and Faingold, 1997). Hence, pentobarbital actions that shorten corticothalamic EPSPs might alleviate N M D A mediated epileptic seizures. Presumed presynaptic actions ofpentobarbital A n anesthetic concentration of pentobarbital enhanced STD at corticothalamic synapses. The increased S T D can be attributed to a presynaptic use-dependent action that produced small size quanta. The effects persisted during blockade o f A M P A receptor desensitization and saturation and were not accompanied by alterations in spontaneous minEPSC size. The rapid fall in quantal size implies a rather small population o f preformed quanta. This was evident in the fast reduction in quantal size after the 3 r d stimulus in the train. The reduction can also be explained by use-dependent inactivation of C a 2 + channels which reduces C a 2 + entry into terminals early in the train and impairs the formation of releasable quanta. Pentobarbital, at a similar concentration, reduces the amplitude of miniature EPSPs by impairing C a 2 + entry into nerve terminals (Baudoux et. al., 2003). B y reducing C a 2 + entry into corticothalamic terminals, pentobarbital could have interfered with glutamate release by selectively affecting a population of small size and/or partially refilled quanta. General Discussion I Ran - 173 -Similar effects of tetrodotoxin and pentobarbital Tetrodotoxin mimicked the pentobarbital enhancement of short-term depression. The effects of T T X to reduce apparent quantal size (Q') in a use-contingent manner differed from that of pentobarbital in that recovery from T T X between iterated trains was incomplete (Table 3.2.10B). T T X had a progressive effect between trains, reducing the number of participating release sites. The same interpretations of a lowered Q' (see Section 3.3.8.4., page 159), apply for T T X as for pentobarbital, except that a specific kind of TTX-induced desensitization seems implausible. The effects of T T X on Q ' occurred at half the concentration which blocked action potentials, and likely resulted from reduced N a + entry into the nerve terminal, rather than from blockade of depolarization. The similar amplitude of pre-stimulation and evoked minEPSCs observed here during T T X application suggests a close resemblance to minEPSCs, or true minEPSCs, observed in other studies during complete blockade of action potentials at T T X concentrations > 1 p M (cf. Edwards et al., 1990). Given the high selectivity of T T X , these results may imply that these drugs modulate Na+-dependent processes that pump glutamate into releasable packets at corticothalamic terminals. There was a large difference in the concentrations of T T X and pentobarbital that affected transmitter release. Unlike pentobarbital, T T X would likely block axonal invasion of the presynaptic action potential. In axons, such blockade by pentobarbital is evident only at millimolar concentrations (Blaustein, 1968). Pentobarbital depression of thalamic firing occurs at much lower, subanesthetic doses and is attributable to a K + conductance shunt (Wan et al., 2003). Hence, pentobarbital in the present experiments General Discussion I Ran -174 -could have shunted the presynaptic action potential by activating K + conductances in the nerve terminal. Effects of glucose deprivation on short-term depression The similar effects of glucose deprivation imply that enhancement of short-term depression by pentobarbital was, in part, attributable to its depressant actions on metabolism (see Introduction). The reduction in fractional release during glucose deprivation could be explained by a drop in the A T P content, which impairs the maintenance of a release-competent pool of transmitter vesicles (Heidelberger et al. 2002). B y reducing the energy supply, pentobarbital may impair both release, per se, and the maintenance of the readily releasable pool of transmitter packets, particularly during periods of increased activity-dependent release of transmitter. Apparent redundancy in pentobarbital effects on transmitter release The mimicry o f action potential blockade and energy deprivation as well as multiple actions on quantal release (cf. Wan et al., 2003), suggests that pentobarbital may have reduced the size of the store and transmitter content of quantal packets. Pentobarbital inhibits glycolysis which provides most of the A T P at synaptic terminals (Crane et al., 1978). A decrease in the A T P content at nerve terminals impairs processes that maintain the content (Dcemoto et al., 2003) and availability (Heidelberger et al. 2002) o f readily-releasable quanta. It is also possible that conditions of glucose deprivation or blockade o f N a + entry by T T X affected separate processes, each impaired by high micromolar concentrations of pentobarbital. General Discussion I Ran - 175 -The effects of pentobarbital, T T X , and energy deprivation on transmitter release have an alternative explanation. These effects may involve impairment of action potentials at specific sites, sparing distal sites of transmitter release. Selective blockade would result in a preferred release of the remaining, highly releasable small quanta, and reduced quantal size as observed postsynaptically. This process is compatible with the observations that pentobarbital depletes docked packets at synaptic active zones (Jones and Devon, 1978; Hajos et al. 1978). Other studies have suggested that heterogeneity in contents of transmitter packets contributes to variation in quantal amplitude (Edwards, 1990). This proposal is tantamount to the proposition that release of partially filled quanta occurs normally. The lowered quantal size due to incomplete filling in the low glucose condition would then be seen as an exaggeration of something that also occurs normally. However, it is also possible that small quanta are preferentially released when energy supply is low and the same could be true in the presence of T T X and pentobarbital. Mechanisms ofpentobarbital enhancement of short-term depression There are at least four distinct mechanisms that could explain the ability o f pentobarbital to reduce quantal size: 1) a selective process whereby pentobarbital depleted large size quanta, sparing the smaller size quanta. For example, pentobarbital can interfere with C a 2 + entry (Charlesworth et al., 1995) into active zones critical for transmitter release at nerve terminals. The selective process is consistent with known actions o f pentobarbital that deplete a subset of docked packets in active zones (Jones and Devon, 1978; Hajos et General Discussion I Ran - 176 -al. 1978); 2) a pentobarbital induced suppression of glycolysis and associated A T P - and N a + -dependent glutamate uptake (see Introduction) which may reduce transmitter content per packet (cf. Gueldry et. al. 1987; Ikemoto et al. 2002); 3) a pentobarbital interference with C a 2 + entry (Charlesworth et al., 1995) that alters the kiss-and-run (Pocock and Richards, 1987; A n and Zenisek, 2004) or porocytotic (Kriebel et al., 2001) type openings of fusion pores. This blockade would limit the amount of released transmitter; and, 4) a pentobarbital-evoked desensitization of postsynaptic cyclothiazide-insensitive receptors. This desensitization would reduce the postsynaptically observed quantal size, in response to a given amount of transmitter (see Introduction). A t present, there is no evidence for such receptors on thalamic neurons (cf. Chen et. al. 2002). In principle, there are many known pre- and postsynaptic sites or targets for barbiturate action that could be involved in these experiments on corticothalamic S T D (Figure 4.1). For example, one might expect pentobarbital to activate a K + conductance-mediated shunt of presynaptic action potentials (Wan et al., 2003), limiting the activation o f voltage-gated C a channels and therefore release probability (P 0). Also a priori, one might expect that, by interfering with metabolism, pentobarbital might inhibit the transfer of resting packets to the releasable store (Heidelberger et al. 2002). The present results, however, contradict these expectations, particularly those from experiments with raised [ K + ] e . In raised [ K + ] e , pentobarbital completely inhibited a novel action o f [ K + ] e to increase quantal release only after the first stimulus in the train. This finding suggests that a major effect of pentobarbital at lower concentrations might be to General Discussion I Ran -177 -suppress normal modulation of transmitter release by changes in local [ K + ] e at synapses, produced by activity at nearby synapses. A n elevated [ K + ] e would arise from release sites/boutons that have not released quanta after the first stimulus. Then, it follows from the depletion model that the transmitter released from nearby synapses would have contributed to the observed greater responses to the second stimuli in a raised K + condition (Figure 3.28 A ; Table 3.2.8A). The inferred 'cross-talk' between neighbouring synapses, and its modification by drugs, represent an interesting area for further study. Limitations The methods and experimental approaches in this thesis possess a number of limitations. The in vitro oscillations induced by pentobarbital did not include concomitant measurement of the intracellular changes in electrical membrane properties that participate in thalamocortical oscillogenesis (von Krosigk et al., 1993). Simultaneous intra- and extracellular recording during pentobarbital application and oscillogenesis would enable identification of ionic conductances that are essential for producing synchronous rhythmic activity. The study of extracellular application of spermine on excitability did not exclude possible intracellular interactions. Spermine may have gained access to the cytoplasm through membrane transporters (Fage et al., 1992; Dot et a l , 2000). For example, an intracellular action of spermine occurs at Ca 2 +-permeable A M P A receptors and inward rectifier K + channels (Williams, 1997), which boosts thalamic excitability. Unlike inward rectifier General Discussion I Ran - 178 -K channels, however, Ca -permeable A M P A receptors have not been identified in the thalamus. The present study involved stimulation of multiple converging corticothalamic fibres which could result in conduction failure at axonal branches, especially late in the stimulus train. Future studies using focal stimulation (Dunant and Muller, 1986; Quastel and Saint, 1988) may attempt to minimize conduction failure which also may be assessed by simultaneous or independent extracellular recording of fibre volleys (Poolos et al., 1987). The binomial model is limited by an assumption of independence between release sites (cf. Vere-Jones, 1966). Not surprisingly, many studies have not reported the existence o f negative covariances. The lack of negative covariances in these studies may be explained by a positive correlation between active release sites which nullifies the negative covariance that would occur early in the train. Hence, a high level of branch conduction failure may explain the lack of negative covariances in other studies. The negative covariances, here, implicates a minimal contribution of conduction failure to the data. In the present thesis, the variance and covariance were obtained from sequential repeats of responses. This procedure minimized slow trends that developed between trains throughout the experiment. On the other hand, this approach prevented assessment o f slow changes due to drug effects. The procedure may have resulted in an increased sampling error or caused an underestimation of quantal parameters. General Discussion I Ran - 179 -The identification of minEPSC size was difficult to estimate within a train and hence, restricted to pre- and post-train periods. The difficulty is attributable to the low signal-to-noise ratio o f intra-train minEPSCs. A method that would correct for the time course o f the decay of E P S C may enable accurate detection of intra-train minEPSCs. The bin size in the present study resulted in a signal resolution with a minimum value o f 5 p A . Studies that have an improved signal-to-noise ratio may provide information on smaller changes in quantal size. This information then could be combined with rise times for estimating the distance between release sites and the synapse. Unequivocal demonstration of a drug effect requires observations of full reversibility. In most cases, the drug effects were reversible. However, T T X application at concentrations > 64 n M or the omission of glucose from the A C S F , produced effects that were irreversible during the period of observation, followed by loss o f neuronal viability. A n inability to observe excitation did not allow a more exact determination of pentobarbital's effects on spontaneous minEPSCs, normally studied in the presence of > 1 | j M [ T T X ] . Future studies that incorporate techniques for differentiating evoked and spontaneous transmitter release (cf. X u and Sastry, 2003), may address this issue. Extensive studies are required to examine reversible effects of glucose deprivations over a much wider concentration range than used here. The new studies may facilitate a distinction between pentobarbital effects on spontaneous and evoked release, for an assessment o f its long-term effects on transmission. General Discussion I Ran -180-Conclusion The present studies have provided several new contributions to understanding pentobarbital actions on thalamic neurons. A subanesthetic concentration of pentobarbital induced network oscillations in vitro, which would modulate thalamic inhibition and facilitate corticothalamic excitation in vivo. Another original contribution is the validation of the binomial depletion model of transmitter release at a conventional C N S synapse. The analysis has led to novel interpretations about pre- and postsynaptic anesthetic actions of pentobarbital on short-term plasticity during repetitive corticothalamic excitation. These studies also showed interesting similarities in its actions that reduced quantal size, to imposed conditions of impaired N a + entry into nerve terminals and energy shortage due to glucose deprivation. It seems significant that pentobarbital reversed the facilitating effect of elevated [ K + ] e which typically promotes a plastic change in transmission. This action represents a new type of synaptic modulation by barbiturate, complementing known anesthetic actions on thalamic neurons. Many of the actions of pentobarbital, including actions on quantal parameters, are summarized in the schematic diagram of Figure 4.1. The method and analysis technique described for pentobarbital actions in this thesis provide a model for examining the pre- and postsynaptic effects of drugs on transmission in the C N S . General Discussion I Ran - 181 -Na* channel Kr-s* j fa, _ . / " . . fa ^ c h a n n e l channel P <-~ S L ~ Pentobarbital • f e . Vesicular selection.. / = > \ ^ \ n h i n n ^ / / s \ . T n > * w Glucose ./ .-a.-! ...-^  ,,A j \ ATP 4"--> Glucose — \ ADP+P, ' ~~" if '"• \ fa) ^ X / J C * -St. / " H * ^ " ^ ^ - Na* A / I <** ( glut ^ / \ A ^ I " A T P glu Reduced pore opening^ Ca' AMPA channel NMDA receptors » * - receptors Pentobarbital Figure 4.1. Possible synaptic targets for pentobarbital actions during corticothalamic S T D . Pentobarbital enhancement of STD might involve ion channel modulation, inhibition of Na+-dependent glutamate uptake into readily releasable packets, suppression o f glycolytic A T P , and impairment o f glutamate exocytosis. Actions on ion channels include: induction of K + channel-mediated shunt, blockade of voltage-gated C a 2 + channels, blockade of N a + channel-mediated action potentials. Pentobarbital might also inhibit carrier-mediated glucose and N a + transport into the terminal cytoplasm. Pentobarbital suppression of glycolysis could reduce ATP-dependent uptake o f glutamate into packets. In the case of a heterogenous population of packets, penotbarbital's inhibition o f voltage gated C a 2 + channels might select for small size quanta. Pentobarbital may also shorten the formation of fusion pores thereby reducing exocytotic release of glutamate. Postsynaptically, pentobarbital could promote desensitization of A M P A or N M D A receptors. Bibliography 1 Ran -182 -Bibliography Aldridge W N , Parker V H . Barbiturates and oxidative phosphorylation. Biochem J 1960, 7 6 : 4 7 - 5 6 . Aimers W , Neher E . Gradual and stepwise changes in the membrane capacitance of rat peritoneal mast cells. J Physiol 1987, 386: 205 - 217. Alonso A , Llinas R R . Subthreshold Na+-dependent theta-like rhythmicity in stellate cells o f entorhinal cortex layer II. Nature 1989, 342: 175 - 177. A n S, Zenisek D . Regulation of exocytosis in neurons and neuroendocrine cells. Curr Opin Neurobiol. 2004, 14:522-530. Andersen P, Andersen S A . Evidence for thalamic control of cortical spindle activity. In: Andersen P and Andersen S A (Eds) Physiological Basis of the Alpha Rhythm, Appleton Century Crofts, New York, 1968, p. 43. Astman N , Gutnick M J , Fleidervish IA. Activation of protein kinase C increases neuronal excitability by regulating persistent N a + current in mouse neocortical slices. J Neurophysiol 1998,80:1547-1551. Auger C, Marty A . Heterogeneity of functional synaptic parameters among single release sites. Neuron 1997, 19: 139 - 150. Auger C, Marty A . Quantal currents at single-site central synapses. J Physiol 2000, 526: 3 - 11. Ba l T, von Krosigk, M , McCormick D A . Synaptic and membrane mechanisms underlying synchronized oscillations in the ferret lateral geniculate nucleus in vitro. / Physiol (London) 1995a, 483: 641-663. Ba l T, von Krosigk M , McCormick D A . Role of the ferret perigeniculate nucleus in the generation of synchronized oscillations in vitro. J Physiol (London) 1995b, 483: 665-685. Barnes-Davies M , Forsythe ID. Pre- and postsynaptic glutamate receptors at a giant excitatory synapse in rat auditory brainstem slices. J Physiol 1995, 488: 387 - 406. Bartlett E L , Smith P H . Anatomic, intrinsic, and synaptic properties of dorsal and ventral division neurons in rat medial geniculate body. J Neurophysiol 1999, 81: 1999-2016. Bartlett E L , Smith P H . Effects of paired-pulse and repetitive stimulation on neurons in the rat medial geniculate body. Neuroscience 2002, 113: 957 - 974. Bibliography 1 Ran - 183 -Baskaya M K , Rao A M , Dogan A , Donaldson D , Gell in G , Dempsey R J . Regional brain polyamine levels in permanent focal cerebral ischemia. Brain Res 1997, 744: 302 - 308. Baudoux S, Empson R M , Richards C D . Pentobarbitone modulates calcium transients in axons and synaptic boutons of hippocampal C A 1 neurons. Br J Pharmacol 2003, 140: 971 - 979. Bekkers J M , Clements JD. Quantal amplitude and quantal variance o f strontium-induced asynchronous EPSCs in rat dentate granule neurons. J Physiol 1999, 516: 227 - 248. Belanger A , Lavoie N , Trudeau F, Massicotte G , Gagnon S. Preserved L T P and water maze learning in hyperglycaemic-hyperinsulinemic Z D F rats. Physiol Behav 2004, 83: 483-494. Benveniste M , Mayer M L . Multiple effects of spermine on A7-methyl-D-aspartic acid receptor responses of rat cultured hippocampal neurones. J Physiol 1993, 464: 131-163. Bhayana V , Al to L E , Dhalla N S . Effects of pentobarbital and pentothal on rat heart contractile force and oxidative phosphorylation activities. Gen Pharmacol 1980, 11: 375 -377. Blaustein M P . Barbiturates block sodium and potassium conductance increases in voltage-clamped lobster axons. J Gen Physiol 1968, 51: 293 - 307. Borst J G , Sakmann B . Effect of changes in action potential shape on calcium currents and transmitter release in a calyx-type synapse of the rat auditory brainstem. Philos Trans R Soc LondBBiol Sci 1999, 354: 347 - 355. Boyd LA, Martin A R . The quantal composition of the mammalian end-plate potential. J Physiol 1955, 129: 14-15P. Bracci E , Ballerini L , Nistri A . Localization of rhythmogenic networks responsible for spontaneous bursts induced by strychnine and bicuculline in the rat isolated spinal cord. J Neurosci 1996, 16: 7063 - 7076. Brigant JL , Mallart A . Presynaptic currents in mouse motor endings. J Physiol 1982, 333: 619-636 . Brody D L , Yue DT. Release-independent short-term synaptic depression in cultured hippocampal neurons. J Neurosci 2000, 20: 2480 - 2494. Brown T H , Perkel D H , Feldman M W . Evoked neurotransmitter release: statistical effects of nonuniformity and nonstationarity. Proc Natl Acad Sci USA 1976, 73: 2913 - 2917. Bibliography I Ran - 184 -Budde T, White J A , Kay A R . Hyperpolarization-activated Na+ -K+ current (Ih) in neocortical neurons is blocked by external proteolysis and internal T E A . J Neurophysiol 1994, 72: 2737-2742. Burke SP, Nadler J V . Effects of glucose deficiency on glutamate/aspartate release and excitatory synaptic responses in the hippocampal C A 1 area in vitro. Brain Res 1989, 500: 333-342. Cahalan M D , Aimers W . Interactions between quaternary lidocaine, the sodium channel gates, and tetrodotoxin. Biophys J, 1979, 27:39-55. Carlen P L , Gurevich N , Davies M F , Blaxter TJ , O'Beirne M . Enhanced neuronal K + conductance: a possible common mechanism for sedative-hypnotic drug action. Can J Physiol Pharmacol 1985, 63: 831 - 837. Carlier E , Dargent B , De Waard M , Couraud F. N a + channel regulation by calmodulin kinase II in rat cerebellar granule cells. Biochem Biophys Res 2000, 274: 392-399. Castro-Alamancos M A , Calcagnotto M E . Presynaptic long-term potentiation in corticothalamic synapses. JNeurosci 1999, 19: 9090 - 9097. Chance B , Hollunger G . Inhibition of electron and energy transfer in mitochondria. J Biol Chem 1963,278:418-431. Charlesworth P, Jacobson I, Richards C D . Pentobarbitone modulation of N M D A receptors in neurones isolated from the rat olfactory brain. Br J Pharmacol 1995, 116: 3005 -3013. Charlton M P , Smith SJ, Zucker RS . Role of presynaptic calcium ions and channels in synaptic facilitation and depression at the squid giant synapse. J Physiol 1982, 323: 173 -193. Chen C, Regehr W G . Contributions of residual calcium to fast synaptic transmission. J Neurosci 1999 19: 6257 - 6266. Chen C, Regehr W G . Developmental remodeling of the retinogeniculate synapse. Neuron 2000, 28: 955-966. Chen C , Bl i tz D M , Regehr W G . Contributions of receptor desensitization and saturation to plasticity at the retinogeniculate synapse. Neuron 2002, 33: 779 - 788. Chida N , Saito H , Abe K . Spermine facilitates the generation of long-term potentiation o f evoked potential in the dentate gyrus of anesthetized rats. Brain Res 1992, 593: 57- 62. Bibliography I Ran -185-Clamann H P , Rioult-Pedotti M S , Luscher H R . The influence of noise on quantal EPSP size obtained by deconvolution in spinal motoneurons in the cat. J Neurophysiol 1991, 65: 67 - 75. Clements JD. Variance-mean analysis: a simple and reliable approach for investigating synaptic transmission and modulation. JNeurosci Methods 2003, 130: 115-125. Collins, V J (ed). Physiologic and Pharmacologic Bases of Anesthesia. Wil l iams &Wil l i ams , Baltimore, M D , 1996. Connor J, Wadman W J , Hockberger PE , Wong R K . Sustained dendritic gradients of C a 2 + induced by excitatory amino acids in C A 1 hippocampal neurons. Science 1988, 240: 649-653. Contreras D , Steriade M . Spindle oscillation in cats: the role of corticothalamic feedback in a thalamically generated rhythm. J Physiol 1996, 490: 159 - 179. Cooke JD, Quastel D M . The specific effect of potassium on transmitter release by motor nerve terminals and its inhibition by calcium. J Physiol 1973, 228: 435-458. Crane P D , Braun L D , Cornford E M , Cremer JE, Glass J M , Oldendorf W H . Dose dependent reduction of glucose utilization by pentobarbital in rat brain. Stroke 1978, 9: 12 - 18. C r i l l W E . Persistent sodium current in mammalian central neurons. Annu Rev Physiol 1996, 58: 349-362. Curro Dossi R, Nunez A , Steriade M . Electrophysiology of a slow (0.5^4 Hz) intrinsic oscillation of cat thalamocortical neurons in vivo. J Physiol (London) 1992, 447: 215 -234. Daniell L C . Alteration of general anesthetic potency by agonists and antagonists o f the polyamine binding site of the N-methyl-D-aspartate receptor. J Pharmacol Exp Ther 1992, 261: 304-310 . De Sarro G B , Bagetta G , Spagnolo C, Nistico G . Antagonists o f N-methyl-D-aspartate receptors block seizures induced by putrescine in the deep prepiriform cortex. Neuropharmacology 1993, 32: 43 - 50. Debanne D , Guerineau N C , Gahwiler B H , Thompson S M . Action-potential propagation gated by an axonal I(A)-like K + conductance in hippocampus. Nature 1997, 389: 286 -289. Debarbieux F, Brunton J, Charpak S. Effect of bicuculline on thalamic activity: a direct block of TAHP in reticularis neurons. J Neurophysiol 1998, 79: 2911-2918. Bibliography I Ran - 186 -Del Castillo J, Katz B . Quantal components of the end-plate potential. J Physiol 1954, 124: 560- 573. Delgado R, Maureira C, Ol iva C, Kidokoro Y , Labarca P. Size of vesicle pools, rates o f mobilization, and recycling at neuromuscular synapses of a Drosophila mutant, shibire. Neuron 2000, 28: 941 - 953. Diefenbach R J , Diefenbach E , Douglas M W , Cunningham A L . The heavy chain o f conventional kinesin interacts with the S N A R E proteins SNAP25 and S N A P 2 3 . Biochemistry 2002, 41:14906 - 14915. DiGregorio D A , Nusser Z , Silver R A . Spillover of glutamate onto synaptic A M P A receptors enhances fast transmission at a cerebellar synapse. Neuron 2002, 35: 521 - 533. DiScenna P G , Ferchmin P A , Eterovic V A , Teyler TJ . Spermine depresses N M D A , K / A M P A and GABAA-media ted synaptic transmission in the rat hippocampal slice preparation. Brain Res 1994, 647: 353 - 356. Dodson P D , Forsythe ID. Presynaptic K + channels: electrifying regulators of synaptic terminal excitability. Trends Neurosci 2004, 27: 210 - 217. Dogan A , Rao A M , Hatcher J, Rao V L , Baskaya M K , Dempsey R J . Effects o f M D L 72527, a specific inhibitor of polyamine oxidase, on brain edema, ischemic injury volume, and tissue polyamine levels in rats after temporary middle cerebral artery occlusion. JNeurochem 1999, 72: 765 - 770. Dot J, Lluch M , Blanco I, Rodriguez-Alvarez J. Polyamine uptake in cultures astrocytes: characterization and modulation by protein kinases. JNeurochem 2000, 75: 1917-1926. Duman J G , Forte J G . What is the role of S N A R E proteins in membrane fusion? Am J Physiol Cell Physiol 2003, 285: C237 - 249. Dunant Y , Muller D . Quantal release of acetylcholine evoked by focal depolarization at the Torpedo nerve-electroplaque junction. J Physiol 1986, 379: 461 - 478. Edwards F A , Konnerth A , Sakmann B . Quantal analysis of inhibitory synaptic transmission in the dentate gyrus of rat hippocampal slices: a patch-clamp study. J Physiol 1990, 430: 213 - 249. Elmqvist D , Hofmann W W , Kugelberg J, Quastel D M . A n electrophysiological investigation o f neuromuscular transmission in myasthenia gravis. J Physiol 1964, Nov;174:417-34. Elmqvist D , Quastel D M . A quantitative study of end-plate potentials in isolated human muscle. J Physiol 1965a, 178: 505 - 529. Bibliography 1 Ran - 187 -Elmqvist D , Quastel D M . Presynaptic action of hemicholinium at the neuromuscular junction. J Physiol 1965b, 177: 463 - 482. Emptage N J , Reid C A , Fine A . Calcium stores in hippocampal synaptic boutons mediate short-term plasticity, store-operated Ca2+ entry, and spontaneous transmitter release. Neuron 2001,29: 197-208. Emri Z , Antal K , Crunelli V . The impact of corticothalamic feedback on the output dynamics of a thalamocortical neurone model: the role of synapse location and metabotropic glutamate receptors. Neuroscience 2003, 117: 229 - 239. Erulkar SD, Weight FF . Extracellular potassium and transmitter release at the giant synapse of squid. J Physiol 1977, 266: 209-218. Eterovic V A , Torres E , Ferchmin P A . Spermine does not compete with omega-conotoxin G V I A in the striatum radiatum of the hippocampal slice. Brain Res 1997, 772: 191 - 202. Fage D , Vol tz C , Scatton B , Carter C. Selective release of spermine and spermidine from the rat striatum by N-methyl-D-aspartate receptor activation in vivo. J Neurochem 1992, 58:2170-2175. Faingold C L , Mi l l an M H , Boersma Anderson C A , Meldrum B S . Induction o f audiogenic seizures in normal and genetically epilepsy-prone rats following focal microinjection of an excitant amino acid into reticular formation and auditory nuclei. Epilepsy Res 1989, 3: 199-205. Fan P, O'Regan P A , Szerb JC. Effect of low glucose concentration on synaptic transmission in the rat hippocampal slice. Brain Res Bull 1988, 21: 741-747. Ferchmin P A , Perez D , Biello M . Spermine is neuroprotective against anoxia and N -methyl-D-aspartate in hippocampal slices. Brain Res 2000, 859: 273 - 279. ffrench-Mullen J M , Barker JL, Rogawski M A . Calcium current block by (-)-pentobarbital, phenobarbital, and C H E B but not (+)-pentobarbital in acutely isolated hippocampal C A 1 neurons: comparison with effects on GABA-activated C l ~ current. J Neurosci 1993, 13: 3211-3221. Forsythe ID, Tsujimoto T, Barnes-Davies M , Cuttle M F , Takahashi T. Inactivation o f presynaptic calcium current contributes to synaptic depression at a fast central synapse. Neuron 1998, 20: 797 - 807. Forsythe ID, Westbrook G L , Mayer M L . Modulation of excitatory synaptic transmission by glycine and zinc in cultures of mouse hippocampal neurons. J Neurosci 1988, 8: 3733 -3741. Bibliography 1 Ran -188-Forti L , Bossi M , Bergamaschi A , V i l l a A , Malgaroli A . Loose-patch recordings of single quanta at individual hippocampal synapses. Nature 1997, 388: 874 - 878. Fortune ES , Rose GJ . Short-term synaptic plasticity contributes to the temporal filtering o f electrosensory information. J Neurosci 2000, 20: 7122 - 7130. Fowler JC. Glucose deprivation results in a lactate preventable increase in adenosine and depression of synaptic transmission in rat hippocampal slices. J Neurochem 1993, 60: 572-576. Franceschetti S, Taverna G , Sancini G , Panzica F, Lombardi R, Avanzini G . Protein kinase C-dependent modulation of Na+ currents increases the excitability of rat neocortical pyramidal neurons. J Physiol 2000, 528: 291-304. Franks N P , Lieb W R . Molecular and cellular mechanisms of general anaesthesia. Nature 1994, 367: 607 - 614. Frerking M , Borges S, Wilson M . Variation in G A B A mini amplitude is the consequence of variation in transmitter concentration. Neuron 1995, 15: 885 - 895. Frerking M , Wilson M . Effects of variance in mini amplitude on stimulus-evoked release: a comparison of two models. Biophys J1996, 70: 2078 - 2091. Gabbiani F, Koch C. Principles of spike train analysis. In: Koch, C. and Segev, I., (Eds), Methods in Neuronal Modeling (2 n d ed.), The M I T Press, Cambridge, M A , 1998, p. 313 -360. Ghavanini A A , Mathers D A , Pui l E . Glycinergic inhibition in thalamus revealed by synaptic receptor blockade. Neuropharmacology 2005, (in press). Gonzalez-Lima F, Scheich H . Functional activation in the auditory system of the rat produced by arousing reticular stimulation: a 2-deoxyglucose study. Brain Res 1984, 299: 201-214. Gueldry S, Rochette L , Bralet J. Comparison of the effects of valproate, ethosuximide, phenytoin, and pentobarbital on cerebral energy metabolism in the rat. Epilepsia 1987, 28: 160- 168. Haghighi A P , Cooper E . A molecular link between inward rectification and calcium permeability of neuronal nicotinic acetylcholine alpha3beta4 and alpha4beta2 receptors. J Neurosci 2000, 20: 529 - 541. Hajos F, Csil lag A , Kalman M . The effect of pentobarbital, chloralhydrate, ether and protoveratrine on the distribution of synaptic vesicles in rat cortical synaptosomes. Exp Brain Res 1978,33:91-99. Bibliography I Ran - 189 -Halonen T, Sivenius J, Miettinen R, Halmekyto M , Kauppinen R, Sinervirta R, Alakuijala L , Alhonen L , MacDonald E , Janne J. Elevated seizure threshold and impaired spatial learning in transgenic mice with putrescine overproduction in the brain. Eur J Neurosci 1993, 5: 1233 - 1239. Harman R J , Shaw G G . High-affinity uptake of spermine by slices of rat cerebral cortex. J Neurochem 1981, 36: 1609 - 1615. Hayashi Y , Hattori Y , Hori Y . Involvement of putrescine in the development of kindled seizure in rats. JNeurochem 1992, 58: 562 - 566. He J, H u B . Differential distribution of burst and single-spike responses in auditory thalamus. J Neurophysiol 2002, 88: 2152 - 2156. He Y , Zorumski C F , Mennerick S. Contribution of presynaptic Na(+) channel inactivation to paired-pulse synaptic depression in cultured hippocampal neurons. J Neurophysiol 2002, 87: 925 - 936. Heidelberger R, Sterling P, Matthews G . Roles of A T P in depletion and replenishment of the releasable pool of synaptic vesicles. J Neurophysiol 2002, 88: 98 - 106. Heierli P, de Ribaupierre F, de Ribaupierre Y . Functional properties and interactions o f neuron pairs simultaneously recorded in the medial geniculate body of the cat. Hear Res 1987, 25: 209-225. Herberg L J , Rose IC, de Belleroche JS, Mintz M . Ornithine decarboxylase induction and polyamine synthesis in the kindling of seizures: the effect of alpha-difluoromethylornithine. Epilepsy Res 1992, 11:3 - 7 . Hermansen K , Johannsen L G , Rasmussen O B . Hypoglycaemic coma in severe primary hypothyroidism. Acta Med Scand 1985, 218: 345 - 346. Hernandez-Cruz A , Pape H C . Identification of two calcium currents in acutely dissociated neurons from the rat lateral geniculate nucleus. J Neurophysiol 1989, 61: 1270- 1283. Hertz E , Shargool M , Hertz L . Effects of barbiturates on energy metabolism by cultured astrocytes and neurons in the presence of normal and elevated concentrations of potassium.Neuropharmacology 1986, 25: 533 - 539. Hertz L , Zielke H R . Astrocytic control of glutamatergic activity: astrocytes as stars of the show. Trends Neurosci 2004, 27: 735 - 743. Hsieh C Y , Leslie F M , Metherate R. Nicotine exposure during a postnatal critical period alters N R 2 A and N R 2 B m R N A expression in rat auditory forebrain. Brain Res Dev Brain Res 2002, 133: 1 9 - 2 5 . Bibliography I Ran - 190-H u B . Cellular basis of temporal synaptic signalling: an in vitro electrophysiological study in rat auditory thalamus. J Physiol 1995, 483: 167 - 182. H u B , Senatorov V , Mooney D . Lemniscal and non-lemniscal synaptic transmission in rat auditory thalamus. J Physiol 1994, 479: 217 - 231. Hubbard JI, Jones SF, Landau E M . The relationship between the state of nerve-terminal polarization and liberation of acetylcholine. Ann N Y Acad Sci 1967, 144: 459 - 470. Hughes SW, Blethyn K L , Cope D W , Crunelli V . Properties and origin of spikelets in thalamocortical neurones in vitro. Neuroscience 2002, 110: 395 - 401. Dcemoto A , Bole D G , Ueda T. Glycolysis and glutamate accumulation into synaptic vesicles. Role of glyceraldehyde phosphate dehydrogenase and 3-phosphoglycerate kinase. J Biol Chem 2003, 278: 5929 - 5940. Isaacson JS. G A B A B receptor-mediated modulation of presynaptic currents and excitatory transmission at a fast central synapse. JNeurophysiol 1998, 80: 1571 - 1576. Isaacson JS. Glutamate spillover mediates excitatory transmission in the rat olfactory bulb. Neuron 1999, 23: 377 - 384. Ishikawa T, Sahara Y , Takahashi T. A single packet of transmitter does not saturate postsynaptic glutamate receptors. Neuron 2002, 34: 613 - 621. Jacobsen R B , Ulr ich D , Huguenard JR. G A B A B and N M D A receptors contribute to spindle-like oscillations in rat thalamus in vitro. J Neurophysiol 2001, 86: 1365 - 1375. Jackson, M F , Joo D T , Al-Mahrouki A A , Orser B A , MacDonald JF. Desensitization o f alpha-amino-3-hydroxy-5-methyl-4-isoxazoleproprionic acid ( A M P A ) receptors facilitates use-dependent inhibition by pentobarbital. Mol Pharmacol 2003, 64: 395 -406. Jahnsen H , Llinas R. Ionic basis for the electro-responsiveness and oscillatory properties of guinea-pig thalamic neurons in vitro. J Physiol 1984, 349: 227 - 247. Jahr C E . High probability opening of N M D A receptor channels by L-glutamate. Science 1992, 255 :470-472 . Jones D G , Devon R M . A n ultrastructural study into the effects of pentobarbitone on synaptic organization. Brain Res 1978,147: 47-63. Jones E G . Thalamic organization and function after Cajal. Prog Brain Res 2002, 136: 333 -357. Bibliography I Ran - 191 -Jones E G , Powell TP. A n electron microscopic study of the mode of termination o f corticothalamic fibres within the sensory relay nuclei of the thalamus. Proc R Soc Lond B Biol Sci 1969, 172: 173 - 185. Jueptner M , Weiller C. Review: does measurement of regional cerebral blood flow reflect synaptic activity? Implications for P E T and fMRI . Neuroimage 1995, 2: 148 - 156. Juhasz G , Kekesi K , Emri Z , Soltesz I, Crunelli V . Sleep-promoting action of excitatory amino acid antagonists: a different role for thalamic N M D A and n o n - N M D A receptors. Neurosci Lett 1990, 114: 333 - 338. Kajiwara R, Sand O, Kidokoro Y , Barish M E , Iijima T. Functional organization o f chromaffin cells and cholinergic synaptic transmission in rat adrenal medulla. Jpn J Physiol 1997, 47: 449 - 464. Kamiya H , Zucker R S . Residual Ca2+ and short-term synaptic plasticity. Nature 1994, 371: 603 -606. Kandel A , Buzsaki G . Cellular-synaptic generation of sleep spindles, spike-and-wave discharges, and evoked thalamocortical responses in the neocortex of the rat. J Neurosci 1997, 17: 6783 - 6797. Keifer JC, Baghdoyan H A , Becker L , Lydic R. Halothane decreases pontine acetylcholine release and increases E E G spindles. Neuroreport 1994, 5: 577 - 580. K e w J N , Kemp J A . A n allosteric interaction between the N M D A receptor polyamine and ifenprodil sites in rat cultured cortical neurons. J Physiol 1998, 512: 17 - 28. Khan A M , Stanley B G , Bozzetti L , Chin C, Stivers C, Curras-Collazo M C . AA-methyl-D-aspartate receptor subunit N R 2 B is widely expressed throughout the rat diencephalon: an immunohistochemical study. J Comp Neurol 2000, 428: 428 - 449. Kiel land A , Heggelund P. A M P A and N M D A currents show different short-term depression in the dorsal lateral geniculate nucleus of the rat. J Physiol 2002, 542: 99 -106. Kitano Y , Kuwabara S, Misawa S, Ogawara K , Kanai K , Kikkawa Y , Yagui K , Hattori T. The acute effects of glycemic control on axonal excitability in human diabetics. Ann Neurol 2004, 56: 462 - 467. Koerner C, Danober L , Boehrer A , Marescaux C, Vergnes M . Thalamic N M D A transmission in a genetic model of absence epilepsy in rats. Epilepsy Res 1996, 25: 11 -19. Bibliography I Ran - 192 -K o h D S , Bumashev N , Jonas P. Block of native Ca(2+)-permeable A M P A receptors in rat brain by intracellular polyamines generates double rectification. J Physiol 1995, 486: 305 -312. Konishi T. Voltage-gated potassium currents in myelinating Schwann cells in the mouse. JPhysiol 1990, 431: 123 - 139. Korn H , Burnod Y , Faber D S . Spontaneous quantal currents in a central neuron match predictions from binomial analysis of evoked responses. Proc Natl Acad Sci USA 1987, 84: 5981 - 5985. Kotlinska J, Liljequist S. Oral administration of glycine and polyamine receptor antagonists blocks ethanol withdrawal seizures. Psychopharmacol 1996,127: 238-244. Kraus N , Smith D l , McGee T, Stein L , Cartee C. Development o f the middle latency response in an animal model and its relation to the human response. Hear Res 1987, 27: 165-176. Kriebel M E , Vautrin J, Holsapple J. Transmitter release: prepackaging and random mechanism or dynamic and deterministic process. Brain Res Brain Res Rev 1990, 15: 167 - 178. Kriebel M E , Keller B , Silver R B , Fox G Q , Pappas G D . Porocytosis: a new approach to synaptic function. Brain Res Brain Res Rev 2001 38: 20 - 32. Kullmann D M , Martin R L , Redman SJ. Reduction by general anaesthetics of group l a excitatory postsynaptic potentials and currents in the cat spinal cord. J Physiol 1989, 412: 277 - 296. Kuromi H , Kidokoro Y . Two synaptic vesicle pools, vesicle recruitment and replenishment of pools at the Drosophila neuromuscular junction. J Neurocytol 2003, 32: 551 -565. Kuwabara N , Suga N . Delay lines and amplitude selectivity are created in subthalamic auditory nuclei: the brachium of the inferior colliculus of the mustached bat. J Neurophysiol 1993, 69: 1713-1724. Larrabee M G , Posternak J M . Selective action of anesthetics on synapses and axons in mammalian sympathetic ganglia. JNeurophysiol 1952, 15:91 -114. Laube G , V e h R W . Astrocytes, not neurons, show most prominent staining for spermidine/spermine-like immunoreactivity in adult rat brain. Glia 1997, 19: 171 - 179. Legendre, P. The glycinergic inhibitory synapse. Cell Mol Life Sci 2001, 58: 760 - 793. Bibliography I Ran - 193 -Leresche N , Lightowler S, Soltesz I, Jassik Gerschenfeld D , Crunelli V . Low-frequency oscillatory activities intrinsic to rat and cat thalamocortical cells. J Physiol (London) 1991,441: 155- 174. Lerma J. Spermine regulates N-methyl-D-aspartate receptor desensitization. Neuron 1992, 8:343 - 352. Lessmann V , Gottmann K . Fast desensitization of glutamate activated AMPA/kaina te receptors in rat thalamic neurones. Neuroreport 1994, 5: 2253 - 2256. L i J, Guido W , Bickford M E . Two distinct types of corticothalamic EPSPs and their contribution to short-term synaptic plasticity. J Neurophysiol 2003, 90: 3429 - 3440. L i P A , Shamloo M , Smith M L , Katsura K , Siesjo B K . The influence o f plasma glucose concentrations on ischemic brain damage is a threshold function. Neurosci Lett 1994, 177: 63 -65 . Li ley A W , North K A . A n electrical investigation of effects of repetitive stimulation on mammalian neuromuscular junction. J Neurophysiol 1953, 16: 509 - 527. L u H , X u T L . The general anesthetic pentobarbital slows desensitization and deactivation of the glycine receptor in the rat spinal dorsal horn neurons. J Biol Chem 2002, 277: 41369-41378. Llinas R, Urbano FJ , Leznik E , Ramirez RR, van Marie H J . Rhythmic and dysrhythmic thalamocortical dynamics: G A B A systems and the edge effect. Trends Neurosci 2005, 28:325 -333. Lukkarinen J, Grohn O H , Sinervirta R, Jarvinen A , Kauppinen R A , Janne J, Alhonen LI . Transgenic rats as models for studying the role of ornithine decarboxylase expression in permanent middle cerebral artery occlusion. Stroke 1997, 28: 639 - 645. Luthi A , D i Paolo G , Cremona O, Daniell L , De Camil l i P, McCormick D A . Synaptojanin 1 contributes to maintaining the stability of G A B A e r g i c transmission in primary cultures of cortical neurons. JNeurosci 2001, 21: 9101 -9111. M a W , Juliano SL . The relationship between thalamocortical connections and stimulus-evoked metabolic activity in the ventroposterior nucleus of the monkey. Somatosens Mot Res 1991, 8: 7 7 - 8 6 . Macdonald R L , Olsen R W . G A B A A receptor channels. Annu Rev Neurosci 1994, 17: 569 -602. M a d l JE, Royer S M . Glutamate in synaptic terminals is reduced by lack of glucose but not hypoxia in rat hippocampal slices. Neuroscience 1999, 94: 417 - 430. Bibliography I Ran - 194-Malosio, M L , Marqueze-Pouey B , Kuhse J, Betz H . Widespread expression of glycine receptor subunit m R N A s in the adult and developing rat brain. EMBO J 1 9 9 1 , 10: 2401 -2409. Mathers D A . The G A B A A receptor: new insights from single-channel recording. Synapse 1987, 1: 9 6 - 101. Matthews E K , Quill iam JP. Effects of central depressant drugs upon acetylcholine release. Br J Pharmacol 1964, 22: 415 - 440. Matthews G . Synaptic exocytosis and endocytosis: capacitance measurements. Curr Opin Neurobiol 1996, 6: 358 - 364. Matveev V , Wang X J . Implications of all-or-none synaptic transmission and short-term depression beyond vesicle depletion: a computational study. J Neurosci 2000, 20: 1575 -1588. McCann PP, Pegg A E . Ornithine decarboxylase as an enzyme target for therapy. Pharmacol Ther 1992, 54: 195 - 215. McCormick D A , Pape H C . Properties of a hyperpolarization-activated cation current and its role in rhythmic oscillation in thalamic relay neurones. J Physiol (London) 1990, 431: 291 -318. Metherate R, Cruikshank SJ. Thalamocortical inputs trigger a propagating envelope o f gamma-band activity in auditory cortex in vitro. Exp Brain Res 1999, 126: 160 - 174. Mi l ed i R, Thies R E . Post-tetanic increase in frequency of miniature end-plate potentials in calcium-free solutions. J Physiol 1967, 192: 54 - 55. More l A , Imig TJ . Thalamic projections to fields A , A l , P, and V P in the cat auditory cortex. J Comp Neurol 1987, 265: 119 - 144. Morrison L D , Becker L , A n g L C , K i s h SJ. Polyamines in human brain: regional distribution and influence of aging. JNeurochem 1995, 65: 636 - 642. Morison R S , Basset D L . Electrical activity of the thalamus and basal ganglia in decorticate cats. J Neurophysiol 1945, 8: 309 - 314. Muhlethaler M , Serafin M . Thalamic spindles in an isolated and perfused preparation in vitro. Brain Res 1990, 524: 1 7 - 2 1 . M u i r K W , Lees K R . Clinical experience with excitatory amino acid antagonist drugs. Stroke 1995, 26: 503 -513. Murthy V N , Sejnowski TJ , Stevens C F . Heterogeneous release properties of visualized individual hippocampal synapses. Neuron 1997 18: 599 - 612. Bibliography I Ran -195-Naas E , Zilles K , Gnahn H , Betz H , Becker C M , Schroder H . Glycine receptor immunoreactivity in rat and human cerebral cortex. Brain Res 1991, 561: 139 - 146. Naves L A , V a n der Kloot W . Repetitive nerve stimulation decreases the acetylcholine content o f quanta at the frog neuromuscular junction. J Physiol 2001, 532: 637 - 647. Nelson L E , Guo T Z , L u J, Saper C B , Franks N P , Maze M . The sedative component of anesthesia is mediated by G A B A A receptors in an endogenous sleep pathway. Nat Neurosci 2002, 5: 979 - 984. N 'Gouemo P, Faingold C L . Audiogenic kindling increases neuronal responses to acoustic stimuli in neurons of the medial geniculate body of the genetically epilepsy-prone rat. Brain Res 1997, 761: 217 - 224. Nicholls J, Wallace B G . Quantal analysis of transmitter release at an inhibitory synapse in the central nervous system of the leech. J Physiol 1978, 281: 171 - 185. Nishimura M , Shimizu Y , Satoh E , Yokoyama T, Yagasaki O. Factors influencing the twin-pulse facilitation of the release of transmitter at the mouse neuromuscular junction. Gen Pharmacol 1993, 24: 1241 - 1247. Ohara PT. Synaptic organization of the thalamic reticular nucleus. J Electron Microsc Tech 1988, 10: 283 -292. Olsen O, Nadim F, H i l l A A , Edwards D H . Uniform growth and neuronal integration. J Neurophysiol 1996, 76: 1850-1857. Otsu Y , Shahrezaei V , L i B , Raymond L A , Delaney K R , Murphy T H . Competition between phasic and asynchronous release for recovered synaptic vesicles at developing hippocampal autaptic synapses. J Neurosci 2004, 24: 420 - 433. Parri H R , Crunelli V . Sodium current in rat and cat thalamocortical neurons: role of a non-inactivating component in tonic and burst firing. J Neurosci 1998,18: 854 - 867. Petralia R S , Rubio M E , Wang Y X , Wenthold RJ . Differential distribution of glutamate receptors in the cochlear nuclei. Hear Res 2000, 147: 59 - 69. Pincus J H , Insler N F . Phenobarbital increases spontaneous transmitter release at the frog neuromuscular junction. Yale J Biol Med 1978, 51: 185 - 189. Pincus J H , Insler N F . Barbiturates and transmitter release at the frog neuromuscular junction. Brain Res 1981, 213:127 - 137. Pocock G , Richards C D . The action of pentobarbitone on stimulus-secretion coupling in adrenal chromaffin cells. Br J Pharmacol 1987, 90: 71 - 80. Bibliography I Ran - 196-Porta R, Doyle R L , Tatter S B , Wilens T E , Schatz R A , Sellinger O Z . The biosynthesis of polyamines in the brain of audiogenic seizure-susceptible and -resistant deermice. J Neurochem 1981, 37: 723 - 729. Poodry C A , Edgar L . Reversible alteration in the neuromuscular junctions o f Drosophila melanogaster bearing a temperature-sensitive mutation, shibire. J Cell Biol 1979, 81: 520 - 527. Poolos N P , Mauk M D , Kocsis JD. Activity-evoked increases in extracellular potassium modulate presynaptic excitability in the C A 1 region of the hippocampus. J Neurophysiol 1987,58:404-416. Pritchard G A , Fahey J M , Minocha SC, Conaty C, Mi l le r L G . Polyamine potentiation and inhibition of NMDA-mediated increases of intracellular free C a 2 + in cultured chick cortical neurons. Eur J Pharmacol 1994, 266: 107 - 115. Proctor W R , Weakly J N . A comparison of the presynaptic and post-synaptic actions o f pentobarbitone and phenobarbitone in the neuromuscular junction of the frog. J Physiol 1976, 258 :257-268 . Pussinen R, Sirvio J, Alhonen L , Larson J, Halmekyto M , Koivisto E , Janne J. Preserved induction of long-term potentiation in the stratum radiatum in the C A 1 field o f hippocampal slices from transgenic mice overexpressing ornithine decarboxylase and overproducing putrescine. Synapse 1998, 28: 288 - 293. Quastel D M . The binomial model in fluctuation analysis of quantal neurotransmitter release. Biophys J1997, 72: 728 - 753. Quastel D M , Hackett JT, Cooke JD. Calcium: is it required for transmitter secretion? Science 1971, 172: 1034 -1036. Quastel J H , Wheatley A H M . Oxidations by the brain. Biochem J 1932, 26: 725 - 744. Quastel D M , Saint D A . Transmitter release at mouse motor nerve terminals mediated by temporary accumulation of intracellular barium. J Physiol 1988, 406: 55 - 73. Rahamimoff R, Erulkar SD, Alnaes E , Mei r i H , Rotshenker S, Rahamimoff H . Modulation of transmitter release by calcium ions and nerve impulses. Cold Spring Harb Symp Quant Biol 1976, 40: 107 - 116. Ran I, Mathers D A , Pui l E . Pentobarbital induces thalamic oscillations in brain slices, modulated by G A B A and glycine receptors. Neuropharmacology 2004, 47: 985 - 993. Ran I, Miura R M , Pui l E . Spermine modulates neuronal excitability and N M D A receptors in juvenile gerbil auditory thalamus. Hear Res 2003, 176: 65 - 79. Bibliography I Ran - 197 -Rehberg B , Bennett E , Xiao Y H , Levinson SR, Duch DS . Voltage- and frequency-dependent pentobarbital suppression of brain and muscle sodium channels expressed in a mammalian cell line. Mol Pharmacol 1995, 48: 89 - 97. Reichova I, Sherman S M . Somatosensory corticothalamic projections: distinguishing drivers from modulators. J Neurophysiol 2004, 92: 2185 - 2197. Reinker S, Pui l E , Miura R M . Membrane resonance and stochastic resonance modulate firing patterns of thalamocortical neurons. J Comput Neurosci 2004, 16: 1 5 - 2 5 . Reynolds IJ. Arcaine uncovers dual interactions of polyamines with the /V-methyl-D-aspartate receptor. J Pharmacol Exp Ther 1990, 255: 1001 - 1007. Richards C D . The selective depression of evoked cortical EPSPs by pentobarbitone. J Physiol \91\,2\1:A\ - 43 . Richards C D . On the mechanism of barbiturate anesthesia. J Physiol (London) 1972, 227: 749 - 767. Roberts E B , Ramoa A S . Enhanced N R 2 A subunit expression and decreased N M D A receptor decay time at the onset of ocular dominance plasticity in the ferret. J Neurophysiol 1999, 81: 2587 - 2591. Rock D M , Macdonald R L . Polyamine regulation of N-methyl-D-aspartate receptor channels. Anna Rev Pharmacol Toxicol 1995, 35: 463 - 482. Rosenmund C , Stevens C F . Definition of the readily releasable pool of vesicles at hippocampal synapses. Neuron 1996, 16: 1197 - 1207. Rossi D J , Hamann M . Spillover-mediated transmission at inhibitory synapses promoted by high affinity alpha6 subunit G A B A ( A ) receptors and glomerular geometry. Neuron 1998, 20: 783 - 795. Roy S A , Al loway K D . Coincidence detection or temporal integration? What the neurons in somatosensory cortex are doing. J Neurosci 2001, 21: 2462 - 2473. Sailer C A , Kaufmann W A , Marksteiner J, Knaus H G . Comparative immunohistochemical distribution of three small-conductance Ca2+-activated potassium channel subunits, S K I , S K 2 , and SK3 in mouse brain. Mol Cell Neurosci 2004, 26: 458 -469. Sakaba T, Neher E . Calmodulin mediates rapid recruitment of fast-releasing synaptic vesicles at a calyx-type synapse. Neuron 2001, 32: 1119-1131. Sato S, Koshiro A , Kakemi M , Fukasawa Y , Katayama K , Koizumi T. Pharmacokinetic and pharmacodynamic studies of centrally acting drugs in rat: effect of pentobarbital and chlorpromazine on electroencephalogram in rat. BiolPharm Bull 1995,18: 1094 - 1103. Bibliography I Ran - 198-Sawada S, Yamamoto C. Blocking action of pentobarbital on receptors for excitatory amino acids in the guinea pig hippocampus. Exp Brain Res 1985, 59: 226 - 231. Scheuss V , Neher E . Estimating synaptic parameters from mean, variance, and ovariance in trains of synaptic responses. Biophys J200\, 81: 1970 - 1989. Scheuss V , Schneggenburger R, Neher E . Separation of presynaptic and postsynaptic contributions to depression by covariance analysis of successive EPSCs at the calyx of held synapse. J Neurosci 2002, 22: 728 - 739. Schneggenburger R, Meyer A C , Neher E . Released fraction and total size of a pool of immediately available transmitter quanta at a calyx synapse. Neuron 1999, 23: 399 - 409. Schuber F. Influence of polyamines on membrane functions. Biochem J 1989, 260: 1 -10. Schwartz W J , Smith C B , Davidsen L , Savaki H , Sokoloff L , Mata M , Fink D J , Gainer H . Metabolic mapping of functional activity in the hypothalamo-neurohypophysial system o f the rat. Science 1979, 205: 723 - 725. Schweitzer L , Casseday J H , Sjoerdsma A , McCann PP, Bartolome J V . Identification o f polyamines in the cochlea o f the rat and their potential role in hearing. Brain Res Bull 1986, 16: 215 -218. Schwindt P C , Cr i l l W E . Amplification of synaptic current by persistent sodium conductance in apical dendrite of neocortical neurons. J Neurophysiol 1995, 74: 2220 -2224. Schwindt P C , Spain W J , C r i l l W E . Effects of intracellular calcium chelation on voltage-dependent and calcium-dependent currents in cat neocortical neurons. Neuroscience 1992, 47: 571 -578. Selzer M E , David G , Yaari Y . On the mechanism by which phenytoin blocks post-tetanic potentiation at the frog neuromuscular junction. J Neurosci 1985, 5: 2894 - 2899. Seutin V , Scuvee-Moreau J, Dresse A . Evidence for a non-GABAergic action o f quarternary salts of bicuculline on dopaminergic neurones. Neuropharmacology 1997, 36: 1653 - 1657. Seyama I, Narahashi T. Mechanism of blockade of neuromuscular transmission by pentobarbital. J Pharmacol Exp Ther 1975, 192: 95 - 104. Sherrington, C.S. Integrative Actions of the Nervous System. Yale University Press: New Haven, C T , 1906. Bibliography I Ran - 199-Siekevitz P. Possible role for calmodulin and the Ca2+/calmodulin-dependent protein kinase II in postsynaptic neurotransmission. Proc Natl Acad Sci USA 1991, 88: 5374 -5378. Siesjo B . Brain Energy Metabolism . John Wiley and Sons, Chichester, 1978. Silver R A , Momiyama A , Cull-Candy SG. Locus of frequency-dependent depression identified with multiple-probability fluctuation analysis at rat climbing fibre-Purkinje cell synapses. J Physiol 1998, 510: 881 - 902. Smith D O . Variable activation of synaptic release sites at the neuromuscular junction. Exp Neurol 1983, 80: 520 - 528. Smith M L , Kal imo H , Warner DS , Siesjo B K . Morphological lesions in the brain preceding the development of postischemic seizures. Acta Neuropathol (Berl) 1988, 76: 253 - 264. Spreafico R, Frassoni C, Arcel l i P, Battaglia G , Wenthold R J , De Biasi S. Distribution o f A M P A selective glutamate receptors in the thalamus of adult rats and during postnatal development. A light and ultrastructural immunocytochemical study. Brain Res Dev Brain Res 1994, 82: 231 - 244. Stafstrom C E , Schwindt P C , Cr i l l W E . Negative slope conductance due to a persistent subthreshold sodium current in cat neocortical neurons in vitro. Brain Res 1982, 236: 221 -226. Steriade M . Coherent oscillations and short-term plasticity in corticothalamic networks. Trends Neurosci 1999, 22: 337 - 345. Steriade M . The corticothalamic system in sleep. Front Biosci 2003, 8: 878 - 899. Steriade M , Domich L , Oakson G , Deschenes M . The deafferented reticular thalamic nucleus generates spindle rhythmicity. J Neurophysiol 1987, 57: 260 - 273. Steriade M , Curro Dossi R, Nunez A . Network modulation of a slow intrinsic oscillation of cat thalamocortical neurons implicated in sleep delta waves: cortically induced synchronization and brainstem cholinergic suppression. J Neurosci 1991, 11: 3200 -3217. Steriade M , Jones E G , McCormick D A . Neurotransmitter actions in the thalamus. In: Steriade, M . , Jones, E . G . and McCormick, D A . (Eds) Thalamus vol . 1, Elsevier Science Ltd, U K , 1997. p. 687-751. Sterman M B , Howe R C , Macdonald R L . Facilitation of spindle-burst sleep by conditioning of electroencephalographic activity while awake. Science 1970, 167: 1146 -1148. Bibliography I Ran -200-Sucher N J , Akbarian S, Chi C L , Leclerc C L , Awobuluyi M , Deitcher D L , W u M K , Yuan JP, Jones E G , Lipton S A . Developmental and regional expression pattern o f a novel N M D A receptor-like subunit ( N M D A R - L ) in the rodent brain. J Neurosci 1995, 15: 6509 - 6520. Takahashi T, Forsythe ID, Tsujimoto T, Barnes-Davies M , Onodera K . Presynaptic calcium current modulation by a metabotropic glutamate receptor. Science 1996, 274: 594 - 597. Takahashi T, Hori T, Kajikawa Y , Tsujimoto T. The role of GTP-binding protein activity in fast central synaptic transmission. Science 2000, 289: 460 - 463. Takamori S, Rhee JS, Rosenmund C, Jahn R. Identification of differentiation-associated brain-specific phosphate transporter as a second vesicular glutamate transporter ( V G L U T 2 ) . J Neurosci 2001, 21: R C I 82. Tallroth G , Lindgren M , Stenberg G , Rosen I, Agardh C D . Neurophysiological changes during insulin-induced hypoglycaemia and in the recovery period following glucose infusion in type 1 (insulin-dependent) diabetes mellitus and in normal man. Diabetologia 1990,33: 319-323 . Tanaka E , Yasumoto S, Hattori G , Niiyama S, Matsuyama S, Higashi Ff. Mechanisms underlying the depression of evoked fast EPSCs following in vitro ischemia in rat hippocampal C A 1 neurons. JNeurophysiol 2001, 86: 1095-1103. Tang Y P , Shimizu E , Dube G R , Rampon C, Kerchner G A , Zhuo M , L i u G , Tsien JZ. Genetic enhancement of learning and memory in mice. Nature 1999, 401: 63 - 69. Tancredi V , Biagini G , D'Antuono M , Louvel J, Pumain R, A v o l i M . Spindle-like thalamocortical synchronization in a rat brain slice preparation. J Neurophysiol 2000, 84: 1093 - 1097. Taverna F A , Cameron B R , Hampson D L , Wang L Y , MacDonald JF. Sensitivity o f A M P A receptors to pentobarbital. Eur J Pharmacol 1994, 267: R3-5. Tebecis A K . 1974. Diencephalon. In: Transmitters and Identified Neurons in the Mammalian Central Nervous System, Scientechnica, Bristol, U K , p. 116 -167. Tekkok S, Krnjevic K . Diabetes mellitus preserves synaptic plasticity in hippocampal slices from middle-aged rats. Neuroscience 1999, 91: 185 - 191. Telgkamp P, Padgett D E , Ledoux V A , Woolley C S , Raman I M . Maintenance of high-frequency transmission at purkinje to cerebellar nuclear synapses by spillover from boutons with multiple release sites. Neuron 2004, 41: 113 - 126. Bibliography I Ran -201 -Tennigkeit F, Schwarz D W , Puil E . Mechanisms for signal transformation in lemniscal auditory thalamus. J Neurophysiol 1996, 76: 3597 - 3608. Tennigkeit F, Schwarz D W , Puil E . Postnatal development o f signal generation in auditory thalamic neurons. Brain Res Dev Brain Res 1998, 109: 255 - 263. Thomson T D , Turkanis S A . Barbiturate-induced transmitter release at a frog neuromuscular junction. Br J Pharmacol 1973, 48: 48 - 58. Toth K , Suares G , Lawrence JJ, Philips-Tansey E , M c B a i n C J . Differential mechanisms o f transmission at three types of mossy fiber synapse. J Neurosci 2000, 20: 8279 - 8289. Trout JJ, Koenig H , Goldstone A D , Iqbal Z , L u C Y , Siddiqui F. N-methyl-D-aspartate receptor excitotoxicity involves activation of polyamine synthesis: protection by alpha-difluoromethylornithine. JNeurochem 1993, 60: 352 - 355. Tsurusaki M , Akasu T, Shoji S. Patch-clamp analysis of hypoglycemia-induced inhibition o f synaptic transmission in the rat dorsolateral septal nucleus. Kurume Med J 1994, 41: 65 - 72. Uchida I, Cestari IN, Yang J. The differential antagonism by bicuculline and SR95531 o f pentobarbitone-induced currents in cultured hippocampal neurons. Euro J Pharmacol 1996, 307: 8 9 - 9 6 . V a n der Kloot W . Pretreatment with hypertonic solutions increases quantal size at the frog neuromuscular junction. J Neurophysiol 1987, 57: 1536 - 1554. V a n der Kloot W , Molgo J. Quantal acetylcholine release at the vertebrate neuromuscular junction. Physiol Rev 1994, 74: 899 - 991. Varadi A , Tsuboi T, Johnson-Cadwell LI , Al lan V J , Rutter G A . Kinesin I and cytoplasmic dynein orchestrate glucose-stimulated insulin-containing vesicle movements in clonal M I N 6 beta-cells. Biochem Biophys Res Commun 2003, 311: 272 - 282. Vautrin J, Barker JL . Presynaptic quantal plasticity: Katz's original hypothesis revisited. Synapse 2003, 47: 184 - 199. Velte TJ , Y u W , Mi l l e r R F . Estimating the contributions of N M D A and n o n - N M D A currents to EPSPs in retinal ganglion cells. Vis Neurosci 1997, 14: 999-1014. Vere-Jones D . Simple stochastic models for the release of quanta of transmitter from nerve terminals. AustJ Stat 1966, 8: 53 - 63. Villablanca J, Marcus R. Sleep-wakefulness, E E G and behavioral studies of chronic cats without neocortex and striatum: the 'diencephalic' cat. Arch Ital Biol 1972, 110: 348 -382. Bibliography I Ran -202-Villablanca J, Salinas-Zeballos M E . Sleep-wakefulness, E E G and behavioral studies of chronic cats without the thalamus: the 'athalamic' cat. Arch Ital Biol 1972,110: 383 -411 . Vital-Durand F, Gerin P, Pernier J, Revol M . Sleep spindles and wave spikes caused by penicillin in cats. J Physiol (Paris) 1972, 65: 313 - 314. V i z i ES , Somogyi GT. Prejunctional modulation of acetylcholine release from the skeletal neuromuscular junction: link between positive (nicotinic)- and negative (muscarinic)-feedback modulation. Br J Pharmacol 1989, 97: 65 - 70. von Gersdorff Ff, Borst JG . Short-term plasticity at the calyx of held. Nat Rev Neurosci 2002,3:53 -64 . von Gersdorff H , Schneggenburger R, Weis S, Neher E . Presynaptic depression at a calyx synapse: the small contribution of metabotropic glutamate receptors. J Neurosci 1997, 17: 8137- 8146. von Krosigk M , Ba l T, McCormick D A . Cellular mechanisms of a synchronized oscillation in the thalamus. Science 1993, 261: 361 - 364. Voronin L L . Quantal analysis of hippocampal long-term potentiation. Rev Neurosci 1994, 5: 141 - 170. Wahl L M , Stratford K J , Larkman A U , Jack JJ. The variance of successive peaks in synaptic amplitude histograms: effects of inter-site differences in quantal size. Proc R Soc Lond B Biol Sci 1995, 262: 77 - 85. Walmsley B , Edwards FR, Tracey DJ . Nonuniform release probabilities underlie quantal synaptic transmission at a mammalian excitatory central synapse. J Neurophysiol 1988, 60: 889 - 908. Wan X , Pui l E . Pentobarbital depressant effects are independent of G A B A receptors in auditory thalamic neurons. J Neurophysiol 2002, 88: 3067 - 3077. Wan X , Mathers D A , Pui l E . Pentobarbital modulates intrinsic and GABA-receptor conductances in thalamocortical inhibition. Neuroscience 2003, 121: 947 - 958. Wang L Y , Kaczmarek L K . High-frequency firing helps replenish the readily releasable pool of synaptic vesicles. Nature 1998, 394: 384 - 388. Wartenberg H C , Urban B W , Duch DS . Distinct molecular sites of anaesthetic action: pentobarbital block of human brain sodium channels is alleviated by removal of fast inactivation. £ r JAnaesth 1999, 82: 74 - 80. Bibliography I Ran -203 -Weakly J N . Effect of barbiturates on 'quantal' synaptic transmission in spinal motoneurones. J Physiol 1969, 204: 63 - 77. Weakly J N , Proctor W R . Barbiturate-induced changes in transmitter release independent o f terminal spike configuration in the frog neuromuscular junction. Neuropharmacology 1977, 16: 507 - 509. Weis S, Schneggenburger R, Neher E . Properties of a model of Ca(++)-dependent vesicle pool dynamics and short term synaptic depression. Biophys J1999, 77: 2418 - 2429. Werz M A , Macdonald R L . Barbiturates decrease voltage-dependent calcium conductance of mouse neurons in dissociated cell culture. Mol Pharmacol 1985, 28: 269 - 277. Wesseling JF, Lo D C . Limit on the role of activity in controlling the release-ready supply of synaptic vesicles. J Neurosci 2002, 22: 9708 - 9720. Wil l iams K . Interactions of polyamines with ion channels. Biochem J 1997, 325: 289 -297. Wil l iams K , Zappia A M , Pritchett D B , Shen Y M Molinoff P B . Sensitivity o f the N-methyl-D-aspartate receptor to polyamines is controlled by N R 2 subunits. Mol Pharmacol 1994, 45: 803 - 809. W u L G , Borst JG . The reduced release probability of releasable vesicles during recovery from short-term synaptic depression. Neuron 1999, 23: 821 - 832. W u L G , Borst J G , Sakmann B . R-type Ca2+ currents evoke transmitter release at a rat central synapse. Proc Natl Acad Sci USA 1998, 95: 4720 - 4725. J. X u , B . Sastry. A postsynaptic mechanism for L T P of IPSC in hippocampal C A 1 neurons. Program No . 690.13. Abstract Viewer/Itinerary Planner. Washington, D C : Society for Neuroscience, 2003. Yamada K A , Rensing N , Izumi Y , De Erausquin G A , Gazit V , Dorsey D A , Herrera D G . Repetitive hypoglycemia in young rats impairs hippocampal long-term potentiation. PediatrRes 2004, 55: 372 -379. Yat in S M , Yat in M , Varadarajan S, A i n K B , Butterfield D A . Role of spermine in amyloid beta-peptide-associated free radical-induced neurotoxicity. J Neurosci Res 2001, 63: 395 -401 . Y o o n K W , Wotring V E , Fuse T. Multiple picrotoxinin effect on glycine channels in rat hippocampal neurons. Neuroscience 1998, 87: 807 - 815. Zalin A M , Lant A F . Anorexia nervosa presenting as reversible hypoglycaemic coma. J R Soc Med 1984, 77: 193 -195. Bibliography I Ran -204-Zarbin M A , Wamsley J K , Kuhar M J . Glycine receptor: light microscopic autoradiographic localization with [3H] strychnine. J Neurosci 1981, 1: 532 - 547. Ziakopoulos Z , Brown M W , Bashir ZI. G A B A B receptors mediate frequency-dependent depression of excitatory potentials in rat perirhinal cortex in vitro. Euro J Neurosci 2000, 12: 803 - 809. Zimmermann H , Whittaker V P . Morphological and biochemical heterogeneity o f holinergic synaptic vesicles. Nature 1977, 267: 633 -635. Zucker R S , Regehr W G . Short-term synaptic plasticity. Anriu Rev Physiol 2002, 64: 355 -405. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0092402/manifest

Comment

Related Items